{"kind":"task","effective_mode":"full","benchmark":{"kind":"benchmark","effective_mode":"full","slug":"longbench-v2","formal_name":"LongBench v2","introduction":"長い資料の深い理解と推論を、多肢選択問題で評価するベンチマークです。公式紹介では503問を収録し、単一・複数文書の質問応答やコードリポジトリ理解などを扱います。\n\nLongBench v2 evaluates deep understanding and reasoning over long contexts through multiple-choice questions. Its official description lists 503 questions spanning tasks such as single-document and multi-document QA and code-repository understanding.","introduction_ja":"","introduction_en":"","category":"Category not supplied","task_count":null,"acquisition_status":"Acquisition status not supplied","official_url":"https://huggingface.co/datasets/zai-org/LongBench-v2","indexing_mode":"noindex"},"task_id":"37fd2409-a3a6-5156-b200-3d3adb253227","task_key":"train--66f28ede821e116aacb29a27","task_revision_id":"1","upstream_id":"66f28ede821e116aacb29a27","short_description":"How do Charmaz and Thornberg in The Pursuit of Quality in Grounded Theory and…","config":"","split":"train","body":"{\"choice_A\":\"Charmaz and Thornberg acknowledge researcher subjectivity as a natural and useful part of the coding process, enriching theoretical insights, while Holton and Walsh emphasize that subjectivity must be controlled through rigorous coding to ensure the theory remains grounded in the data rather than influenced by personal interpretations.\",\"choice_B\":\"Both sets of authors acknowledge the presence of researcher subjectivity, but Charmaz and Thornberg see it as a source of insight that can deepen theoretical understanding, while Holton and Walsh stress that researcher subjectivity must be controlled through rigorous coding to ensure that the theory accurately reflects the data rather than personal interpretations.\",\"choice_C\":\"Charmaz and Thornberg argue that researcher subjectivity provides valuable perspectives that can enhance theory co-construction with participants, while Holton and Walsh focus on minimizing subjectivity to prevent it from distorting the theory, advocating for systematic coding that keeps the theory grounded in the data.\",\"choice_D\":\"Both Charmaz and Thornberg and Holton and Walsh recognize that researcher subjectivity influences coding, but Charmaz and Thornberg see it as a vital aspect of shaping theoretical insights, while Holton and Walsh maintain that subjectivity should be minimized to let the data drive the theoretical process more objectively.\",\"context\":\"The pursuit of quality in grounded theory\\nKathy Charmaza and Robert Thornberg\\nb\\naDepartment of Sociology, Sonoma State University, Rohnert Park, Sonoma, USA; bDepartment of \\nBehavioural Sciences and Learning, Linköping University, Linköping, Sweden\\nABSTRACT\\nThis article introduces grounded theory and places this method \\nin its historical context when 1960s quantitative researchers \\nwielded harsh criticisms of qualitative research. The originators \\nof grounded theory, sociologists Glaser and Strauss, sought to \\ndefend the quality of qualitative research and argued that \\ngrounded theory increased its quality by providing a method \\nof theory construction. Our major foci include: (1) introducing \\nthe logic of grounded theory, with emphasis on how research­\\ners can use it to construct theory, (2) detailing criteria for quality \\nin the major forms of grounded theory advocated by Glaser and \\nStrauss and augmented by Glaser, Strauss and Corbin and \\nrefined by Corbin, and Charmaz and (3) providing an analysis \\nof how constructivist grounded theorists Thornberg, Halldin, \\nBolmsjö and Petersson attended to the interviewing process, \\ncoding, and developing their theoretical concept of double \\nvictimizing. Students and researchers new to the method can \\nuse our concluding guidelines as a checklist to assess the quality \\nof their constructivist grounded theory research.\\nKEYWORDS \\nGrounded theory; \\nconstructivist grounded \\ntheory; quality; qualitative \\nresearch; credibility\\nIntroduction\\nAmerican sociologists Barney Glaser and Anselm Strauss’s (1967) The \\nDiscovery of Grounded Theory: Strategies for Qualitative Research not only \\nproposed a new method of analysis, but also led the charge of defending the \\nquality of qualitative research. Grounded theory is a systematic method of \\nconducting research that shapes collecting data and provides explicit strategies \\nfor analyzing them. The defining purpose of this method is to construct \\na theory that offers an abstract understanding of one or more core concerns \\nin the studied world.\\nIn grounded theory studies, the researcher’s analytic focus emerges during \\nthe research process, rather than being determined before empirical inquiry \\nbegins. Increasingly, grounded theorists assume that the method is a way of \\nthinking about, constructing, and interacting with data throughout the \\nresearch process (Morse et al., forthcoming). Using grounded theory now \\nmeans more than openness to learning about the participants’ lives. It also \\nCONTACT Robert Thornberg \\nrobert.thornberg@liu.se \\nDepartment of Behaviorial Sciences and Learning, \\nLinköping University, Linköping SE-58183, Sweden\\nQUALITATIVE RESEARCH IN PSYCHOLOGY          \\nhttps://doi.org/10.1080/14780887.2020.1780357\\n© 2020 The Author(s). Published with license by Taylor & Francis Group, LLC. \\nThis is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives \\nLicense (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in \\nany medium, provided the original work is properly cited, and is not altered, transformed, or built upon in any way.\\n\\n\\nmeans making what the researchers learn transparent by showing how the \\nresearch has been conducted thoroughly and systematically.\\nThe grounded theory method offers useful strategies to develop researchers’ \\ntheoretical analyses. This method helps them to generate new concepts in their \\ndiscipline and the larger research literature. These concepts may have direct \\napplication for professional policies and practices in psychology and beyond.\\nWe begin by briefly explaining the logic of the method and then sketch the \\nhistorical context of the development of grounded theory. From its beginning \\nto the present, grounded theory has addressed questions about quality in \\nqualitative research. Glaser and Strauss (1967) tied quality to making new \\ntheoretical contributions. Throughout our discussion, we detail how concerns \\nabout quality pervade the research process in grounded theory studies. We \\nalso compare notions of quality among grounded theorists with the larger \\nliterature and then detail development of a concept and its relation to \\na theoretical framework. We end by synthesizing our key points and offering \\na checklist and suggestions for conducting quality in constructivist grounded \\ntheory research, although most of the suggestions are applicable to other \\nversions of grounded theory as well.\\nBefore outlining quality in grounded theory research, we wish to point out \\nthat many researchers, psychologists among them, use this method for \\na variety of worthy purposes instead of or in conjunction with theory con­\\nstruction. Such purposes include exploring a new area of study (Bronk 2012) \\nexplicating and understanding a major process (Qin and Lykes 2006), illumi­\\nnating the situations of people denied a public voice (Ayón et al. 2017; Gibson \\n2016; Scull, Mbonyingabo, and Kotb 2016; Tuason 2013), developing policy \\n(Faija et al. 2017), and implications for professional practice (Song and de Jong \\n2013; Yakushko 2010). Numerous researchers use several grounded theory \\nstrategies for conducting thematic analyses (cf. Braun and Clarke 2006). In this \\narticle, we emphasize quality in grounded theory studies that aim for theory \\nconstruction, despite whether they also serve other goals.\\nThe logic of grounded theory\\nGlaser and Strauss (1967) developed grounded theory by explaining the \\nmethods they used to construct their remarkable qualitative studies of death \\nand dying in hospitals (Glaser & Strauss, 1965, 1968). In this methodological \\ntreatise, they introduced the innovative and systematic strategy of simulta­\\nneous data collection and analysis. It became a hallmark of grounded theory \\nand has become widely adopted throughout qualitative inquiry. Glaser and \\nStrauss argued that by engaging in an iterative process of data collection and \\nanalysis, researchers would sequentially focus on the most significant issues in \\nthe field of study. Simultaneous data collection and analysis help researchers to \\nsteadily focus on developing concepts about the data and to gather further data \\n2\\nK. CHARMAZ AND R. THORNBERG\\n\\n\\nthat flesh out their nascent concepts. Glaser and Strauss opposed the separa­\\ntion of data gathering and analysis that characterized ethnographic research of \\ntheir time. Ethnographers of the 1960s commonly discovered their consider­\\nable quantities of data lacked depth and richness. But by the time they made \\nthis unwelcome discovery, they had left the research site and often could not \\nreturn. If, however, researchers systematically interrogate their data as they \\ngather it, their early ideas can then inform subsequent data collection.\\nTo begin analyzing data, Glaser and Strauss advocated comparing and \\ncoding the data. For them, initial coding meant labeling snippets of data to \\ntake them apart while being attentive to the meanings and actions suggested by \\nthese data. We advocate line-by-line coding as a first step, because it forces the \\nresearcher to take a fresh look at the data, compare fragments of these data, \\nand ask analytic questions about them. The grounded theory method not only \\nhelps researchers to synthesize data but, moreover, to move beyond descrip­\\ntion through constructing new concepts that explicate what is happening.\\nLine-by-line coding helps grounded theorists to understand their research \\nparticipants’ experiences and perspectives. Gaining this understanding can \\nlead to rethinking or relinquishing cherished disciplinary concepts that \\nresearchers might have believed would fit their data. For example, psycholo­\\ngists have conducted many important studies about resilience. However, the \\nusefulness of this concept may not fit the stance of desperate people who \\nexperience constant crises. Kapriskie Seide (in Charmaz 2020, 166) conducted \\nfirsthand research on the situation in Haiti during the cholera epidemic. She \\nfound herself face-to-face with homelessness, immense poverty, starvation, \\nisolation, hopelessness, and death. Seide reported that ‘For a young woman, \\nthe word resilient was patronizing and seemed to normalize structural vio­\\nlence as it is surprising to some that more Haitians are not ‘losing their shit’ \\nfrom being tested and challenged by their everyday lives’.\\nGlaser and Strauss’s (1965, 1968) studies were based on extensive field \\nresearch in multiple sites. Today, most grounded theory studies are based on \\ninterviews. By following up on codes in the early interviews, the grounded \\ntheorists develop more pointed questions and ask about areas that they had \\nnot anticipated would be important. Tacking back and forth between data and \\nanalysis helps to check the pertinence of their nascent ideas, raise the analytic \\nlevel of those that hold up, and support them more fully. While grounded \\ntheorists are coding, they write memos about their codes and the questions \\nthey have about them.\\nMemo writing is the intermediate step between coding and writing the first \\ndraft of the paper, including the provisional analysis. Early memos may \\ninclude discussions of grounded theorists’ codes, analytic and methodological \\nquestions as well as comparisons between fragments of data. Other memos are \\nmore analytic because grounded theorists take codes apart. We suggest that \\nyou first define a code by its properties or characteristics. These properties are \\nQUALITATIVE RESEARCH IN PSYCHOLOGY\\n3\\n\\n\\nwhat constitute the code and differentiate it from other codes. Look for \\nunstated assumptions and examine your own as well as those of your research \\nparticipants. Record where the code takes you and how it is linked to other \\ncodes. Not all codes are equal. Some codes are concrete and descriptive. Others \\nare more analytic and account for a wider range of data. As you proceed with \\nyour research, your memos become more definitive and analytic. In this way, \\nyou are building quality into your study.\\nHow much line-by-line coding should you do? After seeing how your codes \\ncoalesce and identifying which are the most important, you can stop line-by- \\nline coding. Define the most important codes as focused codes. These codes \\naccount for more data than other codes, subsume related codes, and raise the \\nanalytic level of your work. Then use the focused codes to examine how they \\nwork with large batches of data. Focused coding expedites the analytic process \\nwhile retaining your strong foundation in the data. Focused codes help to \\ngenerate tentative analytic categories, which you then pursue. Do these cate­\\ngories hold up? Can they account for these data, or is something else going on?\\nGrounded theory builds checks into the research process that contribute to \\nits quality, such as engaging in focused coding. Simultaneously, using focused \\ncoding helps you to expedite your analysis and streamline your subsequent \\ndata collection to gather targeted data that answer questions in your emerging \\nanalysis. Hence, grounded theory gives the researcher more analytic power \\nwith fewer data. Nonetheless, this point does not excuse having very small \\nsamples (cf., Thomson 2011).\\nQualitative research allows researchers to discern explicit and implicit \\nprocesses in their data. Grounded theory offers a useful way of studying \\nprocesses. To make processes explicit, grounded theorists study actions as \\nwell as meanings and show how they are connected. For example, psycholo­\\ngists Qin and Lykes (2006) provide an excellent analysis of the convergence of \\nmeanings and actions. They studied the situations of Chinese women graduate \\nstudents in the United States and discovered the basic psychological process of \\ngaining self-understanding through reweaving a fragmented self. These stu­\\ndents often experienced a fragmented self before their arrival in the US. In \\nChina, their self-understanding became fragmented due to gender discrimina­\\ntion, political corruption, and inability to manage the web of social relation­\\nships in which they were embedded. The students’ quest for higher education \\ncontradicted cultural norms for women. In the US their selves became further \\nfragmented, as they experienced ‘discrimination, disrespect, loneliness, and \\nself-doubt’ (p. 192). Yet many gained more complex selves as they rewove \\ncross-cultural insights into their self-understandings.\\nAt every stage of the research process grounded theorists make comparisons \\nbeginning with data and ending with comparisons between their categories.1 \\n1For a first-hand depiction of the comparative process, see Minas, Anglin, and Ribeiro (2018).\\n4\\nK. CHARMAZ AND R. THORNBERG\\n\\n\\nSubsequently, they compare their final category or categories with the existing \\nliterature. Because they do not know in advance where grounded theory may \\ntake them, they need to do a systematic search of the literature as they \\ncomplete their study.\\nGlaser and Strauss realized that the iterative process of increasing the \\nanalytic power of the researcher’s conceptual categories gave grounded theory \\nmuch of its strength. They proposed a new type of sampling, ‘theoretical \\nsampling’, to fill out these categories with new data. In this type of sampling, \\nresearchers look for data that will inform their categories. The term is confus­\\ning because it is often mistaken for representative population sampling in \\nquantitative research. However, theoretical sampling has nothing to do with \\nrepresentative sampling. Glaser and Strauss are aiming for what they call \\ntheoretical saturation of the emerging conceptual categories. Grounded the­\\norists achieve theoretical saturation, when they seek more data while theore­\\ntical sampling, but find no new properties or characteristics of their categories. \\nTheoretical sampling and theoretical saturation depend upon pursuing the \\niterative process and thoroughly checking the constructed categories against \\ndata. If, however, researchers only ask the same questions of each participant, \\nthey are likely to elicit similar stories about the topic. That is data saturation, \\nand it occurs with relatively few interviews but seldom contributes to the \\nanalytic power of the study.\\nTheoretical sampling and theoretical saturation contribute to the quality of \\nthe study by strengthening researchers’ analysis and giving them material for \\nmaking explicit claims about it. In short, grounded theorists are building the \\nkind of ‘trustworthiness,’ of their research that counseling psychologists \\nMorrow (2005) and Williams and Morrow (2009) rightly argue is imperative \\nto achieve in research practice.\\nConcerns about quality in the development of grounded theory\\nAt the time of Glaser and Strauss’s (1967) cutting edge statement, they were \\nvery much speaking to sociologists. They never dreamt that their method \\nwould cross many disciplinary and professional borders such as in psychology, \\nnursing, medicine, education, computer science, and urban planning. Glaser \\nand Strauss’s defense of the quality of qualitative research in sociology stood as \\na valiant rebuttal to quantitative sociologists. By the 1960s, qualitative research \\nhad come under attack by US quantitative sociologists who viewed qualitative \\ninquiry as lacking objectivity, validity, reliability, and replicability (Bryant & \\nCharmaz 2007; Charmaz, 1995, Charmaz 2006). Quantitative research had \\nalready been systematically codified and textbooks were increasingly available \\nfor learning how to do it. In contrast, qualitative research had largely been \\ntaught through an oral tradition of mentoring and immersion in the field of \\ninquiry. The results were small descriptive studies. Until Glaser and Strauss’s \\nQUALITATIVE RESEARCH IN PSYCHOLOGY\\n5\\n\\n\\n(1967) statement, little had been written, much less codified into a systematic \\nstatement, about how to construct a qualitative theoretical analysis of high \\nquality.\\nWhen answering quantitative social scientists’ sharp criticisms of its quality, \\nGlaser and Strauss rejected the common practice of evaluating qualitative \\nresearch according to the tenets of the quantitative tradition. Glaser and \\nStrauss rejected quantitative researchers’ criticisms of qualitative inquiry as \\nbeing unsystematic idiosyncratic, biased, and intuitive. Instead, they argued \\nthat qualitative research must be evaluated on its own canons, not on those \\nimposed by the dominant quantitative tradition.\\nGlaser and Strauss avowed that deductive quantitative research assumed \\nthat researchers knew the ‘right’ questions to ask before beginning their \\nstudies. Such questions were typically derived from speculative theorizing far \\nremoved from the empirical world. Glaser and Strauss proposed that (1) \\nresearch questions had to arise from studying empirical situations, (2) deduc­\\ntive hypothesis testing undermined developing new theoretical approaches, \\nand (3) thus, reduced the quality, relevance and usefulness of the research \\nproducts.\\nIn short, Glaser and Strauss declared that inductive qualitative research with \\nrich first-hand data could lead to theory construction and that adhering to \\ncanons of objectivity, validity, reliability and replicability would inhibit theo­\\nrizing. Hence, they argued that immersion in the research and theoretical \\nliteratures before conducting research would sway researchers and subse­\\nquently, preconceive their studies. For Glaser and Strauss, theorizing–and \\ncreating high quality research–needed to be predicated on direct knowledge \\nof the studied phenomenon. Moreover, they contended that studies of direct \\nexperience could be conducted with rigor.\\nTo place Glaser and Strauss’s position in context, it helps to know more \\nabout what was happening in US sociology departments of the 1960s. \\nSociology and anthropology have had long histories of first-hand qualitative \\nresearch from the beginning of their disciplines at the turn of the 20th century. \\nBut by the 1960s, quantitative researchers had attacked and marginalized \\nqualitative inquiry in sociology by gaining control over leading departments, \\njournals, research institutes, and funding agencies. They believed only a few \\ntalented stars could produce qualitative studies worthy of academic attention. \\nGlaser and Strauss rejected this view and democratized qualitative inquiry. \\nThey promised that grounded theory would give ordinary researchers the tools \\nto construct solid theories in their respective areas.\\nGlaser and Strauss’s (1967) book inspired generations of graduate students. \\nThe book legitimized inductive qualitative research and made coding and \\nmemo-writing integral strategies of qualitative inquiry. However, few \\nresearchers engaged in theoretical sampling and saturation and constructed \\ntheory. The appearance of Strauss and Corbin (1990, 1998) immensely popular \\n6\\nK. CHARMAZ AND R. THORNBERG\\n\\n\\nmanuals claimed to give researchers instructions about how to do grounded \\ntheory. However, their books came across as rigid and prescriptive in sharp \\ncontrast to the open-ended, fluid approach of Strauss’s (1959) earlier work, \\nand evident in the Discovery book.2 Strauss and Corbin’s manuals also \\ndiverged from Glaser’s (1978) elaboration of the method, to Glaser’s (1992) \\nfurious response and avowals to have the only ‘classic’ grounded theory \\nmethod.\\nPerhaps ironically, early grounded theory researchers and proponents (e.g., \\nGlaser 1978; Strauss and Corbin 1990, 1998) often treated the method as \\na mechanistic application of procedures to gather and analyze data. The \\nresidues of an epistemology of positivism are also evident in both Glaser’s \\nand Strauss and Corbin’s texts because they presuppose neutral observers who \\n(1) assume an external world, which can be discovered by meticulous observa­\\ntion, (2) separate themselves from the research process, (3) form an objective \\nview of the data, and (4) treat their representation of research participants and \\nthe research process as unproblematic. In addition, Glaser defends small \\nsamples, presumes that grounded theory can be used with any epistemology \\n(a philosophical theory of the nature of knowledge and reasoned justifications \\nfor it) and pursues parsimonious theoretical generalizations that transcend \\ntime, place, individuals, and circumstances. Epistemological stances are, how­\\never, significant because they shape how researchers gather their data and \\nwhether they acknowledge their influence on these data and the subsequent \\nanalysis.\\nPsychologists Henwood and Pidgeon (2003) raised a crucial point 17 years \\nago about grounded theory that still holds today. They contended that rather \\nthan seeing grounded theory as a unitary method, we can view it as a useful \\nnodal point around which we can debate significant issues in qualitative \\ninquiry. Unlike earlier manuals describing how to do grounded theory, con­\\ntemporary versions of the method including Bryant (2017, 2019), Charmaz \\n(2006, 2014), Corbin and Strauss (2008, 2015), and Thornberg and Charmaz \\n(2014) take a flexible approach to grounded theory and adopt more recent \\nepistemologies. These methodologists subscribe to acknowledging multiple \\nrealities, seeking diverse perspectives, and engaging in critical analysis \\nthroughout the research process. This approach appeals to researchers who \\n(1) engage in reflexivity throughout the research process, (2) aim to make their \\nstandpoints, starting points, and research actions as transparent as possible, \\n(3) read theoretical and substantive literatures on their topics before engaging \\nin research, but do not necessarily take these literatures as true or final \\nstatements, and (4) assume every methodological approach, including \\ngrounded theory, assumes an epistemology.\\n2The open-ended character of Strauss’s earlier work is more clearly represented in constructivist grounded theory \\nthan in Strauss and Corbin's books in the 1990 and 1998. (Charmaz 2008).\\nQUALITATIVE RESEARCH IN PSYCHOLOGY\\n7\\n\\n\\nGlaser and Strauss’s statement of grounded theory was revolutionary in the \\n1960s. By 2007, Kenneth Gergen (in Cisneros-Puebla 2007) dismissed Glaser \\nand Strauss’s (1967) grounded theory as passé because of its reliance on \\nempiricism, an epistemological approach ‘holding that all knowledge of mat­\\nters of fact either arises from experience or requires experience for its valida­\\ntion’ (APA Dictionary of Psychology). Gergen pointed out that all empirical \\nviews are interpretive, an epistemological view most 21st century grounded \\ntheorists share. Language and experience shape how researchers see the world. \\nIn contrast, Glaser (2013) still assumes that data are unproblematic and insists \\nresearchers must not preconceive analyzing them by invoking earlier concepts, \\nperspectives, and studies.\\nQualitative research and psychology\\nCriteria about the quality in qualitative research remain unsettled. Although \\nquality is crucial in qualitative research, there is, as Corbin states, ‘little \\nconsensus about what constitutes an appropriate set of evaluation criteria \\nfor qualitative research’ (Corbin and Strauss 2015, 341). This can, at least in \\npart, be explained by the multiplicity of qualitative research traditions (e.g., \\ndiscourse analysis, ethnography, phenomenological research, conversation \\nanalysis) and diversities within them (e.g., critical discourse analysis, inter­\\npretative phenomenological analysis, critical ethnography, constructivist \\ngrounded theory), rooted in various ontological and epistemological beliefs \\nand research ideals. Psychologists have subscribed to this wide range of \\nqualitative methods and contributed to their development.\\nConcerns about objectivity, validity, reliability and replicability in qualita­\\ntive research still pervade academic psychology. In a recent example, Anczyk \\net al. (2019) plea for replication in qualitative research. Rubin, Bell, and \\nMcCleland (2018) document the above concerns in their mixed methods \\nstudy of graduate psychology programs. They found negative judgments \\nabout qualitative inquiry that had long been evident in the discipline. They \\nreported that only 13% of the responding programs required a course in \\nqualitative research. Echoing sociological criticisms of qualitative studies in \\nthe 1960s, Rubin et al.’s respondents stated that “qualitative research was \\neither imagined to be or believed to be ‘inaccurate,’ ‘subjective,’ and ‘lacking \\nrigor’ (p. 41). However, the respondents described qualitative research as \\nproviding the ‘best methods for feminist, community-based, and social justice \\nresearch’ (p. 43). The critics are correct on this point. Psychologists have made \\nenormous contributions in these areas, and many of them use grounded \\ntheory (see, for example, Ayón et al. 2017; Faija et al. 2017; Gibson, 2016; \\nLee 2018; Scull, Mbonyingabo, and Kotb 2016; Tsai, L. 2017; Tuason 2013).\\nYet critics appear to ignore other important considerations about the \\ncontributions of psychologists who conduct qualitative research. Qualitative \\n8\\nK. CHARMAZ AND R. THORNBERG\\n\\n\\npsychologists study an amazing array of problems and are affiliated with \\na wide range of programs, including those in medicine and psychiatry, educa­\\ntion, religion, ethnic and gender studies, and business, to name a few. Their \\nlocations and diverse publication outlets may mask both the quality of their \\nresearch and the scope of their contributions (e.g. Bronk 2012; Mazzone et al. \\n2018; Tsai, Seballos-Llena, and Castellano-Datta 2017; Wójcik 2018). Also, \\nqualitative psychologists often work in specialized areas on multidisciplinary \\nteams (e.g. Khankeh et al. 2015).3 Furthermore, it is not unusual for psychol­\\nogists to hold multi-disciplinary expertise, although they may not be employed \\nin psychology programs.4\\nAs in the 1960s, neither qualitative research, in general, nor grounded \\ntheory, in particular, fit judgments of quality based on quantitative research. \\nWe agree with Tracy’s (2010) ‘big tent’ markers of quality in qualitative \\nresearch: a worthy topic, rich rigor, sincerity, credibility, resonance, significant \\ncontribution, ethics, and meaningful coherence (p. 837). However, what these \\nterms mean in research practice may differ among researchers. Nonetheless, \\ngrounded theorists can meet these markers of quality when researchers con­\\nduct thorough studies. Increasingly, researchers ask for transparency and \\nreflexivity as criteria of quality in qualitative research (e.g. Treharne and \\nRiggs 2014).\\nQuality criteria in various versions of grounded theory\\nGrounded theory needs its own set of criteria for evaluating quality due to its \\nunique features (cf., Berthelsen, Grimshaw-Aagaard, and Hansen 2018; \\nChiovitti and Piran 2003; Elliott and Lazenbatt 2005; Hutchison, Johnston, \\nand Breckon 2011), although grounded theorists cannot escape more general \\nguidelines concerning their use of data gathering methods to ensure quality in \\ndoing interviews, fieldwork etc. The principle of ‘garbage in, garbage out’ is \\nvery much applicable to grounded theory, particularly considering its empha­\\nsis on data and groundedness. Criteria for evaluating quality in grounded \\ntheory can vary due to which version of grounded theory is of concern. In the \\noriginal but less developed one, Glaser and Strauss (1967) focused on cred­\\nibility and applicability. They connected credibility to researchers’ confidence \\nin their own knowledgeability based on carefully studying and analyzing the \\nactual field, and gaining systematic knowledge of the data. Glaser and Strauss \\n33.A quick search of Elsevier publications in Science Direct on the Internet showed that over 10,000 authors or co- \\nauthors of grounded theory articles were linked to psychology departments. Elsevier represents only one major \\npublishing venue.\\n4Exemplars have made stellar contributions to grounded theory. Methodologist Uwe Flick (2014, 2019) holds \\nadvanced degrees in both psychology and sociology; Lillemor Hallberg (2006, Hallberg & Strandmark, 2006) served \\nas the first editor of the International Journal of Qualitative Studies on Health and Well-Being and conducts research \\nas an expert in nursing research and psychology, and psychiatrist Suzan Joon Song ( Song and de Jong 2013) has \\nspecial training in the psychology of trauma and in medicine.\\nQUALITATIVE RESEARCH IN PSYCHOLOGY\\n9\\n\\n\\nsuggested a few criteria for judging credibility: (1) a detailed and vivid descrip­\\ntion of data so the readers feel that they have been in the field as well, and \\nliterally can hear and see the participants; (2) readers’ assessments of how the \\nresearchers came to their conclusions (what are the data and how have they \\nbeen gathered and analyzed?); and (3) multiple comparison groups to increase \\nthe scope and generality of the theory, and to correct and adjust the emerging \\ntheory to diverse conditions. Applicability, in turn, can be connected to \\ngeneralizability (cf., Larsson 2009), and Glaser and Strauss (1967) proposed \\nfour interrelated criteria: (1) Fitness: The ‘theory must fit the substantive area \\nto which it will be applied’ (p. 238), and they emphasized that the theory needs \\nto fit the data under study. (2) Understanding: The theory should make sense \\nand be understandable to non-researchers working or living in the substantive \\narea. (3) Generality: The theory needs to be sufficiently general to increase its \\napplicability – ‘categories should not be so abstract as to lose their sensitizing \\naspect, but yet must be abstract enough to make . . . theory a general guide to \\nmulti-conditional, ever-changing daily situations’ (p. 242). (4) Control: The \\ntheory should help the users to understand and analyze their social reality, to \\ninclude enough explanatory power on what is going on in situations and how \\nto bring about change in them.\\nIn his further development of grounded Glaser (1978, 1998) suggested four \\ncriteria in judging quality: (1) Workability: Does the theory work to explain \\nrelevant behavior in the substantive area of the study? (2) Relevance: Does it \\nhave relevance to the people in the substantive field? (3) Fit: Does the theory fit \\nthe substantive area? The theory and its categories must fit the data. (4) \\nModifiability: Is the theory readily modifiable as new data emerge? In their \\ndevelopment of grounded theory, Strauss and Corbin (1990; Corbin and \\nStrauss 1990; Strauss and Corbin 1998) emphasized four general criteria: (1) \\nquality of the data, in other words, grounded theorists need to carefully \\naddress and adopt quality criteria related to methods used for data gathering \\nto secure credibility and trustworthiness of data; (2) the plausibility and value \\nof the theory itself; (3) adequacy of the research process (sampling procedures, \\ntheoretical sampling, coding, categorization, development of hypotheses or \\nconceptual relations between categories, and selection of core category \\nthrough which the theory is developed); and (4) the issue of the empirical \\ngrounding of the theory. A comprehensive description of checklists and guide­\\nlines for evaluation of the two later criteria is offered by Strauss and Corbin \\n(1990; Corbin and Strauss 1990; Strauss and Corbin 1998).\\nAccording to Strauss and Corbin, questions to ask regarding empirical \\ngrounding of the study are: (1) Are concepts generated? (2) Are the concepts \\nsystematically related? (3) Are there many conceptual linkages, and are the \\ncategories well developed? Do categories have conceptual density? (4) Is \\nvariation built into the theory? Have the concepts been examined under \\na broad range of conditions and do they offer several dimensions? (5) Are \\n10\\nK. CHARMAZ AND R. THORNBERG\\n\\n\\nthe conditions under which variation can be found built into the study and \\nexplained? (6) Has process been considered and identified? (7) Do the theore­\\ntical findings seem significant, and to what extent? (8) Does the theory stand \\nthe test of time and become part of the discussions and ideas exchanged \\namong relevant social and professional groups? (For further reading, see \\nStrauss and Corbin 1998, pp. 270–272.)\\nIn later editions (Corbin and Strauss 2008, 2015), Corbin also added more \\ngeneral criteria of quality in qualitative research: methodological consistency, \\nclarity of the purpose, self-awareness, training in how to conduct qualitative \\nresearch, sensitivity to participants and data, willingness to work hard, ability \\nto connect with the creative self, methodological awareness, and strong desire \\nto do research. Note that in the original grounded theory (Glaser and Strauss \\n1967), Glaserian grounded theory (Glaser 1978, 1998) and Straussian \\ngrounded theory (Strauss and Corbin 1990, 1998), explanatory power (i.e., \\nthe power to explain and predict) is a significant quality criterion of the \\nconstructed theory. These approaches of grounded theory share what \\nCharmaz (2006, 2014) terms as positivist definitions of theory: theories that \\nseek causes, and stress explanation, prediction, generality, and universality.\\nFrom a constructivist view of grounded theory, Charmaz (2006, 2014) \\ncontrasts such definitions of theory with interpretive definitions of theory, in \\nwhich theory ‘emphasizes interpretation and gives abstract understanding \\ngreater priority than explanation’ (Charmaz 2014, 230). Such theories aim to \\nunderstand meanings and actions and how people construct them, and they \\n‘bring in the subjectivity of the actor and may recognize the subjectivity of the \\nresearcher’ (p. 231). They are situated in their social, historical, local, and \\ninteractional contexts. Thus, different epistemologies, ideals, aims etc. between \\ndifferent versions of grounded theory have to be considered when evaluating \\nquality and discussing quality criteria for grounded theory (for overviews of \\nsimilarities and differences between Glaserian, Straussian and constructivist \\ngrounded theory approaches, see Berthelsen, Grimshaw-Aagaard, and Hansen \\n2018; Thornberg 2017; for further discussions about the link between episte­\\nmology and quality criteria, see for example; Healy and Perry 2000; Lincoln, \\nLynham, and Guba 2018; Lützhöft, Nyce, and Petersen 2010; Santiago- \\nDeleofosse et al. 2015).\\nWithin a constructivist grounded theory, Charmaz (2006, 2014) proposes \\nfour main criteria for grounded theory studies: credibility, originality, reso­\\nnance, and usefulness. Credibility begins with having sufficient relevant data \\nfor asking incisive questions about the data, making systematic comparisons \\nthroughout the research process, and developing a thorough analysis. The \\nmore controversial the topic and/or analysis is, the more data the researcher \\nneeds to persuade skeptical audiences.\\nCredibility also involves the researcher’s views and actions. Constructivist \\ngrounded theory requires strong reflexivity throughout the research process. \\nQUALITATIVE RESEARCH IN PSYCHOLOGY\\n11\\n\\n\\nThis means researchers must explicate their taken-for-granted assumptions, \\nwhich requires gaining ‘methodological self-consciousness’ (Charmaz 2017) of \\nhow hidden beliefs can enter the research process. During her fieldwork in \\nBrazil, Elizabeth Hordge-Freeman (in Charmaz 2020, pp. 168–169) acciden­\\ntally discovered the plight of black adopted daughters, filhas de criação, in \\nwhite families. Hordge-Freeman writes:\\nWhen I discovered that one woman was sleeping on the floor in their family’s home and \\nseveral others had been working for no pay in adoptive families for over 40 years, I was \\noverwhelmed with a sense of anger over this injustice. As a critical researcher, my initial \\nimpulse was a desire to mete out justice by exposing these ‘monstrous’ families; however, \\nvery early on the sensibilities of constructivist grounded theory with its emphasis on \\nscrutinizing our interpretations of data and engaging in sustained reflexivity persuaded \\nme to analyze the data in more nuanced ways and to even interrogate my affective \\nresponses. By doing so, I realized that my quest to ‘right’ this injustice and free the \\ninterviewees sounded eeringly similar to the savior complex for which I had critiqued \\nwhite transnational researchers. Ultimately, using constructivist grounded theory, I re- \\ndirected my questions to be able to collect data that provided me a greater understanding \\nof how structural and affective constraints shape filhas de criação lives and used reflex­\\nivity to ensure that my interviewees’ interpretations were taken seriously.\\nHordge-Freeman’s statement represents the type of strong reflexivity most \\nadvocated in constructivist grounded theory. The strong reflexivity under­\\ngirding methodological self-consciousness (Charmaz 2017) means more than \\nexamining the researcher's methodological decisions. It means openness to \\nscrutinizing who the researcher is.\\nOriginality can take varied forms such as offering new insights, providing \\na fresh conceptualization of a recognized problem, and establishing the sig­\\nnificance of the analysis. Resonance demonstrates that the researchers have \\nconstructed concepts that not only represent their research participants’ \\nexperience, but also provide insight to others. To gain resonance, researchers \\nmust fit their data-gathering strategies to illuminate their participants’ experi­\\nence. Once Jennifer Lois (2010) realized her questions of homeschooling \\nmothers focused on quantity of time instead of quality, she gained resonance \\nby hearing their stories from a new standpoint and revising her questions. Her \\nsubsequent codes: ‘Sequencing: eliciting nostalgia and anticipating regret,’ (p. \\n434) and ‘Savoring: staying present and creating quality time’ (p. 437) not only \\nled her to understand these mothers’ actions but also to construct the concept \\nof temporal emotions. She reveals how certain present feelings like regret or \\nhope can only be felt by crossing timeframes from the present to the past or \\nfuture. Lois theorizes that how people handle temporal emotions influences \\ntheir relative continuity of self over time.\\nUsefulness includes clarifying research participants’ understanding of their \\neveryday lives, forming a foundation for policy and practice applications, \\ncontributing to creating new lines of research, as well as revealing pervasive \\n12\\nK. CHARMAZ AND R. THORNBERG\\n\\n\\nprocesses and practices. Gregory Hadley (in Charmaz 2020, 168) describes the \\nusefulness of Glaser and Strauss’s analysis of death and dying in hospitals.\\nGlaser’s work with Strauss in his book, “An Awareness of Dying”, for example, unlocked \\nmuch of the meaning of what was “going on” in a Japanese hospital during the time \\nwhen my father-in-law was on the trajectory of terminal cancer. That this grounded \\ntheory could have salience a half century later in a culture thousands of miles away is \\na testament to the genius behind what Glaser and Strauss did using Classical Grounded \\nTheory.\\nIn this case, Glaser and Strauss’s extensive research shows how they \\nachieved a useful level of generality. Using constructivist grounded theory, \\nSnow and Moss (2014) delineate the conditions, including those shaping \\nbehavior and emotions, in which seemingly spontaneous collective action \\narises. Sensitivity to these conditions can assist social movement organizers \\nin mobilizing crowds.\\nDouble victimizing: an example of developing a new concept with \\ngrounded theory\\nIn their constructivist grounded theory study, Thornberg et al. (2013) exam­\\nined how individuals, who had been victims of school bullying, perceived their \\nbullying experiences, and how these had affected them. The project began with \\ndistributing 523 open-ended questionnaires to students in three secondary \\nschools and two universities in Sweden. The questionnaire asked about their \\npast experiences with bullying and their willingness to be interviewed about it. \\nOf 511 responses, 168 students reported having been bullied in the past and 36 \\nvolunteered to be interviewed. The researchers interviewed the 21 students \\nwho had been bullied for more than one year. Halldin, Bolmsjö and Petersson, \\nwho conducted all the interviews, were instructed and trained in qualitative \\ninterviewing. Quality criteria from the literature guided their interviewing, \\nsuch as establishing a comfortable, private, and quiet interview setting build­\\ning rapport and emphasizing with the interviewee; taking the role of an \\ninterested learner; active listening; using open-ended questions and avoiding \\nleading questions; probing; and adopting a non-judgmental approach (e.g., \\nKing & Horrocks, 2010; Kvale & Brinkmann, 2009; Seidman, 2006).\\nIn constructivist grounded theory (Charmaz 2014), interviewing is not \\nconsidered as efforts to mirror reality but as emergent interactions through \\na mutual exploration of the interviewee’s experiences and perspectives. \\nTherefore, the interviewer’s approach and way of asking questions, listening \\nand following up what the interviewee is telling are crucial in the co- \\nconstruction and quality of data. Such an approach, based on the qualitative \\ninterview literature, helped Thornberg et al. (2013) to gather vivid, rich, and \\ncomprehensive data. In accordance with grounded theory, they moved back \\nQUALITATIVE RESEARCH IN PSYCHOLOGY\\n13\\n\\n\\nand forth between gathering and analyzing data (iterative process). This \\napproach has four advantages. It prevented them from (1) gathering data in \\na superficial and random way; (2) feeling overwhelmed due to a huge amount \\nof unanalyzed data; (3) being unfocused for lengthy periods; and (4) uncriti­\\ncally adopting the participants’ view or stock disciplinary categories.\\nThornberg et al. (2013) constructed codes that fit the data by engaging in \\ninitial coding, in which they compared data with data, data with codes, and \\ncodes with codes, and stayed close to and remained open to exploring what \\nthey interpreted was happening in the data. The provisional, initial codes were \\ncarefully compared with each other and with data, further elaborated and \\ngrouped together based on similarities and differences, leading to fewer but \\nmore focused and comprehensive codes. As a result of the iterative process, \\ncoding and constant comparison, their constructed focused codes fit tightly \\nwith their data. For example, the focused code ‘self-inhibiting’ had its roots in \\ninitial codes like ‘trying to not stand out’, ‘becoming passive out of social fear’, \\n‘keeping oneself down’, ‘believing social invisibility prevents bullying’, ‘inhi­\\nbiting the social presence of self’, and ‘becoming silent’. Another focused code, \\n‘self-doubting’, had its roots in initial codes like ‘becoming insecure’, ‘feeling \\nself-worthlessness’, ‘loss of self-confidence’, and ‘getting bad self-confidence \\nfrom being bullied’, as indicated in Tables 1 and 2.\\nThornberg and colleagues’ comparative approach to data collection and \\nanalysis (2013) during their focused coding included theoretical sampling as \\na prominent part of the iterative process, which guided their later interviews \\nand helped them to fill out emerging categories. As noted by Charmaz (2014), \\ntheir focused codes provided the makings of a frame for the later analysis. \\n‘Rather than using these focused codes as the final frame, however, Thornberg \\net al. (2013) developed an analysis of processes involved in becoming a victim \\nof school bullying and in extricating self from it’ (Charmaz 2014, 141).\\nGrounded theory methods of focused coding, theoretical coding, constant \\ncomparison, memo-writing and theoretical sampling guided Thornberg et al. \\n(2013) to merge focused codes such as ‘self-inhibition’, ‘self-isolation’, and \\n‘turning off emotions’ into an even more comprehensive focused code devel­\\noped into a category that they labeled ‘self-protecting’, in which the former \\nfocused codes that constituted this category were included as subcategories \\nrepresenting different self-protecting strategies. Other groups of focused codes \\nwere further merged into other more comprehensive categories such as ‘a \\nsense of not fitting in’ and ‘self-blaming’. The interplay between co- \\nconstructing rich data, constant comparison, focused coding, theoretical cod­\\ning, memo-writing and theoretical sampling was crucial in order to develop \\na set of conceptual categories anchored in the data.\\nThis iterative process helped Thornberg et al. (2013) to construct a major \\ncategory – what Glaser (1998) and Corbin and Strauss (2015) would refer to as \\nthe ‘core concept’ of the study – namely ‘double victimizing’. This major \\n14\\nK. CHARMAZ AND R. THORNBERG\\n\\n\\ncategory refers to an interplay and cycling process between external victimiz­\\ning (i.e., a social process of repeated harmful acts directed at the victims that \\nconfirmed their victim role among the peers) and internal victimizing (i.e., \\nvictims incorporated the victim-image produced by external victimizing at the \\nsame time as they tried to develop strategies in order to protect themselves). \\nLike the term core concept (Corbin and Strauss 2015; Glaser 1998), ‘double \\nvictimizing’ was indeed the most significant and frequent code, and was \\nrelated to as many other codes as possible and accounted for more data than \\nother categories.\\nA grounded theory of this double victimizing was finally constructed in \\nwhich the other categories were parts of the double victimizing or in other \\nTable 1. Initial coding.\\nInitial Coding\\nInterview Data\\nBecoming insecure; \\nSelf-doubting; Loss of self-confidence; Thinking \\nbullying depends on wrongness with self; \\nBelieving bullies’ negative image of you; \\nGetting bad self-confidence from being bullied; \\nBecoming passive out of social fear;    \\nBelieving of the wrongness with self as a result of \\nbeing bullied; Feeling self-worthlessness; Being \\nglobally disliked;    \\nBeing bullied because of being different; \\nThe constant message of being nerdish; \\nA sense of not fitting in as a result of being \\nbullied; Inferring social deviance of self from \\nthe experiences of peer victimization; \\nA lingering sense of being different;      \\nAvoiding bullying;  \\nInhibiting the social presence of self; Believing \\nsocial invisibility prevents bullying; Inaction \\nprotects self from embarrassment and teasing     \\nStanding out leads to more bullying; \\nBecoming silent; \\nAvoiding attention;\\nInterviewer: How did the bullying affect you during this \\nperiod? \\nEric: I started to feel very insecure. In other words, I started to \\ndoubt myself more and more. I lost my self-confidence. \\nI thought there has to be something wrong with me, \\nbecause otherwise they wouldn’t have picked me as \\na victim. I believed all the stupid things they said about \\nme. So, I really got very bad self-confidence from all the \\nbullying. I really didn’t dare to do things I wanted to do \\nwhen other people were nearby. \\nInterviewer: The bullying gave you bad self-confidence? \\nEric: Yes, and it made me believe there was something wrong \\nwith me, that I was stupid. I felt worthless, that no one \\nwould like to be with me. \\nInterviewer: You said before that you thought they bullied \\nyou because there was something wrong with you. Can \\nyou tell me more about that? \\nEric: Because I was a different or a bit odd, I wasn’t like them. \\nInterviewer: You became bullied because you were different? \\nEric: Yeah, that was what I was told all the time, that I was \\na nerd, I wore ugly clothes and stuff like that. But it was \\nonly when the bullying started that I began to feel \\ndifferent, that I didn’t fit in. I didn’t think like that before. \\nBut when they started to tease me, push me around, and \\nwhen I was frozen out all the time, I began to \\nunderstand that I was different. I can still remember that \\nfeeling. \\nInterviewer: What did you do when you got bullied at school? \\nEric: I tried to avoid it. \\nInterviewer: How? \\nEric: For example, by not putting my hand up during the \\nlessons, being quiet and not standing out. I thought if \\nI didn’t stand out, if they wouldn’t notice me, then they \\nwouldn’t bully me. If I didn’t say or do things when other \\npeople were around, nothing embarrassing would \\nhappen, no one would tease me. \\nInterviewer: What do you mean? \\nEric: Well, if I said something, if I tried to take some space, \\nthen they would just say, ‘We have to put him down! We \\nhave to bully him even more!’ So, the best thing was to \\nbe quiet and not be noticed.\\nQUALITATIVE RESEARCH IN PSYCHOLOGY\\n15\\n\\n\\nways related to it. For instance, whereas external victimizing consists of the \\nsocial processes of stigmatizing by being labelled as different and social \\nexcluding, internal victimizing includes a sense of not fitting in, distrusting \\nothers, self-protecting, self-doubting, self-blaming, and resignation. Initial \\nattacks antedated double victimizing, and when double victimizing ended in \\nterms of bullying exit, after-effects of bullying included a lingering internal \\nvictimizing. In this study, theoretical sampling was necessary to reach theore­\\ntical saturation in terms of having constructed a grounded theory with theo­\\nretical completeness (Charmaz 2014; Glaser, 2001).\\nThe carefully conducted interviews and the systematic use of grounded \\ntheory methods made sure that the final grounded theory of double victimiz­\\ning in school bullying reached empirical grounding (Strauss and Corbin 1998) \\nas the theory and its concepts clearly fitted with the data (Glaser 1998), and \\ndemonstrated credibility (Charmaz 2014; Strauss and Corbin 1998) and trust­\\nworthiness (Strauss and Corbin 1998). In addition, the empirical grounding of \\nthe theory led to: (a) workability (Glaser 1998) as the developed middle-range \\ntheory contributes to explain the link between bullying victimization and \\nmental health problems in the literature (e.g., Klomek, Sourander, and \\nElonheimo 2015) from the voice and perspectives of the participants; (b) \\nabstract understanding (Charmaz, 2017) as it contributes to increase our \\nunderstanding of meanings and actions and how victims construct them; \\nand (c) relevance (Glaser 1998) and resonance (Charmaz 2014) as participants \\nin the study and other victims of school bullying could recognize most or all \\nprocesses included in the grounded theory of double victimizing. Due to the \\nopen and exploratory approach, the developed grounded theory showed \\noriginality (Charmaz 2014) as it contributes to the literature by offering new \\ninsights and proving a fresh conceptualization of victims’ experiences and path \\nthrough school bullying and beyond. As a result of empirical grounding, fit, \\nTable 2. Focused coding.\\nFocused Coding\\nInterview Data\\nSelf-inhibiting         \\nSelf-doubting\\nEric: For example, by not putting my hand up during the lessons, being quiet and not standing \\nout. I thought if I didn’t stand out, if they wouldn’t notice me, then they wouldn’t bully me. \\nIf I didn’t say or do things when other people were around, nothing embarrassing would \\nhappen, no one would tease me. \\nInterviewer: What do you mean? \\nEric: Well, if I said something, if I tried to take some space, then they would just say, ‘We have to \\nput him down! We have to bully him even more!’ So, the best thing was to be quiet and not \\nbe noticed.  \\nAnn: I felt that there had to be something very wrong with me because everyone picked on me. \\nI felt that I was worthless. I felt that I really must be a boring–, a very boring person because \\neveryone avoided me and because they teased me and because of all things they did to me. \\nI never thought that I didn’t want to live anymore. I didn’t think that way. I don’t think I did. \\nAt least I can’t recall I did. I just felt that I must be messed up in my head, and that I was \\nmuch more inferior to the others.\\nThese tables first appeared in Thornberg and Charmaz (2014).\\n16\\nK. CHARMAZ AND R. THORNBERG\\n\\n\\ncredibility, relevance, workability and resonance, Thornberg et al. (2013) \\ngrounded theory of double victimizing also demonstrated usefulness \\n(Charmaz 2014) as it contributes to creating new lines of research, clarifies \\nbullying victim experiences, and reveals that an inner victimizing seems to \\ncontinue several years after bullying has ended, which has anti-bullying \\nimplications. In accordance with the pragmatistic view of knowledge and \\nknowing, underlying the constructivist grounded theory approach (Charmaz \\n2006, 2014; Thornberg 2012), Thornberg et al. (2013) grounded theory of \\ndouble victimizing is considered to be situated, fallible and provisional, and \\ntherefore embraces modifiability (Glaser 1998). New data may revise and \\nelaborate the theory further.\\nQuality in constructivist grounded theory: checklists and guidelines\\nWe conclude by offering guidelines that graduate and postgraduate students \\nand beginners in constructivist grounded theory can use as a checklist while \\nconducting their work. Use the list as a resource, not as a substitute for \\nrigorous research. As Barbour (2001) observes, using checklists risks becoming \\na methodological crutch that avoids deep engagement with the method. We \\nstrongly advocate that researchers gain a deep engagement with their method \\nand data.\\nAlthough we can discern similarities and differences between various ver­\\nsions of grounded theory, we offer a general set of guidelines. At the same time, \\nwe recommend you learn more about issues of rigor, credibility and quality \\nemphasized in the methodological discourse of the specific version of \\ngrounded theory that you choose to use. So just take these points as flexible \\nguidelines to consider when conducting qualitative inquiry in general or \\nconstructing a grounded theory study in particular.\\n(1) Strive to achieve methodological self-consciousness (Charmaz 2017). \\nWhy have you chosen the specific topic, methodology and methods, \\nand how do these fit with who you are and your research objectives \\nand questions? What version of grounded theory have you adopted \\nand why? What are the ontological and epistemological assumptions, \\nand what do these mean for the research process, researcher position, \\nfindings, and quality issues, including transferability?\\n(2) Learn everything you can about the type of qualitative inquiry you \\nadopt, whether it’s narrative inquiry, discourse analysis, or a version of \\ngrounded theory. If possible, work with a mentor who is knowledge­\\nable about your approach.\\n(3) Take an open, non-committal, critical, analytic view of the existing \\nliterature in the field. In contrast to Glaserian grounded theory but in \\nline with Straussian and constructivist grounded theory, we \\nQUALITATIVE RESEARCH IN PSYCHOLOGY\\n17\\n\\n\\nrecommend that you review the literature to establish a defensible \\nrationale for the study, to avoid re-inventing the wheel, and to increase \\ntheoretical sensitivity. Treat the literature as provisional and fallible, \\nnot as the Truth (for further reading, see Thornberg 2012; Thornberg \\nand Dunne 2019).\\n(4) Gather rich data. For psychologists, rich data usually means learning \\nand collecting the stories of people who have had or are having \\na specific experience. Rich data means an openness to the empirical \\nworld and a willingness to try to understand the experiences of people \\nwho may be far different from you.\\n(5) Be transparent. Describe how you conducted your study, obtained \\nyour sample and state how and why you have included the partici­\\npants, and how you have used grounded theory and data collection \\nmethods. Include justifications of your choices.\\n(6) Go back and forth between data and your developing analysis to focus \\nyour subsequent data collection and to fill out your emerging analytic \\ncategories.\\n(7) Tolerate ambiguity while you struggle to gain intimate familiarity with \\nthe empirical world and to create an analytic handle to understand it.\\n(8) As you proceed, ask progressively focused questions about the data \\nthat help you develop your emerging analysis.\\n(9) Play with your data and your ideas about it. Look for all possible \\ntheoretical explanations of the data and check them.\\n(10) Collect sufficient data to (a) make useful comparisons, (b) create \\nrobust analytic categories, and (c) convince readers of the significance \\nof your categories.\\n(11) Ask questions about your categories: What are their properties? In \\nwhich ways do they subsume minor categories? How are your main \\ncategories connected? How do they make a theoretical statement? \\nWhat is the significance of this statement?\\n(12) Always treat your codes, categories and theoretical outlines as provi­\\nsional and open for revision and even rejection in the light of new data \\nand further analysis.\\n(13) After you have completed your analysis, compare it with relevant \\nmaterial from the literature, which may well include case studies and \\nperspectives that you did not address during your earlier review. At \\nthis time, your review will be focused on the ideas that you have \\ndeveloped. This review gives you the opportunity to show how your \\nanalysis fits, extends, or challenges leading ideas in your field.\\nDisclosure statement\\nNo potential conflict of interest was reported by the authors.\\n18\\nK. CHARMAZ AND R. THORNBERG\\n\\n\\nNotes on contributors\\nKathy Charmaz is professor emerita of sociology at Sonoma State University. She has written, \\nco-authored, or co-edited several books, including two award-winning works: Good Days, Bad \\nDays: The Self in Illness and Time and Constructing Grounded Theory. Charmaz has recently \\npublished The Social Self and Everyday Life: Understanding the World through Symbolic \\nInteractionism with co-authors Scott R. Harris and Leslie Irvine and is a co-editor of \\nthe Handbook of Current Developments in Grounded Theory with Antony Bryant. She has \\nalso written or co-authored numerous articles and chapters about qualitative research.\\nRobert Thornberg is professor of education at Linköping University. His current research is on \\nschool bullying, especially with a focus on social and moral processes involved in bullying, \\nbystander rationales, reactions and actions, and students' perspectives, experiences and expla­\\nnations. He is also engaged in research on student teachers' experiences of distressed situation \\nduring teacher training, and research on values education in everyday school life. Thornberg \\nhas published numerous empirical articles in a range of peer-reviewed journals and written or \\nco-authored chapters on grounded theory in various method books.\\nORCID\\nRobert Thornberg \\nhttp://orcid.org/0000-0001-9233-3862\\nReferences\\nAnczyk, A., H. Grzymała-Moszczyńska, A. Krzysztof-Świderska, and J. Prusak. 2019. The \\nreplication crisis and qualitative research in the psychology of religion. The International \\nJournal for the Psychology of Religion 29:278–91. doi:10.1080/10508619.2019.1687197.\\nAyón, C., J. Messing, M. Gurrola, and D. Valencia-Garcia. 2017. The oppression of Latina \\nmothers: Experiences of exploitation, violence, marginalization, cultural imperialism, and \\npowerlessness in their everyday lives. Violence against Women 24 (8):879–900. doi:10.1177/ \\n1077801217724451.\\nBarbour, R. S. 2001. Checklists for improving rigour in qualitative research: A case of the tail \\nwagging the dog? British Medical Journal 322:1115–17. doi:10.1136/bmj.322.7294.1115.\\nBerthelsen, C. B., S. Grimshaw-Aagaard, and C. Hansen. 2018. Developing a guideline for \\nreporting and evaluating grounded theory research studies (GUREGT). International \\nJournal of Health Sciences 6 (2):64–76.\\nBraun, V., and V. Clarke. 2006. Using thematic analysis in psychology. Qualitative Research in \\nPsychology 3 (2):77–101. doi:10.1191/1478088706qp063oa.\\nBronk, K. 2012. A grounded theory of the development of noble youth purpose. Journal of \\nAdolescent Research 27 (1):78–109. doi: 10.1177/0743558411412958.\\nBryant, A., and K. Charmaz 2007. Grounded theory in historical perspective: An epistemolo­\\ngical account. In The Sage handbook of grounded theory, A Bryant and K Charmaz (Eds.), pp. \\n31–57. London: Sage.\\nBryant, A. 2017. Grounded theory and grounded theorizing: Pragmatism in research practice. \\nNew York, NY: Oxford University Press.\\nBryant, A. 2019. The varieties of grounded theory. London: Sage.\\nCharmaz, K. 1995. Between positivism and postmodernism: implications for methods. In \\nStudies in symbolic interaction, vol. 17, N. K. Denzin (Ed.), pp. 43–72. Greenwich, CT: JAI \\nPress.\\nQUALITATIVE RESEARCH IN PSYCHOLOGY\\n19\\n\\n\\nCharmaz, K. 2006. Constructing grounded theory: A practical guide through qualitative analysis. \\nLondon: Sage.\\nCharmaz, K. 2008. The legacy of Anselm Strauss for constructivist grounded theory. In Studies \\nin symbolic interaction, ed. N. K. Denzin., vol. 32, 127–41. Bingley, UK: Emerald Publishing \\nGroup, Ltd.\\nCharmaz, K. 2014. Constructing grounded theory. 2nd ed. Thousand Oaks, CA: Sage.\\nCharmaz, K. 2017. The power of constructivist grounded theory for critical inquiry. Qualitative \\nInquiry 23 (1):34–45. doi:10.1177/1077800416657105.\\nCharmaz, K. 2020. ”With constructivist grounded theory you can’t hide”: Social justice \\nresearch and critical inquiry in the public sphere. Qualitative Inquiry 26 (2):165–76. \\ndoi:10.1177/1077800419879081.\\nChiovitti, R. F., and N. Piran. 2003. Rigour and grounded theory research. Journal of Advanced \\nNursing 44 (4):427–35. doi:10.1046/j.0309-2402.2003.02822.x.\\nCisneros-Puebla, C. A. 2007. The deconstructive and reconstructive faces of social construc­\\ntion: Kenneth Gergen in conversation with César A. Cisneros-Puebla. With an introduction \\nby Robert B. Faux [83 paragraphs]. Forum Qualitative Sozialforschung/Forum: Qualitative \\nSocial Research 9 (1):Art. 20. http://nbn-resolving.de/urn:nbn:de:0114-fqs0801204.\\nCorbin, J., and A. Strauss. 1990. Grounded theory research: Procedures, canons, and evaluative \\ncriteria. Qualitative Sociology 13 (1):3–21. doi:10.1007/BF00988593.\\nCorbin, J., and A. Strauss. 2008. Basics of qualitative research. 3rd ed. Thousand Oaks, CA: \\nSage.\\nCorbin, J., and A. Strauss. 2015. Basics of qualitative research. 4th ed. Thousand Oaks, CA: Sage.\\nElliott, N., and A. Lazenbatt. 2005. How to recognize a “quality” grounded theory research \\nstudy. Australian Journal of Advanced Nursing 22 (3):48–52.\\nFaija, C. L., S. Tierney, P. A. Gooding, S. Peters, and J. R. Fox. 2017. The role of pride in women \\nwith anorexia nervosa: A grounded theory study. Psychology and Psychotherapy: Theory, \\nResearch and Practice 90 (4):567–85. doi:10.1111/papt.12125.\\nFlick, U., ed.. 2014. The SAGE handbook of qualitative analysis. London: Sage.\\nFlick, U. 2019. Grounded theory. London: Sage.\\nGibson, P. R. 2016. Women growing older with environmental sensitivities: A grounded theory \\nmodel of meeting one’s needs. Health Care for Women International 37 (12):1289–303. \\ndoi:10.1080/07399332.2016.1191495.\\nGlaser, B. G. 1978. Theoretical sensitivity. Mill Valley, CA: Sociology Press.\\nGlaser, B. G. 1992. Basics of grounded theory analysis. Mill Valley, CA: Sociology Press.\\nGlaser, B. G. 1998. Doing grounded theory: Issues and discussions. Mill Valley, CA: Sociology \\nPress.\\nGlaser, B. G. 2001. The grounded theory perspective: Conceptualization contrasted with \\ndescription. Mill Valley, CA: Sociology Press.\\nGlaser, B. G. 2013. No preconceptions: The grounded theory dictum. Mill Valley, CA: Sociology \\nPress.\\nGlaser, B. G., and A. L. Strauss. 1967. The discovery of grounded theory: Strategies for qualitative \\nresearch. New York, NY: Aldine de Gruyter.\\nGlaser, B. G., and Strauss, A. L. 1965. Awareness of dying. Chicago: Aldine.\\nGlaser, B. G., and Strauss, A. L. 1968. Time for dying. Chicago: Aldine.\\nHallberg, L. R. M. 2006. The “core category” of grounded theory: Making constant compar­\\nisons. International Journal of Qualitative Studies on Health and Well-being 1 (3):141–48. \\ndoi:10.1080/17482620600858399.\\nHallberg, L. R. M., and M. K. Strandmark. 2006. Health consequences of workplace bullying: \\nExperiences from the perspective of employees in the public service sector. International \\n20\\nK. CHARMAZ AND R. THORNBERG\\n\\n\\nJournal of Qualitative Studies on Health and Well-being 1 (2):109–19. doi:10.1080/ \\n17482620600555664.\\nHealy, M., and C. Perry. 2000. Comprehensive criteria to judge validity and reliability of \\nqualitative research within the realism paradigm. Qualitative Market Research 3 \\n(3):118–26. doi:10.1108/13522750010333861.\\nHenwood, K., and N. Pidgeon. 2003. Grounded theory in psychological research. In Qualitative \\nresearch in psychology: Expanding perspectives in methodology and design, ed. P. M. Camic, \\nJ. E. Rhodes, and L. Yardley, 131–55. Washington: American Psychological Association.\\nHutchison, A. J., L. Johnston, and J. Breckon. 2011. Qualitative research in psychology, \\nGrounded theory-based research within exercise psychology: A critical review. Qualitative \\nResearch in Psychology 11 (3):245–72.\\nKhankeh, H. R., S. A. Hosseini, L. Rezaie, J. Shakeri, and D. C. Schwebel. 2015. A model to \\nexplain suicide by self-immolation among Iranian women. Burns 41, (7):1562–71. \\ndoi:10.1016/j.burns.2015.03.015.\\nKlomek, A. B., A. Sourander, and H. Elonheimo. 2015. Bullying by peers in childhood and \\neffects on psychopathology, suicidality, and criminality in adulthood. Lancet Psychiatry \\n2:930–41. doi:10.1016/S2215-0366(15)00223-0.\\nKing, N., and Horrocks, C. 2010. Interviews in qualitative research. Sage, London.\\nKvale, S., and Brinkmann, S. 2009. Interviews: learning the craft of qualitative research inter­\\nviewing. Sage, London.\\nLarsson, S. 2009. A pluralist view of generalization in qualitative research. International Journal \\nof Research & Method in Education 32 (1):25–38. doi:10.1080/17437270902759931.\\nLee, S. 2018. Understanding the dynamics among acculturative stress, coping, and growth: \\nA grounded theory of the Korean immigrant adolescent experience. Children and Youth \\nServices Review 94:105–14. doi:10.1016/j.childyouth.2018.09.030.\\nLincoln, Y. S., S. A. Lynham, and E. G. Guba. 2018. Paradigmatic controversies, contradictions, \\nand emerging confluences, revisited. In The SAGE handbook of qualitative research, ed. \\nN. K. Denzin and Y. S. Lincoln, 108–50. 5th ed. Thousand Oaks, CA: Sage.\\nLois, J. 2010. The temporal emotion work of motherhood: Homeschoolers' strategies for \\nmanaging time shortage. Gender & Society 24 (4): 421–446. doi: 10.1177/0891243210377762.\\nLützhöft, M., J. M. Nyce, and S. Petersen. 2010. ’Epistemology in ethnography: Assessing the \\nquality of knowledge in human factors research. Theoretical Issues in Ergonomic Science 11 \\n(6):532–45. doi:10.1080/14639220903247777.\\nMazzone, A., R. Thornberg, S. Stefanelli, C. Cadei, and S. Caravita. 2018. ”Judging by the \\ncover”: A grounded theory of same country and immigrant bullying. Child and Youth \\nServices Review 91:403–12. doi:10.1016/j.childyouth.2018.06.029.\\nMinas, M., J. Anglin, and M. K. Ribeiro. 2018. Building reciprocity: The dialectic processes of \\ncreating a grounded theory and the emergence of a theoretical framework. Qualitative \\nResearch in Psychology 15 (1):41–67. doi:10.1080/14780887.2017.1392669.\\nMorrow, S. L. 2005. Quality and trustworthiness in qualitative research in counseling \\npsychology. Journal of Counseling Psychology 52 (2):250–60. doi:10.1037/0022- \\n0167.52.2.250.\\nQin, D., and M. B. Lykes. 2006. Reweaving a fragmented self: A grounded theory of \\nself-understanding among Chinese women students in the United States of America. \\nInternational Journal of Qualitative Studies in Education 19 (2):177–200. doi:10.1080/ \\n09518390600576087.\\nRubin, J. D., S. Bell, and S. I. McCleland. 2018. Graduate education in psychology: Current \\ntrends and recommendations for the future. Qualitative Research in Psychology 15 (1):29–50. \\ndoi:10.1080/14780887.2017.1392668.\\nQUALITATIVE RESEARCH IN PSYCHOLOGY\\n21\\n\\n\\nSantiago-Deleofosse, M., C. Bruchez, A. Gavin, S. L. Stephen, and P. Roux. 2015. Complexity of \\nthe paradigms present in quality criteria of qualitative research grids. SAGE Open 5 (5). doi:  \\n10.1177/2158244015621350.\\nScull, N., C. Mbonyingabo, and M. Kotb. 2016. Transforming ordinary people into killers: \\nA psychosocial examination of Hutu participation in the Tutsi genocide. Peace and Conflict: \\nJournal of Peace Psychology 22 (4):334–44. doi:10.1037/pac0000158.\\nSeidman, I. 2006. Interviewing as qualitative research: a guide for researchers in education and \\nsocial sciences (3rd ed.), Teachers College, New York.\\nSnow, D. A., and D. M. Moss. 2014. Protest on the fly: Toward a theory of spontaneity in the \\ndynamics of protest and social movements. American Sociological Review 79 (6):1122–43. \\ndoi:10.1177/0003122414554081.\\nSong, S. J., and J. de Jong. 2013. Silence and disclosure: Intergenerational indero between \\nBurundian former child soldiers and their children. International Journal for the \\nAdvancement of Counseling 36:84–95. doi:10.1007/s10447-013-9192-x.\\nStrauss, A. 1959. Mirrors and masks: The search for identity. Mill Valley, CA: Sociology Press.\\nStrauss, A., and J. Corbin. 1990. Basics of qualitative research: Grounded theory procedures and \\ntechniques. Newbury Park, CA: Sage.\\nStrauss, A., and J. Corbin. 1998. Basics of qualitative research: Grounded theory procedures and \\ntechniques. 2nd ed. Thousand Oaks, CA: Sage.\\nThomson, S. B. 2011. Sample size and grounded theory. JOAAG 5 (1):45–52.\\nThornberg, R. 2012. Informed grounded theory. Scandinavian Journal of Educational Research \\n56:243–59. doi:10.1080/00313831.2011.581686.\\nThornberg, R., and K. Charmaz. 2014. Grounded theory and theoretical coding. In The SAGE \\nhandbook of qualitative analysis, ed. U. Flick, 153–69. London: Sage.\\nThornberg, R. 2017. Grounded theory. In The BERA/SAGE handbook of educational research, \\ned. D. Wyse, N. Selwyn, E. Smith, and L. E. Suter., vol. 1, 355–75. London: Sage.\\nThornberg, R., and C. Dunne. 2019. Literature review in grounded theory. In The SAGE \\nhandbook of current developments in grounded theory, ed. A. Bryant and K. Charmaz, \\n206–21. 2nd ed. London: Sage.\\nThornberg, R., K. Halldin, N. Bolmsjö, and A. Petersson. 2013. Victimising of school bullying: \\nA grounded theory. Research Papers in Education 28 (3):309–29. doi:10.1080/ \\n02671522.2011.641999.\\nTracy, S. J. 2010. Qualitative quality: Eight “big-tent” criteria for excellent qualitative research. \\nQualitative Inquiry 16:837–51. doi:10.1177/1077800410383121.\\nTreharne, G. J., and D. W. Riggs. 2014. Ensuring quality in qualitative research. In Qualitative \\nresearch in clinical and health psychology, ed. P. Rohleder and A. Lyons, 57–73. London: Red \\nGlobe Press.\\nTsai, L. 2017. The process of managing family financial pressures upon community reentry \\namong survivors of sex trafficking in the Philippines: A grounded theory study. Journal of \\nHuman Trafficking 3 (3):211–30. doi:10.1080/23322705.2016.1199181.\\nTsai, L., I. Seballos-Llena, and R. Castellano-Datta. 2017. Participatory assessment of a matched \\nsavings program for human trafficking survivors and their family members in the \\nPhilippines [52 paragraphs]. Forum Qualitative Sozialforschung/Forum: Qualitative Social \\nResearch 18 (2):Art. 11. http://nbn-resolving.de/urn:nbn:de:0114-fqs1702116.FQS.\\nTuason, M. T. G. 2013. Those who were born poor: A qualitative study of Philippine poverty. \\nQualitative Psychology 1 (S):95–115. doi:10.1037/2326-3598.1.S.95.\\nWilliams, E. N., and S. L. Morrow. 2009. Achieving trustworthiness in qualitative research. \\nPsychotherapy Research 19 (4–5):576–82. doi:10.1080/10503300802702113.\\nWójcik, M. 2018. The parallel culture of bullying in Polish secondary schools: A grounded \\ntheory study. Journal of Adolescence 69: 72–79.\\n22\\nK. CHARMAZ AND R. THORNBERG\\n\\n\\nYakushko, O. 2010. Stress and coping strategies in the lives of recent immigrants: A grounded \\ntheory model. International Journal for the Advancement of Counseling 32 (4):256–73. \\ndoi:10.1007/s10447-010-9105-1.\\nQUALITATIVE RESEARCH IN PSYCHOLOGY\\n23\\n\\n\\n\\n\\nClassic Grounded Theory\\n\\n\\nSAGE was founded in 1965 by Sara Miller McCune to support \\nthe dissemination of usable knowledge by publishing innovative \\nand high-quality research and teaching content. Today, we \\npublish over 900 journals, including those of more than 400 \\nlearned societies, more than 800 new books per year, and a \\ngrowing range of library products including archives, data, case \\nstudies, reports, and video. SAGE remains majority-owned by \\nour founder, and after Sara’s lifetime will become owned by \\na charitable trust that secures our continued independence.\\nLos Angeles | London | New Delhi | Singapore | Washington DC | Melbourne\\n\\n\\nClassic Grounded \\nTheory\\nApplications With Qualitative \\nand Quantitative Data\\n\\n\\nJudith A. Holton\\nMount Allison University, Canada\\nIsabelle Walsh\\nSKEMA Business School, France\\n\\n\\nFOR INFORMATION:\\nSAGE Publications, Inc.\\n2455 Teller Road\\nThousand Oaks, California 91320\\nE-mail: order@sagepub.com\\nSAGE Publications Ltd.\\n1 Oliver’s Yard\\n55 City Road\\nLondon, EC1Y 1SP\\nUnited Kingdom\\nSAGE Publications India Pvt. Ltd.\\nB 1/I 1 Mohan Cooperative Industrial Area\\nMathura Road, New Delhi 110 044\\nIndia\\nSAGE Publications Asia-Pacific Pte. Ltd.\\n3 Church Street\\n#10–04 Samsung Hub\\nSingapore 049483\\nCopyright © 2017 by SAGE Publications, Inc.\\nAll rights reserved. No part of this book may \\nbe reproduced or utilized in any form or by \\nany means, electronic or mechanical, including \\nphotocopying, recording, or by any information \\nstorage and retrieval system, without \\npermission in writing from the publisher.\\nPrinted in the United States of America\\nLibrary of Congress Cataloging-in-Publication \\nData\\nNames: Holton, Judith A., author. | Walsh, \\nIsabelle, author.\\nTitle: Classic grounded theory : applications \\nwith qualitative and quantitative data / Judith \\nA. Holton, Mount Allison University, Canada, \\nIsabelle Walsh, SKEMA Business School, \\nFrance.\\nDescription: Los Angeles : SAGE, [2017] | \\nIncludes bibliographical references and index.\\nIdentifiers: LCCN 2015041977 | \\nISBN 9781483372549 (pbk. : alk. paper)\\nSubjects:  LCSH: Grounded theory.\\nClassification: LCC H61.24 .H65 2017 | DDC \\n001.42--dc23 LC record available at http://\\nlccn.loc.gov/2015041977\\n16  17  18  19  20  10  9  8  7  6  5  4  3  2  1\\nAcquisitions Editor: Helen Salmon\\nEditorial Assistant: Anna Villarruel\\nProduction Editor: Veronica Stapleton Hooper\\nCopy Editor: Colleen Brennan\\nTypesetter: Hurix Systems Pvt. Ltd.\\nProofreader: Wendy Jo Dymond\\nIndexer: Sheila Bodell\\nCover Designer: Alexa Turner\\nMarketing Manager: Susannah Goldes\\n\\n\\nBrief Contents\\nPreface\\b\\nxi\\nAbout the Authors\\b\\nxxi\\nAn Overview of Classic Grounded Theory\\b\\nChapter 1: \\t\\nSituating Grounded Theory on the \\nResearch Landscape\\b\\n1\\nChapter 2: \\t\\nDiscovering New Theory as the End \\nPurpose of Classic Grounded Theory\\b\\n15\\nChapter 3: \\t\\nFoundational Pillars of Classic Grounded Theory\\b\\n28\\nChapter 4: \\t\\nClarifying Common Sources of Confusion in \\nGrounded Theory\\b\\n46\\nClassic Grounded Theory in Practice\\b\\nChapter 5: \\t\\nFinding Your Data\\b\\n57\\nChapter 6: \\t\\nAnalyzing Your Data\\b\\n76\\nChapter 7: \\t\\nShaping the Theory\\b\\n99\\nWriting and Evaluating Classic Grounded Theory\\b\\nChapter 8: \\t\\nWriting as an Important Part of Doing Classic GT\\b\\n120\\nChapter 9: \\t\\nWriting Classic GT for Publication\\b\\n134\\nChapter 10: \\t Evaluating Classic Grounded Theory\\b\\n152\\n\\n\\nConclusion\\b\\n159\\nAppendix A: Doing Qualitative Research Does Not Mean \\nYou Are Doing Classic GT\\b\\n162\\nAppendix B: Classic GT Using Qualitative Data\\b\\n167\\nAppendix C: Doing Classic GT Using Mixed Qualitative \\nand Quantitative Data\\b\\n190\\nAppendix D: Some Examples of Theoretical Coding \\nUsing Mixed Qualitative and Quantitative Data\\b\\n201\\nAppendix E: Examples of Diagrams as Precursors of a Theory\\b\\n206\\nGlossary\\b\\n210\\nReferences\\b\\n215\\nIndex\\b\\n228\\n\\n\\nDetailed Contents\\nPreface\\b\\nxi\\nOur Philosophical Perspective  xii\\nSome Definitions  xiii\\nHow We Have Organized This Book  xvi\\nAcknowledgments  xvii\\nAbout the Authors\\b\\nxxi\\nAn Overview of Classic Grounded Theory\\b\\nChapter 1: \\t Situating Grounded Theory on the \\nResearch Landscape  1\\nThe Early Development of Grounded Theory  1\\nBarney Glaser’s Early Influences and Ideas  2\\nGlaser and Strauss: From Collaboration to Difference  4\\nAre Glaserian and Straussian GT Reconcilable?  6\\nThe Nature and Power of Classic GT  10\\nEmergence of the Classic GT Paradigm  11\\nIn Summary  12\\nTest Your Knowledge  12\\nFurther Reading  13\\nChapter 2: \\t Discovering New Theory as the End Purpose of \\nClassic Grounded Theory  15\\nWhat Is a Theory?  16\\nThe Drive From Substantive to Formal GT  20\\nTheorizing in Rupture With Existing Literature  25\\nIn Summary  26\\nTest Your Knowledge  26\\nFurther Reading  27\\n\\n\\nChapter 3: \\t Foundational Pillars of Classic Grounded Theory  28\\nEmergence  30\\nConstant Comparative Analysis  33\\nTheoretical Sampling  37\\nDeveloping Your Stance as a Grounded Theorist  41\\nIn Summary  43\\nTest Your Knowledge  43\\nFurther Reading  44\\nChapter 4: \\t Clarifying Common Sources of Confusion in \\nGrounded Theory  46\\nEmergence Versus Preconception  47\\nDescription Versus Conceptualization  50\\nSubstantive Versus Theoretical Coding  52\\nIn Summary  54\\nTest Your Knowledge  55\\nFurther Reading  55\\nClassic Grounded Theory in Practice\\b\\nChapter 5: \\t Finding Your Data  57\\nWhy “All Is Data”  57\\nOverlooked Sources of Available Data  62\\nCollecting Your Own Data  64\\nGaining Access  64\\nInstilling a Spill  67\\nField Notes  70\\nData Collection and Analysis as an Iterative Process  72\\nIn Summary  73\\nTest Your Knowledge  74\\nFurther Reading  74\\nChapter 6: \\t Analyzing Your Data  76\\nConceptual Analysis as Distinct From Descriptive Analysis  78\\nUsing Constant Comparative Analysis  78\\nTypes and Stages of Coding in Grounded Theory  80\\nOpen Coding  81\\nSelective Coding  83\\nTheoretical Coding  86\\nCore Category Emergence  87\\nThe Key Role of Analytic Memoing in GT Analysis  89\\nThe Iterative Nature of Data Analysis in Grounded Theory  94\\nTo Software or Not to Software  95\\n\\n\\nIn Summary  96\\nTest Your Knowledge  97\\nFurther Reading  98\\nChapter 7: \\t Shaping the Theory  99\\nTheoretical Sampling  100\\nTheoretical Saturation  103\\nTheoretical Coding  104\\nTheoretical Sorting  109\\nAnalytic Rules  113\\nIn Summary  118\\nTest Your Knowledge  118\\nFurther Reading  119\\nWriting and Evaluating Classic Grounded Theory\\b\\nChapter 8: \\t Writing as an Important Part of Doing \\nClassic GT  120\\nThe Process Leading to the Readiness to Write Moment  121\\nApproaching the Literature in a GT Study  124\\nWriting GT With Mixed Quantitative and Qualitative Data  125\\nThe First Draft  127\\nWriting Blocks  128\\nPersonal Blocks  129\\nMethodological Blocks  130\\nExternal/Contextual Blocks  131\\nIn Summary  132\\nTest Your Knowledge  132\\nFurther Reading  133\\nChapter 9: \\t Writing Classic GT for Publication  134\\nDoing Versus Writing  134\\nRhetorical Issues: The Way We Write  135\\nEthical Issues: Adjusting Your Work to Fit Established \\nConventions  137\\nMethodological Issues: Writing to Be Read and Published  138\\nWriting Classic GT Using Mixed Data  143\\nWriting Conventions  146\\nSome Guidelines for Publishing Classic GT in Top Tier Research \\nJournals  148\\nIn Summary  150\\nTest Your Knowledge  150\\nFurther Reading  151\\n\\n\\nChapter 10: \\t\\nEvaluating Classic Grounded Theory  152\\nThe Consequences of the Paradigm War  152\\nCriteria for Evaluating Grounded Theory  154\\nIn Summary  157\\nTest Your Knowledge  157\\nFurther Reading  158\\nConclusion\\b\\n159\\nWhat Is GT, and Why the Continual Remodeling?  161\\nAppendix A: Doing Qualitative Research Does Not Mean \\nYou Are Doing Classic GT\\b\\n162\\nAppendix B: Classic GT Using Qualitative Data\\b\\n167\\nAppendix C: Doing Classic GT Using Mixed Qualitative \\nand Quantitative Data\\b\\n190\\nAppendix D: Some Examples of Theoretical Coding Using \\nMixed Qualitative and Quantitative Data\\b\\n201\\nAppendix E: Examples of Diagrams as Precursors of a Theory\\b\\n206\\nGlossary\\b\\n210\\nReferences\\b\\n215\\nIndex\\b\\n228\\n\\n\\nxi\\nPreface\\nS\\nince the publication of The Discovery of Grounded Theory  (Glaser \\n& Strauss, 1967), grounded theory (GT) has grown in popularity \\nto become one of the most frequently cited research methodolo­\\ngies. GT researchers come from a wide range of professional fields, \\nsuch as social work, nursing, medicine, allied health, education, man­\\nagement, and business. They are practitioners and scholars seeking to \\nexplain patterns of behavior from within their disciplines. Grounded \\ntheories offer a “sensitizing recognition” (Glaser, 1998, p. 62) through \\nhigh-impact variables that succinctly yet eloquently explain experi­\\nenced reality. The wide interest in GT has resulted in many journal \\narticles and texts intended to guide researchers in using the methodol­\\nogy. This interest in studying the nature and practice of GT—what \\nGlaser has so aptly termed “the rhetorical wrestle” (Glaser, 1998, Chap. \\n3)—extends to established scholars as well as the novice researcher \\n(Walsh, Holton, Bailyn, Fernandez, Levina, & Glaser, 2015a).\\nThe “draw and grab” (Glaser, 1998, p. 62) of GT is highly motivat­\\ning, but learning and doing GT is a delayed action learning process \\n(Glaser, 1978, p. 6; 2001, p. 1; 2003, p. 78). The importance of learning \\nthrough doing cannot be overestimated. As Glaser has often com­\\nmented in his troubleshooting seminars, a grounded theory (a theory \\nthat is grounded in data) is asymptotic; it is very close to what is hap­\\npening but never fully captures reality. Its propositional nature allows \\nfor its modification with additional data and with further skill develop­\\nment on the part of the theorist. What one misses in the first grounded \\ntheory study can be realized and modified in subsequent studies.\\nNovice classic grounded theorists often find that they are breaking \\nnew territory at their institutions. They are striving to master a \\n­\\nmethodology that not only is new to them but also is often unfamiliar \\nto supervisors and colleagues or misinterpreted as a qualitative \\nresearch approach situated within an interpretivist paradigm. While it \\n\\n\\nCLASSIC GROUNDED THEORY\\nxii\\nhas become common practice for many qualitative researchers to apply \\nsome GT techniques in their data analysis (e.g., constant comparison, \\ntheoretical sampling)—drawing on the “tenets of GT” so to speak—it \\nis important to differentiate such practices from studies that employ \\nfull classic GT methodology.\\nOur goal in this text is twofold. First, we take the reader back to the \\nroots of GT to illustrate how those roots are important to a full appreciation \\nof GT as much more than a qualitative method. Our purpose is in no way \\nintended to suggest that the immensely productive collaboration between \\nGlaser and Strauss was not an important milestone in the development of \\nGT. However, we feel strongly that failing to recognize the significance of \\nGT’s early roots has led some scholars to accord to GT a paradigmatic \\nposition that not only has restricted its use to qualitative data but, in \\ndoing so, has remodeled and reduced its theory-generating power.\\nSecond, we offer practical guidance in undertaking a research \\nstudy staying true to the classic GT practice paradigm and using the \\nfull classic GT methodology package. The topics covered in this text are \\ngrounded in the experiences of many novice researchers who have \\nparticipated in GT troubleshooting seminars or who have contacted us \\nwith questions and problems. The challenges faced are universal in \\nthat they are routinely encountered as part of the GT learning curve. \\nOffering some practical advice and examples for troubleshooting these \\nchallenges will help readers develop their own GT expertise.\\nl\\nl\\nOUR PHILOSOPHICAL PERSPECTIVE\\nIn this text, we emphasize the philosophical flexibility of classic GT as a \\n“full package” approach that can be applied to any study and any type \\nof data where the goal is to discover and generate a conceptually inte­\\ngrated theory. While GT is philosophically unbiased, we believe that no \\nmethodological book can be fully epistemologically and ontologically \\nneutral. Hence, it is essential to state from the outset that both authors of \\nthis text are critical realists, and this has probably influenced the exam­\\nples that we have chosen to illustrate various aspects of the GT process.\\nA discussion about critical realism is beyond the purpose of this \\nbook. However, it might be useful to stress that, for a critical realist, real­\\nity is multifaceted and may be perceived differently by different indi­\\nviduals and in different contexts. Critical realists consider, however, that \\nan intransitive domain exists independent of our perceptions. In the \\nontological domain, critical realism holds a realist view and, in the epis­\\ntemological domain, it accepts the “relativism of knowledge as socially \\n\\n\\nPreface\\nxiii\\nand historically conditioned” (Mingers, 2004, p. 91). Through retroduc­\\ntion, critical realism aims to discover the underlying structures that \\ngenerate specific patterns of events or non-events, and welcomes quan­\\ntitative and/or qualitative data, methods, and techniques. The critical \\nrealist retroductive approach to knowledge creation, well elaborated by \\nZachariadis, Scott, and Barrett (2013), is fully congruent with the neces­\\nsary emergent quality of grounded theories.\\nIt is also essential to clarify what causality means for a critical realist, \\nas theories are often conceived as centered on causality, that is, the rela­\\ntionship between a cause and an event (Gregor, 2006). The notion of \\ncausality as a generative mechanism is a core and defining feature of criti­\\ncal realism (Bhaskar, 2002). Generative mechanisms are best understood \\nas tendencies, as their activation is highly context dependent (Bhaskar, \\n2002). In contrast with the Humean vision of causality (A causes B) com­\\nmonly accepted in traditional positivist quantitative circles, a generative \\nmechanism can be reformulated as “A generates B in context C” \\n(Cartwright, 2003; Smith, 2010). For a critical realist, causality is, thus, a \\nprocess of how causal powers are actualized in some particular context; \\na process in which the generative mechanisms of that context shape \\n(modulate, dampen, etc.) the particular outcomes. For instance, wood in \\na camp fire (A) will have a tendency to generate heat (B) if somebody has \\nput together the right elements for the fire to burn adequately and has \\nput a light to it and if it does not rain on the fire (C). Thus, even though \\nsome regularity in events may be empirically derived, critical realism \\ndoes not look for universal laws and recognizes the significant role of \\ncontext, social structures, and individual agencies in causal explanations.\\n\\t\\nSOME DEFINITIONS\\t\\nl\\nWe refer the reader to Walsh (2014c) for a full discussion on mixed \\ndesign GT and the various works that may be found in the literature to \\nfit such a description. Some of these works use, within the same \\nresearch project, an exploratory GT stance with qualitative data and a \\nconfirmatory hypothetical-deductive stance with quantitative data. In \\nthis book we are concerned with adopting an exploratory GT stance \\nwith any data, both qualitative and/or quantitative, and from any \\nphilosophical/epistemological perspective that may be adopted by a \\nresearcher. When we mention “mixed GT” or “GT using mixed data,” \\nit implies that all data are analyzed as one set and that the constant \\ncomparative analyses of qualitative and quantitative data are not sepa­\\nrated but embedded.\\n\\n\\nCLASSIC GROUNDED THEORY\\nxiv\\nAt the outset, it is also essential to propose some definitions for \\nconstitutive elements that may be found in the design of any \\nresearch project. As highlighted by Mingers (2001), certain words \\nmay be interpreted in many different ways by different researchers; \\nthis sometimes leads to misunderstandings. To avoid this, it is \\nimportant to define some terms. We do not claim these definitions \\nto be correct across domains, but they will be used consistently \\nthroughout this book.\\nData are the necessary basis on which we build theories (Evermann \\n& Tate, 2011, p. 634), and in most cases, we need good data to produce \\ngood theory (Lyytinen, 2009). Creswell (2014) refers to qualitative data \\nas consisting of text and images and quantitative data as numbers. The \\nterm quantitative data is mostly used to describe a type of information \\nthat may be counted or expressed numerically and is amenable to sta­\\ntistical analysis whereas qualitative data provides explanation for and \\ninformation about something in the form of words and involves \\n­\\nin-depth description (Monette, Sullivan, & Delong, 2011). However, a \\nresearcher may choose to quantitize qualitative data (Sandelowski, \\nVoils, & Knafl, 2009) or to qualify (Creswell, 2007), or qualitize \\n(Sandelowski, 2000), quantitative data. A researcher may also use sec­\\nondary data in various ways, with different techniques, regardless of \\nthe methods used to collect these data.\\nMethods are the data-collection methods used in the research \\nproject, such as interviews, observation, filming, or surveys. \\nContrary to commonly accepted beliefs, we do not consider that \\nspecific methods will produce specific data: Methods traditionally \\nunderstood as quantitative may produce qualitative data (e.g., one \\nmay include open-ended questions in a survey that will yield \\nqualitative data), and methods traditionally understood as \\nqualitative may produce data that may subsequently be quantitized \\n(Sandelowski et al., 2009), for example, the number of times a term \\nis found in an interview.\\nTechniques are the instruments used in the research project to help \\nanalyze and make sense of the collected data, such as text analysis, \\ncluster analysis, or structural equation modeling. Here again, we \\ndepart from traditional beliefs considering qualitative and quantitative \\ninstruments as completely distinct. Some techniques traditionally \\nbelieved to belong to the qualitative realm may be perceived to be \\nextremely close to the quantitative realm. For instance, Glaser’s (1978) \\nsubstantive and theoretical coding may be realized through text \\nanalysis but also through elaboration analysis, factor/cluster analyses, \\n\\n\\nPreface\\nxv\\nand soft structural equation modeling—traditionally considered as \\nquantitative techniques and often used in quantitative studies.\\nMethodology is the specific combination of research methods and \\ntechniques used in a research project. Each research project may be \\nconsidered as having a specific methodology, as we understand \\nresearchers as important stakeholders in their own research, who may \\nadapt the methods and techniques they adopt.\\nFramework is the general set of guidelines proposed by some \\nauthors that a researcher may choose to follow in a given project, such \\nas action research (Baskerville & Pries-Heje, 1999), case-study research \\n(Eisenhardt, 1989), quantitative grounded theory (Glaser, 2008), or \\nStraussian grounded theory (Strauss, 1987).\\nParadigm has been understood and used in social sciences at \\ndifferent levels of generality (Morgan, 2007). The four versions of \\nparadigms found in the literature and highlighted by Morgan (2007) \\nare worldviews, epistemological stance, shared beliefs in a research \\nfield, and model examples. These are not mutually exclusive but \\n“nested within each other” (p. 54). In all four versions, paradigms are \\n“shared belief systems that influence the kinds of knowledge researchers \\nseek and how they interpret the evidence they collect” (p. 50).\\nAnother issue that has to be addressed up front is the meaning of \\nthe term grounded theory itself, because it leads to misunderstanding as \\nalready highlighted by Bryant and Charmaz (2007). When asked the \\nquestion, “What is grounded theory?” Barney Glaser has responded \\nthat GT is a set of “interrelated grounded concepts based on an \\nemergent theoretical code.”1 Walsh, Holton, Bailyn, Fernandez, Levina, \\nand Glaser (2015b) elaborate on this definition:\\nGT is the systematic generation of theory from data that has \\nitself been systematically obtained (Glaser, 1978, p. 2). GT’s \\nexploratory approach may be adopted irrespective of the \\nresearcher’s philosophical positioning. It may include \\nqualitative and quantitative data, or both.\\nThe term grounded theory describes at the same time both the \\nresearch process (the way the research is conducted) and the end result \\n(a new theory that is empirically grounded in data); it also serves as a \\nguide for the research design. In this book, and while also contextual­\\nizing our discourse, we use the abbreviation GT when we mean the \\nresearch process and the full term grounded theory when we mean \\n\\nthe end result of a GT study.\\n\\n\\nCLASSIC GROUNDED THEORY\\nxvi\\nChapters 1 to 4 provide an overview of classic GT by situating GT on \\nthe research landscape (Chapter 1), discussing theory discovery as the \\nend purpose of GT (Chapter 2), elaborating the foundational pillars of \\nclassic GT (Chapter 3), and clarifying common sources of confusion in \\nGT (Chapter 4).\\nChapters 5 to 7 offer practical guidance in the application of a \\n­\\nclassic GT framework using both qualitative and quantitative data. \\nChapter 5 explores the process of data collection, Chapter 6 guides the \\nreader through classic GT’s data analysis procedures, and Chapter 7 \\nelaborates the various procedures applied in classic GT for shaping the \\nemergent theory.\\nChapters 8 to 10 offer the reader advice on writing as an important \\npart of the classic GT process (Chapter 8) as well as on writing for an \\naudience, that is, for publication (Chapter 9). Chapter 10 addresses the \\nissue of evaluating grounded theories.\\nEach chapter features a short quiz (Test Your Knowledge) to help \\nyou quickly check your knowledge and understanding of the mate­\\nrial covered. We identify additional works (Further Reading) at the \\nend of each chapter that may also assist in enhancing your knowl­\\nedge of classic GT. Also included are five appendices, to which we \\nregularly refer as illustrations of some issues examined in several \\nchapters.\\nAppendix A provides comparative examples of data collection and \\nanalysis using a preformulated, descriptive qualitative approach with \\nclassic GT’s emergent, conceptual approach.\\nAppendix B provides an overview and selected excerpts from \\nHolton’s (2006) thesis, illustrating the application of classic GT to \\nqualitative data.\\nAppendix C provides selected methodological excerpts from \\nWalsh (2014a) illustrating the use of a classic GT approach with mixed \\ndata, methods, and techniques.\\nAppendix D is extracted and adapted from Walsh (2014c). It provides \\nsome examples of theoretical coding using mixed qualitative and quanti­\\ntative data. It also illustrates how the literature may be used as secondary \\ndata, once the core category of a grounded theory has emerged.\\nAppendix E is extracted from Walsh, Renaud, and Kalika (2013) \\nand shows an example of diagrams that helped the emergence of a \\ntheory and guided its writing up.\\nFinally, while we have no intention in this book to replicate or \\nsummarize Barney Glaser’s many publications on classic grounded \\nl\\nl\\nHOW WE HAVE ORGANIZED THIS BOOK\\n\\n\\nPreface\\nxvii\\nThis book has been “in the works” for more than 10 years. In 2004, \\nBarney Glaser suggested that I write a “troubleshooting book” based \\non the many questions and blocks to doing classic GT that the novice \\nresearcher encounters. I have since had the privilege of attending \\nmany of Barney’s troubleshooting seminars as well as offering several \\nseminars myself internationally. Many of these seminars were collabo­\\nrations with Fellows of the Grounded Theory Institute, including Tom \\nAndrews, Toke Barfod, Ólavur Christiansen, Foster Fei, Astrid \\nGynnild, Tina Johnston, Andy Lowe, Vivian Martin, Antoinette \\nMcCallin, Alvita Nathaniel, Anna Sandgren, Helen Scott, Odis \\nSimmons, Michael Thomas, and Hans Thulesius. The seminars and \\nmy 8 years as editor in chief of The Grounded Theory Review have pro­\\nvided me with a greater appreciation for the power of doing classic GT \\nas well as offering valuable insights into the challenges met by those \\nnew to the GT experience. In conversations with Barney over the ensu­\\ning years, I have had my perspective at times validated and at other \\ntimes challenged! All has been wonderful learning for which I owe \\nBarney a great debt of gratitude.\\nA chance e-mail exchange in October 2012 (in response to an \\nAcademy of Management electronic mailing list posting titled “Whoever \\nsaid that grounded theory (GT) was a qualitative methodology?”) has \\nled to another fruitful collaboration with my coauthor, Isabelle Walsh. \\nIn a very short period of intense exchange, our collaboration has \\nincluded co-organizing both a symposium (2013) and a professional \\ndevelopment workshop (2014) on GT through the Academy of \\nManagement, as well as the publication of a special issue on grounded \\ntheory in a top-tier journal, Organizational Research Methods.\\nIsabelle and I have had very different GT learning journeys but \\nshare a common philosophical perspective as critical realists and as \\nmanagement scholars. In this collaboration as well, I have at times had \\nmy perspective on GT both validated and challenged. Our collaboration \\nhas further stretched my learning and helped to firm up the unique \\nconcept for this book—a guide to the fundamentals of classic GT \\nhighlighting applications with both qualitative and quantitative data.\\n—Judith Holton, September 2015\\ntheory, we make frequent reference to his work. Studying the original \\nwork of Barney Glaser is a must for anyone seeking to do classic \\ngrounded theory (CGT).\\n\\t\\nACKNOWLEDGMENTS\\t\\nl\\n\\n\\nCLASSIC GROUNDED THEORY\\nxviii\\nI was not directly trained by Barney Glaser as such. I only attended \\none of his seminars, which I organized in Paris during the spring of \\n2014. Beyond the very limited aspects of GT that I learned from some \\nmethodological seminars I attended during my doctoral years, my \\nknowledge of GT is empirically grounded (I learned about GT as I was \\ndoing it). However, I have to highlight that I was guided by Barney’s \\nnumerous publications in my methodological endeavors. Through the \\nyears and as I was conducting my various research projects, every time \\nI wondered what I should do methodologically, which methods or \\ntechniques I could use, I just referred to the seminal book about GT \\n(Glaser & Strauss, 1967) and then to Barney’s subsequent writings. \\nThese freed my creativity, allowed me to produce some theories con­\\nsidered of some worth, and sent me on the way to becoming a recog­\\nnized scholar in my own research field.\\nI started collaborating with Judith and Barney only 3 years ago, after \\nputting, out of desperation and loneliness, a post on the Academy of \\nManagement electronic mailing list: “Whoever said that grounded \\ntheory (GT) was a qualitative methodology?” This post was like a bottle \\nthrown from my own little intellectual island into the ocean of knowledge \\nthat was out there. It was sent out of desperation as I had been working \\non a GT methodological article for the previous 5 years and was, at the \\ntime, fighting to have my vision of GT recognized by a top-tier journal \\nof my research field. Going against the mainstream in one’s research \\nfield is difficult enough. Doing it alone is a thankless task that I was \\nspared, at least for the past couple of years, thanks to Judith’s help. She \\nwas one of the many people who picked up and answered the message \\nI had posted. Our subsequent meeting and the start of our friendship \\nand work together happened at a time when I needed support to be able \\nto express my findings. I could not have done it without her. I needed \\nreassurance that what I understood as the scope and reach of classic GT \\nwere indeed what I saw them to be. And then, I wanted to share what \\n\\nI had found through the methodological article I had been working on \\nfor many years and which I finally published in 2014.\\nThis post on the Academy of Management listserv marked a \\nturning point in my intellectual journey to the land of theories. It \\nbrought within my reach fabulously rich intellects and opened the way \\nto some deep friendships. The past few years have been interwoven \\nwith sharp intellectual challenges, and I certainly would not have \\nachieved what I have done during these years without Judith and \\nBarney’s encouragement and reassurance.\\nJudith is so close to Barney’s work and Barney’s thinking that, as \\nwe were cowriting this book, I would call her “Barnette” during our \\n\\n\\nPreface\\nxix\\nintellectual fights! She has the unique asset of a thorough knowledge \\nnot only of Barney’s work but also of Barney himself as an exceptional \\nman and of his thoughts and their evolution over the years.\\nThe present book aims to give a clear and savory taste of classic GT \\nand its fundamentals. Ultimately, it should motivate readers to start \\nreading, or to read again with fresh eyes, Barney Glaser’s seminal \\npublications.\\n—Isabelle Walsh, September 2015\\nFinally, we would be remiss if we did not acknowledge the guid­\\nance, encouragement, and support that we have received from the \\neditorial team of Sage Publications, in particular, Helen Salmon, Anna \\nVillarruel, Veronica Stapleton Hooper, and Colleen Brennan. We are \\nalso grateful for the feedback of the following SAGE reviewers: Wayne \\nA. Babchuk, University of Nebraska–Lincoln; Vivian B. Martin, Central \\nConnecticut State University; Sebastien Point, EM Strasbourg Business \\nSchool; and Karthigeyan Subramaniam, University of North Texas.\\nENDNOTE\\n1.\\t E-mail exchange with second author, February 12, 2015.\\nPowerPoint slides to accompany the text are available to instructors at \\nstudy.sagepub.com/holton.\\n\\n\\n\\n\\nxxi\\nAbout the Authors\\nJudith A. Holton completed her Masters in Leadership at Royal Roads \\nUniversity, Victoria (Canada) and her PhD in Management Studies at \\nthe University of Northampton (UK). During her PhD research, she \\nwas closely mentored by Barney Glaser. She has written a number of \\nmethodological papers and coedited books about classic grounded \\ntheory and was the founding editor of The Grounded Theory Review as a \\npeer-reviewed journal dedicated to classic grounded theory research. \\nShe is currently Associate Professor of Management in the Ron Joyce \\nCentre for Business Studies at Mount Allison University, Sackville \\n(Canada), where she teaches strategy, leadership, and organizational \\nchange.\\nIsabelle Walsh defended her PhD and HDR (Habilitation à Diriger des \\nRecherches: Habilitation to Supervise Research) in Paris–Dauphine \\nUniversity (France). Beside methodological and research design issues, \\nher research themes include information systems usage, cultural issues, \\nand change management. Her research has been published in several \\ninternational top tier outlets (the European Journal of Information Systems, \\nthe Journal of Strategic Information Systems, Organizational Research Methods, \\nSystèmes d’Information & Management, etc.). Beyond her research achieve­\\nments, she has extensive corporate and consultancy experience and is \\ncurrently a full professor at SKEMA Business School in France and Head \\nof the Project Management, Information Systems and Supply Chain \\ndepartment; her teaching is aligned with her research themes.\\n\\n\\n\\n\\n1\\n1\\nSituating Grounded Theory \\non the Research Landscape\\nAfter studying this chapter, you will:\\n• understand the origins of grounded theory\\n• understand grounded theory as a general research paradigm\\n\\t\\nTHE EARLY DEVELOPMENT OF GROUNDED THEORY\\t\\nl\\nBarney Glaser has often commented that grounded theory (GT) was \\ndiscovered, not invented (Glaser, 1992, p. 7; 1998, p. 21). He explains \\nthat after completing their famous awareness of dying study (Glaser & \\nStrauss, 1965a), he and Anselm Strauss were frequently asked to \\nexplain how they had developed their theory. Glaser ­\\nsuggested to \\nStrauss that they write up their methodology; the result was the publi­\\ncation of The Discovery of Grounded Theory (Glaser & Strauss, 1967).1 \\nAlthough the book is generally acknowledged as the seminal work on \\nGT, Glaser (1992, 1998) reveals that he had begun developing the meth­\\nodological ideas that would form the foundation of GT during his \\ndoctoral work at Columbia University. What were these ideas, and \\nwhat precipitated their emergence?\\n\\n\\nAN OVERVIEW OF CLASSIC GROUNDED THEORY\\n2\\nBarney Glaser’s Early Influences and Ideas\\nGlaser completed his doctoral degree at Columbia University in \\n1961, later publishing his dissertation (Glaser, 1964b). It is worth noting \\nthe societal context of the time as one of tension and turmoil with \\ntaken-for-granted ideas and accepted norms being challenged every­\\nwhere, including the scientific community where the preeminence of \\nthe positivist, hypothetical-deductive paradigm in scientific research \\nwas being persistently questioned, particularly in terms of its effective­\\nness for the development of theory. This debate reflects the debate \\nbetween the logic of discovery and the logic of verification.2 Gibson \\nand Hartman (2014) suggest that Discovery clearly developed from this \\ndebate with Glaser and Strauss’s particular innovation being the speci­\\nfication of procedures for the discovery of theory from data (p. 75).\\nHans Zetterberg, Glaser’s doctoral supervisor, has described the \\nclimate at Columbia at this time as a long-standing debate between \\nthose favoring the advancement of sociology through theory, led by \\nRobert Merton, and those favoring its advancement through methodol­\\nogy, led by Paul Lazarsfeld (Zetterberg, 1954, as cited in Holton, 2011). \\nTheory was logically elaborated with empirical data used simply to \\nverify conjectured hypotheses. In this environment, Glaser was exposed \\nto departmental tensions and methodological innovations through the \\nwork of Lazarsfeld, Merton, and contemporaries (Glaser, 2008). He \\ndescribes being steeped in Lazarsfeld’s elaboration analysis and the \\nuse of secondary data for discovering latent patterns and unanticipated \\nrelationships among variables. Key elements of GT would later emerge \\nfrom quantitative methodology and “qualitative mathematics,” both \\ntaught by Lazarsfeld (Glaser, 1998, pp. 22–27).\\nPaul Lazarsfeld’s focus on methodology over methods and data and \\nover ideology—his “quantomania” (Glaser, 2008, p. 4)—would inspire \\nGlaser’s focus on theory discovery over verification. His methodological \\nfocus would aid in systematizing Glaser’s inherent propensity to iden­\\ntify and integrate ideas, to recognize and conceptualize patterns and \\nstructures in data. Lazarsfeld’s interest in identifying unobserved vari­\\nables through latent structure analysis taught Glaser to avoid precon­\\nceiving a research focus and to instead look for the unexpected, the \\nunanticipated in the data.\\nGlaser has specifically acknowledged Lazarsfeld’s influence as \\n“seeding” him with important methodological contributions to the \\ndevelopment of GT: index formation, interchangeability of indicators, \\nconstant comparative analysis, and core variable analysis (Glaser, \\n2005b).3 The first two methods come directly from Lazarsfeld’s work \\n\\n\\nChapter 1    Situating Grounded Theory on the Research Landscape\\n3\\n(Lazarsfeld & Thielsens, 1958, pp. 402–407) while Glaser used Lazars­\\nfeld’s method of comparative analysis as a starting point for his own \\ndiscovery and development of constant comparative analysis, a foun­\\ndational pillar of GT methodology. Glaser has commented that Lazars­\\nfeld “missed this one.  .  .  . The power of this procedure to generate \\ntheory is phenomenal. What a theoretical yield of discovery. What a \\nmiss. The constant comparative technique became the influential \\n\\nanalytic procedure of GT to generate and discover theory” (Glaser, \\n2005b, in Holton, 2011, p. 208). Gibson and Hartman (2014) suggest that \\n“rather than specifying a concept and then seeking out indicators to \\nillustrate it, in the manner of Lazarsfeld, Glaser (1978) reversed the \\nrelationship and argued that indicators indicate concepts” (p. 75). This \\nswitch from comparative analysis for testing and verifying theory to \\ntheory discovery, they suggest, was a major departure in research \\nmethodology at the time and marked a significant methodological \\ninnovation in Glaser’s work.\\nRobert Merton’s approach to theory construction was also a con­\\nsiderable influence on Glaser’s methodological development (Glaser, \\n1998, pp. 29–31). From Merton, Glaser learned to read for ideas, under­\\nlining and noting concepts as they emerged from careful reading of \\ntext (Glaser, 1998, pp. 29–30). This close reading approach comple­\\nmented his earlier training in explication de texte (meaning literally, in \\nFrench, “explanation of a text”), which he had acquired during a year \\nof study at the Sorbonne in Paris. Merton’s influence on Glaser’s meth­\\nodological development is perhaps most apparent through Merton’s \\nemphasis on conceptual integration (i.e., theoretical coding in classic \\nGT) as essential to theory development (Merton, 1968, p. 143).\\nIt was not Merton or Lazarsfeld, however, who supervised Glaser’s \\ndoctoral studies at Columbia but rather the quiet-spoken Hans Zetter­\\nberg (1954, 1961, 1962, 1965). Glaser was particularly taken with Zetter­\\nberg’s focus on the practical value of social theory and the importance of \\nempirical research as the basis for theory development. Glaser’s concern \\nfor the practical value of sociology would surface in the later articulation \\nof criteria for evaluating a good grounded theory—that is, a theory that \\nfits, works, has relevance, and is readily modifiable on the basis of new \\ndata (Glaser, 1978). Zetterberg admonished the definitional and descrip­\\ntive nature of contemporary sociology and called for more effort at mov­\\ning past simply setting forth findings and simple propositional statements \\nin favor of integrating propositions into systems, or theories (Zetterberg, \\n1954). He railed against conjectured theory and called instead for \\n“research-grounded theories” (Zetterberg, 1954, p. 20); albeit, like \\nLazarsfeld, his primary concern was with theory verification.\\n\\n\\nAN OVERVIEW OF CLASSIC GROUNDED THEORY\\n4\\nWhile Glaser has frequently acknowledged the influence of Lazars­\\nfeld, Merton, and Zetterberg on his conceptual ideation of GT, it was \\nGlaser’s autonomous creativity (Holton, 2011) that enabled him to \\nstand outside the shadow of these great men of science to develop his \\nown ideas about science and social behavior, ideas that would progress \\nand develop throughout his scholarly career and foster his determined \\nadvancement of GT. This autonomous creativity enabled him to tran­\\nscend the dominant positivist tradition, effectively rejecting that para­\\ndigm’s premise of preconceived theoretical bases in favor of developing \\ntheory from empirical data. Glaser’s ability to conceptually transcend \\nthe dominant positivist view with respect to theory development \\nwould later enable him to approach Anselm Strauss’s qualitative \\nrealm, with its rich possibilities for data, open to discovery without \\nimposing the structures of positivism’s logic of justification; this was \\nan intellectual stance that not only facilitated the emergence of GT but \\nalso laid the foundation for GT as a general methodology for research \\nopen to any epistemological perspective and using any type of data \\n(Glaser, 2003, 2005a; Holton, 2008) and as an integrative research “para­\\ndigm for discovery” (Glaser, 2005a, p. 145).\\nGlaser and Strauss: From Collaboration to Difference\\nHaving completed his PhD studies in 1961, Glaser moved back to the \\nSan Francisco Bay area, where he met and began to work with Anselm \\nStrauss (1916–1996). Strauss had moved from Chicago to the University \\nof California at San Francisco (UCSF) in 1960. His expertise was in quali­\\ntative methods with a symbolic interactionist perspective. The shift in the \\nphilosophy of science in the 1960s (Gibson & Hartman, 2014) had consid­\\nerable momentum at the Chicago school during Strauss’s time there, with \\nefforts focused on displacing the privileged status of quantitative meth­\\nodologies (e.g., variable analysis). A leading proponent of the \\n\\nChicago school, Herbert Blumer asserted that these methodologies pro­\\nduced “disconnected findings” absent of the context that characterizes a \\nsociological variable as “an intricate and inner-moving complex” (1956, \\npp. 684, 688). As such, both Glaser and Strauss were sensitive to this \\nemerging tension in methodological approaches to theory development.\\nGlaser would join Strauss at UCSF to commence work on what would \\nbecome their famous study of dying in American hospitals (Glaser & \\nStrauss, 1965a).4 The study relied largely on interviews and observational \\ndata captured through field notes. Their collaboration afforded distinct \\nyet complementary roles for the new research partners. While Strauss led \\nthe fieldwork, Glaser’s focus was methodological as he applied his \\n\\n\\nChapter 1    Situating Grounded Theory on the Research Landscape\\n5\\nattention to conceptual coding of the data. What emerged fairly quickly in \\nthe process was his recognition that the study was that of another index \\n(i.e., a core category)—an awareness index—methodologically similar to \\nLazarsfeld’s apprehension index (Lazarsfeld & Thielens, 1958) and his \\nown recognition index (Glaser, 1961, 1964a, 1964b).\\nBy affording Glaser a first opportunity to work extensively with \\nqualitative data, the Awareness study offered an opportunity to further \\nrefine methodological ideas that he had begun to develop through his \\ndoctoral work. An early paper (Glaser, 1965) sets down his procedures \\nfor taking explicit coding strategies from quantitative methodology \\nand combining these with methods for generative theory development \\nin contrast to more conventional approaches to qualitative analysis \\nthrough analytic induction. Thus, Awareness became another arena for \\nGlaser’s elaboration and refinement of earlier methodological ideas as \\nseeded during his days at Columbia—reading carefully the data, con­\\nceptually labeling ideas, comparing them for interchangeable indica­\\ntors, theoretically saturating and integrating them to explain a pattern \\nof social behavior. “The process was the same: the data was a little \\nsofter, but no less rigorously dealt with” (Glaser, 1998, p. 27).\\nAwareness generated significant interest at the time, not simply for \\nits theoretical contribution to understanding the process of dying in \\nhospitals but also for its methodological innovation. This interest \\nprompted Glaser to suggest publishing their methodology (Glaser, \\n1998, pp. 21–22). Strauss had not been trained in methodology but \\nencouraged Glaser to proceed. Glaser claims that he authored the first \\ndraft of Discovery while Strauss was on sabbatical in Europe, later shar­\\ning it with Strauss, who then added comments and wrote an additional \\nthree chapters (pp. 21–22). Suddaby (2006) suggests that “the genius of \\nGlaser and Strauss’s original methodology is that it outlines a proce­\\ndure by which formerly tacit approaches are made explicit” (p. 640). \\nHowever, this first seminal book was more an introduction than a \\n­\\nprocedural guide. Hence, Glaser went on to write a second book, \\n­\\nTheoretical Sensitivity (1978), detailing more specifically what has \\nbecome the classic methodology for doing GT.\\nAs is well documented, Glaser and Strauss were to later disagree \\nabout the precise nature of GT and would discontinue their profes­\\nsional collaboration. Consistent with their respective backgrounds, \\nStrauss would continue his research and writing from within the \\nqualitative realm while Glaser denied holding any particular perspec­\\ntive, instead espousing his relentless dictum: “All is data.” Glaser, \\nhowever, is generally recognized as having retained both the spirit \\nand the ­\\nsubstance of their original work (Locke, 2001, p. 64).\\n\\n\\nAN OVERVIEW OF CLASSIC GROUNDED THEORY\\n6\\nGlaser and Strauss (1967) start from a common position of GT as a \\ngeneral method of comparative analysis for the generation of theory \\nfrom empirical data (p. 1) in opposition to other approaches where data \\nwere collected and analyzed to verify speculative theory. Indeed, \\n­\\nGlaser (1991) reflected on their common quest and achievement in a \\ntribute to Anselm Strauss in which he credits their respective \\ncontributions—Strauss’s ability for sociological conceptualization and \\nhis own for theoretical integration and formulation via the central role \\nof core categories—as fundamental to the emergence of GT.\\nJuliet Corbin, the other individual widely recognized for her col­\\nlaboration with Anselm Strauss, acknowledges the contribution that \\nBarney Glaser made to Strauss’s methodological advancement in her \\ntribute to Strauss in the same 1991 festschrift. She comments that, early \\nin his career, Strauss struggled to analyze collected data and that his \\nfrustration was in how to penetrate the surface of data, what proce­\\ndures to use to identify and label concepts, and how to relate concepts \\nin theoretical formulations. She suggests that it is doubtful that Strauss \\nwould ever have been able to clarify procedures if he had not worked \\nwith Barney Glaser.\\nIt is generally accepted that Discovery is more an outline of the \\napproach than it is a procedural guide. We can, however, look to sub­\\nsequent texts to explore more specifically each author’s approach to \\nteaching GT, offering insight into their eventual methodological diver­\\ngence. Glaser’s 1978 publication of Theoretical Sensitivity was based in \\nlarge measure on his experiences of teaching GT to graduate students \\nat UCSF. Nine years later, Strauss published Qualitative Analysis for \\nSocial Scientists (1987), which draws heavily on his experiences of \\nteaching qualitative methodology to graduate students, including his \\napproach to GT. In these two texts, we glimpse a marked contrast in \\nhow each author “teaches” GT that may offer us a view to their respec­\\ntive perspectives.\\nGlaser’s (1978) text extends the discussion commenced in Discovery \\nby setting forth a systematic approach to analysis and also by explicating \\nthe general and specific nature of GT, integrating and elaborating on \\nmethodological advances discovered and developed through his years \\nof teaching. Beginning with a chapter on the importance of theoretical \\npacing, Glaser leads the reader through the full package of GT \\ntechniques—sampling, coding, memoing, integrating, and writing up. \\nIndeed, his explication of GT methods is so well articulated that 9 years \\nlater, Strauss quotes extensively from the 1978 text in the introduction \\nl\\nl\\nARE GLASERIAN AND STRAUSSIAN GT RECONCILABLE?\\n\\n\\nChapter 1    Situating Grounded Theory on the Research Landscape\\n7\\nto one of his own texts (Strauss, 1987). As Strauss comments in his \\nPreface, “indeed the second half of that chapter is essentially his \\n[Glaser’s] except for some amplification” (1987, p. xiv). However, we \\nbelieve that what Strauss proceeds to offer in his text illustrates a \\nfundamental difference in how each author perceives GT and what each \\nconsiders essential to doing GT.\\nGlaser (1978) explains the various GT procedures largely absent \\n\\nof specific detailed examples. He writes conceptually, giving the reader \\na sense of the methods and how they might be employed without pre­\\nscribing precisely how they need to be employed. He does so without \\ndetailed procedural specification, leaving the researcher more scope for \\ncreativity and flexibility. Partington (2000) suggests that “Glaser was \\nless keen to see grounded theorists following an orthodoxy of \\napproach, preferring to direct his attention to ways of enhancing \\nresearchers’ latent creativity” (p. 94). Options are offered, as in theoreti­\\ncal coding families, but none is prescribed as required. Apart from sug­\\ngesting questions for substantive coding of data, the actual mechanics \\nof ­\\ncoding, for example, are left largely to the researcher’s discretion. \\nExamples are footnoted for the reader’s convenience, but they do not \\ninterfere with general explanations offered in the text.\\nBy contrast, Strauss (1987) emphasized a need for structure to aid \\nanalysis—in particular, his coding paradigm and axial coding with \\nemphasis on conditions, interactions, strategies, tactics, and ­\\nconsequences \\n(pp. 27–28, 32–33). He described GT as\\na style of doing qualitative analysis that includes a number of \\ndistinct features, such as theoretical sampling, and certain \\nmethodological guidelines, such as the making of constant \\ncomparisons and the use of a coding paradigm, to ensure con­\\nceptual development and density. (p. 5)\\nHis emphasis in theory generation is on change as a constant in \\nsocial life, social interaction and social processes at the center of atten­\\ntion (p. 6). The approach is much more descriptive. While Strauss \\n(1987) embraced the general ideas and the power in doing GT, unlike \\nGlaser he was clearly less comfortable trusting in the creative ideation \\nthat emerges from trusting in the systematic and sustained analysis of \\ndata. His insistence on structure in advance and throughout the pro­\\ncess preempts encouraging the researcher to remain open and to trust \\nin emergence. Corbin (1991) affirms Strauss’s more structured approach \\nwith its emphasis on action, interaction, conditions, constraints, and \\nconsequences.\\n\\n\\nAN OVERVIEW OF CLASSIC GROUNDED THEORY\\n8\\nAs a guide to doing qualitative research, Strauss (1987) offers some \\nexcellent advice and many rich descriptive examples. However, he \\ntends to overthink the process, potentially undermining researcher \\nautonomy by restraining the creativity and flexibility inherent in GT’s \\ntheory-generating power. Perhaps Strauss’s many years as a supervisor \\nof students had produced in him a need for cautionary control of the \\nnovice researcher. Apart from the Introduction chapter, which, as noted \\nearlier, draws largely on Glaser’s (1978) text, subsequent chapters con­\\ntain extensive detailed descriptive excerpts from Strauss’s own research \\nand that of various students. It would seem that each man valued \\nsomething different as crucial to developing the ability to do GT. For \\nGlaser, it was presenting the methodological ideas and trusting in the \\nresearcher’s ability to understand and work with those ideas, develop­\\ning skill over time. For Strauss, what was most important was ensuring \\nsufficient descriptive examples, perhaps feeling that with good exam­\\nples “how qualitative analysis is actually done is made vivid” and that \\n“one can literally see it being taught” (Strauss, 1987, p. xii).\\nThe Straussian approach suggests that the experience, knowledge, \\nand interest of the researcher should come into play in the initial focus­\\ning of the study (see Strauss, 1987, pp. 10–11). While acknowledging \\nthat the researcher will be motivated to explore a general area of inter­\\nest, the Glaserian approach challenges the researcher to set aside pre­\\nconceived professional ideas and remain open to what emerges as the \\nmain concern in the area under study. While some may simply reject \\nGlaser’s call for openness as impossible, opting instead for the Strauss­\\nian approach with its specified analytic tools and coding frameworks, \\nthe latter can be a slippery slope, particularly for those trained in quali­\\ntative methods. As is illustrated in many published papers (that \\nwrongly and with a very limitative perspective) identify GT as a quali­\\ntative method, authors mistakenly apply standard precepts of estab­\\nlished research procedures traditionally applied within academia (e.g., \\ninitial literature reviews, theoretical frameworks, specified research \\nquestions, interview protocols, taping and transcribing) as standard \\naspects of the GT package.\\nThis emphasis on preconceiving the research process, a need to \\nknow in advance and to map out in detail the research journey, became \\nso evident in the work of Strauss and Corbin (1990) that it would mark \\na clear methodological divergence from their earlier collaboration. \\n­\\nGlaser’s (1992) response sets forth in great detail the inconsistencies \\nwith the basic tenets of GT in Strauss and Corbin’s (1990) text, declar­\\ning their work to be a completely different methodology. Gibson and \\n­\\nHartman (2014) affirm Glaser’s concerns, suggesting that\\n\\n\\nChapter 1    Situating Grounded Theory on the Research Landscape\\n9\\nin the need to make things simple, [Strauss and Corbin] pre­\\nsented the name of the things that made up grounded theory \\nbut changed the notion of what grounded theory was. They \\ndid so without carefully justifying their changes with reference \\nto the original texts. (p. 97)\\nReferring to Strauss and Corbin (1998), Hernandez (2010) asserts \\nthat\\nthe method described in that book, published after Strauss’s \\ndeath in 1996, deviated completely from original grounded \\n­\\ntheory methodology (Glaser & Strauss, 1967) because of its \\ndescriptive, deductive, and verificational focus—as opposed \\nto a grounded theory’s explanatory, inductive, and discovery \\nfocus. (p. 152)\\nHernandez continues: “Corbin (2007) has called their method \\nStraussian Grounded Theory and has asserted that it is not a research \\nmethodology but rather a qualitative data analysis approach” (p. 152). \\nGlaser would conceptualize this lexical takeover in his 2009 book, \\n­\\nJargonizing: Using the Grounded Theory Vocabulary, in which he suggests \\nthat, perhaps ironically, vocabulary is GT’s greatest gift to the research \\nworld.\\nGiven the strong positions that appear to exist within the research \\ncommunity regarding this famous split, it is worth noting that Basics \\nof Grounded Theory Analysis: Emergence vs. Forcing (1992)5 was written \\nonly after Glaser had made several attempts to convince Strauss to \\nwithdraw the earlier book that Strauss had coauthored with Juliet \\nCorbin (Strauss & Corbin, 1990). Rather than withdraw, Strauss \\nencouraged Glaser to publish his objections. While much has been \\nmade of the famous schism, with supporters on both sides defending \\nclearly divergent positions, it is worth noting that Glaser’s criticism \\nwas confined to asserting an absence of good scholarship on the part \\nof Strauss and Corbin in not differentiating their methods for qualita­\\ntive analysis from the original GT methodology and for not acknowl­\\nedging the roots of GT in quantitative methodology and qualitative \\nmathematics.6\\nIn the opening paragraphs of Strauss and Corbin’s (1998) second \\nedition, they relate a story about two famous French artists, Cezanne \\nand Monet, in which Cezanne is purported to have said of Monet, “He \\nis only an eye—but what an eye!” They suggest that although Monet’s \\nway was different from Cezanne’s way, it was indeed just as insightful \\n\\n\\nAN OVERVIEW OF CLASSIC GROUNDED THEORY\\n10\\nand valuable. This story perhaps epitomizes the “great divide” \\nbetween the Glaserian (i.e., classic GT) and Straussian approaches. \\nHere, Corbin (writing after Strauss’s death in 1996) subtly acknowl­\\nedges that what they (Strauss and Corbin) had produced was indeed a \\ndifferent methodology—just as insightful and valuable but different. \\nHere we agree with Gibson and Hartman (2014) that Strauss and \\nCorbin’s approach had fundamentally changed the notion of GT as it \\nhad been intended.\\nl\\nl\\nTHE NATURE AND POWER OF CLASSIC GT\\nGT is the systematic generation of theory from data that has itself been \\nsystematically obtained (Glaser, 1978, p. 2). A grounded theory, the \\nresult of a GT study is not the reporting of facts but the generation of \\nprobability statements about the relationships between concepts, a set \\nof conceptual hypotheses/propositions developed from empirical data \\n(Glaser, 1998, pp. 3, 22). GT offers a transcending view of a main con­\\ncern in a substantive area and the social behavior that explains how \\nthat concern is processed, managed, or resolved.\\nTo understand the nature of the classic GT methodology, one must \\nunderstand the distinction between conceptualization and description. \\nClassic GT is not about accurate, descriptive reporting of findings in a \\nstudy, nor is it an act of interpreting meaning as ascribed by the partici­\\npants in a study; rather, it is an act of conceptual abstraction. For a clas­\\nsic grounded theorist, what matters are the concepts, the relationships \\namong those concepts, and their power to offer the reader a conceptual \\nexplanation of significant behavior within the social setting under \\nstudy, free of the particularistic details of specific empirical incidents \\n(indicators). “Without abstraction from time, place and people, there \\ncan be no multivariate, integrated theory based on conceptual, hypo­\\nthetical relationships—descriptions cannot be related to each other as \\nhypotheses as there is no conceptual handle” (Glaser, 2002, p. 267).\\nWhile descriptive findings in qualitative research are most certainly \\nvaluable, they do not provide conceptual abstraction. A GT must explain, \\nnot merely describe, what is happening in a social setting. This ability to \\nabstract the conceptual idea from empirical indicators (incidents in the \\ndata under analysis), without the burden of descriptive detail, is what \\ndistinguishes classic GT methodology from qualitative data analysis. This \\nabstraction to a conceptual level theoretically explains rather than \\ndescribes behavior that occurs conceptually and generally in many \\ndiverse groups with a common concern (Glaser, 2003).\\n\\n\\nChapter 1    Situating Grounded Theory on the Research Landscape\\n11\\nWith this goal of conceptual abstraction, classic GT methodology \\nuses data of all types and media and accommodates a range of episte­\\nmological and ontological perspectives without having to espouse any \\none perspective; that is, the methodology is epistemologically and \\nontologically flexible. Whether data are viewed as interpreted or objec­\\ntive is immaterial in classic GT methodology, as it is not the descriptive \\ndetail or the way in which the data are “constructed” that concerns the \\ngrounded theorist; rather, it is the discovery of abstract concepts that \\nlie within the data. Conceptual abstraction frees the researcher from the \\nemphasis on detailed description and elucidation of multiple perspec­\\ntives. The skill of the grounded theorist is to abstract concepts by leav­\\ning the detail of the data behind, lifting the concepts above the data and \\nintegrating them into a theory that explains the latent social pattern \\nunderlying the behavior in a substantive area (Locke, 2001). Research­\\ners’ philosophical stance will certainly come into play in whatever \\nsubstantive area of interest they may choose as their initial focus and \\nwhat they consider to be appropriate data sources. This aside, how­\\never, the full package of GT can be applied to any data, and indeed, this \\nfull package needs to be applied to merit its distinction as a grounded \\ntheory.\\n\\t\\nEMERGENCE OF THE CLASSIC GT PARADIGM\\t\\nl\\nDespite the widespread embrace of GT by qualitative researchers, a \\ngrowing number of theorists are adopting Glaser’s view that GT is a \\ngeneral research methodology occupying its own distinct paradigm on \\nthe research landscape (Holton, 2008). The use of any and all types of \\ndata in classic GT, together with its ontological and epistemological \\nflexibility, clearly distinguishes GT from the precepts of other research \\nparadigms. Many qualitative studies labeled as grounded theories sim­\\nply fall short of the criteria of a classic grounded theory. Such studies \\nhave been “remodeled” to meet the criteria of a qualitative framework \\n(Glaser, 2003; Glaser & Holton, 2004). In the process, researchers sacri­\\nfice the power of emergence that is fundamental to classic GT.\\nAs a general research methodology, GT uses any and all types of \\ndata. For grounded theorists, using both qualitative and quantitative \\ndata opens a vast realm of additional empirical possibilities for gener­\\nating theory. Embracing GT with mixed qualitative and quantitative \\ndata (Walsh, 2014a, 2014b, 2014c) will also encourage further scholar­\\nship among established GT researchers eager to advance their analytic \\nskills by learning to work with quantitative indices as sources of data.\\n\\n\\nAN OVERVIEW OF CLASSIC GROUNDED THEORY\\n12\\nPerhaps it is inevitable that GT’s tremendous power and its inher­\\nent flexibility invite methodological evolution and plurality. That being \\nsaid, when evolution and its consequent plurality create “drift” \\nwhereby the key tenets of GT are not sustained as they were intended, \\ncan we still accept claims to grounded theory? While much excellent \\nresearch is done under the banner of some qualitative frameworks, only \\nstudies that have employed the full GT methodological package should \\nbe accepted as grounded theory. It is our position in this text that recog­\\nnizing this distinction is important to advancing methodological schol­\\narship. This distinction, of course, raises a key question that we address \\nin this text: What are the essential elements of methodology that must be evi­\\ndent for a study to claim that it is a grounded theory in the classic sense?\\nIN SUMMARY\\nGT is about the discovery of theory from data systematically collected \\nand analyzed without structured forcing through predetermined theo­\\nretical frameworks. Contrary to popular belief, GT is not a qualitative \\nresearch method. It is a general research methodology that uses any \\nand all types of data, both quantitative and qualitative. As such, GT is \\nnot confined to any one philosophical perspective. It is ontologically \\nand epistemologically flexible, accommodating any philosophical per­\\nspective as espoused by the researcher. Classic GT is a “full package” \\nmethodology that progresses a study from initial data collection and \\nanalysis through to the presentation of an integrated theoretical expla­\\nnation of a latent pattern of social behavior. Classic GT originated with \\nthe early work of Barney Glaser using quantitative data and was fur­\\nther developed and articulated through Glaser’s famous collaboration \\nwith Anselm Strauss. More recently, a growing community of classic \\nGT scholars has espoused GT as a distinct research paradigm with its \\nown distinct perspective and practices.\\nTEST YOUR KNOWLEDGE\\nIn the following questions, one or several answers may be correct.\\n1.\\t GT is about\\na.\\t Theory verification\\nb.\\t Theory building\\nc.\\t Deductive reasoning\\n\\n\\nChapter 1    Situating Grounded Theory on the Research Landscape\\n13\\nd.\\t Inductive reasoning\\ne.\\t None of the above\\n2.\\t GT originated\\na.\\t From Glaser and Strauss’s 1967 book\\nb.\\t From Glaser and Strauss’s collaboration\\nc.\\t From Glaser’s doctoral work\\nd.\\t From Strauss’s doctoral work\\ne.\\t None of the above\\n3.\\t GT can be applied\\na.\\t Only with qualitative data\\nb.\\t Only with quantitative data\\nc.\\t With any type of data\\nd.\\t Only with data collected from interviews\\ne.\\t None of the above\\n4.\\t In classic GT it is important to\\na.\\t Describe in detail a substantive area\\nb.\\t Report on facts in a detailed manner\\nc.\\t Maintain an interpretive stance\\nd.\\t Conceptualize from empirical indicators\\ne.\\t None of the above\\nFURTHER READING\\nGlaser, B. G. (1998). The roots of grounded theory. In Doing grounded theory: \\nIssues and discussions (pp. 21–33). Mill Valley, CA: Sociology Press.\\nHolton, J. A. (2011). The autonomous creativity of Barney G. Glaser: Early influ­\\nences in the emergence of classic grounded theory methodology. In \\n­\\nA. Gynnild & V. Martin (Eds.), Grounded theory: The philosophy, method and \\nwork of Barney Glaser (pp. 201–223). Boca Raton, FL: BrownWalker Press.\\nWalsh, I. (2014). Using quantitative data in mixed-design grounded theory \\nstudies: An enhanced path to formal grounded theory in information sys­\\ntems. European Journal of Information Systems. Advance online publication. \\ndoi:10/1057/ejis.2014.23\\nENDNOTES\\n1.\\t Hereafter referred to as Discovery.\\n2.\\t Gibson and Hartman (2014) offer valuable insight into this context. See \\nin particular pp. 5–13.\\n\\n\\nAN OVERVIEW OF CLASSIC GROUNDED THEORY\\n14\\n3.\\t See also Glaser and Strauss (1967, Chap. 8).\\n4.\\t Hereafter referred to as Awareness.\\n5.\\t Hereafter referred to as Basics.\\n6.\\t Despite Glaser and Strauss’s obvious methodological divergence, \\nGlaser has continued to acknowledge and respect Anselm Strauss as a \\nclose personal friend.\\n7.\\t The \\nCreative \\nCommons-Attribution-NonCommercial-ShareAlike \\nLicense 4.0 International applies to all works published by the \\nInternational Journal of Qualitative Methods. Copyright for articles \\npublished in the International Journal of Qualitative Methods remains \\nwith the first author.\\n\\n\\n15\\n2\\nDiscovering New Theory as \\nthe End Purpose of Classic \\nGrounded Theory1\\nAfter studying this chapter, you will:\\n• be aware of the different meanings attached to the word theory as they \\nrelate to different philosophical assumptions\\n• understand GT as a process aimed at theory discovery and congruent \\nwith any philosophical perspective\\nM\\nany social sciences are still lacking in the area of theorizing, \\nand we need a plurality of theories in order to make sense of \\nthe complex and dynamic world that surrounds us (Junglas \\n\\net al., 2011). “Developing theory is what we are meant to do as \\n­\\nacademic researchers and it sets us apart from practitioners and \\n­\\nconsultants” (Gregor, 2006, p. 613). Hence, the ultimate aim of GT \\nbeing theory building, there is little doubt of its importance and fit in \\nany academic career. However, it is first essential to stop and reflect on \\nthe meaning of the word theory.\\n\\n\\nAN OVERVIEW OF CLASSIC GROUNDED THEORY\\n16\\nAs we saw in the introduction, we recently asked Barney Glaser, \\n“What is a grounded theory?” His answer was “interrelated \\ngrounded concepts based on an emergent theoretical code” (per­\\nsonal communication, February 12, 2015). However, many different \\nviews of what a theory is have been given in the literature from etic \\n(i.e., nomothetic) or emic (i.e., idiographic) perspectives; we need to \\nexamine these views in detail in order to truly understand the vari­\\nous ways in which GT is being applied all over the world and in \\nmany different disciplines. These perspectives are often linked to \\nthe researcher’s background, training, and philosophical assump­\\ntions. For instance, Bacharach (1989) views a theory “as a system of \\nconstructs and variables in which the constructs are related to each \\nother by propositions and the variables are related to each other by \\nhypotheses”’ (p. 498), whereas Weick (1995) views theories as \\n“approximations” of a complex reality that one has to make sense \\nof. However, these definitions are not philosophically neutral. We \\naccept the all-encompassing definition provided by Gregor (2006), \\nwhich is as philosophically neutral as possible and which views \\n“theories as abstract entities that aim to describe, explain, and \\nenhance understanding of the world and, in some cases, to provide \\npredictions of what will happen in the future and to give a basis for \\nintervention and action” (Gregor, 2006, p. 616).\\nGregor investigates the structure and form of theories. She argues \\nthat the four primary goals of theory are analysis and description, \\nexplanation, prediction, and prescription (Gregor, 2006). She shows \\nthat combinations of these goals lead to five types of theories that are \\ninterrelated, answer different types of questioning, and which should \\nbe combined: analytic (type I: What is?), explanatory (type II: What \\nis? How? Why? When? Where?), predictive (type III: What is? What \\nwill be?), explanatory and predictive (type IV: What is? How? \\n­\\nWhy?  When? Where? What will be?), and theory for design and \\naction (type V: How to do?). Usually, one must analyze before one can \\nexplain, predict, or act. And, preferably, one must attempt to explain \\nand ­\\npredict, if at all possible, before one actually acts. This is \\n­\\nsummarized in Table 2.1.\\nWhen doing classic GT, type I is not sufficient, and at least type II \\nmust be achieved. As we emphasize in this book, classic GT may be \\napplied with any data, qualitative and/or quantitative, and is compati­\\nble with any philosophical assumptions. Different data may be collected \\nl\\nl\\nWHAT IS A THEORY?\\n\\n\\n17\\nChapter 2    Discovering New Theory as the End Purpose of Classic Grounded Theory\\nTable 2.1    Different Types of Theory (Adapted from Gregor, 2006)\\nTheory type\\nAnswers the following \\nquestions . . .\\nDescription\\nI\\nAnalytical\\nWhat is it? \\nWhat is happening?\\nThe theory analyzes and \\ndescribes phenomena \\nbut no causal \\nrelationship is \\nhighlighted.\\nII\\nExplanatory\\nWhat is it? What is \\nhappening? How? Why? \\nWhen? Where?\\nThe theory explains but \\ndoes not predict. It does \\nnot highlight \\npropositions.\\nIII\\nPredictive\\nWhat is it? What is \\nhappening? What will be? \\nWhat will happen?\\nThe theory predicts and \\nhighlights testable \\npropositions but does \\nnot develop justified \\ncausal explanations.\\nIV\\nExplanatory \\nand \\npredictive\\nWhat is it? What is \\nhappening? How? Why? \\nWhen? Where? What will \\nbe? What will happen?\\nThe theory predicts and \\nalso highlights testable \\ncausal explanations.\\nV\\nPrescriptive\\nWhat should be done? \\nHow should it be done?\\nThe theory explicitly \\nprescribes (methods, \\ntechniques, principles).\\nthrough various methods (e.g., interviews, focus groups, photograph­\\ning, filming, surveys, etc.) and analyzed with the help of different tech­\\nniques (e.g., text analysis, cluster analysis, etc.) through the constant \\ncomparative method. The resulting theory may differ from one researcher \\nto the next, depending on researchers’ philosophical assumptions, the \\ndata they choose as appropriate, and their previous training.\\nOver the past 20 years, an important debate has ensued about the \\nrationale for combining methods and techniques previously consid­\\nered as incompatible due to philosophical assumptions and world­\\nviews presumed to be linked to these (Bryman, 1998; Hall, 2012). This \\ndispute has led to philosophical caricatures (Bryman, 1998) at each end of \\na continuum (see Figure 2.1).\\n\\n\\nAN OVERVIEW OF CLASSIC GROUNDED THEORY\\n18\\nAt one end, quantitative purists espouse a positivist philosophy \\nand adopt the “natural science model” (Bryman, 1998) with indepen­\\ndent and dependent variables and quantitative data (Lee & Hubona, \\n2009): The observer is separate from the object of study; science is \\nobjective and aims to uncover laws that are time- and context-free; and \\nthe focus is on deduction, confirmation, hypothesis testing, and quan­\\ntitative analysis (Johnson & Onwuegbuzie, 2004). Causality is under­\\nstood from a Humean perspective (“A causes B”).\\nAt the other end of the continuum, qualitative purists espouse an \\ninterpretive philosophy associated with ethnography, hermeneutics, and \\nsome forms of case research, interpreting settings and contexts to take \\nthe “natives’ point of view” (Lee & Hubona, 2009, p. 238): Realities are \\nmultiple and socially constructed so that knower and known cannot be \\nseparated; research is subjective and value-bound; causes and effects \\ncannot be differentiated; and the focus is on induction, exploration, \\n­\\ndiscovery, and qualitative analysis (Johnson & Onwuegbuzie, 2004).\\nHowever, many researchers have now reached an agreement on \\nsome major issues (Johnson & Onwuegbuzie, 2004) regarding theoriz­\\ning, whether they use qualitative or/and quantitative data and \\n­\\ntechniques and even if their philosophical assumptions may vary \\n\\n1- Data and techniques:\\n \\nQuantitative\\n2- Reality: Governed \\n \\nby laws that are\\n \\ntime- and context- free\\n3- Focus:\\n \\nDeduction,\\n \\nconfirmation, \\n \\nhypothesis testing\\n4- Causality:\\n \\nA causes B\\nInterpretivism\\n1- Data and techniques: \\n \\nQualitative\\n2- Reality: Multiple\\n \\nand Socially\\n \\nconstructed\\n3- Focus: \\n \\nInduction, exploration, \\n \\nand discovery\\n4- Causality: \\n \\nCauses and effects \\n \\ncannot be separated\\nPost Positivism\\nCritical Realism\\nPragmatism\\nA\\nz\\nA\\nB\\np\\nY\\nw\\nc\\nA\\na\\nA\\nn\\no\\nw\\nB\\nw\\nU\\n2\\n1\\n9\\n7\\n5\\n4\\n8\\n3\\n5\\n6\\n4\\n3\\n8\\n8\\n9\\n2\\n6\\n1\\n3\\n4\\n3\\n2\\n5\\n6\\n7\\n3\\n1\\nA\\nU\\nc\\nz\\nn\\nA\\nA\\nM\\nM\\nT\\nPhilosophical Continuum\\nCaricature\\nPositivism\\nCaricature\\nFigure 2.1    \\u0007\\nCaricatures at Each End of the Philosophical Continuum \\n(Adapted from Walsh, Holton, Bailyn, Fernandez, Levina, \\n& Glaser, 2015a)\\n\\n\\n19\\nDifferent befitting theories\\nfor one set of data\\nProbabilistic evidence\\nis not final proof\\nRelativity of reasoning\\nObservation as an\\napproximation of reality\\nBeliefs are embedded\\nin assumptions\\nFigure 2.2    Consensual Perspective\\n(see Figure 2.2): Reasoning is relative and varies among people; obser­\\nvation is an approximation of reality; a single set of empirical data can \\nyield different befitting theories; hypotheses are linked to assumptions; \\nprobabilistic evidence is not final proof; and researchers’ beliefs are \\nembedded in the assumptions of their respective communities.\\nOne reason these caricatures developed is that philosophical assump­\\ntions have been mapped mainly as an opposition between nomothetic \\nperspectives and more idiographic perspectives. This element, and the \\ncoupling of quantitative data and techniques with a nomothetic approach \\nversus qualitative data and techniques with an idiographic approach, \\ncompletely ignores another dimension that opposes confirmatory, theory-\\ndriven research and exploratory data-driven research.\\nThe Possibility of Using Any Type of Data With Any Philosophical \\nStance to Do GT\\nIn Awareness (Glaser & Strauss, 1965a), and through the use of qualitative \\ndata, Glaser and Strauss highlight an awareness index that explains the social \\ninteractions surrounding a dying person. This index makes sense at the \\nsame time in Glaser’s post-positivist worldview and in Strauss’s symbolic \\ninteractionist perspective. Walsh (2010, 2014a) uses mixed qualitative and \\nquantitative data/techniques in a critical realist stance to propose a digital \\nculture index that explains information technology usage; that is, why people \\nuse (or do not use) information technologies, for what purpose, and how \\nthey do so through the emergence of their digital needs.\\nChapter 2    Discovering New Theory as the End Purpose of Classic Grounded Theory\\n\\n\\nAN OVERVIEW OF CLASSIC GROUNDED THEORY\\n20\\nWhen theorizing, researchers produce theories at different levels of gen­\\neralization; these are sometimes linked to the theory’s level of abstrac­\\ntion. For instance, a metatheory is at a high level of abstraction and \\nprovides “a way of thinking about other theories” (Gregor, 2006, p. 616); \\na grand theory is abstract, scarce in observational details, and \\n“‘unbounded in space and/or time’ (p. 616), whereas some empirical \\ngeneralizations are rich in observational details but often “bounded in \\nspace and time” (Bacharach, 1989, p. 500). There are, however, different \\nconceptions of generalizability, which “refers to the validity of a theory \\nin a setting different from the one where it was empirically tested and \\nconfirmed” (Lee & Baskerville, 2003, p. 221). Lee and Baskerville (2003) \\nidentify different types of generalizability and generalizing beyond sta­\\ntistical, sampling-based generalizability: generalizing from empirical \\nstatement to empirical statement (i.e., from data to description), from \\nempirical statement to theoretical statement (i.e., from description to \\ntheory), from theoretical statement to empirical statement (i.e., from \\ntheory to description), and from theoretical statement to theoretical \\nstatement (i.e., from concepts to theory). This is summarized in Table 2.2.\\nIn the case of a grounded theory, the generalization is done from \\ndata to theory through conceptualization, even though one often starts \\nfrom data to description in order to see through and understand the \\ndata, that is, what they are telling us (i.e., indicating) in order to con­\\nceptualize. It must, however, be highlighted that one cannot claim to \\nhave done classic GT if one remains at the descriptive level, short of \\nproducing a conceptual theory.\\nl\\nl\\nTHE DRIVE FROM SUBSTANTIVE TO FORMAL GT\\nGeneralization\\nToward empirical elements\\nToward theoretical \\nelements\\nFrom empirical \\nstatements\\nDescription of data\\nFrom description to theory\\nThe researcher measures, \\nobserves, describes\\nThe researcher \\nconceptualizes from her/\\nhis data\\nFrom theoretical \\nstatements\\nFrom theory to description\\nFrom concepts to theory\\nThe researcher applies to a \\ngiven context a theory \\ndeveloped and confirmed \\nin another context\\nThe researcher highlights \\npropositions linking \\nconcepts\\nTable 2.2    \\u0007\\nDifferent Types of Generalization (Adapted from Lee and \\nBaskerville, 2003)\\n\\n\\n21\\nGrounded theories may be developed at both the substantive and \\nformal levels. A substantive grounded theory reaches beyond observed \\nincidents and analyzed data but applies to the substantive area of inquiry \\n(Urquhart, Lehmann, & Myers, 2010). A formal grounded theory is \\nabstract in terms of time, place, and people until it is applied (Glaser, \\n2007); it has been defined by Glaser (2007) as “a theory of SGT [substan­\\ntive GT] core category’s general implications, using, as widely as possible \\nother data and studies in the same substantive area and in other substan­\\ntive areas” (p. 99). A formal GT has no predetermined level of abstraction, \\nas this will be determined by the theoretical sampling and the data used. \\nIt allows generalizing on a core category from different substantive areas \\n“with more multivariate conceptual complexity” (Glaser, 2007, p. 1). \\nHence, and in order to reach formal GT, a substantive GT first has to be \\ndeveloped. Formal GT makes substantive findings “meaningful on a \\nlarger scale” (Kearney, 2007, p. 131) and extends the applicability of a \\nsubstantive grounded theory across contexts, settings, or both.\\nThe formalization of a substantive grounded theory involves start­\\ning with an existing substantive theory; its aim, however, is to “enhance \\nthe theory, widen its scope or in other ways improve it—but not to \\nverify or falsify it” (Urquhart et al., 2010, p. 4). The formalization of a \\nsubstantive grounded theory will therefore always involve the full GT \\nprocess. To sum up, theorizing is a continuum (Runkel & Runkel, 1984; \\nWeick, 1995): A substantive grounded theory can be understood as what \\nWeick (1995) terms an “interim struggle,” an early stage of formal theo­\\nretical development leading toward further development and stronger \\nFrom Substantive to Formal Theory\\nIn their famous collaboration, Glaser and Strauss first developed a general \\nsubstantive theory related to dying patients (Awareness, 1965). They then \\nextended their core category (i.e., patients’ changing status from living to \\ndying) to build a formal theory of status passage (Glaser & Strauss, 1971) \\nrelated to many different settings (e.g., career changes, mobility within \\norganizations, etc.).\\nWalsh, Kefi, and Baskerville (2010) proposed a substantive theory related \\nto the process of acculturation to information technologies and a digital \\nculture index developed from qualitative data collected in some given con­\\ntexts (corporate and societal) but bounded by qualitative data. This index \\nwas then amended and the theory formalized into a general substantive \\ntheory by extension to other substantive contexts through supplementary \\nquantitative data slices (Walsh, 2014a; Walsh & Gettler-Summa, 2010).\\nChapter 2    Discovering New Theory as the End Purpose of Classic Grounded Theory\\n\\n\\nAN OVERVIEW OF CLASSIC GROUNDED THEORY\\n22\\nformal theory at a “higher gradation of abstractness and generality” \\n\\n(p. 385). Glaser (2011) stresses that the procedures required to generate \\nformal GT are the same as those to generate substantive GT. The \\n­\\nmultiple differences between the two are in the theoretical sampling:\\nIn SGT [substantive GT], one samples within a substantive \\nchosen site or population. In doing FGT [formal GT], one \\nsamples widely in other substantive sites and populations both \\nwithin and outside the substantive area in order to make the \\ntheory more general, as one constantly compares, adding new \\nproperties and categories to the core category being general­\\nized. (Glaser, 2011, p. 257)\\nDoing GT does not imply an idiographic (i.e., local) or nomothetic \\n(i.e., universal) perspective. Both are possible, depending solely on the \\nresearcher’s philosophical stance. Idiographic, substantive theories \\nlead to different types of nomothetic, formalized theories, using any \\nkind of data (qualitative and/or quantitative) while always remaining \\nin an exploratory, data-driven stance (Figure 2.3). As highlighted by \\nGlaser (2007), a formal theory may be a so called middle-range theory \\nor “grand theory” in Merton’s (1967) sense.\\nGlaser and Strauss (1967) differentiated between substantive and \\nformal GT through their differing degrees of abstraction and generality; \\nhowever, they cautioned that they lacked hindsight and experience in \\nthis area. Glaser (1978) suggested that between a theory (e.g., the \\nchange of status of patients in one hospital) in a specific area (e.g., \\ndying patients) and a formal theory (e.g., status passage) in its full \\ngenerality (e.g., career changes, mobility within organizations), “there \\nis a type of theory in between which we call general substantive \\ntheory” (p. 52), for example, the awareness of dying related to patients \\nin any hospital. To fully understand the process of formalization of a \\nsubstantive theory, and in a critical realist stance, we would propose \\nto replace the term generality as used by Glaser and Strauss (1967) by \\nsubstantive formalization and “abstraction” by ­\\nconceptual formalization. \\nThe substantive formalization of a substantive theory may be obtained \\nthrough sampling different substantive groups, contexts, and/or \\nsocial units within the same setting/­\\nsubstantive area; it increases the \\nscope of a grounded theory. The conceptual formalization, which can \\nmostly be achieved once the substantive formalization has been \\nensured, may be obtained through sampling in different settings/\\nsubstantive areas; it increases the reach of a grounded theory. This is \\nsummarized in Figure 2.4. The two examples provided in the previous \\nboxed text are illustrated in Figure 2.5.\\n\\n\\n23\\nNomothetic\\nPerspective\\nTheory-driven\\nData-driven\\nIdiographic\\nPerspective\\nUniversal laws\\nExploration\\nConfirmation\\nSettings and\\ncontexts\\nGROUNDED\\nFormalizing substantive theories\\nTHEORIES\\nFigure 2.3    \\u0007\\nThe Land of Theories (Adapted from Walsh et al., 2015a)\\nSubstantive\\nTheory\\n–\\nReach and scope\\n+\\nGeneral Substantive Theory\\n(Middle Range)\\nData\\nConceptual formalization\\nSubstantive formalization\\nEmergence of core\\ncategory and\\nrelated categories \\nSlices of data in\\na given group/social\\nunit/context\\nFormalizing a theory\\nTheoretical sampling to formalize a theory\\nSlices of data in different settings\\nSlices of data in different\\nsubstantive groups/social units/contexts\\nwithin the same setting\\nFormal Theory\\n(Grand) \\nSubstantive formalization\\nConceptual formalization\\nFigure 2.4     \\u0007\\nFrom Substantive to Formal Grounded Theory (Translated \\nand adapted from Walsh, 2015)\\nChapter 2    Discovering New Theory as the End Purpose of Classic Grounded Theory\\n\\n\\n24\\nSubstantive Formalization\\nConceptual Formalization\\nGeneral substantive\\ntheory about the\\nawareness of dying \\nData related to\\ndying patients \\nData related to \\ndifferent\\nsettings/domains\\nConceptually formalized \\ntheory about status\\npassage\\nSubstantive theory about\\nIT culture creeping. \\nDigital culture index \\n(Walsh et al., 2010)\\nGeneral substantive theory of IT\\nculture. Amendment of the digital\\nculture index (Walsh, 2010; 2014a)\\nData (qualitative)\\nrelated to IT culture\\nData (qualitative\\nand quantitative)\\nrelated to IT culture\\nFigure 2.5    \\u0007\\nExamples of Substantive and Conceptual Formalization\\n\\n\\n25\\nBeyond the various classifications aimed at describing theories \\nproposed by different authors, we propose a further consideration that \\nis acutely needed to understand fully the contribution of classic GT to \\nthe scientific research process. When researchers theorize, they can do \\nso through what we term incremental theorizing (theorizing while using \\nexisting concepts/constructs) or rupture theorizing (theorizing while \\nusing new concepts/constructs; Walsh, 2014c). Incremental theorizing, \\nalthough essential to help a research field mature and grow, represents \\nadditional information and gradual developments on existing \\nconcepts/constructs. Rupture theorizing uses nascent concepts/\\nconstructs, which were previously unrevealed and unstudied in the \\nliterature, or which were previously applied in, and adapted to, \\ncompletely different domains.\\nRupture theorizing may involve defining and specifying these new \\nconcepts/constructs and/or investigating relationships between these \\nand other previously established and studied concepts/constructs. For \\ninstance, in a hypothetical-deductive stance, the literature is usually \\nfirst investigated for clues to lay down hypotheses in a linear research \\napproach (Creswell, 2014; Zachariadis, Scott, & Barrett, 2013). Hypo­\\nthetical-deductive research has led to data poverty (as data collected \\nare used for theory testing and not for theory building; Evermann & \\nTate, 2011; Lyytinen, 2009) and, consequently, to incremental theoriz­\\ning. In the case of rupture theorizing, if a concept/construct has never \\nbeen previously studied in a field, the literature will be of little help to \\ndescribe or explain it or to develop propositions or hypotheses involv­\\ning this new construct. As all preconceptions are to be set aside when \\ndoing GT (Glaser, 1978), adopting a classic GT exploratory stance natu­\\nrally leads to the emergence of new concepts/constructs and to theo­\\nrizing in rupture with existing literature. Thus, as the main concern \\nemerges while doing GT, it may lead the researcher to investigate \\neither new phenomena, never before investigated in the literature, or, \\nalternatively, the GT process will lead to studying phenomena already \\ninvestigated in past literature but from different and alternative \\nperspectives.\\n\\t\\nTHEORIZING IN RUPTURE WITH EXISTING LITERATURE\\t\\nl\\nChapter 2    Discovering New Theory as the End Purpose of Classic Grounded Theory\\n\\n\\nAN OVERVIEW OF CLASSIC GROUNDED THEORY\\n26\\nIN SUMMARY\\nClassic GT may accommodate any philosophical perspective. The ulti­\\nmate purpose of classic GT is to go beyond description and to discover \\nnew theories grounded in empirical data; either a new theoretical per­\\nspective on a well-established and theoretically developed phenome­\\nnon or a theory in rupture with existing literature. A grounded theory \\nmay be substantive and framed in an idiographic perspective, or it may \\nbe formal (i.e., with greater reach and scope) and framed in a more \\nnomothetic perspective. Depending on a researcher’s philosophical \\nstance, the potential resulting grounded theories may vary widely.\\nTEST YOUR KNOWLEDGE\\n\\t\\nIn the following questions one or several answers may be correct.\\n1.\\t A theory\\na.\\t Is the same thing whatever your philosophical stance\\nb.\\t May be only descriptive in classic GT terms\\nc.\\t Always involves causal elements whatever your philosophi­\\ncal stance\\nd.\\t None of the above\\n2.\\t A substantive grounded theory\\na.\\t Reaches beyond analyzed data and observed incidents\\nb.\\t Applies to the substantive area of inquiry\\nc.\\t Generalizes from empirical data to theoretical statement\\nd.\\t None of the above\\n3.\\t A formal grounded theory\\na.\\t Has no predetermined level of abstraction\\nb.\\t Allows generalizing on a core category from different sub­\\nstantive areas\\nc.\\t Develops further a substantive GT\\nd.\\t Is at a greater level of abstraction and/or generality than a \\nsubstantive GT\\ne.\\t None of the above\\n\\n\\n27\\nFURTHER READING\\nGibson, B., & Hartman, J. (2014). Rediscovering grounded theory (Chapter 2). \\nLondon: Sage.\\nGlaser, B. G. (1978). Theoretical sensitivity (Chapter 1). Mill Valley, CA: Sociology \\nPress.\\nGlaser, B. G. (2007). Doing formal grounded theory. Mill Valley, CA: Sociology \\nPress.\\nGlaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies \\nfor qualitative research (Chapters 1 & 2). New York: Aldine de Gruyter.\\nGregor, S. (2006). The nature of theory in information systems. MIS Quarterly, \\n30(3), 611–642.\\nENDNOTE\\n1.\\t Some materials in this chapter were previously published in Walsh \\n(2014c).\\nChapter 2    Discovering New Theory as the End Purpose of Classic Grounded Theory\\n\\n\\n28\\n3\\nFoundational Pillars of \\nClassic Grounded Theory\\nAfter studying this chapter, you will:\\n• understand the foundational pillars of grounded theory\\n• appreciate the necessary frame of mind for doing grounded theory\\nI\\nt is common to read in published accounts of grounded theory that \\ndoing GT is not easy, that it is a “difficult subject” (Suddaby, 2006, \\np. 633), “worryingly messy” (Partington, 2002, p. 155), “problematic” \\n(Backman & Kyngas, 1999, p. 148), and a source of “methodological con­\\nfusion” (Goulding, 1999, p. 866). Reading such accounts would be \\nenough to discourage any novice from adopting GT as a methodological \\nchoice! If one looks beyond such assessments, it is generally the case that \\nthe authors have conflated GT with some of the precepts of traditional \\nqualitative research. Such precepts are often misaligned with GT’s pre­\\ncepts, and therein frequently rest the struggles. Partington (2002) suggests \\nthis conflation has resulted in a loss of attention to key principles of GT.\\nSuddaby’s (2006) editorial in the Academy of Management Journal is \\na leading example of this qualitative assumption applied to GT. He \\nconfuses GT’s emergent paradigm with a lack of systematic procedure \\nwhen he suggests that “totally unstructured research produces totally \\nunstructured manuscripts that are unlikely to make it past the desk \\n\\n\\nChapter 3    Foundational Pillars of Classic Grounded Theory\\n29\\neditor at any credible journal of social science” (p. 633). Suddaby also \\nconfuses the way GT research is actually conducted and the way it is \\nwritten to fit within the constraints of established academic publishing \\nstandards (de Vaujany, Walsh, & Mitev, 2011). Similar to Partington \\n(2002), Suddaby advocates for a clear research question and theoretical \\nframework up front. Furthermore, he misinterprets Glaser and Strauss \\n(1967, p. 79) as calling for grounded theory to be generated from extant \\ntheory. Here, he confuses their reference to generating formal theory as \\napplying more broadly to substantive theory generation. What Glaser \\nand Strauss are actually proposing is that a formal grounded theory is \\nbest generated when it follows from a substantive grounded theory \\nusing its core category as the basis for more formal theory develop­\\nment. They are not suggesting that to develop a grounded theory, one \\nmust start from existing theory.\\nWhat Glaser and Strauss (1967) actually said about the use of extant \\ntheory:\\nA substantive theory generated from the data must first be formulated \\nin order to see which of diverse formal theories are, perhaps, applicable \\nfor ­\\nfurthering substantive formulations . . . allowing substantive concepts \\nand hypotheses to emerge first, on their own, enables the analyst to \\nascertain which, if any, existing formal theory may help him to generate \\nhis ­\\nsubstantive theories. (p. 34)\\nWe agree with Suddaby’s advice that GT is not an excuse to ignore \\nthe literature. GT is not about ignoring the literature but rather about \\nstaying open by holding the literature in abeyance until the core cate­\\ngory of the theory has emerged. Doing so avoids the potential of forc­\\ning of the study’s theoretical direction; instead, it enhances the potential \\nto discover a true main concern in the area under study and a novel \\nway to explain it through a core category. As a prominent scholar, \\nSuddaby is influential, but his conception of GT as a qualitative \\nmethod only adds to the confusion for those wishing to pursue the \\nclassic GT approach.\\nThe purpose of this chapter is to help clarify some of this confusion \\nand hopefully relieve the angst of “doing it wrong” by being very clear \\nabout the foundational pillars of classic GT and how these shape the \\nnecessary mindset for doing GT. We offer three components of GT as \\nfoundational and essential to its application: emergence, constant \\n\\n\\nAN OVERVIEW OF CLASSIC GROUNDED THEORY\\n30\\ncomparative analysis, and theoretical sampling. By emergence, we refer to \\nthe result of GT’s open and exploratory stance in relation to the research \\nfield. Constant comparative analysis refers to the process through which \\nall data are analyzed as they are collected and together with all previously \\ncollected data. Theoretical sampling is the process through which empirical \\ndata are selected and collected as guided by the emerging theory.\\nl\\nl\\nEMERGENCE\\nThe GT approach is a general methodology for the discovery of theory. \\nThe researcher enters the field and explores a substantive area by \\nallowing the chief concerns of those actively engaged therein to guide \\nthe emergence of a core issue or problem. The conceptualization of that \\nissue (or problem) becomes the basis for the articulation of a grounded \\ntheory that explains how the issue is processed, managed, or resolved. \\nAs such, a defining characteristic of GT is its rejection of preconceived \\ntheorizing. Contrary to the traditional quantitative approach, which \\nimplies the articulation of hypotheses in advance, or the traditional \\nqualitative approach, which implies a theoretical framework to be \\napplied, GT does not require as a starting point the identification of a \\ngap in the literature from which precise research questions are framed \\nto guide the study. A GT study begins with the researcher being open \\nto discovering latent patterns that can be theorized to explain social \\nbehavior.\\nThose who have been trained in other research methodologies \\noften find it a challenge to set aside learned habits, but doing so is \\nfundamental to undertaking a grounded theory study where \\nthe focus of the study and its theorization cannot be decided at \\nthe outset of the study by preconceived notions resident within the \\nresearcher’s worldview, an initial professional problem, or an extant \\ntheory and framework. Such preconception is simply forced \\nconjecture—what Glaser and Strauss (1965b) refer to as “logical \\nspeculation” (p. 11). Preconceiving a study based on extant theory \\nblocks the emergent potential with “categories that do not seem to \\nemerge in the data but, rather, are imposed on it as existing theory \\nwhich ought to be applicable. Their relevance is implicitly assumed” \\n(Glaser, 1966, p. 332).\\nOur philosophical stance does preconceive our research approach \\nto some degree as it can limit what we consider to be appropriate data \\nand how we approach the analysis of those data. As an example, \\nconstructivism privileges interview and focus group data and would \\n\\n\\nChapter 3    Foundational Pillars of Classic Grounded Theory\\n31\\ncommonly reject the use of observational or documentary data as not \\nappropriate for conveying the perspectives of the participants. GT’s \\nopenness to all data creates a form of triangulation in data sources, \\nthereby reducing the potential for a perspective bias (whether that of the \\nresearcher or the participants) being introduced through restricted use \\nof data types.\\nIn a GT study, if a researcher chooses to use quantitative data and/\\nor techniques traditionally considered by some as confirmatory, it is \\nnot the technique in itself that is important but how it is used and \\napplied with an exploratory purpose. As early as 1980, Tukey \\nhighlighted that “exploratory (quantitative) data analysis is an attitude \\nand a flexibility” (p. 24). For instance, factor analysis that helps to \\ngroup emerging variables, and/or cluster analysis that helps to classify \\nobjects/people/concepts based on their similarities and differences, \\ncould fruitfully be used in an exploratory fashion. Structural equation \\nmodeling (SEM) in the soft exploratory version of the partial least \\nsquares (PLS) approach can also help theorizing. Researchers may use \\nmany other techniques, if they always take care not to impose \\nhypotheses on the data. When one uses quantitative data, it may be \\nuseful to remember that the numbers themselves are not what matters \\nin social sciences: It is the meaning behind the numbers that is \\nimportant. Hence, quantitative techniques may be used; they should, \\nhowever, never be used in a mechanical fashion without fully \\nunderstanding the meaning behind the techniques (Glaser, 2008). It \\nmight be useful for the reader to refer to Appendix C as an illustration \\nof some quantitative techniques used in an exploratory GT stance.\\nWhile doing GT, concepts come from the data; they are not \\npreconceived or imposed on data. This is particularly important when \\ndata, methods, and techniques are mixed. The quantitative and \\nqualitative components are not used for verification or testing but for \\nfurther elaboration of the theory. There should be no forcing, where it is \\nassumed that particular constructs will occur in the data. Thus, and \\nwhatever the data that are being used, exploration and emergence of \\nthe theory from the data have to remain central.\\nOur preconceptions about the social worlds we wish to study can \\ncreep in unawares as this statement in one methodology guide \\nillustrates: “In management research, you will often find that the more \\nsenior the manager the more useful the interview data” (Partington, \\n2002, p. 144). Concepts in vogue (e.g., context) are often assumed and \\nforced upon a study, but in GT they are simply potential variables that \\nmay or may not emerge as relevant through constant comparative \\nanalysis. How such variables may influence a latent pattern within \\n\\n\\nAN OVERVIEW OF CLASSIC GROUNDED THEORY\\n32\\nyour data must emerge as relevant to the substantive area that is \\ninvestigated and not be assumed and forced into the analysis (Glaser, \\n1992, 2013). Studies in disciplines that are highly theorized (e.g., \\nmanagement, economics, psychology) are particularly susceptible to \\npreconception; this is what Alvesson and Sandberg (2014) refer to as \\n“‘boxed-in research’. . . encouraging specialization, incremental adding-\\nto-the-literature contributions and a blinkered mindset” (p. 967). \\nUndertaking a grounded theory study in such disciplines necessitates \\nthat researchers appreciate from the outset the undermining effect of \\npreconceiving the research focus through the lens of their professional \\nresearch interests, extensive preliminary literature reviews, precise \\nresearch questions, defined sample populations, structured interview \\nprotocols, and other factors. Emergence as a foundational pillar \\nnecessitates that researchers remain open to discovering a main \\nconcern motivating the interest or action within the area under study \\nand offers a conceptual theory to explain the behavior.\\nGlaser (2013) has identified several types of preconception that \\nfrequently derail GT intentions. Among those commonly adopted are \\nthe following:\\n• Conducting an extensive preliminary literature review to \\n­\\nidentify a gap in the research around which to focus a proposed \\nstudy\\n• Assuming the relevance of a “pet theoretical code” (Glaser, \\n2005a) such as social network, basic social process, conditional \\nmatrix, and so on, as a lens through which the data are \\nanalyzed\\n• Specifying research questions around which all data collection is \\nfocused\\n• Assuming the relevance of face sheet data (e.g., gender, age, \\nincome, etc.) to the analysis and theory development\\nMisconceptions such as all grounded theories are basic social \\nprocesses (Langley, 1999) or actions (Creswell, 2013) lead to forcing a \\ntheory to fit the misconception. While Glaser (1978) highlighted basic \\nsocial processes as an important theoretical code, he did so to empha­\\nsize the patterned nature of grounded theory in contrast to unit \\nanalysis, which invariably leads to descriptive findings rather than \\nconceptual theory. His inclusion in the same text of several alternative \\ntheoretical coding families should clarify the misconception of \\ngrounded theory as always being a basic social process. Forcing the use \\n\\n\\nChapter 3    Foundational Pillars of Classic Grounded Theory\\n33\\nof a specific theoretical code, such as using a basic social process to \\norganize and model a theory, often results in rather flat and thin theory \\nand prevents the theorist from seeing a more interesting, novel, \\nemergent theoretical code around which to integrate the theory. \\nSimilarly, preconceiving the study from the outset with a “skeletal \\nframework . . . focusing the inquiry at an early stage . . . provid[ing] \\ninternal structure to the study” (Morse & Mitcham, 2008, p. 32) is \\nanother preempting of GT’s emergent potential. Such strategies can be \\nparticularly problematic for an inexperienced theorist who lacks the \\nconfidence to stay open and reverts to the assurance of a structure in \\nadvance, especially under supervisor pressures to adopt the precepts \\nof qualitative research.\\nSimilarly, extensive literature reviews in advance of a study, often \\nwith the goal of identifying a gap in the literature, also present the \\npotential for preconception—what Morse and Mitcham (2008) refer to \\nas “the pink elephant syndrome.” At best, identifying such a gap is a \\nstab in the dark because the literature reviewed may not correlate to \\nthe emergent main concern and may even inhibit the researcher’s \\nability to see the main concern. Such practices, however, are often \\nencouraged by PhD supervisors unfamiliar with classic GT and can \\neven influence standard requirements of the PhD thesis at many \\nuniversities and of peer reviewer expectations for mainstream journal \\npublication. As such, the undermining potential for GT studies is \\npersistent and significant but can be overcome with a solid \\nunderstanding and appreciation for emergence as foundational to GT \\ndevelopment.\\nIf, in accordance with classic GT precepts, the literature review is \\nconducted after the core category has emerged and around this core \\ncategory, the literature may be used as data. It may help in theoretical \\nsampling for full saturation of emergent concepts as well as in \\ntheoretical coding for the emergence of relationships between concepts \\nand overall integration of the theory.\\n\\t\\nCONSTANT COMPARATIVE ANALYSIS\\t\\nl\\nThe goal of GT is the generation of theory that conceptually explains \\nlatent patterns of social behavior:\\nConceptualization is the medium of GT for a simple reason: \\nwithout the abstraction from time, place and people, there \\ncan be no multivariate, integrated theory based on conceptual, \\n\\n\\nAN OVERVIEW OF CLASSIC GROUNDED THEORY\\n34\\nhypothetical relationships  .  .  .  Because GT operates on a \\nconceptual level, relating concept to concept, it can tap the \\nlatent structure which is always there and drives and organizes \\nbehavior and its social psychological aspects, all of which are \\nabstract of objective fact. (Glaser, 2002, p. 241)\\nThus, the conceptual abstraction of empirical data using constant \\ncomparative analysis is fundamental to doing GT. Constant comparative \\nanalysis is a strategy for directing the collection and analysis of data in \\ntandem with theoretical sampling as a means of guiding the direction \\nof further data sampling (Figure 3.1).\\nThe analytical processes that underlie the constant comparative \\nmethod have their roots in latent structure analysis as developed at the \\nUniversity of Columbia’s School of Applied Arts under the leadership \\nof Paul Lazarsfeld (Lazarsfeld & Henry, 1968). As a student at \\nColumbia, Glaser saw the practical value in theory generation using \\nempirical data that had been collected as preparatory to large-scale \\nsurvey work. His own doctoral work employed secondary analysis of \\nsuch data. As a student of Lazarsfeld, Glaser was exposed to Lazarsfeld’s \\nindex formation process but saw the stripped-down, summing-up \\nquality as a loss (Glaser, 1998, p. 23). One had an index but without \\nmeaning. Glaser’s idea was rather than simply summing up indicators, \\nhe would instead compare indicator to indicator thereby generating \\nconceptual properties and dimensions; this would be an index with \\nTheoretical\\nSampling:\\nData Slice\\nTheoretical\\nSampling:\\nData Slice\\nTheoretical\\nSampling:\\nData Slice\\nDiscovery of\\nPatterns:\\nEmerging\\nTheory \\nFigure 3.1    \\u0007\\nThe Constant Comparative Analysis Process (Walsh, \\n2014c)2\\n\\n\\nChapter 3    Foundational Pillars of Classic Grounded Theory\\n35\\nmeaning (Glaser, 1998, p. 24). Another aspect of constant comparative \\nmethod is explication de texte,3 which Glaser studied at the Sorbonne. \\nExplication de texte refers to “reading closely line by line to ascertain \\nwhat exactly the author is saying without imputing what is said, \\ninterpreting it or reifying its meaning” (Glaser, 1998, p. 24). Glaser \\nexplains how explication de texte stirred his interest in latent patterns of \\nbehavior in social settings. Applying explication de texte to constant \\ncomparative analysis helps to generate concepts that closely fit what is \\ngoing on in the substantive area without imputing meaning to the \\nindividual action; one simply names the concept as an abstraction.\\nThe concept-indicator model requires that concepts and their \\nproperties and dimensions earn their way into a theory by systematic \\ngeneration of data. GT’s concept-indicator model begins with substan­\\ntive coding in which empirical incidents are compared, incident \\n­\\nto  incident, to generate concepts. When many indicators become \\ninterchangeable, they lead to a pattern that is named (concept). \\nAdditional empirical incidents are compared to the emergent concepts, \\nto build and substantiate these with property and dimensional definition. \\nWhen further data yield no new concepts or additional elaboration of \\ntheir properties and dimensions, theoretical saturation has been \\nachieved. Then the concepts are related to each other to generate \\nhypotheses. Constant comparison continues through theoretical coding \\nwhereby concepts are compared to other concepts to establish conceptual \\nlevels and their theoretical integration. Thus, the constant comparative \\nanalysis of data threads throughout the entire GT process, taking full \\nadvantage of the interchangeability of indicators to develop a broad \\nrange of acceptable indicators of concepts and their properties (Glaser & \\nStrauss, 1971, p. 184), from the initial data collected through to the full \\nintegration of the generated theory that is “integrated, consistent, \\nplausible and close to the data” (Glaser, 1965, pp. 437–438).\\nGlaser’s (1965) seminal paper on the constant comparative method \\noutlines four stages to the process—comparing incidents applicable to \\neach category, integrating categories and their properties, delimiting \\nthe theory, and writing the theory—while reminding the reader that \\nthe process is not lockstep and linear but rather that the stages, once \\ncommenced, continue in tandem as the theorist cycles back and forth \\nthrough the various stages. In addition, he specifies two defining rules \\nof constant comparison: (1) “while coding an incident for a category, \\ncompare it with the previous incidents in the same and different \\ngroups coded in the same category” (p. 439), and (2) “stop coding and \\nrecord a memo on your ideas” (p. 440). This latter rule establishes the \\ncentral role of memoing in constant comparative analysis.\\n\\n\\nAN OVERVIEW OF CLASSIC GROUNDED THEORY\\n36\\nMemos are theoretical notes about the data and the conceptual \\nconnections between categories. By memoing throughout the constant \\ncomparative process, the researcher captures her emergent ideation of \\nsubstantive and theoretical categories:\\nBits of data and early codes are systematically examined, \\nexplored and elaborated on. Through memo writing, the \\ntheorist takes his emerging ideas apart, checks them and \\noutlines further data collection  .  .  .  looks at the data from a \\nvariety of perspectives and analyzes them  .  .  .  The theorist \\ngains further insights and creates more ideas about the data \\nwhile writing. Hence, writing and rewriting are crucial phases \\nof the analytic process. (Charmaz, 1990, p. 1169)\\nLater in the process, having achieved theoretical saturation of the \\ncore, its properties, and its related concepts, the researcher proceeds to \\nreview, sort and integrate the numerous memos captured along the \\nway. The sorted memos generate a theoretical outline (a conceptual \\nframework) for the full articulation of her grounded theory through an \\nintegrated set of hypotheses, thus achieving the final stage of the \\nconstant comparative process.\\nConstant comparative analysis shifts the theorist’s attention from \\nverification of facts to ideas generation, thereby enabling maximum \\ncreativity in following emerging theoretical leads, unconstrained by \\ninterview protocols or a preconceived theoretical framework. “Since no \\nproof is involved, the constant comparative method requires only \\nsaturation of data” (Glaser & Strauss, 1967, p. 104). Comparing \\nadditional incidents in the data to emerging concepts has four \\npurposes:\\n(1) It verifies the concept as a category denoting a pattern in \\nthe data. (2) It verifies the fit of the category nomenclature to \\nthe pattern. (3) It generates properties of the category. (4) It \\nsaturates \\nthe \\ncategory \\nand \\nits \\nproperties \\nby \\nthe \\ninterchangeability \\nof indicators. (Glaser, 1998, p. 139)\\nAs such, GT’s verification is conceptual and propositional, not \\nfactual. Categories suggested are always tentative, and although they \\ncannot be disproved, they can be modified by further data collection. \\nData collection continues until concepts are theoretically saturated, \\nmeaning that nothing new is being added to the elaboration of the \\nconcept or property. If conceptual ideas suggested in initial data do not \\n\\n\\nChapter 3    Foundational Pillars of Classic Grounded Theory\\n37\\npattern out with further data collection, then the theorist must decide \\nwhether the initial ideas have earned relevance in the emerging theory \\nand decide what prominence, if any, they hold as the theory continues \\nto develop. In so doing, the constant comparison method ensures the \\nearned relevance of emergent conceptual ideas, helping to overcome \\nwhat Morse and Mitcham (2008) refer to as “conceptual tunnel vision” \\nin assigning more data to a category than actually belong.\\nMemoing in tandem with coding using Glaser’s (1978, p. 57) \\nquestions4 will control this tendency as the theorist must explain how \\nthe data indicate a concept or a property of a concept and how they \\nrelate to other concepts, and so on. Earned relevance and theoretical \\nsaturation of emergent concepts also help to control for “pink elephant \\nparadox” (Morse & Mitcham, 2008) of thin data sets in which \\ninsufficient incidents fail to offer up enough conceptual indicators. The \\nconstant comparative method requires the theorist to keep moving \\nthrough the data to see conceptual indicators over and over, constantly \\ncomparing and memoing her ideas and deciding what further data are \\nneeded to elaborate and saturate emerging conceptual patterns.\\nDifferent slices of data, including quantitative and qualitative \\nslices, feed information into each other. If one uses mixed data, one \\nshould not analyze or interpret them in isolation but rather as data \\n\\nare collected and together with all previously collected data. As an \\nillustration the reader may refer to Sections 1.2 and 1.3 in Appendix C \\nand to Appendix D to see how all data (primary and secondary, \\nqualitative and quantitative) were constantly compared in order to \\ntheorize. As one theoretically samples quantitative data to supplement \\nqualitative data (or vice versa), one should not wait until all quantitative \\ndata are collected but keep analyzing the data set as new data are \\ncollected in order to systematically check for possible emerging new \\npatterns and/or categories.\\n\\t\\nTHEORETICAL SAMPLING\\t\\nl\\nAs noted earlier, the explicit coding and analytic procedures of constant \\ncomparative analysis work in tandem with the third foundational \\npillar of GT—theoretical sampling. In GT, theoretical sampling refers to \\nthe collection of data as guided by the emerging theory as opposed to \\nsome predefined sample or unit. In theoretical sampling, the theorist \\n“jointly collects, codes and analyzes data and decides what data to \\ncollect next and where to find them in order to develop his theory as it \\nemerges” (Glaser & Strauss, 1967, p. 45). Glaser (1978) refers to the \\n\\n\\nAN OVERVIEW OF CLASSIC GROUNDED THEORY\\n38\\nprocess as deductive (conceptual) elaboration as opposed to logical \\nelaboration found in hypothesis testing research. He defines conceptual \\nelaboration as\\nthe systematic deduction from the emerging theory of the \\ntheoretical possibilities and probabilities for elaborating the theory \\nas to explanations and interpretations. These become hypotheses \\nwhich guide the researcher back to locations and comparative \\ngroups in the field to discover more ideas and connections from \\ndata. The data constantly check deductions that lead nowhere, as \\nthe analyst takes his directions from the emerging relevancies. \\nDeduced hypotheses are not forced on the data when they fit \\npoorly, they are discarded and others emerge in their place by \\nconstant comparative analysis. (p. 40)\\nGlaser and Strauss (1971) on selecting comparison groups:\\nThe basic criterion governing the selection of comparison groups for \\ngenerating theory is their theoretical relevance for furthering the devel­\\nopment of emerging categories, properties, hypotheses, and integration \\nof the theory. Any groups may be selected that will help generate these \\nelements of the theory. (p. 183)\\nTheoretical sampling sharpens the definition of concepts, helps to \\ndefine their properties, and tests tentative theoretical propositions \\n(Draucker, Martsolf, Ross, & Rusk, 2007). Sampling is driven by concep­\\ntual emergence and relevance and limited by theoretical saturation, not \\nby a preconceived theoretical framework or by statistical significance. \\nConsequently, selection of data sources is neither random nor totally \\npredetermined. Additional data are selected in service to the emerging \\ntheory as the theorist gathers additional data on emerging patterns by \\nasking specific questions in subsequent data collection or by returning \\nto data already collected and reviewing it, selectively sampling there \\n\\nfor the emerging pattern. As Glaser (1978, pp. 37–38) explains, the \\ndeductive nature of theoretical sampling complements GT’s overall \\ninductive approach.\\nTheoretical sampling as a means of guiding the direction of the \\nemerging theory has little efficacy if it is separated from the processes \\nof coding and analysis. Memoing is part of the decision process around \\ntheoretical sampling, as it assists theorists to identify gaps in their \\n\\n\\nChapter 3    Foundational Pillars of Classic Grounded Theory\\n39\\nanalysis and points to possible new directions for further data \\ncollection, coding, and analysis. If the data are collected by theoretical \\nsampling at the same time that they are analyzed, the integration of the \\ntheory is more likely to emerge by itself as this taps to the fullest extent \\nthe in vivo patterns of integration in the data itself. Questions guide the \\ncollection of data to fill in gaps and to extend the theory as a further \\nintegrative strategy. Theoretical sampling involves an iterative process \\n(Figure 3.2) and requires time to learn.\\nRemaining analytical and flexible while also\\nmaking theoretically sensitive judgments about saturation is \\nnever precise. The researcher’s judgment becomes confidently \\nclear only toward the close of his joint collection and analysis, \\nwhen considerable saturation of categories in many groups to \\nthe limits of his data has occurred, so that his theory is \\napproaching stable integration and dense development of \\nproperties. (Glaser & Strauss, 1967, p. 64)\\nTheoretical sampling also reduces the amount of data needed \\n(Glaser & Strauss, 1967, pp. 109–111). At the same time, inadequate \\nsampling will be evident in theory that is thin and not well integrated \\n(p. 63). As Glaser and Strauss (1967) state,\\nThe basic question in theoretical sampling . . . is: what groups \\nor subgroups does one turn to next in data collection? And for \\nwhat theoretical purpose?  .  .  .  The possibilities of multiple \\ncomparisons are infinite, and so groups must be chosen \\naccording to theoretical criteria. (p. 47)\\nData collection\\nand analysis\\nTheoretical\\nsampling\\nFigure 3.2    An Iterative Process \\n\\n\\nAN OVERVIEW OF CLASSIC GROUNDED THEORY\\n40\\nComparisons provide diverse and similar indicators of conceptual \\ncategories and their properties. Discovering similarities and differences \\nare important because they can indicate conditions under which \\ncategories and their properties vary and to what degree they may \\nvary. Theoretical sampling encourages the theorist to employ multiple \\n“slices of data” (Glaser & Strauss, 1967, p. 65) to explore a range of \\nperspectives. Glaser (1978) offers a number of strategies for focusing \\ndata collection in theoretical sampling, including changing the \\nlocation, participants, or interviewing style to elicit different slices of \\ndata; following up on emerging patterns in the data; and selectively \\nsampling for additional data on specific concepts that appear to be \\nsignificant to the emerging theory and, where possible, going back to \\nkey participants to ask for more information on these particular \\nconcepts. As Glaser and Strauss (1967) explain, the variety in data \\nslices should be based on theoretical understanding of the conceptual \\ncategory under diverse conditions and not simply on triangulated \\ndata types (p. 66). Sampling is determined only by the necessity of \\ntheoretical relevance and coverage. If the need to mix different types \\nof data (quantitative and qualitative) emerges during the research \\nprocess, the purpose is not to test or correct what has been found \\npreviously but to extend understanding of the phenomena under \\nscrutiny and of the scope of the emerging theory, as well as to densify \\nthe concepts toward theoretical saturation: “Different kinds of data \\ngive the analyst different views or vantage points from which to \\nunderstand a category and to develop its properties” (Glaser & \\nStrauss, 1967, p. 65).\\nWhen theoretically sampling quantitative data, statistical validity \\nmay be set aside because data collection should be guided only by the \\nemerging theory. All cases, including deviant minority cases, can be \\ntaken into account, as they are all potentially important and might \\nrequire further qualitative analysis in order to discover unobserved \\nvariables and causal statements (Glaser, 2008). When software are used \\nfor some of the analyses, researchers can reflectively ensure that the \\nmathematical algorithms in the software do not eliminate these \\nstatistically insignificant cases or groups of cases without investigating \\nthem, with the help of qualitative data previously collected or \\nspecifically collected for this purpose. Mixing quantitative and \\nqualitative data whenever needed, and analyzing and interpreting \\nthem in an embedded manner, will allow researchers to identify \\nwhether or not so-called statistically insignificant results should in fact \\nbe taken into account. The reader may refer to Appendix C, Section 1.3 \\nfor an illustration of theoretically sampling mixed data.\\n\\n\\nChapter 3    Foundational Pillars of Classic Grounded Theory\\n41\\n\\t\\nDEVELOPING YOUR STANCE AS A GROUNDED THEORIST\\t\\nl\\nAs mentioned at the outset of this chapter, many published accounts \\nsuggest that GT is a difficult methodology to master. Indeed, the \\nexpectation of joint collection, coding, and analysis of data with all \\nthree operations being done together, blurring and intertwining \\ncontinually, from the beginning of the investigation to its end (Glaser & \\nStrauss, 1967, p. 43) may seem a daunting quest to undertake, \\nparticularly for those trained in highly structured, linear, lockstep \\nmethodological approaches. What some may view as exciting and \\nfreeing can be seen by others as a terrifying jump into the abyss of \\nendless and overwhelming data. Glaser (1992) cites the requisite skills \\nfor doing GT as the ability “to absorb the data as data, to be able to step \\nback or distance oneself from it, and then to abstractly conceptualize \\nthe data” (p. 10). Glaser (1998) describes skill development in GT as a \\ndelayed action learning process (p. 220). He cites the importance of \\nthree attributes in the GT researcher: “an ability to conceptualize data, \\nan ability to tolerate some confusion, and an ability to tolerate \\nconfusion’s attendant regression” (Glaser, 2010, p. 4).\\nGlaser also repeatedly insists that GT is often best achieved by novice \\nresearchers, who are not as yet invested in a favored discipline perspective \\nor theoretical code (e.g., power dynamics, social networks, etc.) but who \\nare motivated and open to learning (Glaser, 2003). Over and over, in his \\nwriting, Glaser emphasizes the importance of staying open and not forcing \\npreconceived ideas or frameworks upon data but rather fostering the \\nemergence of conceptual theorizing from the empirical data. This dictum \\nruns throughout all of his writings and must be seen as fundamental to \\ndeveloping the necessary frame of mind for doing GT.\\nGlaser and Strauss (1967) cite, as well, the crucial role of theoretical \\nsensitivity in one’s ability to develop good grounded theories. Much has \\nbeen written about theoretical sensitivity; indeed, Glaser’s second book \\nis titled Theoretical Sensitivity, which may suggest its centrality to doing \\nGT from his perspective. Glaser and Strauss (1967) describe theoretical \\nsensitivity as the capacity to conceptualize and formulate a theory as it \\nemerges from data; a continuous skill development process that involves \\nboth a personal and temperamental bent as well as the ability to “have \\ntheoretical insight combined with ability to make something of these \\ninsights” (p. 46). Suddaby (2006) also recognizes the subtle but significant \\ninterplay between technique and creativity in doing GT:\\nSuccessful grounded theory research has a clear creative \\ncomponent. Glaser and Strauss were aware of this component \\n\\n\\nAN OVERVIEW OF CLASSIC GROUNDED THEORY\\n42\\nand the tension it would create with those who find comfort in \\ntrusting an algorithm to produce results. Glaser (1978) used the \\nterm “theoretical sensitivity” to describe the essential tension \\nbetween the mechanical application of technique and the \\nimportance of interpretive insight. (p. 636)\\nKlag and Langley (2013) describe theoretical sensitivity as the \\ncapacity to generate conceptual leaps and suggest that “life events, \\nbackground reading and unique training and work trajectories—may \\noffer resources for insight” (p. 157). Glaser, however, warns not to \\nallow personal perspectives and assumptions to taint one’s openness to \\nwhat may emerge in a GT study. Fundamental to one’s theoretical \\nsensitivity, he suggests, is the ability to “enter the research setting with \\nas few preconceived ideas as possible—especially logically deducted, a \\npriori hypotheses” (Glaser, 1978, pp. 2–3). Hernandez (2010) agrees \\nwith Glaser, suggesting that researcher preconceptions in the form of \\n“personal beliefs or biases regarding a particular substantive area or \\nideas gleaned from published articles within this substantive area” \\n\\n(p. 154) are major blocks to developing theoretical sensitivity.\\nGlaser (1978) describes theoretical sensitivity as the ability to take \\na transcending view of data; to become a “non-citizen” (p. 8) by not \\nprivileging any particular perspective or view but rather letting the \\ndata speak for themselves. This is particularly the case with extant \\ntheory where Glaser suggests one need simply to footnote the work of \\nothers and integrate their ideas into one’s emerging theory where \\nrelevant through “emergent fit” (pp. 8–9). Here, he is not suggesting \\nthat the theorist reject extant theory or that one must acknowledge and \\naddress all possible theoretical perspectives in one’s analysis but rather \\nthat the theorist apply sensitivity and judgment in how to balance \\nscholarship of extant theory and analysis of emergent conceptual ideas \\nin service to theory with parsimony and scope (pp. 10–12).\\nFrom such statements we can see that theoretical sensitivity is not \\nabout bringing one’s knowledge of extant theory into framing a GT \\nstudy through theoretical frameworks and extensive literature reviews \\nat the outset of a study. Glaser (1992) elaborates further on theoretical \\nsensitivity by stating that it “refers to the researcher’s knowledge, \\nunderstanding and skill, which foster his generation of categories and \\nproperties and increases his ability to relate them into hypotheses, and \\nto further integrate the hypotheses, according to emergent theoretical \\ncodes” (p. 27). Kelle (2005), however, complains that Glaser and Strauss \\nhave not provided sufficiently clear methodological rules for developing \\n\\n\\nChapter 3    Foundational Pillars of Classic Grounded Theory\\n43\\none’s theoretical sensitivity suggesting that “it remains unclear how a \\ntheoretically sensitive researcher can use previous theoretical \\nknowledge to avoid drowning in the data” (para. 9). McCallin (2006) \\nsuggests that the grounded theorist must embrace a particular style of \\nthinking that is creative, inquisitive, critical, analytical, and comfortable \\nwith complexity—a style that is not necessarily inherent in all \\nresearchers and one that can be challenging to the novice.\\nDeveloping grounded theory—and one’s stance as a grounded \\ntheorist—takes time and requires that the theorist is comfortable in \\ntolerating confusion (Glaser, 1998, pp. 100–102), developing a rhythm \\nof personal pacing as one cycles through the various analytic stages \\n(Glaser, 1978, pp. 18–33), and assuming autonomy and ownership of \\none’s emerging theory and its contribution to knowledge (Glaser, 1998, \\npp. 98–100). Perhaps most important to developing the stance of a true \\ngrounded theorist, Glaser suggests, is the ability to trust in GT’s \\nemergent power (1998, pp. 235–254).\\nIN SUMMARY\\nGT’s three foundational pillars are exploration and emergence (i.e., con­\\ncepts and relationships come from the data and are not preconceived or \\nimposed on data), constant comparative analysis (i.e., data are continu­\\nously compared with previously collected and analyzed data, looking \\nfor similarities and differences to help toward conceptualization and \\ntheorization), and theoretical sampling (i.e., sampling is directed by the \\nemerging theory and continues until theoretical saturation). Doing GT \\nrequires researchers to acquire and build up their theoretical sensitivity, \\nwhich is the capacity to conceptualize and formulate a theory as it \\nemerges from data and without any preconception.\\nTEST YOUR KNOWLEDGE\\nIn the following questions one or several answers may be correct.\\n1.\\t When you do GT\\na.\\t You have to do a literature review before you start collecting \\ndata\\nb.\\t You analyze your data when you have finished collecting \\nthem\\n\\n\\nAN OVERVIEW OF CLASSIC GROUNDED THEORY\\n44\\nc.\\t If you use quantitative data, your sampling has to be \\n­\\nstatistically significant\\nd.\\t None of the above\\n2.\\t Theoretical sensitivity is\\na.\\t The capacity to adapt an existing framework\\nb.\\t The capacity to apply a theory\\nc.\\t The capacity to conceptualize a theory\\nd.\\t The capacity to formulate a theory\\ne.\\t None of the above\\n3.\\t GT’s foundational pillars are\\na.\\t Emergence, constant comparative analysis, and sampling\\nb.\\t Emergence, content analysis, and theoretical sampling\\nc.\\t Verification, explication de texte, and theoretical sampling\\nd.\\t Emergence, constant comparative analysis, and theoretical \\nsampling\\ne.\\t None of the above\\n4.\\t Memos\\na.\\t Are to be written at the end of the process of doing GT\\nb.\\t Should be written during the GT process\\nc.\\t Are important when you do GT\\nd.\\t Are not important in the GT process\\ne.\\t None of the above\\nFURTHER READING\\nBreckenridge, J. (2009). Demystifying theoretical sampling in grounded theory \\nresearch. Grounded Theory Review, 8(2), 113–126.\\nGlaser, B. G. (1978). Theoretical sampling. In Theoretical sensitivity: Advances in \\nthe methodology of grounded theory (pp. 36–54). Mill Valley, CA: Sociology \\nPress.\\nGlaser, B. G. (2013). No preconceptions: The grounded theory dictum. Mill Valley, \\nCA: Sociology Press.\\nGlaser, B. G., & Strauss, A. L. (1967). Theoretical sampling. In The discovery of \\ngrounded theory: Strategies for qualitative research (pp. 45–77). Hawthorne, \\nNY: Aldine de Gruyter.\\nGynnild, A. (2006). Growing open: The transition from QDA to grounded \\n­\\ntheory. Grounded Theory Review, 6(1), 61–78.\\nRoderick, C. (2009). Learning classic grounded theory: An account of the jour­\\nney and advice for new researchers. Grounded Theory Review, 8(2), 49–63.\\n\\n\\nChapter 3    Foundational Pillars of Classic Grounded Theory\\n45\\nENDNOTES\\n1.\\t The \\nCreative \\nCommons-Attribution-NonCommercial-ShareAlike \\nLicense 4.0 International applies to all works published by the \\nInternational Journal of Qualitative Methods. Copyright for articles \\npublished in the International Journal of Qualitative Methods remains \\nwith the first author.\\n2.\\t The double arrows in Figure 3.1 illustrate the constant comparative \\nanalysis and “inextricable link between data collection and analysis” \\n(Birks, Fernandez, Levina, & Nasirin, 2013, p. 3) leading to the discovery \\nof patterns, which in turn guide further theoretical sampling until \\n­\\ntheoretical saturation is reached.\\n3.\\t Literally translated, this French expression means “explanation of \\n\\na text.”\\n4.\\t What are these data a study of? What category or property of category \\ndoes this incident indicate? What is going on? What accounts for the \\nmain issue, and how is this issue processed or resolved?\\n\\n\\n46\\n4\\nClarifying Common Sources \\nof Confusion in Grounded \\nTheory\\nAfter studying this chapter, you will:\\n• identify, understand, and avoid common sources of preconception that \\nlimit emergent discovery in grounded theory\\n• differentiate between describing research findings and conceptualizing \\ndata\\n• differentiate between substantive and theoretical coding\\nM\\nost people carry assumptions about how to do research. These \\nassumptions result from their university background and PhD \\ntraining and sometimes prove to be a real challenge to recon­\\ncile with those of classic GT. Indeed, some might prove irreconcilable. \\nAmong the common assumptions are the need for a thorough literature \\nreview at the outset of a study to identify a gap that justifies pursuing \\nthe study, to establish a theoretical framework for data collection and \\nanalysis, and to formulate hypotheses to be tested or precise research \\nquestions developed ex ante to be answered in the study. Such assump­\\ntions are a major source of confusion and frustration for those wishing \\n\\n\\nChapter 4    Clarifying Common Sources of Confusion in Grounded Theory\\n47\\nto follow a classic GT approach. In this chapter, we address three com­\\nmon sources of methodological confusion that will undermine your GT \\nefforts: emergence versus preconception, description versus conceptu­\\nalization, and substantive versus theoretical coding.\\n\\t\\nEMERGENCE VERSUS PRECONCEPTION\\t\\nl\\nResearch studies are often initiated to explore or confirm a specific \\nresearch question or to test formulated hypotheses that are of interest to \\nthe researcher. GT’s discovery stance, however, rejects the formulation \\nof preconceived notions regarding the focus of a study in favor of \\nexploring openly (i.e., without preconception) what is actually going on \\nin the area under study rather than what we might decide at the outset \\nshould be the focus for inquiry. “GT listens to participants’ own . . . con­\\ncerns and then subsequently crafts questions from emergent concepts \\nfor subsequent theoretical sampling” (Glaser, 2003, ­\\np. 118). Remaining \\nopen to discovering what is really going on in the field of inquiry is \\noften blocked, however, by what Glaser (1998) refers to as the forcing of \\npreconceived notions resident within the researcher’s worldview (i.e., \\nontological and epistemological perspective), an ­\\ninitial professional \\nproblem to be studied, or an extant theory and framework around \\nwhich to focus the inquiry—all of which preempt the researcher’s \\n­\\nability to suspend preconception. An experienced grounded theorist, \\nÓlavur Christiansen describes professional research problems as\\nproblems “thought-out” and “thought-up” by the researcher, \\nproblems logically derived by the researcher and his/her \\nseniors, problems found methodologically convenient and \\nmethodologically appropriate for the researcher, problems \\nmade apparently relevant by the availability of certain data \\n(frequently quantitative data), problems made apparently \\nrelevant by fitting into some received concepts from preexistent \\ntheory, or problems made apparently relevant by fitting into \\nsome preconceived or spuriously drawn borders between sub-\\ndisciplines. Such “professional problems” are typically not \\ngrounded in the underlying and essential agenda of those being \\nstudied. (Personal communication, July 19, 2004)\\nIn GT the intention is to remain open to what emerges as a main \\nissue in the substantive area of inquiry. To do so requires that the \\nresearcher be theoretically sensitive to discovering this main concern as \\n\\n\\nAN OVERVIEW OF CLASSIC GROUNDED THEORY\\n48\\nopposed to pursuing a specific, predetermined research interest. “The \\nfirst step in gaining theoretical sensitivity is to enter the research setting \\nwith as few predetermined ideas as possible—especially logically \\ndeduced, a priori hypotheses . . . to remain open to what is actually hap­\\npening” (Glaser, 1978, p. 3). To be sure, a researcher’s professional train­\\ning and experience will often stimulate an initial research interest \\nproviding the motivation for pursuing a study, but a GT study must set \\naside the preconceiving practices of traditional training and established \\nresearch methodology that condition the researcher to approach a study \\nthrough an extant theoretical framework, ad hoc hypotheses, defined \\nresearch questions, interview protocols, or precise coding schemes.\\nPreconception: A Classic Example!\\nGlaser recalls how meeting Alvin Gouldner (1920–1980) impressed upon \\nhim just how far off a preconceived interest can be from what is really going \\non in a research setting. He recounts Gouldner’s using the example of a \\nstudy in which a student was sent out to study risk-taking behavior among \\nsteeplejacks in New York City. The student was frustrated by the difficulty \\nhe had in getting the “jacks” to talk about risk taking in their line of work \\nuntil one day, he observed them drawing straws before beginning a job and \\nhe was certain they were using the process to allocate the potential risk. \\nTo his surprise, the tactic had nothing to do with risk; rather, they were \\ndrawing straws for the best vantage points in terms of window peeping \\nwhile on the job! The preconceived professional concern with risk taking \\nwas displaced with a focus on what was really taking place—a study in stra­\\ntegic positioning. (Holton, 2011, p. 213)\\nThe fundamental premise of avoiding preconception has been dis­\\nmissed by some scholars as naive and unrealistic (Dey, 1993, 1999; Kelle, \\n2007). Charmaz (2014) insists that “guiding interests, sensitizing con­\\ncepts, and disciplinary perspectives” can serve as “points of departure \\nfor developing, rather than limiting, our ideas” (p. 31), but this is a slip­\\npery slope. When our inquiry is pre-framed through such heuristics, we \\ncan inadvertently shape how we approach our study and the resultant \\ndata that we use (see “Staying Open” in Appendix B, pp. 168). The adage \\n“start as you mean to go” is particularly relevant to GT. The intention \\nhere is not to dismiss extant knowledge but rather to suspend its privi­\\nleging in the initial design and framing of the study. “To preconceive \\nrelevance is to force data, not to discover from data what really works as \\na relevant explanation” (Glaser & Strauss, 1967, pp. 142–143).\\n\\n\\nChapter 4    Clarifying Common Sources of Confusion in Grounded Theory\\n49\\nYet, to remain truly open to GT’s principle of emergence can be one \\nof the most challenging issues to those new to GT. This is a challenge \\nthat is often complicated by institutional requirements that routinely \\nand without question require such pre-formulation at the study pro­\\nposal stage. Such assumptions are so engrained in the scholarly com­\\nmunity that they have become pro forma requirements in thesis \\nproposals at many universities as well as in applications to ethics \\nreview boards and research funding agencies.1 While compromises are \\nnecessary (Xie, 2009), all too often, mixing GT with other research \\napproaches unfortunately reduces the outcome to conceptual descrip­\\ntion, presented thematically or evidentially, rather than as an inte­\\ngrated conceptual explanation. While such practices can render \\nperfectly acceptable research findings, the absence of a core GT pillar \\n(i.e., emergence) invalidates the claim of grounded theory.\\nA persistent preconception regarding GT is that it is a qualitative \\nresearch method (Clarke, 2005; Goulding, 2002; Locke, 2001; Shah & \\nCorley, 2006). We have already addressed this misconception in previ­\\nous chapters. To reiterate, as a general research paradigm, GT can use \\nany and all data and can adopt any philosophical perspective appro­\\npriate to the data and the ontological stance of the theorist. Consistent \\nwith the misguided assumption that GT is a qualitative method, many \\npublished guides to doing GT advocate the use of a preconceived theo­\\nretical framework (Goulding, 2002; Partington, 2002; Suddaby, 2006). \\nConducting an initial literature review is another dimension of precon­\\nception that is frequently advocated, but that again undermines the \\nbasic premise of classic GT, that is, that theory emerges from the con­\\nceptualization of data and not from extant theory. Such extensive \\nreview of the literature at the outset of a study can stifle a researcher’s \\nefforts to remain open to discovering what concepts are in the data \\nregardless of what extant theories might suggest should be the focus of \\ninquiry.\\nGT does not ignore extant literature, however. Rather, it simply \\nholds review of the literature in abeyance until the core category of the \\ngrounded theory has emerged. Doing so avoids the possibility of forc­\\ning the theoretical direction of the study and instead enhances the \\npotential to discover a true main concern in the area under study and \\nto theorize from a new perspective and in rupture with existing \\nliterature.\\nPractically speaking, reviewing literature at the outset may simply \\nresult in the researcher spending valuable time on an area of literature \\nthat proves to be of little significance to the discovered grounded \\ntheory.\\n\\n\\nAN OVERVIEW OF CLASSIC GROUNDED THEORY\\n50\\nIndeed the analyst may be hard put to know which conceptual \\nfield [to invest in] until a theory [with its core category] \\nemerges. If there is a particularly good theory in the field, one \\nmay . . . look for emergent fits. The result is usually extending \\nand transcending the extant theory. (Glaser, 1978, p. 31)\\nExtensive engagement with the literature prior to data collection \\nand analysis may also constrain the researcher’s ability to remain open \\nto theoretical ideas that may not have figured prominently in the litera­\\nture. In classic GT,\\nthe literature is just more data to be coded and integrated into the \\nstudy through constant comparative analysis but its analysis and \\nintegration happens only after the core category, its properties \\nand related categories have emerged, and the basic conceptual \\ndevelopment is well underway, not in advance as is commonly \\npresented in published research. Unless pre-empted by precon­\\nception, emergence is natural with the resultant grounded theory \\noften charting new theoretical territory. (Holton, 2007a, p. 272)\\nl\\nl\\nDESCRIPTION VERSUS CONCEPTUALIZATION\\nSome Firsthand Experiences of Novice Grounded Theorists\\n“At times I feel as if I am lingering on the ‘ground level of description’ wait­\\ning for the wings of conceptualization to pop up on my back and enable me \\nto fly.” (“The Conceptualization of Data,” 2010)\\n“I have a lot of categories and relationships between them but I am \\nafraid that I am using too much of the wording from existing theory.” \\n(E-mail excerpt from PhD student)\\n“My supervisor had her own preconceived ideas and was bent on full \\ndescription. She kept asking me which group said what and to give percent­\\nages on the initial coding results.” (E-mail excerpt from PhD student)\\n“Quite early in the research process, however, a ‘story’ presented by \\na participant was so compelling, and original, that just to ‘code’ the inter­\\nview transcript without a full inclusion of this particular participant’s story \\nwould have seriously diminished the research outcome.” (Bryant & Lasky, \\n2007, p. 179)\\n\\n\\nChapter 4    Clarifying Common Sources of Confusion in Grounded Theory\\n51\\nAs the comments in the previous box suggest, those new to GT \\noften struggle to transcend the descriptive level. Transcending descrip­\\ntive detail to achieve a conceptual explanation rather than simply \\ndescribing research findings can be a particular challenge for those \\nwho are research-trained in evidence-based professional disciplines \\nthat value solely empirical detail. However, moving to explanation and \\nthe subsequent explanatory power of conceptual theory has much to \\ncontribute to these fields where GT’s theoretical completeness offers a \\n“powerful, predictive ability” (Guthrie, 2000, in Glaser, 2002, p. 342).\\nEven though some studies might start with description before con­\\nceptualization is brought into play, unlike descriptive methods, the \\ngoal of GT is to discover and conceptually explain a main concern in a \\nsubstantive area of inquiry and how that concern is addressed or \\nresolved. Grounded theory is not about the accuracy of descriptive \\nfindings. It is about tapping into the latent structures in social settings. \\nIn abstracting these patterns to a conceptual level, GT theoretically \\nexplains rather than describes behavior that occurs generally in many \\ndiverse groups with a common concern (Glaser, 2003). While empiri­\\ncally grounded in one substantive area, the conceptual nature of a \\ngrounded theory offers general implications as theoretical propositions \\nthat extend beyond its empirical grounding.\\nIn contrast, some qualitative research produces descriptive find­\\nings from which generalizations may be suggested but which are \\nalways limited by the sample (unit, case, etc.). At the other end of the \\nspectrum, some quantitative research aiming at uncovering universal \\nlaws, considered as true until proved otherwise, overgeneralizes with­\\nout taking structures, context, and human agency into consideration. \\n“The concern is always accuracy, that is, the researcher has worrisome \\naccuracy to constantly contend with. Generalizations are difficult to \\nmake from one unit to another and even within the unit studied” (Gla­\\nser, 2007, p. 42). Description bogs down conceptualization and reduces \\nthe level of abstraction with the concepts buried within the detailed \\ndescription. As such, it is difficult, if not impossible, to see the simple \\nelegance in the relationships between concepts that constitute the \\npropositions or hypotheses that together integrate the (grounded) the­\\nory into a rich, multivariate conceptual theory (Glaser, 2003).\\nA further limitation of qualitative description is that it is confined \\nto the unit studied, with emphasis on accurately depicting that unit but \\noffering no claims to generalizing findings beyond the study sample. \\n“The theory remains ‘trapped’ in the unit and is read as description” \\n(Bigus, Hadden, & Glaser, 1994, p. 44). Description is also stale dated \\nbecause as the unit of analysis changes over time, the description \\n\\n\\nAN OVERVIEW OF CLASSIC GROUNDED THEORY\\n52\\nbecomes less accurate and relevant. By contrast, GT’s focus on latent \\npatterns of social psychological and social structural processes, action, \\nand interaction gives it an enduring quality, and its propositional (not \\nevidentiary) presentation suggests general implications of the theory \\nthat apply beyond the unit(s) theoretically sampled (and more gener­\\nally, with some modification based on further theoretical sampling, \\nthus eliminating the risk of hasty generalization).\\nA certain mystery seems to surround the notion of conceptualization. \\nKlag and Langley (2013) refer to the “conceptual leap” when theorizing \\nfrom empirical data as “a consciously realized and abstract theoretical \\nidea in an empirical study . . . beyond the specific context . . . connect[ing] \\nsituated particularities with conceptual understandings . . . grounded \\nin abductive reasoning” (p. 150) and theoretical sensitivity as the \\ncapacity to generate such conceptual leaps (p. 157). Although they \\nrecognize GT’s creative and iterative nature, they conflate GT with \\nqualitative research and characterize the process as one of wading \\nknee-deep in data while leaping from insight to insight. In so doing, \\nthey capture the qualitative researcher’s experience of “data \\noverwhelm” (Glaser, 2003, p. 24) but appear to miss the systematic \\nmethodology inherent in classic GT through constant comparative \\nanalysis, theoretical sampling (for core category emergence), and \\ntheoretical coding (for theoretical integration). GT’s full package goes \\nwell beyond simply leaping from data to theorizing.\\nl\\nl\\nSUBSTANTIVE VERSUS THEORETICAL CODING\\nCoding is a rather ubiquitous term, often applied to describe the itera­\\ntive and complex procedures at the heart of GT’s constant comparative \\nmethod of analysis. In GT, coding refers to both the substantive coding \\nof empirical data and, later, to the theoretical integration of emergent \\nconcepts through theoretical coding. While coding and memoing in GT \\nproceed in tandem, substantive and theoretical coding do not. Theo­\\nretical coding is subsequent to substantive coding.\\nSubstantive coding is the process of conceptualizing the empirical \\nsubstance of the area under study, that is, the data in which the theory is \\ngrounded. Incidents in the empirical data are coded for indicators of \\nconcepts from which a grounded theory is then generated. Substantive \\ncoding begins with the initial open coding of data where the analyst \\ncodes each incident for as many concepts as possibly relevant. Open \\ncoding proceeds (in tandem with memoing) until the analyst perceives \\na pattern emerging that may indicate a potential core category. Once a \\n\\n\\nChapter 4    Clarifying Common Sources of Confusion in Grounded Theory\\n53\\ncore category has been tentatively identified, the analyst switches from \\nopen coding to selective coding by delimiting data collection and \\nanalysis to just the core category and any potentially related concepts. \\nAt this point, the analyst focuses on theoretically saturating the core \\ncategory and related concepts and then proceeds to the final stage of \\nGT coding: theoretical coding. Theoretical codes conceptualize how the \\nsubstantive codes may relate to each other as hypotheses to be \\nintegrated into the final theory, reminding us that the essential \\nprinciple of earned relevance in GT is not restricted to conceptual \\nemergence but is equally important in the conceptual integration of the \\ntheory through the emergence of relevant theoretical codes.\\nIn theoretical coding, the analyst constantly compares and ques­\\ntions the many analytic memos that have been generated through the \\nearlier stages of substantive coding. This conceptual sorting (Glaser, \\n1978, p. 116) generates further memoing as the analyst explores the \\nrelationships among the core category and other categories. This pro­\\ncess is done manually and thoughtfully with the analyst exploring \\nconnections and conceptual orderings among the concepts until a pat­\\ntern emerges that organizes the concepts as a theoretical model with \\nfull explanatory power of the main concern. The underlying impera­\\ntive, however, is that theoretical coding is not selected and imposed on \\nthe theory but that it must emerge and fit to earn its relevance as a theo­\\nretical integrator of the core category and related concepts.\\nCommon misunderstandings regarding theoretical coding include \\n(a) that it refers to the use of theoretical labels when coding empirical \\ndata and (b) that the analyst selects a specific theoretical code and \\napplies it to the overall analysis of the data (Kelle, 2005). Various schol­\\narly accounts of GT methodology have offered confusing and ultimately \\nmisleading advice regarding coding. Locke (2001), for example, describes \\nGT coding as substantive and theoretical but then suggests that theoreti­\\ncal codes are discipline-related and applied by the analyst “through a \\nkind of top down process” (p. 66). She goes on to suggest that while the \\ngeneral rule is that substantive coding occurs early in the study and \\ntheoretical coding at a later point, “these are not hard and fast \\nrules . . . researchers will be working with a mix of in vivo and theoretical \\ncategories throughout their analysis” (p. 66). It would appear that she \\nhas not understood the transition from substantive coding to theoretical \\ncoding through core category emergence and theoretical saturation.\\nGibson and Hartman (2014) offer a somewhat vague account of \\ntheoretical codes as selected and applied during selective coding, with \\ntheir particular preference being the use of typologies to explore relation­\\nships between concepts. They make no mention of the hand sorting \\n\\n\\nAN OVERVIEW OF CLASSIC GROUNDED THEORY\\n54\\nprocess in theoretical coding that Glaser has repeatedly emphasized as \\ncrucial to ensuring that theoretical codes are emergent and not forced \\n(Glaser, 1978, pp. 116–127; 1998, pp. 187–192; 2005a, pp. 33–40). Dey \\n(1999) suggests that the distinction between substantive and theoretical \\ncoding is unclear. He poses three questions that suggest his level of con­\\nfusion is considerable: “Is theoretical coding an aspect of substantive \\ncoding or a separate activity? How do we select from among theoretical \\ncodes that all fit the data? Are the empirical cues guiding selection based \\nonly on observation?” (p. 109). Here again, such questions indicate that, \\nwhile offering advice on GT, the author has not understood the progres­\\nsive nature of GT coding or the emergence and appropriate fit of \\n­\\ntheoretical codes through hand sorting of conceptual memos.\\nAnother common misconception regarding theoretical coding is \\nthat grounded theory is always a basic social process (Clarke & Friese, \\n2007). To be sure, the basic social process is undoubtedly the most \\nwidely used theoretical code in grounded theory. Glaser (2005a) \\nacknowledges this popularity and frequency but at the same time \\nnotes, “I have seen many GTs that have no process in them, so process \\nmust be earned, NOT assumed” (p. 96). As another example of over­\\nreliance on one theoretical code, Glaser (1978) notes the tendency of \\nsome theorists to conflate behavioral consequences with intentional \\nstrategies and to rely too frequently on the strategy family of theoreti­\\ncal codes (p. 76). He has offered a wide range of potential theoretical \\ncodes (Glaser, 1978, pp. 72–82; 1998, pp. 163–175; 2005a, pp. 17–30) and \\nencourages researchers to read widely from theories in many disci­\\nplines as a means of identifying and adding new theoretical models to \\ntheir repertoire. The purpose here is not to select (i.e., preconceive) a \\nmodel but to simply enhance theoretical sensitivity by opening the \\nresearcher up to the numerous ways in which theory can be modeled.\\nIN SUMMARY\\nCommon sources of confusion in grounded theory can be streamlined \\nto three main areas: emergence (i.e., remaining open to what is actually \\ngoing on in the data) versus preconception (i.e., forcing preconceived \\nnotions on the data), description (i.e., describing research findings con­\\nfined to the unit studied) versus conceptualization (i.e., conceptually \\nexplaining how a concern is addressed and solved in a substantive \\narea), and substantive (i.e., conceptualizing the empirical substance of \\nthe area under study) versus theoretical coding (i.e., conceptualizing \\nrelationships between substantive codes).\\n\\n\\nChapter 4    Clarifying Common Sources of Confusion in Grounded Theory\\n55\\nTEST YOUR KNOWLEDGE\\nIn the following questions one or several answers may be correct.\\n1.\\t When you conduct a research using classic GT\\na.\\t You start with a research question\\nb.\\t You must use qualitative data\\nc.\\t You must conduct an extensive literature review ex ante\\nd.\\t None of the above\\n2.\\t When you conduct research using classic GT\\na.\\t Extensive description is necessary\\nb.\\t You must make theoretical propositions\\nc.\\t You should not read about theories in other domains\\nd.\\t None of the above\\n3.\\t Substantive coding\\na.\\t Includes open and selective coding\\nb.\\t Explains the main concern\\nc.\\t Is about searching the core category\\nd.\\t None of the above\\n4.\\t Theoretical coding\\na.\\t Helps the researcher to sort conceptually\\nb.\\t Explains the main concern\\nc.\\t Involves making propositions linking concepts\\nd.\\t None of the above\\nFURTHER READING\\nGibson, B., & Hartman, J. (2014). Rediscovering grounded theory (Chapter 12). \\nLondon: Sage.\\nGlaser, B. G. (1978). Theoretical sensitivity: Advances in the methodology of grounded \\ntheory (pp. 2–3, 31–33, 44–46). Mill Valley, CA: Sociology Press.\\nGlaser, B. G. (1998). Doing grounded theory: Issues and discussions (Chapters \\n\\n5 & 6). Mill Valley, CA: Sociology Press.\\nGlaser, B. G. (2001). The grounded theory perspective: Conceptualization contrasted \\nwith description (Chapters 3, 4, & 6). Mill Valley, CA: Sociology Press.\\nGlaser, B. G. (2011). Getting out of the data: Grounded theory conceptualization. Mill \\nValley, CA: Sociology Press.\\nGlaser, B. G. (2013). No preconceptions: The grounded theory dictum. Mill Valley, \\nCA: Sociology Press.\\n\\n\\nAN OVERVIEW OF CLASSIC GROUNDED THEORY\\n56\\nENDNOTE\\n1.\\t For some helpful advice on managing ethics review, see http://www\\n.groundedtheoryonline.com/getting-started/ethical-review-irb.\\n2.\\t The \\nCreative \\nCommons-Attribution-NonCommercial-ShareAlike \\nLicense 4.0 International applies to all works published by the \\nInternational Journal of Qualitative Methods. Copyright for articles \\npublished in the International Journal of Qualitative Methods remains \\nwith the first author.\\n\\n\\n57\\n5\\nFinding Your Data\\nAfter studying this chapter, you will:\\n• understand grounded theory’s use of data\\n• know how to collect data for a grounded theory study\\n• understand the iterative nature of data collection and analysis in doing \\ngrounded theory\\n\\t\\nWHY “ALL IS DATA”\\t\\nl\\nAs grounded theorists, we have the wonderful freedom to identify and \\nutilize in our analyses any and all data obtained from any relevant, \\navailable source. Doing so enhances the power of our theories to \\nexplain what is really going on. We use the data as we choose and find \\nit to explain the pattern of behavior at issue. This explanatory power is \\na primary reason that many of us are drawn to GT. Its conceptual \\npower renders GT valuable not only for theorists but also for practitio­\\nners and the participants in our studies through its focus on explaining \\n“high impact main concerns” (Glaser, 1995, p. 4). The GT process \\nensures that our research, when finished, will be “actionable” (Schön, \\n1983).\\nIn many methodological books, and as highlighted in Walsh (2014c), \\nmuch is said about methods and techniques and little about data. \\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n58\\nHowever, methods and techniques do not mean much without data, as \\n“data is a necessary basis on which to build theories” (Evermann & Tate, \\n2011, p. 634) and, in most cases, you need good data to produce good \\ntheory (Lyytinen, 2009). Creswell (2014) describes qualitative data as text \\nand images and quantitative data as numbers. The term quantitative data \\nis mostly used to describe a type of information that may be counted or \\nexpressed numerically and is amenable to statistical analysis (Monette, \\nSullivan, & Delong, 2011), whereas qualitative data provide explanation \\nfor, and information about, something in the form of words that describe \\nthat phenomenon in depth (Monette et al., 2011). However, a researcher \\nmay choose to quantitize qualitative data (Sandelowski et al., 2009) or to \\nqualify (Creswell, 2007) or qualitize (Sandelowski, 2000) quantitative data.\\nA researcher may also use secondary data in various ways, with \\ndifferent techniques, regardless of the methods used to collect these \\ndata. While many GT studies rely largely on qualitative data and inter­\\nviews as a primary method for data collection, GT uses any and all data \\nand data collection methods, including observations, visual and audi­\\ntory media, or surveys from any sources (e.g., reports, newspapers, or \\nquestionnaires). As Glaser and Strauss (1967) explain,\\nin field studies, theoretical sampling usually requires reading \\ndocuments, interviewing, and observing all at the same time, \\nsince all slices of data1 are relevant. There is little, if any, \\nsystematic interviewing of a sample of respondents, or \\ninterviewing that excludes observation. At the beginning of \\nthe research, interviews usually consist of open-ended \\nconversations during which respondents are allowed to talk \\nwith no imposed limitations of time. Often the researcher sits \\nback and listens while the respondents tell their stories. Later \\nwhen interviews and observations are directed by the emerging \\ntheory, he can ask direct questions bearing on his categories. \\nThese can be answered sufficiently and fairly quickly. Thus, \\nthe time for any one interview grows shorter as the number of \\ninterviews increases, because the researcher now questions \\nmany people, in different positions and different groups, about \\nthe same topics. Although the time taken by most interviews \\ndecreases as the theory develops, the sociologist still cannot \\nstate how long all his interviews will take because a new \\ncategory might emerge at any time; this emergence will call for \\nlengthy open-ended conversations and prolonged observations \\nwithin some groups. Also, theoretical sampling aimed at \\nfollowing an incident or observing over a period of time \\n\\n\\nChapter 5    Finding Your Data\\n59\\nrequires sequential interviews, with no clear notion of when \\nthe sequence will be terminated. (pp. 75–76)\\nFurthermore, when developing emerging concepts/constructs that \\nhave not been extensively investigated in past literature, it is usually \\nimportant to start your investigations with qualitative data. Additional \\nslices of quantitative data may be needed to enhance theoretical satura­\\ntion for generalizability purposes and for formal theory generation. A \\ngood example of this may be found in Leonardi and Bailey (2008). \\nAlthough the authors never clearly state they conducted their research \\nin a GT stance, an informal interview with the second author of this text \\nconfirmed that they had adopted a GT stance; they did not state this \\nclearly in their work because the journal in which they wished to publish \\nwas not familiar with mixed data GT. You will find more about this issue \\nof fulfilling research outlets’ conventions, while not forfeiting your basic \\nphilosophical assumptions and methodological choices, in Chapter 9.\\nSo this idea of “all is data” (Glaser, 2001) helps us to get away from \\nan overemphasis on interview data and opens us up to fully exploring \\na wide range of data sources and slices. For example, one GT study of \\ninteragency working was completed by observing group meetings and \\nusing meeting minutes (Wood, 2009). Classic GT studies such as Jones \\nand Kriflik (2006) and Sandgren, Thulesius, Fridlund, and Petersson \\n(2006) explain their use of personal interviews, informal conversations, \\nobservation, and document analysis. Hämäläinen (2014) describes \\nusing opportunistic sampling to access entrepreneurs through his \\nsocial networks and through chance encounters when traveling. \\nSnowballing sampling through referrals led to further opportunities \\nfor data collection. His self-interviews and field observations over a \\nperiod of years as a practicing entrepreneur provided yet more data. \\nGlaser and Strauss (1965a) relied extensively on field observation and \\ninterviews from comparison groups across a range of hospital settings \\nin their formulation of awareness theory, including geriatric, preterm \\nbirth, cancer, intensive care, pediatric, neurology, urology, and \\nemergency wards. Walsh (2014a) used secondary qualitative data from \\na previously published GT and from the literature, primary quantitative \\ndata collected through a survey, and primary qualitative data collected \\nthrough in-depth interviews (see Appendix C, Section 1.3). Focus \\ngroups are another valuable way to collect data from a number of \\nindividuals. They can be especially valuable for exploring a potential \\ncore concept as they allow you to gain the perspectives of a number of \\nindividuals at one time. Stjernsward and Ostman (2008) used both \\ninterviews and focus groups in their study of families living close to a \\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n60\\ndepressed individual. Focus groups allowed them to capture data \\naround mutual experiences and identities and to explore different \\nperspectives through group interaction (p. 359). Focus groups can also \\nallow you to gain a perspective by observing and noting the dynamics \\nof the group conversation.\\nAs with interviews, it is important not to preconceive the direction of \\nthe group’s conversation by applying a set of preplanned questions; \\nrather, open the group up with a “grand tour question” (Simmons, 2010). \\nFor example, if you were interested in exploring the impending status \\npassage (Glaser & Strauss, 1971) of university students in their final year \\nof study, you might open the discussion by saying, “I am supposed to be \\nstudying how students in their final year of study prepare for what’s next, \\nbut I don’t know what to ask you. What do you think we should talk \\nabout?” and then wait to see how individuals respond, how the conversa­\\ntion builds. You might probe by asking what they are doing to prepare for \\nthe next stage (job hunting, grad school, etc.), how it feels, and so on. \\nAgain, you can observe whether probing strikes a chord and generates a \\nlively exchange or whether something else seems to be of more interest. \\nDo different concerns emerge? What appears to be the main concern?\\nHelpful Hint: Focus groups are more challenging in that a lot can be \\nhappening at once given the number of individuals involved. Therefore, \\nit is usually recommended to have a second person help you facilitate \\nand make field notes.\\nWhile for grounded theorists, all can serve as data, in conceptualiz­\\ning, it is important that we are able to recognize what kind of data we are \\ngetting. Here we are not referring to the methods of data collection \\n\\n(i.e., interviews, observations, surveys) but rather to the intention of the \\ndata—what Glaser (1998) refers to as types of data—baseline data \\n\\n(i.e., honestly describing what is), proper line accounts (i.e., filtered data \\nwhere only what is considered proper or correct is revealed), profession­\\nally interpreted (i.e., a third-party expertise perspective), and vague \\nresponses (i.e., concealing or evading; p. 9). Recognizing the type of data \\nthat you are getting is also data and is important to discovering what is \\nreally going on in the situation at hand. Glaser’s (1978) questions for cod­\\ning data will help you to think about what types of data you are gathering; \\nin particular, his question, “What is actually happening in the data?” (p. 57).\\n\\n\\nChapter 5    Finding Your Data\\n61\\nMemoing as you code your data will also help you to discover \\nwhat is actually happening in the situation at hand; for example, is the \\ninterviewee offering an account of his or her own experience (baseline \\ndescription), a reflection on that experience (vague or distanced), a \\nprojection of someone else’s experience (professionally interpreted), or \\na massaged corporate version (proper line)? All are possibilities, and it \\nis important to know what type of data you are getting and why you \\nare getting this type of data. What is this really telling you about the \\nphenomenon you are studying, and, more precisely, what may be \\nemerging as the main concern?\\nFor instance, in the focus group example presented earlier, you can \\nobserve who appears to dominate the conversation, how others react, \\nwhat the tone is, and what the nature is of the interactions among \\ngroup members; in other words, what is really going on with the group \\nin their discussion. Is there a pattern in the way the group interacts? It \\nis quite conceivable that using a series of focus groups, you might dis­\\ncover an interesting GT around, for instance, surfacing sensitive topics, \\nsilencing, or influencing and persuasion. If you are open to what \\nemerges, you may have more than one GT study from the same focus \\ngroup data.\\nIndeed, observation is an important source of data for doing GT. \\nWe are surrounded by data but often neglect to observe and note what \\nis happening. As one student wrote, “I am a nurse on leave (50%) to do \\nmy PhD research. I have discovered, however, that when I go into work \\nto pick up my mail or to just visit, I find myself observing what I see \\nand hear and making note in my memos.” Here is someone who has \\ngrasped GT’s all is data dictum for leveraging valuable data about \\nwhat’s really going on; these are data that she may not have gained in \\na formal interview setting.\\nPerhaps You Are Wondering . . . \\nHow many interviews do I need for a GT study?\\nThere is no magic number of interviews required. While it is common \\nfor some qualitative or quantitative methods to specify sample size at \\nthe outset of the study, for a GT study the extent of data \\n\\ncollection is governed by the emergence and theoretical saturation of \\na core concept. Students are often required, however, to state a specific \\nnumber in their research proposals, and some supervisors may have a \\n(Continued)\\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n62\\nspecific number or range in mind to ensure adequate coverage of a \\npopulation or sufficient data depth. So, how should you address this \\nin your proposal? You might consider including a statement like this: \\n\\nI expect my field work to involve approximately 30 participants. In accor­\\ndance with classic grounded theory methodology, however, sample size \\ncannot be definitively established in advance of the field work; rather, it \\nemerges as the researcher engages in the constant comparison of concep­\\ntual indicators/incidents articulated by interviewees and proceeds through \\nfurther theoretical sampling to theoretical saturation of emergent concepts. \\nHere, you’ve satisfied the requirement by offering a proposed sample \\nsize while also qualifying the specific number to allow for what \\nemerges as necessary for theoretical saturation which you can later \\njustify in the methods section of your research paper or thesis.\\n(Continued)\\nA frequently overlooked source of data for doing GT is secondary data, \\nthat is, data that have been collected for some other purpose. Data collec­\\ntion can be expensive as well as time-consuming. Accessing extant data \\ncan accelerate and densify conceptual elaboration. As one example, Barney \\nGlaser’s doctoral research was based on the secondary analysis of survey \\ndata conducted by the Survey Research Center at the University of Michi­\\ngan. Glaser explains that the data consisted of responses to 1,952 survey \\nquestionnaires completed by research staff at a large government medical \\nresearch organization from which he selected a subset of 332 respondents \\nand 100 questions that seemed relevant to his research interest of recogni­\\ntion in scientific careers (Glaser, 2008, pp. 34–35).2 In addition, Glaser and \\nStrauss (1967) devote a full chapter to the use of available qualitative data \\nfor generating theory. Furthermore, in our times of big data, the Web is also \\na valuable source of already collected data (secondary data, both qualita­\\ntive and quantitative) that are mostly free and available to all, through \\ndata mining. Thus, readily available data can be a rich source for theoreti­\\ncal sampling as a GT study progresses.\\nAnalysis of secondary data can yield unexpected emergent core \\nconcepts that trigger theorizing. As one example, while conducting a \\nqualitative thematic analysis to assess leadership development needs \\nin an organization, one author of this text was struck by what she \\nl\\nl\\nOVERLOOKED SOURCES OF AVAILABLE DATA\\n\\n\\nChapter 5    Finding Your Data\\n63\\nobserved to be a persistent disconnect between what managers \\nexpressed about the organization’s efforts to enhance healthy work­\\nplace practices (proper line data) and what they shared about their \\nown experiences of the organization’s culture (what was really going \\non). This disconnect emerged in interviews as an unexamined inner \\ndialogue where managers struggle to resolve the tension they experi­\\nence while still attempting to honor espoused organizational values \\n(Holton & Grandy, 2015). While this emergent concept of an inner dia­\\nlogue was outside the objectives of the original research project, it \\noffered important insight and explanation as to how middle managers \\nin organizations struggle to process conflicts in attempting to honor \\nand reconcile personal and organizational values.\\nOf course, extant literature can also be used as data, particularly if \\nthe literature is empirically grounded. A major issue in using literature \\nas data for a GT study is when to use it. While it is common in many \\nresearch methodologies to initiate a study with a literature review, as \\ndiscussed in Chapter 4, in a GT study we suspend our engagement \\nwith extant theory so as to avoid the risk of preconceiving our work, \\nthereby missing important concepts that may emerge from our data \\nsimply because we are too focused on what the literature has deemed \\nas important. In GT, we start with empirical data and remain open to \\nwhat emerges rather than being influenced by what comes from the \\nliterature. Once our core concept has emerged, we can use the literature \\nas an additional data source to theoretically sample and constantly \\ncompare concepts from the literature with emergent concepts. As such, \\nliterature based on empirical data can help to elaborate our emerging \\nconcepts and assist us in knowing where to direct our attention in fur­\\nther theoretical sampling.\\nPerhaps You Are Wondering . . . \\nMy university (or funding agency) requires that I do an ini­\\ntial literature review as part of my research proposal. How \\ndo I manage this without undermining my GT study?\\nStandardized approaches to research proposals frequently follow the \\ntraditional hypothetical-deductive approach where extant literature is \\nsearched and assessed so that the proposed study can be situated \\nwithin, and contribute to, that body of knowledge. The literature review \\nis intended to point to a gap in that knowledge and the proposed study \\n(Continued)\\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n64\\nGaining Access\\nGT studies that rely on qualitative data often feature interviews as a \\nprimary source of data. Individuals, however, often exhibit reticence to be \\ninterviewed, especially if you wish to tape-record the interview. Some \\nmay politely agree but then guardedly engage. It is often much easier to \\nengage in informal conversations with participants, simply making note \\nof key points, incidents, and responses that you can later write up, code, \\nand memo. It is often enough to simply capture a few key words during \\nthe conversation; you may not even need to write them down at the time \\nbut simply store key words or phrases in your memory. They will serve as \\ntriggers for you to sit down as soon as possible after the conversation to \\nelaborate in a more extensive field note that you can then code and memo.\\nAudio taping often sets up a completely different dynamic between \\nthe researcher and participant. Taping can change the data as the \\nresearcher relies on the assurance of a taped interview that can be con­\\nsulted later and, as such, the researcher may relax his or her listening \\nrather than staying fully engaged in capturing key words and thoughts \\nin situ. In addition, interviewees may respond very differently to ques­\\ntions when they know their answers are being taped. Many become \\nnervous at the prospect of being formally interviewed and taped. Such \\nnervousness can be a form of performance anxiety (e.g., Will I be able to \\nanswer the questions? Will I be asked something I prefer not to answer? If so, \\nhow do I handle the situation without feeling embarrassed? What if I make the \\nresearcher unhappy? And so on . . .).\\nResearchers often worry that by not taping, they will miss impor­\\ntant data. What you are discovering in a GT study is a pattern that is \\ngoing on all the time, so there are endless opportunities to capture it; \\nto contribute to addressing that gap. One strategy for addressing this \\nrequirement is to keep the initial literature review broad in scope, per­\\nhaps by offering a contextual or thematic overview of the substantive \\narea of research interest, setting the stage for an exploratory study \\nrather than focusing on a specific gap to be addressed. The latter \\nassumes a focus that may not be relevant to the proposed GT study \\nand, as such, not only consumes researcher time and energy but may \\nwell derail the potential to discover an emergent grounded theory.\\n(Continued)\\nl\\nl\\nCOLLECTING YOUR OWN DATA\\n\\n\\nChapter 5    Finding Your Data\\n65\\nno one interview is that precious. If something is relevant to your the­\\nory, you will find multiple indicators in subsequent data as your study \\nprogresses. Furthermore, taping and then later transcribing leads to a \\nvolume of data that may render little of relevance in terms of identify­\\ning a core concept. Researchers are often required, however, to tape \\ninterviews as a condition of ethics review or, in the case of doctoral \\nstudents, to demonstrate competence in interviewing. The experience \\nof Sandgren, Thulesius, Petersson, and Fridlund (2007) describes an \\napproach that accommodates such requirements without unduly \\nundermining the GT research approach:\\nThe first ten interviews were tape recorded and transcribed, \\nbut for the later interviews, only field notes were taken accord­\\ning to classic grounded theory (Glaser, 1998, 2001). Interviews \\nlasted between 45 and 90 min. By the end of the study, inter­\\nviews were shorter owing to the delimiting properties of \\ngrounded theory. (p. 228).\\nConducting a research study within an organizational context can \\ncarry its own challenges to gaining access and can take considerable \\ntime. There are, of course, the ethics review requirements of universities \\nand granting agencies that must be met in terms of full and informed \\nconsent and voluntary participation of the individual participants. To \\ngain access to individual participants within an organizational context, \\nhowever, also necessitates determining the appropriate organizational \\ngatekeeper, ascertaining their concerns with your request and considering \\nwhat you may be able to do to address these concerns.\\nOne way to initiate the relationship is to request a meeting to dis­\\ncuss your research project and to seek their advice in how their organi­\\nzation might be involved in your study. Be humble, polite, and \\nrespectful in your interactions with the organization and the gate­\\nkeeper, as you will most certainly be shut out if you respond otherwise. \\nOnce you have been able to allay, or at least reduce, any apprehension \\nthrough an open discussion, you will have a better chance to gain the \\ndesired access. The more prepared you are to allay any concerns, the \\nmore likely your chances will be to proceed with data collection with­\\nout undue delays.\\nHelpful Hint: Be sure to make notes during the conversation with the \\norganizational gatekeeper, as your notes may reveal early indicators of \\na main concern within the organization.\\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n66\\nOf course, there will be times when you simply will not be granted \\naccess or your access will be restricted in such a way that you may not \\ngain the data you had hoped for. In such situations, what you gain may \\nbe proper line data, where the individuals you are permitted to inter­\\nview and the situations you are permitted to observe will be carefully \\nselected and controlled to present the organization in a favorable light. \\nThese are still data that you can code and memo, but it is important to \\nremember what these data are and to capture this in your memos. What \\nare you really hearing in these data? What is the context in which you have \\ngained these data? What data are you not being given access to? What’s really \\ngoing on in this situation? One way to encourage a breakthrough when \\nyou think you are being proper-lined is to probe with questions such as \\nWhat do you mean by that? I hear a lot of managers say that, but I’m not sure \\nwhat they really mean. What does it mean to you? Why is that important?\\nCollecting Quantitative Data\\nWhile doing GT, you may feel compelled to use quantitative data slices and \\nwish to collect these through a survey, for instance, in order to theoretically \\nsaturate your concepts. Then, it is often essential during pretests to also inter­\\nview some of the respondents in order to investigate how the questions asked \\nin your survey are understood and if they address the category that you are \\ntrying to saturate.  While administering the final survey, it is often necessary \\nto informally reassure participants and confirm that anonymity will, indeed, \\nbe maintained and that they will not be traced through their answers. Beyond \\nethical issues, if respondents are convinced you will keep their information \\nconfidential, it will increase the chances that you are collecting baseline data as \\nopposed to proper-line data, more especially if data are self-reported by par­\\nticipants and hence subjective. (See Appendix C, Section 1.4 as an illustration \\nof the careful choice of measures in a quantitative survey within a GT study.)\\nWhile collecting data for a survey, it might also be important to carefully \\nnote the context of the survey administration. (As an illustration, you may \\nrefer to Appendix C, Section 1.3.) For instance, if students are respondents \\nto a questionnaire, one could legitimately inquire and note if both the ques­\\ntionnaire administered and the research core investigations are related to a \\ncourse attended by participants. Along the same lines, it might also be neces­\\nsary to note if student participation to the survey is mandatory as a required \\npart of a course, based on partially voluntary participation (extra grades for \\nthose who answer the questionnaire) or totally voluntary participation (no \\nspecific incentives). \\nY\\nour emergent theory may vary with the social structural \\nconditions under which the data are collected.\\n\\n\\nChapter 5    Finding Your Data\\n67\\nIn GT, it is also often helpful to reinterview some participants as \\nyour study progresses. Reinterviewing is a form of theoretical sampling \\nthat can add important depth to your emerging conceptualization and \\ncan be as important as the breadth offered by interviewing a range of \\nindividuals. Reinterviewing can be especially helpful where your access \\nto data may be limited to a very specific population or a sensitive area \\nof research interest (e.g., transgender youth, marathon runners over the \\nage of 80, illegal immigrants seeking health care, etc.). In particular, \\nonce your core category has emerged, you may wish to theoretically \\nsample by returning to earlier interviewees and probing more specifically \\naround the emergent core. You might say something like “I recall when \\nwe chatted previously that you mentioned . . . (core concept). . . . I’ve heard \\nfrom others some similar experiences so I’d like to know more about what you \\nexperienced.” Or you might say, “In some of my conversations with others, \\nthey have mentioned. . . . Has this been your experience? Can you tell me more \\nabout what happened?” Given that reinterviewing is so often a valuable \\nway of theoretically sampling to elaborate the properties and dimensions \\nof your emergent core concept, it is important to identify your intention \\nto employ multiple interviews for theoretical sampling purposes as part \\nof your research ethics proposal.\\nHelpful Hint: In preparing your research ethics proposal, be sure \\nto consider potential issues related to gaining access and articulate \\nhow you plan to address each to ensure the key principles of ethical \\nconduct in research (i.e., free and informed consent, voluntary partici­\\npation, confidentiality, etc.). It is much easier to address these issues \\nat the outset rather than having to later return to the ethics review \\nboard with amendments. Addressing potential issues at the outset of \\nyour study forces you to think through what you intend to do and, as \\nmuch as possible, problem solve in advance. It also demonstrates to \\nthe review board that you have thought through the research process \\nand can conduct your study in a scholarly and ethical manner, thereby \\nenhancing the confidence of the review board in your overall proposal.\\nInstilling a Spill\\nIn other research approaches, it is generally researcher interests \\nthat shape a study. Researchers explore and answer questions that \\ncome from their own professional interests or, as is often the case with \\nnovice researchers, topics that have been suggested by a supervisor \\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n68\\nbased on their knowledge of the research field and the related litera­\\nture. As we have previously explained, in classic GT the research aim \\nis to discover a main concern that is being processed, managed, or \\nresolved in the substantive area under study. The best way to discover \\nthis main issue is to avoid, as much as possible, any preconceptions \\nregarding what the main issue may be. As Nathaniel (2008) explains, \\n“the researcher must be open to hearing the story from the informant’s \\nperspective” (p. 61). This means setting out to collect your data without \\na theoretical framework or specific research questions, interview proto­\\ncols, or predetermined coding schemes.\\nIn collecting interview or informal conversation data for a GT study, \\nall that is needed is one “grand tour” question (Simmons, 2010), \\nsomething as simple and open as “Tell me about your job/family/hobby/what \\nit’s like to work here, etc.” Sandgren et al. (2006) opened their interviews \\nwith palliative care nurses by asking, “Tell me what it is like to care for \\npalliative cancer patients” (p. 81). Jones and Kriflik (2006) used two \\nbroad questions to open their interviews: “Tell me what leadership \\nmeans to you,” and “How do you feel about your job?” (p. 160). \\nHämäläinen (2014) found that, in many instances, simply starting the \\nconversation with, “Hi. I’m doing [a] PhD about entrepreneurship” \\n\\n(p. 17) was all that was needed. Questions such as these open the space \\nfor interviewees to tell you what they want to talk about. As the researcher, \\nyour challenge is to wait to see where they wish to take the conversation \\nand then to ask follow-up questions that sustain the openness while \\nprobing enough to have the interviewees elaborate on what they have \\nraised. Sandgren et al. (2007) describe using more specific questions such \\nas “Tell me about a difficult caring situation,” “How do you handle \\ndifficult situations?” and “Tell me about a caring situation that went well” \\nto encourage interviewees to elaborate on the range of their caring \\nexperiences (p. 228). As Glaser (1998) points out, the goal in GT is to \\n“instill a spill” (p. 111) to find the trigger that starts the person talking \\nnaturally and exhaustively about what concerns or interests him or her.\\nPerhaps You Are Wondering . . . \\nCan my opening question be too broad? \\nOne novice researcher asked My field of interest is to research emo­\\ntions in organizations. I have conducted 10 interviews so far asking \\npeople how they are feeling about their work. I have coded and memoed \\nafter each interest. The only issue is that I do not seem to derive \\n\\n\\nChapter 5    Finding Your Data\\n69\\nTry to set the climate for the interview as one between equals and \\nnot one of interviewer and interviewee. Your goal is to set the tone for \\nthe conversation so that your interviewee feels comfortable in opening \\nup and sharing with you. Use your intuition; you’ll get a sense from the \\nenergy in the conversation when the interviewee is finished spilling. \\nContinuing to probe may render little of value. Remember, you are not \\nseeking to capture descriptive detail; rather, your data are in service to \\nidentifying concepts.\\nIt may be helpful to begin by talking about yourself and by swap­\\nping stories to reduce the sense of formality. However, it is important \\nnot to dominate the conversation. The less you talk, the more likely you \\nare to get to their main concern. If there is a pause in the conversation, \\ndon’t panic and attempt to fill the void. Offer soft encouragement for \\nthe individual to continue, for example, a nod or a smile and a simple \\n“Please go on.” You might also just repeat the last few words the indi­\\nvidual has said and then wait. Give the interviewee time to process his \\nor her responses. Be patient. As Glaser (2001) says, be a “Big Ear” \\n\\n(p. 175), an active and engaged listener. Probe and encourage your \\ninterviewees to tell you more. Don’t rush them or they may shut down, \\ncommon patterns or processes or even indicators thereof. . . . I am find­\\ning that the variables for feelings are endless and are so tightly coupled \\nwith the individual. I was wondering if this has to do with the fact that \\nmy inquiry is too broad.\\nHere we have a novice who is trying to stay open to emergence of \\na main concern but feeling worried as no potential core category (pat­\\ntern or process) is emerging.  \\nWhat should he do? Well, his area of \\ninterest—emotions in organizations—is clearly a broad topic. \\nThere are \\nseveral options that he might consider: First of all, he might narrow his \\npopulation to a certain type of organization (public service, communi­\\ncations, health care, etc.) or to a certain occupation (auditors, recep­\\ntionists, nurses, etc.) across organizations. He might also narrow his \\nopening question somewhat; for example, “How did you feel when you \\narrived at work this morning?” “Tell me about a time when . . . ,” or “What \\nfeeling would best describe how you feel about your job?” Offering a little \\ncontext for your question should help to elicit focused responses from \\nyour interviewees without preconceiving and encouraging responses \\naround a suggested emotion. Be patient with the first few interviews. \\nStay open and permit your interviewees to decide where they wish to \\ntake the conversation. Follow their lead with more probing questions.\\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n70\\nassuming you are in a hurry or simply not interested in what they want \\nto share. This is especially important if what is being shared is in any \\nway sensitive. Ensure that your interviewees have time to feel comfort­\\nable in opening up to you. As one researcher wrote, “I have found that \\nonce you break the ice with them, they go on and on—literally! They want to \\ntalk and they want to tell you their experience.”\\nStaying Open\\nGlaser (1978) offers useful ideas for staying open during field work. These \\ninclude changing the style of your interviewing (e.g., sharing a walk with the \\ninterviewee rather than sitting down to converse, etc.) and the place(s) where, \\nor time of day when, you normally do interviews (e.g., a park rather than a \\ncoffee shop; weekends rather than weekdays, etc.). Hans Thulesius, a medical \\ndoctor and experienced grounded theorist, draws on the advice of famous \\nCanadian physician \\nWilliam Osler for assessing patients in offering the following:\\nAs a GP, I’ve learned that listening to the patient will give you the diagnosis \\nin most cases. One of the worst mistakes a doctor can make is interrupting \\nthe patient too early because they are already busy formulating a hypothesis \\nabout the patient’s problem. So in GT interviewing, if you’re doing the same \\nthings as doctors, you want to go from one part of the story to the next \\nso you’re busy thinking about the next question to move the interview \\nalong but you’re not focusing on what the interviewee is saying. If you \\njust wait, almost always the interviewee will give you the rest of the story, \\nthe explanation you were looking for, so your “thinking up” energy is just \\nwasted. Let your interviewee talk. Listen carefully and just give encouraging \\nnods or hums, keep eye contact, and eventually repeat the last word in \\nwhat he just said to encourage him to go on with his story—[it] almost \\nalways works! (Personal communication, October 20, 2014)\\nField Notes\\nGT does not require capturing the type of detailed descriptive data \\nthat may be considered as important in some circles. However, you \\nmay find that these descriptive details may be required in order to \\npublish your theories in some top tier research outlets (see Chapter 9). \\nAs far as GT methodology is concerned, however, field notes suffice as \\nGT is a concept-indicator methodology that relies on the patterned \\npresence of conceptual indicators through multiple incidents as cap­\\ntured across your data. What you are attempting to do is capture \\n\\n\\nChapter 5    Finding Your Data\\n71\\nincidents (experiences, stories, gossip, confessions) that may suggest a \\nconcept (i.e., a pattern of behavior) that helps to explain what is going \\non (i.e., a main concern) in the situation that you are studying.\\nNovice researchers will often ask about the right way to do field notes, \\nbut there is no one right way. Variations on field notes are common to \\nmany professions; for example, the proceedings of a business meeting \\nmay be captured as minutes, journalists will make notes during inter­\\nviews, lawyers keep case notes, and doctors and nurses observe and make \\nnote of a patient’s symptoms. Field notes in GT are not intended to serve \\nas detailed evidentiary notes as is common to the health or legal profes­\\nsion; instead, GT field notes are simply in the moment reminders of inci­\\ndents that may indicate potential concepts. There is no need to try to \\ncapture detailed descriptions. You need only capture enough detail about \\nthe incidents being shared or observed so that you will be able to later \\ncode your notes for those concepts and elaborate your ideas in memos. For \\ninstance, when you are interviewing, simply jot down key words as you \\nlisten and complete your field notes as soon as you have an opportunity \\nto code and memo your ideas (see Figure 5.1 for a sample of quick notes \\ntaken in the field). Or, when you are developing a questionnaire based on \\nverbatim data where exploratory factor analysis points at an item as \\ndouble loading on two categories previously identified as different, take \\nsome brief notes as you go along, until you have time to go back to your \\nqualitative data and investigate further. Waiting too long, however, may \\nmean that you lose some of the insights that triggered the initial note.\\nFigure 5.1    Field Notes Taken During an Interview\\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n72\\nWhile doing GT, “any data (qualitative and/or quantitative), films, pho­\\ntographs, literature etc. may be used. They may be collected by methods, \\nand analyzed with the help of techniques that appear suitable . . . to \\nfacilitate theory emergence, while congruent with . . . [your] philosophi­\\ncal stance” (Walsh, 2014c, p. 19). Whether you begin data collection with \\ninterviews, focus groups, observations, social media accounts, organiza­\\ntional documents, or newspapers, it is important to remember that data \\ncollection is incremental and iterative. GT enables the stimulation of \\nparallel processes of analysis, synthesis, and conceptualization of data. \\nData are coded as they are collected, and memos are written in conjunc­\\ntion with coding. Memoing opens up the data to reveal subtle patterns \\nof social behavior. It is also very important not to hinder the emergence \\nPerhaps You Are Wondering . . . \\nWill I miss important data if I use field notes rather than tap­\\ning and transcribing interviews? \\nOne novice researcher wrote, “I know that using field notes helps to \\ndelimit the data for analysis. My supervisor is worried, however, that \\nthis leads to important things being left out. Also, what determines \\nwhat I remember to include in the field notes? \\nWe compared one field \\nnote that I had done with the transcript for that interview and found \\nthat some of the subtleties and nuances from the interview had been \\nlost in the field note. So, should I rely solely on the field notes or start \\nselectively coding sections of the interview transcripts to ­\\ncomplement \\nthem?”\\nThis researcher is worried about missing something important by \\nrelying on field notes. It is important to remember that GT is the study \\nof a concept, that is, a pattern of social behavior. Any social setting will \\ncontain many patterns, but the goal of GT is not to identify and explain \\nevery pattern in the data but, rather, just one pattern that appears to \\naccount for much of what is going on.  As such, worries about ensuring \\nfull and accurate coverage in data collection are inconsistent with GT \\nmethodology.  The full scope of social behavior in any substantive set­\\nting is simply too complex to be fully captured in any one grounded \\ntheory study. Anything that is relevant to that pattern will surface in \\nseveral incidents (i.e., interchangeable indicators), so there is really no \\nneed to worry about missing something in a transcript.\\nl\\nl\\nDATA COLLECTION AND ANALYSIS AS AN ITERATIVE PROCESS\\n\\n\\nChapter 5    Finding Your Data\\n73\\nof theory but rather to allow your creativity some freedom, while at the \\nsame time following classic GT’s essential guidelines.\\nAdhering to the simultaneous and sequential nature of data collec­\\ntion, coding, and memoing is fundamental to classic GT and will save \\nyou in so many ways! GT’s constant comparative analysis and theoreti­\\ncal sampling will save you time as you only need to collect the data \\nnecessary to discover and elaborate your theory. Doing so will save \\nyou from that overwhelming feeling that so many novice researchers \\nexperience when they have spent considerable time collecting moun­\\ntains of data “because it was there” and then have no idea where to \\nstart in making sense of what they have collected. Instead, collect your \\ndata one opportunity at a time and focus your efforts on coding and \\nmemoing as you go, building up your memo bank. Doing so will reveal \\nthe subtle patterns that begin to emerge from your data to explain what \\nis really happening in your area of research interest. If you need quan­\\ntitative data to help you make sense and see more clearly the emerging \\npatterns, then you should not hesitate to collect these data as well.\\nRemember also that GT is a delayed action phenomenon (Glaser, \\n1978, p. 34; 1998, p. 6). Frequently, several days after you have collected \\na slice of data, coded it, and written memos, new thoughts will arise in \\nyour mind that bring moments of clarity to what is emerging from your \\ndata. Glaser (1978) calls this preconscious processing (p. 23). At such \\nmoments, stop and capture your thoughts in another memo! You may \\nwant to explore these new insights by theoretically sampling; you may \\nwant to go back to the last interviewee and ask more questions about this \\nissue, review earlier data to see if there are indicators of this new idea, or \\npursue it in your next interview. The goal is to develop more under­\\nstanding of conceptual ideas as they emerge in your analysis.\\nIN SUMMARY\\nGT is based on a concept-indicator model of theory building. Concep­\\ntual indicators can be found in any and all types of data. The cardinal \\nrule in GT is not to undermine the discovery of latent patterns in data \\nby preconceiving what to look for or what type of data to use. The goal \\nwhen initiating your research is to remain open to what emerges as a \\nsignificant concern in the area under study and to then build an empir­\\nically grounded theoretical explanation of how that main concern is \\nhandled. In GT, data are analyzed as they are collected. Additional \\ndata are theoretically sampled based on emerging conceptual ideas as \\ncaptured in field notes and analytic memos.\\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n74\\nTEST YOUR KNOWLEDGE\\nIn the following questions one or several answers may be correct.\\n1.\\t While doing classic GT\\na.\\t You have to use only qualitative data\\nb.\\t You have to use qualitative and quantitative data\\nc.\\t You may use quantitative data\\nd.\\t You may use any data\\ne.\\t None of the above\\n2.\\t In classic GT, primary data\\na.\\t Are the data that are the most important\\nb.\\t Can only be used once\\nc.\\t Are the data specifically collected for your research\\nd.\\t Should be used first\\ne.\\t None of the above\\n3.\\t While conducting interviews in a classic GT stance\\na.\\t You must tape all interviews\\nb.\\t You should never tape interviews\\nc.\\t You can tape some or all interviews\\nd.\\t None of the above\\n4.\\t In classic GT\\na.\\t Detailed description is essential\\nb.\\t Field notes are essential\\nc.\\t Capturing incidents to identify patterns is essential\\nd.\\t None of the above\\nFURTHER READING\\nAndrews, L., Higgins, A., Andrews, M. W., & Lalor, J. G. (2012). Classic \\ngrounded theory to analyze secondary data: Reality and reflections. \\nGrounded Theory Review, 11(1), 12–26.\\nGlaser, B. G. (2001). The grounded theory perspective: Conceptualization contrasted \\nwith description (Chapters 11 & 12). Mill Valley, CA: Sociology Press.\\nNathaniel, A. (2008). Eliciting spill: A methodological note. Grounded Theory \\nReview, 7(1), 61–65.\\n\\n\\nChapter 5    Finding Your Data\\n75\\nENDNOTES\\n1.\\t For an excellent discussion of “slices of data,” see Gibson and Hartman \\n(2014), pages 125–131.\\n2.\\t For a full elaboration of Glaser’s secondary analysis of data, see Glaser \\nand Strauss (1967), Chapter 8.\\n\\n\\n76\\n6\\nAnalyzing Your Data\\nAfter studying this chapter, you will:\\n• appreciate the distinction between descriptive analysis and conceptual \\nanalysis\\n• articulate how to employ constant comparative analysis\\n• distinguish between the various types and stages of coding used in GT\\n• appreciate the key role of analytic memoing in GT analysis\\n• articulate how to identify a core category\\n• appreciate the iterative nature of data analysis in GT\\n• evaluate the impact of software applications on GT analysis\\nI\\nn this chapter we focus on the techniques of data analysis as \\nemployed in classic GT. The conceptualization of data through cod­\\ning and memoing is the foundation of GT analysis, the goal being the \\ndiscovery of a latent pattern of social behavior that explains a main \\nissue or concern within an area of research interest. This latent pattern \\nanalysis approach unravels the complexity in a social setting to reveal \\na main concern motivating predictable action. Such patterns of behav­\\nior are often subtle and deeply embedded, requiring careful attention \\nto what is revealed in the data. GT focuses on one issue—a main \\n­\\nconcern—and the pattern of behavior that explains how that concern is \\nperpetually processed, managed, or resolved.\\n\\n\\nChapter 6    Analyzing Your Data\\n77\\nThis pattern—the core category—emerges as data are open coded \\nand conceptualized. Once the pattern emerges, the analyst shifts atten­\\ntion to concentrate on theoretically sampling and selectively coding \\nfurther data to elaborate, and eventually saturate, this core category \\nand any concepts that appear to have some relationship to the core. As \\na pattern, the core is modeled via theoretical coding to organize and \\nintegrate the relationships between and among the core and related \\nconcepts. This theoretical coding is not selected but rather emerges \\nthrough intensive hand sorting of analytic memos, aided by creative \\ninsights through the analyst’s preconscious processing of analytic ideas \\nthat have percolated outside conscious efforts to analyze the data—\\nwhat Glaser (1998) calls the drugless trip.\\nIncidents can be described as indicators of phenomena or experi­\\nences as observed or articulated in the data. These are labeled and \\nanalyzed using the constant comparative method to generate initially \\nsubstantive, and later theoretical, codes. The process begins with sub­\\nstantive coding through careful reading of the data to identify inci­\\ndents as indicators of concepts emergent within the data. As the \\ntheorist proceeds to compare incident to incident in the data, then \\nincidents to emerging concepts, a main issue or concern surfaces across \\nthe coded data, and a category, which appears to account for most of \\nthe variation around this issue or concern, becomes the focus of the \\nstudy. This category is referred to as the core category and becomes \\nthe focus of further selective data collection and coding efforts until the \\ntheorist has sufficiently elaborated and integrated the core category, its \\nproperties, and its theoretical connections to other relevant categories. \\nAt this point, the theorist has achieved theoretical saturation of the core \\nconcept and is ready to articulate the emergent theory.\\nWorth Noting: Study participants may not specifically articulate a main con­\\ncern in your collected data.  \\nThe main concern is a latent pattern that they \\nsimply act on without consciously identifying.  \\nThe grounded theorist sees the \\nconcern emerge as a latent pattern through systematic coding and memoing.\\nAs Suddaby (2006) suggests, GT is not simply the routine applica­\\ntion of formulaic technique to data but has a clear creative component. \\nThis creativity is particularly evident in the memoing of conceptual \\nideas that parallels the coding process. Indeed, without this parallel \\nprocessing, key conceptual realizations will be lost and the resulting \\nanalysis flat. It is this grab that offers such explanatory power.\\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n78\\nl\\nl\\nCONCEPTUAL ANALYSIS AS DISTINCT FROM\\nDESCRIPTIVE ANALYSIS\\nBarney Glaser has been known to comment that description runs the \\nworld, with conjecture a close second. His point is that while empiri­\\ncally descriptive and hypothesized research studies outnumber \\ngrounded theories, they often fail to offer high-impact explanatory \\nvariables in the same way that a good grounded theory can offer. As \\nwe have discussed in Chapter 4, description captures a moment in \\ntime, but the essentially linear nature of descriptive writing hinders the \\ntheorist’s ability to produce a complex yet parsimonious multivariate \\ntheory. Concepts remain buried amid detailed empirical accounts, and \\nit is difficult if not impossible to see the simple elegance in the relation­\\nships between concepts that together integrate a grounded theory and \\nprovide its explanatory power. GT provides a systematic way to con­\\nceptualize and integrate a theory to achieve that full explanatory \\npower.\\nAs discussed in Chapter 3, constant comparative analysis is a \\nfoundational pillar of classic GT. After first discussing here the \\nvarious types of coding in GT, we explore constant comparative \\nanalysis, memoing, and core category emergence and development. \\nNext we discuss some cognitive attributes of the classic GT approach \\nthat enhance theorizing. While we touch on theoretical sampling \\nand theoretical coding in this chapter, we reserve their full discussion \\nfor Chapter 7, as both are key techniques in shaping a grounded \\ntheory.\\nl\\nl\\nUSING CONSTANT COMPARATIVE ANALYSIS\\nConceptualization is fundamental to GT for “without the abstraction \\nfrom time, place and people, there can be no multivariate, integrated \\ntheory based on conceptual, hypothetical relationships” (Glaser, 2002, \\n\\np. 261). Coding and memoing are the key heuristic techniques in constant \\ncomparative analysis. While initial attempts at data analysis may very \\nwell be more empirically descriptive than conceptual, the classic grounded \\ntheorist moves past description to fracturing the data for their conceptual \\nessence through constant comparative analysis. The purpose of constant \\ncomparison is to see if the data support and continue to support emerg­\\ning concepts. Moving through the data to see incidents over and over, \\nconstantly comparing and conceptualizing, forces the theorist into\\n\\n\\nChapter 6    Analyzing Your Data\\n79\\nconfronting similarities, differences and degrees of consis­\\ntency of meaning between indicators which generates an \\nunderlying uniformity which in turn results in a coded cate­\\ngory and the beginnings of properties of it. From the com­\\nparisons of further indicators to the conceptual codes, the \\n[conceptual] code is sharpened to achieve its best fit while \\nfurther properties are generated until the [concept] is verified \\nand saturated. (Glaser, 1978, p. 62)\\nBy moving through the data to see conceptual indicators repeated \\nin incident after incident and constantly comparing these, concepts are \\ndeveloped and their properties and dimensions elaborated.\\nConcept validity is established through the interchangeability of \\nindicators in multiple incidents. This means that although incidents are \\ndifferent, they indicate the same concepts. As such, GT’s constant com­\\nparative analysis is a gradual building up of conceptual codes into \\nconcepts and then concepts into categories.2 Doing so requires the theo­\\nrist to grapple with both chaos and control. The chaos is in tolerating \\nthe uncertainty of not knowing in advance but remaining open to what \\nemerges through the diligent application of constant comparison for \\ninterchangeability of indicators and theoretical sampling for core emer­\\ngence and saturation. At the same time, constant comparison helps \\nalleviate the “data overwhelm” (Glaser, 2003, p. 24) that results from \\nsimply collecting data in a linear fashion without taking the time for \\nanalysis and shaping further data collection through theoretical sam­\\npling. Used in conjunction with theoretical sampling, constant com­\\nparison generates theory that is “integrated, consistent, plausible and \\nclose to the data” while also “allow[ing], with discipline, for some of \\nthe vagueness and flexibility that aid the creative generation of theory” \\n(Glaser & Strauss, 1967, p. 103).\\nThe constant comparative approach begins with the first data col­\\nlected and continues throughout the analytic process through to theo­\\nretical saturation and integration, illustrating the nonlinear, spiraling \\nnature of GT with the boundaries between the various steps not clearly \\ndelineated. Glaser (1998) describes the GT process as “subsequent, \\nsequential, simultaneous, serendipitous and scheduled” (p. 15). The \\nprocess begins with the analyst first coding and comparing incidents in \\nthe data—first, incident to incident; later, incidents to concepts (i.e., \\nemergent concepts from earlier comparisons) whereby concepts and \\nthe properties and dimensions3 of those concepts are elaborated and \\neventually saturated (i.e., no new properties or dimensions are emerg­\\ning from additional data); and then, concept to concept to establish \\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n80\\nconceptual levels between relevant concepts in relation to the core \\ncategory.\\nGlaser uses the phrase interchangeability of indicators to describe \\nthe process of comparative analysis of conceptual indicators in data. \\nThe phrase comes from his background in quantitative analysis and the \\nwork of Paul Lazarsfeld in index formation:\\nLazarsfeld used the traditional psychological index formation \\napproach. He would take questions that bore on the same topic \\nand sum up their aggregate values for a group. Then he cuts \\nthe range of values in half or thirds to construct an index with \\nvarious values. For example several indicators on degree of \\nrecognition produced an index of high, medium and low rec­\\nognition. Then every person in the sample got a value. What \\nwas not done was to compare with its meaning each indicator \\nto each indicator . . . When the comparison is made . . . a cate­\\ngory is generated (an index with meaning). To continually \\ncompare each incident to incidents and categories generates \\nmeaningful properties of categories. This constant comparison \\nis a far richer yield of concepts and relationships . . . than is the \\nyield from just summing. (Glaser, 1998, pp. 23–24)\\nConstant comparison requires concepts to earn their way into a \\ngrounded theory through this interchangeability of indicators. This \\nearned relevance through emergence is a fundamental premise of GT \\nand is grounded in the coding of data. (See Appendix B for an illustra­\\ntion of the constant comparative method.)\\nl\\nl\\nTYPES AND STAGES OF CODING IN GROUNDED THEORY\\nAll coding in GT is in aid of conceptualization. There are, however, two \\ndistinct types of coding: substantive and theoretical. By substantive \\ncoding we are referring to the process of conceptualizing the empirical \\nsubstance of our data. Theoretical coding refers to the integration of \\nour emergent concepts as a model that explains the pattern of relation­\\nships between the core category and related concepts. Substantive cod­\\ning begins with the first data collected, whereas theoretical coding is \\nreserved for the later stage of the GT process once a core category has \\nemerged and been fully developed (saturated) and any concepts that \\nhave a relationship to the core category have been fully saturated. At \\nthis stage, it is time to explore the modeling of those relationships as a \\nfully integrated grounded theory.\\n\\n\\nChapter 6    Analyzing Your Data\\n81\\nOpen Coding\\nThe substantive coding process begins with open coding of empir­\\nical data. Open coding means to identify incidents in the data that \\nappear to indicate one or more concepts and labeling these using one \\nor two salient words per code—preferably verbs to capture the action \\nin concept(s) indicated. Glaser (1978) refers to two types of substantive \\ncodes: in vivo and analytic. He describes in vivo codes as capturing \\nexactly what is going on in the incident (e.g., remaining composed) \\nand analytic codes as explaining theoretically what is happening (e.g., \\nidentity maintenance). As such, in vivo codes are a first effort at \\n­\\nconcept identification, and analytic codes raise the level of conceptual \\nabstraction.\\nDescriptive labeling of incidents is fairly straightforward, but it can \\nbe a challenge for some to move from the descriptive to the conceptual \\nlevel. In close reading of the data, concepts are identified and eventually \\nsaturated. Close reading also ensures that no important concepts are \\noverlooked and that concepts earn their relevance rather than being \\naccorded an assumed importance based on extant theory. It is important \\nhere to avoid numerous one incident codes. Constant comparison of inci­\\ndents, noting similarities and differences among incidents and looking \\nfor patterns emerging in the coded data, will help to reduce the potential \\nfor drowning in one incident codes. The more you code, the more you \\nbegin to see the same conceptual indicators (codes) across your data—a \\npattern emerges, raising the conceptual level of your coding.\\nA common source of confusion for novice grounded theorists, \\nespecially those familiar with qualitative methods, arises with Glaser’s \\n(1978) reference to line by line coding (p. 57). Many novice researchers—\\nparticularly PhD students—are required to tape and transcribe \\ninterviews as a demonstration of research skill development and \\npossibly as a condition of graduate studies or research ethics approval. \\nHowever, in most instances, coding this type of data may not lead the \\nresearcher beyond description due to the volume of data produced in \\ntranscripts. However, Glaser (1978) is not referring to the line-by-line \\ncoding of interview transcripts but rather to the coding of field notes—\\nabbreviated notations captured during periods of field observation, \\ninterviews, conversations, focus groups, and so on. Field notes help the \\nresearcher to begin the conceptualization process while in the field as \\nshe listens for the main concern and notes indicators of how that \\nconcern is processed or resolved. Quick notes can then be written up in \\nmore detail and then coded and elaborated through memoing. (See \\nFigure 6.1 for a sample of coded field notes.) Coding of field notes, in \\ncomparison to lengthy transcripts, helps the analyst to stay focused on \\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n82\\nwhat is really happening in the substantive area—what appears to be \\nthe main concern— while the absence of detailed conversational or \\ndescriptive passages facilitates coding and memoing on a higher \\nconceptual level.\\nFigure 6.1    \\u0007\\nCoded Field Notes\\nIt goes without saying that a grounded theorist must code her own \\ndata. Coding unlocks the conceptualization of data and initiates theory \\ngeneration. A fundamental criterion for coding is the suspension of any \\npreconceived notions about what the data should reveal. Accordingly, \\na grounded theorist doesn’t employ a predefined set of codes assumed \\nto be relevant. As Glaser (1992) suggests, “the analyst should just not \\nknow as he approaches the data” (p. 50). Instead, the goal is to fracture \\nthe data conceptually by posing questions such as the following:\\nWhat are these data a study of?\\nWhat category does this incident indicate?\\nWhat is actually happening in the data?\\nWhat is the issue facing the participants?\\nWhat is their main concern, and how do they manage or resolve it?\\n\\n\\nChapter 6    Analyzing Your Data\\n83\\nAsking these questions of the data as you code incident by incident \\nfacilitates the generation of a core category (Glaser, 1978, p. 57).\\nLabeling incidents with conceptual codes and, later, with categories \\n“help[s] the theorist to build an analysis of the data rather than remain at \\nthe level of ethnographic description” (Charmaz, 1990, p.  1167). Glaser \\n(1992) reminds us that in open coding we must guard against “slipping \\ninto preconception instead of listening carefully to each incident in order \\nto figure out what the research is truly a study of” (p. 45). Open coding \\nincident by incident in the data with the list of previously presented ques­\\ntions keeps the analyst close to the data rather than lapsing into ungrounded \\ntheorizing and corrects any forcing of “pet” themes on the data by requir­\\ning “the analyst to verify and saturate categories, and minimiz[ing] miss­\\ning an important category, produc[ing] a dense, rich theory, and giv[ing] a \\nfeeling that nothing has been left out” (Glaser, 1978, p. 58).\\nKeep moving through the data rather than pondering too long on \\nany one incident. Doing so will facilitate your seeing emerging patterns \\nin your coding. As concepts begin to take shape, additional constant \\ncomparisons help to clarify the logic in emerging concepts and their \\nproperties as well as reducing the emerging theory through a smaller \\nset of higher-level interrelated concepts. Given this constant, emergent \\nrefinement of concepts and categories, it is important not to fret about \\nthe specific labels at the open coding stage. Use labels that help to cap­\\nture the idea while keeping in mind that there will be time later in the \\nprocess to refine the labels of your concepts and categories.\\nGlaser on Data Collecting\\nGlaser (1978) suggests that staying too long “in the field” collecting data \\nis a common problem. He recommends no more than half a day at a time \\nin data collection. He also reminds us that while open coding can gener­\\nate many codes very quickly, it is important to interrupt coding to memo \\nideas. Coding and memoing as data are collected is important because \\nthis is when ideas about the data are freshest in your mind. Doing so also \\nfacilitates theoretical sampling by allowing you to see where you need \\nto go next to elaborate on ideas that emerge through your coding and \\nmemoing.\\nSelective Coding\\nIn selective coding, the analyst delimits coding to only those \\nconcepts (i.e., variables) that relate to the core category in sufficiently \\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n84\\nsignificant ways to produce a parsimonious theory. If a concept, \\nregardless of its novelty or the personal preference of the analyst, \\ndoes not have relevance in relation to the core category, it is dropped \\nfrom subsequent analysis and theoretical elaboration. In this way, \\nthe core category becomes a guide to further data collection and \\ntheoretical sampling. During the initial open coding, the analyst will \\nhave coded many incidents with many of these being coded for more \\nthan one concept. Once a potential core category is selected, the \\ncoding technique shifts from open to selective. Because not all data \\npertain to the core category (or related concepts), it is necessary to \\n“delimit” coding to just that which pertains to the core and related \\nconcepts.\\nInitiating Selective Coding\\nDelimiting occurs at two levels. First, as the theory integrates, it solidifies \\nwith fewer modifications needed as the researcher compares the next inci­\\ndents of a category to its properties. Later modifications are mainly about \\nclarifying the logic of the theory and integrating elaborating details of prop­\\nerties into the major outline of interrelated categories. As the researcher \\nbegins to discover an underlying uniformity in the categories and proper­\\nties, the theory is reformulated with a smaller set of higher level concepts. \\nThis second level of delimiting the theory reduces the original list of cat­\\negories for coding.\\nAs the theory develops, becomes reduced, and increasingly works bet­\\nter in ordering a mass of data, the researcher becomes committed to it. \\nThis allows for a delimiting of the original list of categories for subse­\\nquent collecting and selective coding of additional data, according to the \\nnewly established boundaries of the theory. By delimiting the focus to \\none category as the core variable, only those categories related to that \\ncore are now included in the theory. This list of categories, now delim­\\nited for additional selective coding, is subsequently (and continuously) \\ndelimited through theoretical saturation of each category. (Holton, 2007a, \\npp. 280–281)\\nAny incidents that have been coded for several possible \\nconcepts are reviewed and their relevance and relationship to the \\ncore category affirmed in memos. As concepts are saturated \\n\\n\\nChapter 6    Analyzing Your Data\\n85\\n(i.e., ­\\nadditional data fail to yield any new properties or dimensions \\nof the concept), additional data collection and coding are further \\ndelimited as the fully saturated concepts are set aside while the \\nanalyst continues to theoretically sample and saturate the remaining \\nconcepts. Delimiting speeds up the analysis by focusing theoretical \\nsampling and constant comparison on just the core category and \\nrelated concepts and reduces the potential for the analyst to be \\noverwhelmed with excessive caches of data that bear no relevance \\nto the emerging theory.\\nAnother Aspect of Delimiting\\nIn studies with a heterogeneous sample (e.g., senior managers, mid­\\ndle managers, and frontline staff), in which each segment of the sample \\nappears to have a different main concern, the analyst should consider \\ndelimiting the study to focus on one subset of the sample at a time. \\nSubsequent stages of analysis can focus on the other subsets with the \\npotential for different core categories emerging from each subset ana­\\nlyzed. A good example is Sandgren (2010). The author undertook three \\nseparate GT studies in the same substantive area (palliative cancer \\ncare): (1) acute care nurses’ striving for emotional survival, (2) home care \\nnurses’ doing good care, and (3) patients’ and relatives’ living on hold. Sand­\\ngren then used data from the three grounded theories to generate an \\noverarching grounded theory of deciphering unwritten rules. These rules \\nexplain how patients, relatives, and nurses deal with the uncertainty of \\nhow to act and behave within the sensitivities associated with a palliative \\ncare setting.\\nWhen doing open and selective coding, researchers sometimes feel \\nthe need to collect quantitative data. They have done some open cod­\\ning of collected qualitative data and have identified the core category \\nand some related categories through further selective coding. How­\\never, some categories are not saturated, and some data are double \\ncoded. To help produce parsimonious results and fully saturated cate­\\ngories, quantitative data may be collected. A questionnaire may be \\ndeveloped with items grounded in the qualitative data previously \\ncollected. Exploratory factor analysis of resulting quantitative \\n\\ndata may then help the researcher to achieve parsimony and reach \\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n86\\ntheoretical saturation of the various categories previously highlighted \\nwith qualitative data. The process of selective coding is sometimes \\nlimited by the researcher’s interpretation of the data and capacity to \\nsynthesize parsimoniously all the data obtained. In some instances, \\nthis drawback may be considerably lessened by using quantitative data \\nand techniques.\\nMixed Methods Example\\nThe theory of information technology (IT) culture creep (Walsh et al., \\n2010) was grounded in qualitative data. Using qualitative data in a \\nStraussian approach to GT helped provide rich descriptions of IT user \\nideal types and the 18 possible dimensions of the explanatory second-\\norder IT culture construct. However, some of the 18 dimensions \\nwe had qualitatively identified for the IT culture construct appeared \\nredundant, as many items from our qualitative data were double-coded \\nwithin two subcategories. With our qualitative data, we had not been \\nable to eliminate the redundant dimensions among the 18 identified \\nqualitatively. We therefore decided to collect quantitative data and use \\nquantitative techniques to help us solve this issue (Walsh & Gettler-\\nSumma, 2010). With a diversified population, we thus investigated \\nfurther our core category, the users’ IT culture, while linking it to user \\nideal types and eliminated correlated redundant dimensions.  At the \\nsame time, we aimed to assess the relative importance of each first-\\norder construct (the IT culture dimensions) used to describe the ideal \\ntypes. To sum up, mixing data and methods helped us conceptualize \\nwith parsimony; it allowed us to densify the IT culture concept through \\ndimension reduction (from 18 to 8).\\nTheoretical Coding\\nAs discussed earlier, theoretical coding refers to the modeling of \\nthe relationships between and among the core category and related \\nconcepts as a fully integrated theory. It is the final stage in the coding \\nprocess and is largely responsible for the overall “shaping” of the the­\\nory. We turn to a more extensive discussion of theoretical coding in \\nChapter 7.\\nThe different types of coding in classic GT, which we described \\nearlier, are summarized in Table 6.1.\\n\\n\\nChapter 6    Analyzing Your Data\\n87\\n\\t\\nCORE CATEGORY EMERGENCE\\t\\nl\\nTypes of Coding\\nDescription\\nPurpose\\nSubstantive\\nOpen\\nIn vivo codes: \\nCapturing what is \\ngoing on\\nAnalytic codes: \\nConceptualizing what \\nis happening\\nTo identify incidents in \\nthe data that appear to \\nindicate one or more \\nconcepts\\nSelective\\nCoding around the \\ncore category\\nTo identify properties \\nand dimensions of the \\ncore category\\nTheoretical\\nCoding to model the \\nrelationships between \\nand among the core \\ncategory and related \\nconcepts\\nTo shape and integrate \\nthe theory\\nTable 6.1    \\u0007\\nCoding in Classic GT\\nData collection in GT focuses on a general area of interest. Then, as \\ncoding and memoing proceed using the questions listed earlier, the \\nanalyst begins to discern a latent pattern that focuses attention on a \\nmain issue or concern in the area under study. As the analyst memos \\nideas around this main issue, a core category emerges that seems to \\nexplain the most variation in the social behavior around the central \\nissue or concern and how the issue is processed or addressed. This core \\ncategory will relate to several other concepts, providing a focus for \\ntheoretical integration through elaboration of the relationships between \\nand among the emergent core and related concepts. Indeed, this theo­\\nretical integration is the “prime function” (Glaser, 1978, p. 93) of the \\nOne Student’s Experience\\n“When I let go of what I wanted the problem to be, I actually found \\nsome of the answers to what the problem really was, and I actually found \\nsomething much bigger—a theory of who, how, and why.”\\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n88\\ncore category, ensuring that the resultant theory is both conceptually \\ndense and theoretically saturated through the increasing development \\nof conceptual interrelationships that organize the theory. Without a \\ndynamic core variable that cuts across empirical units, theoretical link­\\nages may not be seen and the theorist may opt to simply link up largely \\ndescriptive comparisons (Bigus et al., 1994).\\nDifferentiating Between a Main Concern and a Core Category\\nNovice grounded theorists will often confuse the main concern with the \\ncore category. The distinction is that the main concern highlights the issue \\nor problem that occupies much of the action and attention in the research \\nsetting, whereas the core category explains how that concern or problem is \\nmanaged, processed, or resolved. As such, the main concern and the core \\ncategory are tightly coupled but distinct conceptualizations. As one exam­\\nple, in Holton (2006, 2007b) the main concern of knowledge professionals \\nwas coping with persistent and unpredictable change in their workplaces. \\nThe core category that emerged was rehumanizing, a three-stage process, \\nwhereby knowledge professionals engaged in fluctuating support networks \\nthat served as informal psychic softeners in countering dehumanizing \\nworkplace climates.\\nAs another example, in Walsh, Kefi, and Baskerville (2010), the main \\nmanagerial concern was dealing with different types of IT usage within \\nfirms, and the core category that emerged was the IT acculturation pro­\\ncess investigated through IT culture ideal types.\\nA core category is tentatively identified, often fairly early in the \\nopen coding stage, following which the analyst considers which other \\ncoded concepts appear to have some relationship to the core category. \\nThe core category, as a theoretical code, can be organized or modeled \\nin numerous ways. It is often a process with two or more stages (e.g., \\n“status passage”; Glaser & Strauss, 1971), but it can also be a typology \\n(e.g., “awareness contexts”; Glaser & Strauss, 1965a; Walsh et al., \\n2010), a continuum (Giske & Gjengedal, 2007), strategies (Sandgren et \\nal., 2006), a cycle (Higgins, Barker, & Begley, 2008), or dimensions \\n(Pergert, 2008).\\nGlaser (1978) and Glaser and Holton (2004) offer criteria for select­\\ning and confirming a core category: centrality (i.e., it appears central to \\nthe main concern), frequency (i.e., it reoccurs frequently in the data and \\ncan be seen as a stable pattern), relevance (i.e., it relates meaningfully \\n\\n\\nChapter 6    Analyzing Your Data\\n89\\nand easily with other categories), grab (i.e., imagery and explanatory \\npower with general implications4 beyond the substantive area), and \\nvariability (i.e., although conditions vary, the essential meaning \\nremains constant; see Figure 6.2).\\nCore\\nCategory\\nMain\\nConcern\\nCentrality\\nFrequency\\nRelevance\\nGrab\\nVariability\\nSignificantly\\nexplains \\nFigure 6.2    Core Category and Main Concern\\nAs previously discussed, once a core category has emerged, the \\nanalyst considers other concepts that have patterned out in open cod­\\ning to determine their relevance in relation to the core category. Further \\ndata collection, coding, and memoing are delimited to theoretical sam­\\npling for additional conceptual indicators of the core and these related \\nconcepts, moving the analyst in the selective coding stage. Given its \\ncentrality and significance for ensuring full explanatory power, it does \\ntake longer to saturate the core category than related concepts.\\nOne Student’s Experience\\n“I think I’ve got it . . . one word that has come from my analysis of the \\ndata . . . it fits with so much data that I am amazed. It accounts for dif­\\nferences in emotions in a wide variety of situations; attitudes towards \\na wide variety of shopping types and explains very diverse shopping \\nbehaviors. And it makes sense. It did not come from me; it came from \\nthe data.”\\n\\t\\nTHE KEY ROLE OF ANALYTIC MEMOING IN GT ANALYSIS\\t\\nl\\nMemoing is the core ideational processing of theoretical ideas as they \\nemerge through coding and constant comparative analysis. Memos are \\nthe researcher’s theoretical notes about the data and the conceptual \\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n90\\nconnections between categories. Memo writing is a continual process \\nthat helps to raise the data to a conceptual level and develop the prop­\\nerties of each category. Memos present hypotheses about connections \\nbetween categories and/or their properties/dimensions and begin the \\nintegration of these connections with clusters of other categories to \\ngenerate a theory. The constant comparative reasoning captured in \\nmemos debunks preconceived notions about what the data should \\nreveal (Glaser, 1978). As Glaser (1978) suggests, the goals in memoing \\nare to stimulate and capture conceptual ideas that emerge from coding \\nand constant comparative analysis, to do so with complete freedom \\nfrom the usual constraints of proper writing (i.e., grammar, syntax), \\nand to build up a rich fund of captured ideas in a highly sortable for­\\nmat that facilitates theoretical integration of a rich, dense yet parsimo­\\nnious theory.\\nMemoing your ideas—AS YOU CODE—is essential to transcend­\\ning descriptive analysis. Glaser (1998) notes that “falling out” \\n(p. 177) at the coding stage is the downfall for many who wish to do \\nGT. Without theoretical sensitivity and persistence in using memos \\nto capture ideas as they emerge, such work defaults to conceptual \\ndescription where concepts and their properties are simply identi­\\nfied and described without any integrative theoretical explanation. \\nAs Klag and Langley (2013) suggest, “it is often in the very act of \\nwriting that productive conceptual ideas develop and crystallize” \\n(p. 161).\\nOne Student’s Experience\\n“I read the transcripts line-by-line, asking all the time ‘What is the par­\\nticipant trying to do?’ and making notes on the margins. Curiously these \\nnotes were less important than the memos that this exercise generated. \\nIt was as if writing memos allowed me to better understand what was \\ngoing on—to gradually increase my knowledge of the situation. To see \\nwhat the participants were trying to do and how they did it. After some \\ntime I had so many memos that I was curious to see the whole picture. \\nIt takes time but if you are constantly analyzing the data and looking \\ncarefully at it, there’s no way you can escape learning more and more \\nabout the situation. And the memos are the tool that will allow you to \\nput everything together so you don’t have to worry about keeping it all \\nin your mind all the time.”\\n\\n\\nChapter 6    Analyzing Your Data\\n91\\nLocke (2007) suggests that capturing theoretical ideas in GT neces­\\nsitates dual modes of thinking whereby we acknowledge the tension \\nbetween the irrational free play of ideas as our minds wander and \\nexplore and the rational control required in diligently pursuing those \\nideas in a systematic way: “Thus, staying firmly and single-mindedly \\nin touch with what it is we are trying to understand at the same time \\nas we open and loosen up our thinking to roam freely defines thinking \\nin these dual modes” (p. 570).\\nFine and Deegan (1996) describe serendipity as a unique and contin­\\ngent mix of insight and chance, whereas Klag and Langley (2013) con­\\ntrast serendipity with deliberate application of heuristics to stimulate \\nvariety in theorizing, suggesting that both play an important role in \\nfacilitating conceptual leaps. While acknowledging that GT entails \\nboth the serendipitous and the systematic, Glaser does not view logic \\nand intuition as diametrically opposed but rather as twin foundations \\nfor theory generation. The theorizing process in GT operates at both a \\nconscious and a preconscious level to process and organize empirically \\ngrounded, analytically induced concepts.\\nSkill in memoing, and therefore the conceptual depth of the \\nmemos, is part of the experiential learning curve in GT skill develop­\\nment. Memos are works in progress, intended to capture ideas as they \\nemerge without the worries of writing style, grammar, and spelling \\nthat can cause some writers to freeze their thinking. Many memos writ­\\nten early in the constant comparison process may never find their way \\ninto the final theory because they may not relate to the core category; \\nthus, to spend time perfecting these memos is to waste research time. \\nTime is better spent in letting the ideas flow and capturing them as \\nthey emerge. There will be time later in the writing process to polish \\nthem.\\nEarly in the GT process, writing several short memos on each cat­\\negory facilitates theoretical sorting of ideas from which more substan­\\ntial memos will be written as the process moves toward conceptual \\nelaboration and, later, integration. (See Appendix B for an illustration \\nof the maturation of early shorter memos into more substantial memos \\nas conceptual ideas develop.) Glaser (2014) advocates “free style” \\nmemoing:\\nHans Thulesius, a well-known grounded theory researcher \\nand teacher, confirms my dictum of free style memoing. Hans \\nsays “Memoing is very important in GT research, but you can \\nmemo in whatever way you like. Hand writing, typing or \\ndrawing diagrams but keep them private. When you have a \\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n92\\nbig stack of memos you can begin hand sorting them, placing \\nthem in smaller piles on a big table. Then writing more memos \\nare triggered by the sorted memos which when you reach satu­\\nration will eventually lead to a working paper.” Hans contin­\\nues about free style memoing: “Every person finds his or her \\nown way of writing and organizing one’s memos. The most \\nimportant thing about memos is that you get out your ideas \\nonto paper. Glaser warns against rules for memos since they \\ncan stifle the creative memo writing. Memos should be written \\nat any time to capture the ideas that come when you compare \\nand code data. And stay open to whatever emerges and memo \\nit at all stages of doing GT” (p. 26).\\nPreconscious processing means that ideas can come to mind at \\nany time. Glaser (2014) urges the GT theorist to write continually, \\nmemoing thoughts and ideas as analytic bits. Charmaz (2014) describes \\nwriting and rewriting as crucial phases of the analytic process. Hav­\\ning confidence in your memoing is critical, especially early in the GT \\nprocess when it is unclear as yet where theorizing may be headed. \\nGrounded theorists learn to trust in whatever ideas occur as they are \\nreading through their data and trying to find patterns to explain \\nsocial behaviors in the data. The delayed action nature of GT requires \\ncontinuous memoing of ideas, comparing them to earlier memos to \\ncontinually build a bank of memos. Just as with a bank, your invest­\\nment grows as you continue to deposit and build your account, prom­\\nising a profitable return. As Breckenridge (2013) describes,\\nAs my memo bank grew, my confidence increased and my \\npace quickened as I became less worried about getting the \\ncoding right and more excited by the number of possibilities \\nemerging in the data. The more I analyzed, the longer my \\nmemos became as I was able to see more theoretical connec­\\ntions and identify further questions for theoretical sam­\\npling. This in turn slowed my pace, as I moved back and \\nforth between coding, constant comparison and memo writ­\\ning. At times I found the systematic approach to coding and \\nconstant comparison rather tedious. Thankfully, however, \\nmoments of tedium were only ever temporary and fre­\\nquently gave way to excitement as new concepts emerged, \\nreigniting my motivation as memos became more full and \\nvibrant with new ideas and connections. (p. 9)\\n\\n\\nChapter 6    Analyzing Your Data\\n93\\nSome Advice for Productive Memoing\\nWhile emphasizing that memoing is a highly personal process, Glaser \\n(1978) does offer a number of rules for memoing, including the following:\\n\\t\\n1.\\t \\u0007\\nKeep data and memos separate. Reference memos to field \\nnotes from where they emerge. Use incidents as illustrations \\nof only one code (as much as possible).\\n\\t\\n2.\\t Always interrupt coding to memo whenever an idea arises.\\n\\t\\n3.\\t \\u0007\\nBring on memoing by starting to write on a code—such writing \\nwill open up the output stage of creativity.\\n\\t\\n4.\\t Modify memos as growth and realizations occur.\\n\\t\\n5.\\t Keep a list of emergent codes at hand.\\n\\t\\n6.\\t \\u0007\\nIf too many memos on different codes seem the same, com­\\npare codes or their dimensions for differences that are being \\nmissed between the codes. If they are still the same, collapse \\nthe two into one code.\\n\\t\\n7.\\t \\u0007\\nProblematic digressions should be followed through on a con­\\nceptual, not logical, elaboration basis for the purpose of theo­\\nretical sampling for indicating an area for future research. In \\nsuch cases, memos should make clear distinctions between \\ndata and conjecture.\\n\\t\\n8.\\t \\u0007\\nRun memos open as long as possible to enable development of \\nrich diversity.\\n\\t\\n9.\\t \\u0007\\nWhen writing memos, talk conceptually about the substantive \\ncodes as they are theoretically coded (do not talk about peo­\\nple), and write about the conceptually generated patterns that \\npeople engage in (and not about the people per se).\\n\\t\\n10.\\t \\u0007\\nIf analyst has two burning ideas, he should write them up one \\nat a time to keep them straight and avoid confusion.\\n\\t\\n11.\\t \\u0007\\nIndicate in memos “saturation” when analyst thinks he has sat­\\nurated a category.\\n\\t\\n12.\\t \\u0007\\nNo matter how well memoing techniques are working, always \\nbe flexible with them.\\nA common concern expressed by novice researchers is how to illus­\\ntrate (i.e., verify) the empirical grounding of their concepts and catego­\\nries, as would be expected in a qualitative study. One way to achieve \\nthis is by creating an audit trail that links memos to the data from \\nwhich they were generated. Then, using a simple labeling scheme, \\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n94\\nfootnotes can be used to record the connections (from data to memos) \\nin the final memo sorting and writing up of the theory.\\nAn Example of Labeling to Create an Audit Trail\\nA memo labeled “Memo AU 0904-4” (indicating that it is one of four \\nmemos written on August 9, 2004) and including references to data from \\ninterviews collected on four occasions (October 23, 2001; September 10, \\n2003; November 12, 2001; and November 26, 2001) is footnoted in the \\nfinal write-up of the theory as “Field Interview O 2301-2, S 1003, N 1201, \\nN 2601; Memo AU 0904-4.”\\nWhile such evidentiary detail is not relevant or required as part of \\nthe GT process, it may be important—and expected—to satisfy thesis \\nadvisers or journal reviewers accustomed to full data displays. Initiat­\\ning a similar labeling scheme for all data and memos at the outset of a \\nstudy offers the analyst an opportunity to keep the data-memo connec­\\ntions clear as the study progresses as well as offering an audit trail \\noption for later thesis defense or publication.\\nl\\nl\\nTHE ITERATIVE NATURE OF DATA ANALYSIS IN \\nGROUNDED THEORY\\nEarlier in this chapter we make reference to Glaser’s (1998) description \\nof GT as “subsequent, sequential, simultaneous, serendipitous and \\nscheduled” (p. 15). Glaser’s description is an apt one, and our chal­\\nlenge in writing this chapter has been to attempt to relate linearly a \\nprocess that is anything but linear! Glaser (1978) describes the impor­\\ntance of personal pacing, creativity, and cycling in doing a GT study. \\nThe iterative nature of GT, coupling a rigorous and systematic analytic \\napproach with an open and flexible embrace of creative insights \\nthrough preconscious processing—wherever and whenever they \\nemerge—requires the grounded theorist to be comfortable with both \\nchaos and control:\\nThe chaos is in tolerating the uncertainty and subsequent regres­\\nsion of not knowing in advance and of remaining open to what \\nemerges through the diligent, controlled, often tedious applica­\\ntion of the method’s synchronous and iterative processes of \\n­\\nline-by-line coding, constant comparison for interchangeability \\n\\n\\nChapter 6    Analyzing Your Data\\n95\\nof indicators, and theoretical sampling for core emergence and \\ntheoretical saturation. This discipline is simultaneously comple­\\nmented by requiring the theorist to remain open to the innate \\ncreativity in preconscious processing of conceptual ideation and \\ntheoretical integration. (Holton, 2007a, p. 273)\\nFortunately, GT has built-in motivational phases that carry the theorist \\nonward with energy and the excitement of discovery. Eureka moments \\n(Glaser, 1998) that emerge from preconscious processing of a study in \\nprogress are an important motivator. As Breckenridge (2013) describes,\\nTheoretical realizations sometimes take a while to emerge. Glaser \\nhas called this preconscious processing. Although I was not con­\\nvinced about this idea at first, reflecting on my own experience, \\nI can remember clearly those occasions where I had given up only \\nto find that clarity of thought emerged when I least expected it: as \\nI was cooking, out walking, on the train and even waking in the \\nmiddle of the night to scribble down memos in the dark. (p. 11)\\nSuch experiences are commonly shared among grounded theorists! \\nYet, those new to GT often seek specific procedures or templates for \\nanalyzing their data. There is no one correct way to code and memo. The \\nchoice is a personal one that will evolve as your GT skills develop. Some \\nexperienced grounded theorists have shared their approaches (Brecken­\\nridge, 2013; Holton, 2007a; Scott, 2009). You may find these interesting \\nand possibly helpful in developing your own approach. The best advice \\nis to keep it simple but organized and do whatever you can to stimulate \\nhand-eye-brain cognitive and preconscious processing through manual \\ncoding, memoing, and sorting procedures. And finally, the grounded \\ntheorist must be able to trust in the process every step along the way \\nand to trust that following the GT methodology will yield a theory that \\nis rich in explanatory power and relevant to the area under study.\\n\\t\\nTO SOFTWARE OR NOT TO SOFTWARE\\t\\nl\\nComputer-assisted data analysis is now commonplace in the research \\nworld and novice grounded theorists are often tempted to adopt soft­\\nware for data analysis. Bringer, Johnston, and Brackenridge (2004) sug­\\ngested that QSR*NVIVO is compatible with doing GT; however, in doing \\nso, they conflate GT with qualitative research methodology for which \\ndetailed transcription of interview data is the norm. While software pro­\\ngrams are excellent for searching and sorting through such data, they \\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n96\\ntend to do so in a structured linear and hierarchical manner that is more \\nappropriate to what Glaser would refer to as the full coverage nature of \\nqualitative data analysis. GT quickly leaves the data behind through \\nconstant comparison to conceptualization, a process that relies on lateral \\nthinking (de Bono, 1968) through the conceptual sorting of ideas and \\ncreative realizations through preconscious processing.\\nGT is about the sorting of theoretical ideas captured in memos. The \\ndanger in using qualitative software applications is that the analyst \\nslips into mechanically coding large volumes of data and sorting codes \\nwithout engaging systematically in constantly comparing incidents \\nwithin the data and memoing conceptual ideas that, over time, surface \\nthe subtleties in realizations and relationships between and among \\nincidents and emergent concepts. As Breckenridge (2013) explains,\\nIt was easy to code the data—too easy perhaps—and I quickly \\namassed a long list of codes with no idea how they might be con­\\nnected to one another. Sitting at the computer “dragging and \\ndropping” sections of data became automatic and unthinking. \\nI was simply organizing my data into a fractured list, without \\ncomparing incident to incident or stopping to write memos. \\n(pp. 8–9)\\nThe result of such mindless coding may, at best, lead to a drift into \\nconceptual description without the capacity to generate a conceptually \\ndense and fully integrated grounded theory.\\nA further danger to employing software for data analysis is that the \\nanalyst may be lured into indiscriminate data collection, collecting \\nmore data than is necessary simply because it is there. Practically \\nspeaking, employing qualitative software entails a significant learning \\ncurve; it is not only time consuming but holds the potential for stifling \\nanalytic creativity. Both authors of this book, who have been trained \\nvery differently, find NVIVO software useful for archiving coded quali­\\ntative data and for linking memos to incidents in the data and linking \\nmemos as they cumulatively build ideas. This archiving capacity is \\nparticularly useful in later writing up for publication.\\nIN SUMMARY\\nWhen doing classic GT, you have to go beyond descriptive analysis and \\nreach conceptual analysis. All data—whether qualitative, quantitative, \\n\\n\\nChapter 6    Analyzing Your Data\\n97\\nor both—must be constantly compared and analyzed as they are col­\\nlected through iterative cycles and memoing effected alongside this \\nprocess, as ideas emerge. To help analysis, various steps in coding of \\ndata are applied: Open coding of all data is done until core category \\nemergence, then selective coding around the core category, and, finally, \\ntheoretical coding to highlight causal relationships between concepts. \\nMemoing plays a key role in the analytic process. Memoing begins as \\nsoon as the first data are coded and continues throughout the process to \\nthe final write-up of the theory. Despite its popularity, grounded theo­\\nrists are cautious in their use of analytic software, reserving its use for \\narchiving (and later retrieval) of manually coded and sorted data.\\nTEST YOUR KNOWLEDGE\\nIn the following questions one or several answers may be correct.\\n1.\\t In classic GT\\na.\\t You first have to collect all data and then analyze them\\nb.\\t You analyze your data as you collect them\\nc.\\t You must decide which data to collect next before you start \\nanalyzing collected data\\nd.\\t You should constantly compare all collected data\\ne.\\t None of the above\\n\\t\\n2.\\t In classic GT, coding is\\na.\\t Open, axial, and selective\\nb.\\t Open, selective, and theoretical\\nc.\\t Substantive and theoretical\\nd.\\t None of the above\\n\\t\\n3.\\t The core category\\na.\\t Keeps recurring in the data\\nb.\\t Has grab\\nc.\\t Emerges while you are open coding\\nd.\\t Must emerge before you start selectively coding\\ne.\\t None of the above\\n\\t\\n4.\\t While doing classic GT with qualitative data,\\na.\\t You must code every word\\nb.\\t You must code every sentence\\nc.\\t You must use software to help you code\\nd.\\t None of the above\\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n98\\nFURTHER READING\\nGlaser, B. G. (1965). The constant comparative method of qualitative analysis. \\nSocial Problems, 12, 436–445.\\nGlaser, B. G. (1998). Generating concepts. In Doing grounded theory: Issues and \\ndiscussions (Chapter 9). Mill Valley, CA: Sociology Press.\\nGlaser, B. G. (2001). Constant comparison and core confusion. In The grounded \\ntheory perspective: Conceptualization contrasted with description (Chapter 13). \\nMill Valley, CA: Sociology Press.\\nGlaser, B. G. (2014). Memoing: A vital grounded theory procedure. Mill Valley, CA: \\nSociology Press.\\nHolton, J. A. (2007a). The coding process and its challenges. In A. Bryant & \\nC. Charmaz (Eds.), The SAGE handbook of grounded theory (pp. 265–289). \\nLondon: Sage.\\nENDNOTES\\n1.\\t The \\nCreative \\nCommons-Attribution-NonCommercial-ShareAlike \\nLicense 4.0 International applies to all works published by the \\nInternational Journal of Qualitative Methods. Copyright for articles \\npublished in the International Journal of Qualitative Methods remains \\nwith the first author.\\n2.\\t Glaser has a tendency to use the terms code, concept, and category inter­\\nchangeably. Here we wish to clarify our use of the terms. By code, we \\nrefer to the labels attached to incidents during the substantive coding \\nstage of analysis. Codes are often descriptive, particularly in the initial \\nstages of data analysis. By concept, we refer to the resultant naming of \\na higher level abstraction of a conceptual idea as indicated through the \\ncumulative comparison of codes. Similarly, a category denotes the fur­\\nther abstraction of individual concepts as a latent pattern. As an exam­\\nple, we might note the following codes in data: being myself, being heard, \\nauthentic connecting, quiet voices, comfortable with each other, trusting my \\nown voice, being genuine. These, along with other codes, are later incorpo­\\nrated into the concept authenticity, which later emerges as a property of \\nthe core category of rehumanizing.\\n3.\\t We differentiate here between properties as latent, qualitative character­\\nistics (e.g., authenticity, trust, healing) and dimensions as the measurable \\ncomponent of a concept (e.g., space, time, size).\\n4.\\t It is the general implications of a core category that lead to formal theory \\ndevelopment (Glaser, 2007, p. 4)\\n\\n\\n99\\nAfter studying this chapter, you will be able to:\\n• articulate the role of theoretical sampling, theoretical saturation, and \\ntheoretical coding in shaping an emerging grounded theory\\n• utilize theoretical sampling and theoretical saturation to build emergent \\nconcepts and categories\\n• utilize theoretical coding to integrate a grounded theory\\nO\\nur purpose in this chapter is to explain key techniques for giv­\\ning shape to the theory as it emerges and develops. As we have \\nset out in earlier chapters, GT is an elegant and intricate \\napproach to capturing and conceptualizing latent patterns in social set­\\ntings. While too complex to be approached in a linear fashion, GT is \\nreadily embraced by our innate capacity for pattern recognition \\nthrough preconscious processing (Glaser, 1998, p. 50). In this chapter, \\nwe will explore the techniques of theoretical sampling, theoretical satura­\\ntion, and theoretical coding, which enable the analyst to explicitly build \\nand shape the emerging theory and to present it as an integrated \\n­\\nconceptual explanation.\\n7\\nShaping the Theory\\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n100\\nl\\nl\\nTHEORETICAL SAMPLING\\nTheoretical sampling is the process whereby the analyst decides both \\nwhat data to collect to develop emerging concepts and where those \\ndata may be available. The purpose of theoretical sampling is to \\ndiscover categories, properties, and interrelationships suggestive of a \\ntheoretical whole. As such, theoretical sampling is theory-driven \\nsampling, controlled by the emerging theory and not determined in \\nadvance based on a preconceived theoretical framework. Glaser and \\nStrauss (1971) note that\\nno one kind of data on a category nor any single technique for \\ndata collection is necessarily appropriate. Different kinds of \\ndata give the analyst different views or vantage points from \\nwhich to understand a category and to develop its properties. \\n(pp. 183–184)\\nGlaser and Strauss (1967) speak of these different views or van­\\ntage points as slices of data and suggest that “theory generated from \\njust one kind of data never fits, or works as well, as theory generated \\nfrom diverse slices of data on the same category” (p. 68). Gibson and \\nHartman (2014) distinguish between data slicing and theoretical \\nsampling:\\nData slicing, in keeping with Glaser and Strauss (1971), refers \\nto the use of an individual comparison between different \\nsubstantive groups or different social units for the purposes \\nof developing theory, usually in the form of refining categories. \\nTheoretical sampling refers to a process of combining several \\ndifferent slices of data for the purpose of developing theory. \\n(p. 128)\\nTheoretical sampling of different slices, in parallel with constant \\ncomparative analysis, assists conceptual elaboration with the overall \\nintegration of the theory more likely to emerge organically. By employing \\na range of data slices, the analyst discovers similarities and differences \\nthat bring forth the variability of concepts and their properties in \\nvarious contexts and under varying conditions, thereby enhancing the \\nconceptual density of the emerging theory. The comparisons are \\nideational, not population-specific. As such, “there are no limits to the \\ntechniques of data collection, the way they are used, or the types of \\ndata required” (Glaser & Strauss, 1971, p. 184), provided that GT’s \\nfoundational principle of emergence is respected.\\n\\n\\n101\\nChapter 7    Shaping the Theory\\nAs an illustration of theoretical sampling of different slices of data in \\ndifferent substantive groups and in different social units, you may refer to \\nWalsh, Kefi, and Baskerville (2010). Sampling was first done in a corporate \\nsetting, within different firms (different substantive groups) and then in a \\nsocietal setting (different social unit).\\nDuring the open coding stage, sampling proceeds in all directions \\nthat seem relevant and add to the emerging concepts. Once a potential \\ncore category has emerged, theoretical sampling becomes selective as the \\nanalyst focuses on developing central issues related to the emerging \\ntheory “guided by the logic of the emerging analytic framework” (Glaser \\n& Strauss, 1965a, p. 289). There is a deductive element to theoretical \\nsampling as the analyst memos by probing and questioning emergent \\ntheoretical ideas and deciding where to go next to get the data to fill gaps \\nand elaborate on emerging ideas, but this deduction is in service to \\nfurther inductive analysis as the theory develops. The analyst may go \\nback to existing data to sample for emerging ideas as well as generating \\nnew data from new slices of data. This deductive (conceptual) elaborating \\nis essential for the systematic generation of theoretical possibilities, as \\npropositions or hypotheses, to build and explain the emerging theory. \\nFresh data from new populations and samples assist in extending the \\nscope and coverage of the emerging theory, thus driving toward \\nsubstantively formalized theory and a less idiographic, contextualized \\nperspective. The possibilities for multiple comparisons through theoretical \\nsampling are endless and so the grounded theorist selects comparison \\ngroups according to theoretical criteria. “Groups are, from a theoretical \\nviewpoint, clusters of variables, which are not readily apparent, and \\nmany of which are to be discovered by comparison” (Glaser, 1978, p. 42).\\nThe criteria of theoretical sampling are designed to be applied \\nin the on-going joint collection and analysis of data associated \\nwith the generation of theory. Therefore they are continually \\ntailored to fit the data and are applied judiciously at the right \\npoint and moment in the analysis. The analyst can continually \\nadjust his control of data collection to ensure the data’s relevance \\nto the impersonal criteria of the emerging theory. (Glaser & \\nStrauss, 1967, p. 48)\\nSampling strategies also control the scope of the theory by con­\\nsciously choosing to limit or extend the population sampled. Glaser \\nand Strauss (1967) suggest that at the outset, the process is to establish \\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n102\\nthe basic categories and their properties by minimizing differences (i.e., \\nmaximizing similarities) in theoretically sampling comparative groups \\nso as to verify the relevance of emerging categories, generate their \\nproperties, and identify the conditions and fundamental differences \\nunder which they vary. Once this basic work is achieved, the analyst \\nturns to maximizing differences among comparison groups as a means \\nof identifying the scope of the emerging theory and stimulating the \\ntheoretical density so as to saturate conceptual properties once the \\nbasic conceptual framework has emerged (p. 58).\\nSkill in theoretical sampling develops over time. It requires the ability \\nto be both analytic and flexible. There is no precise formula to be applied \\nand no limitations on the techniques for data collection; rather, sampling \\nrequires sensitive judgments as to the best places to go next to find the data \\nneeded to answer key questions as they emerge in advancing and satisfy­\\ning gaps in the emerging theory. Comparing data as it is collected pro­\\nduces a “cumulatively intensive theoretical sensitivity” (Glaser, 1978, p. 36) as \\nthe analyst gradually sees more and more of theoretical relevance in the \\ndata, creating more abstract levels of theoretical connections as the theory \\nis gradually built up inductively through progressive stages of analysis.\\nJørgensen, Dahl, Pedersen, and Lomborg (2013) explain the theoretical \\nsampling approach in their grounded theory of coping with COPD (chronic \\nobstructive pulmonary disorder):\\nThe aim was to secure data that could capture the multidimensionality \\nof coping . . . Data collection and analysis guided the theoretical sampling \\nphase. For example, the coping behavior of the first included participant \\nwith moderate COPD persistently deviated from the behavior of the \\nfollowing five included participants with severe or very severe COPD. \\nSeveral possible reasons for this deviation were looked for in the data. In \\nthe ensuing data collection and analysis process, this heterogeneity led \\nus to obtain an equal distribution of male and female participants with \\ndiffering degrees of COPD. Additionally, data revealed that spouses con­\\nstituted a significant extra-personal and interpersonal resource, playing \\na vital role in how participants handled a life influenced by breathless­\\nness. Therefore, spouses were invited to participate in the following \\ninterviews with the intention of enriching our understanding of the \\nparticipants’ coping strategies. These discoveries on equal terms with \\nother preliminary findings resulted in an ongoing alteration of guides \\nfor interview, video observation, and selection of video sequences. (p. 5)\\n\\n\\nChapter 7    Shaping the Theory\\n103\\nTheoretical sampling generally means that the researcher may well \\nspend a longer time in the field but will be collecting fewer data as the \\ndirection of data collection is more intentional and focused, thereby \\ngiving the researcher momentum, purpose, and confidence in the cat­\\negories that are emerging from the data (Glaser & Strauss, 1967, p. 76). \\nThe adequacy of the analyst’s theoretical sampling will vary with the \\nlevel of theory proposed; the more formal the theory is, the greater the \\nscope and range of data sampled will be.\\nPerhaps You Are Wondering . . . \\nEarly in the research process, theoretical sampling may involve asking \\nan interviewee about a concept that has emerged in other data that \\nyou have collected. While you don’t want to force the idea on \\ninterviewees, there is no harm in asking a simple question about their \\nexperience of the concept (e.g., “Have you ever experienced \\nharassment in your workplace?” “Are there any taboo topics in this \\ndepartment?” etc.). Your question may spark a response and more \\nindicators for analysis, or it may not—which is also data!\\n\\t\\nTHEORETICAL SATURATION\\t\\nl\\nTheoretical saturation refers to the constant comparison of conceptual \\nindicators in the data to the point where additional indicators yield no \\nfurther theoretical specification or elaboration. Glaser (2001) describes \\ntheoretical saturation as “intense property development” (p. 191), sug­\\ngesting that this stage of theory development is “necessary to lift the \\ntheory above description and enable its integration through theoretical \\npropositions (hypotheses) as abstract conceptual theory” (Holton, \\n2007a, p. 281). By theoretically sampling to maximize the diversity in \\nboth comparative groups and varieties of data, the analyst will enhance \\nsaturation of both the properties and dimensions of concepts. Probing \\nquestions, during data collection and analysis, will assist the densifica­\\ntion of conceptual properties (Glaser, 1978).\\nAs the theory develops, the theorist learns which categories require \\nthe most and the least saturation and which ones can be dropped; thus, \\nthe emerging theory generates its own selectivity (Glaser & Strauss, \\n1967). Core categories should be saturated as completely as possible. \\nFailure to saturate the core and related concepts will leave the theorist \\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n104\\nwith a thin theory of loosely connected concepts and little explanatory \\npower. However, continuing to collect data for concepts that have satu­\\nrated or for those that are not related in some way to the emerging \\ntheory’s core category is a waste of researcher time and resources. “The \\ncriteria for determining saturation then are a combination of the empir­\\nical limits of the data, the integration and density of the theory and the \\nanalyst’s theoretical sensitivity” (Glaser & Strauss, 1967, p. 62).\\nPerhaps You Are Wondering . . . \\nDo I need to saturate my concepts in one location/­\\ncomparative group before I theoretically sample for those \\nconcepts in another location/comparative group?\\nIt is difficult to know when a concept will be theoretically saturated. \\nIt is possible that the concept may saturate in one comparative group \\nand that you may not identify any additional properties or dimensions \\nof that concept in subsequent theoretical sampling of other groups. \\nHowever, theoretical sampling in a number of comparative groups \\nmay well capture new properties or dimensions or possibly even new \\nconcepts with relevance to your emerging theory. As the theorist, it \\nis your decision as to when you feel that concepts have been satu­\\nrated and there are no obvious gaps in the emerging theory that need \\nto be followed up with additional sampling. It is a judgment call based \\non the explanatory power, or explanatory gaps, of the emerging \\ntheory.\\nl\\nl\\nTHEORETICAL CODING\\nOnce the core category has emerged, having identified those concepts \\nthat bear some relationship to the core and having theoretically \\nsampled to saturate the core and related concepts, the analyst is ready \\nto begin the process of theoretical coding as the final shaping of the \\nemergent theory. It is at this point, however, that some theorists fail to \\nachieve the full explanatory power of classic GT by not pursuing its \\nintegration through theoretical coding. Instead, they default to simply \\noffering conceptual description rather than pressing on to this final \\nstage of GT development. As Bacharach (1989) reminds us, “while \\ndescriptions may be the source material of theories, they are not \\nthemselves theoretical statements” (p. 497) and “just as a collection of \\n\\n\\nChapter 7    Shaping the Theory\\n105\\nwords does not make a sentence, a collection of constructs and \\nvariables does not necessarily make a theory” (p. 496).\\nThe integration of a grounded theory through theoretical coding \\ncaptures our innate ability for perceiving latent patterns that organize \\nand integrate relevant related concepts to provide a full explanation of \\nthe emergent theory’s resolving or managing of a main concern. Glaser \\nand Strauss (1971) speak of “the emergence of a natural integration” \\n\\n(p. 190). Theoretical codes conceptualize (i.e., model) how substantive \\ncodes may relate to each other as propositions or hypotheses to be \\nintegrated into a theory.\\nTheoretical coding occurs only after all substantive coding is com­\\npleted and the core category and related concepts have been theoreti­\\ncally saturated: “Theoretical codes give integrative scope, broad \\npictures and a new perspective” (Glaser, 1978, p. 72) and help the \\nanalyst maintain a conceptual level in writing about concepts and their \\ninterrelations. They “conceptually punctuate the emerging theory with \\nan organizing pattern.”1 As discussed in Chapter 6, the core category of \\na grounded theory is always presented (or modeled) as some form of \\ntheoretical code—a process, typology, range, or other. It is during the \\nhand sorting of memos that ideas begin to emerge about the relation­\\nships between the core category and related concepts. The analyst \\nbegins to sense some sort of overall pattern (or model) that organizes \\nthe emerging theory through an appropriate theoretical code (or codes) \\nthat explains the relationships, thereby giving the emerging theory its \\noverall shape.\\nPerhaps You Are Wondering . . . \\nCan I use extant theories in theoretical coding?\\nIt is important here to differentiate between using the substance of \\nextant theories and using how those theories are modeled. Using the \\nsubstance of an extant theory may lead the analyst to “borrow” \\nconcepts from the extant theory and introduce (i.e., potentially force) \\nthem onto the emerging grounded theory. On the other hand, being \\ntheoretically sensitive to how different extant theories are modeled \\nmay help the analyst to see how the emerging grounded theory might \\nbe integrated and modeled in a similar way. Playing with this model in \\nrelation to the emerging grounded theory is fine. It is only when the \\n(Continued)\\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n106\\nWhile the emergence of a theoretical code that works to integrate \\none’s theory is often a flash of realization—what Glaser (1978) has \\ntermed the “Eureka moment”—developing one’s theoretical sensitivity \\nto a wide range of theoretical codes enhances the ability to see their \\nemergent fit to a developing theory. Reading widely opens a researcher \\nto serendipitous discovery of new theoretical codes from other \\ndisciplines, offering novel frameworks and models for integrating \\ngrounded theories. The essential requirement is that the code must \\nemerge from theoretical sorting of memos; it is not preselected and \\nforced on the theory (Glaser, 1992).\\nGlaser (2005a) advocates reading theory in other substantive areas \\nand disciplines as a means of enhancing one’s theoretical sensitivity \\nby building a repertoire of theoretical codes. Reading theory in other \\nfields enhances the theorist’s pattern recognition abilities, opening the \\npossibility of serendipitous discovery of new theoretical codes with \\nemergent fit. Glaser (1978, 1998, 2005a) offers many options for \\ntheoretically coding a grounded theory. The critical criterion is that the \\nappropriate theoretical code(s) must emerge through sorting of \\nmemos; they are not assumed or selected in advance (see, e.g., Walsh \\net al., 2013, who use translation theory [Callon, 1986; Latour, 1987], a \\nsociological theory, to help make sense of their data and propose the \\ntranslated strategic alignment model in rupture with existing literature \\non the subject of strategic alignment in the field of information \\nsystems).\\nanalyst forces the emerging theory to fit a specific theoretical model \\nthat the full integrative potential of GT may be compromised.\\nThe researcher who does not reach outside extant theory for \\ntheoretical coding possibilities runs the risk of producing adequate \\nbut rather mundane conceptual theory. Such theory makes a limited \\ncontribution to knowledge and, although certainly preferable to \\npurely conjectured theory, it will lack the impact that the creative \\nemergence of a novel or nontraditional theoretical code may offer. \\nNo matter how intellectually seductive, fashionable, or discipline-\\ndictated a theoretical code may be, to cross the line from theoretical \\nexploration to forced integration with a preconceived theoretical \\nmodel undermines the generative nature of grounded theory (Holton, \\n2007a, p. 283).\\n(Continued)\\n\\n\\nChapter 7    Shaping the Theory\\n107\\nGlaser (1978) advocates the implicitness of theoretical coding: \\n“One talks substantively and thinks theoretically of the relationship \\nbetween codes” (p. 72). In writing the theory, however, theoretical \\ncodes may be expressed implicitly or explicitly. For example, Sandgren \\net al. (2007) refer explicitly to the theoretical code of typology in \\norganizing their theory of doing good care, whereas Sandgren et al. \\n(2006) never directly refer to a specific theoretical code in their theory \\nof striving for emotional survival but implicitly suggest the theoretical \\ncode of strategies. Holton (2007b) describes her grounded theory of \\nfluctuating support networks as a basic social structural process with its \\ncore category of rehumanizing as a basic social psychological process.\\nPerhaps You Are Wondering . . . \\nIs grounded theory always a basic social process?\\nThe concept of the basic social process is a generic theoretical \\nconstruct in sociology explaining fundamental patterns in the \\norganization of social behavior as it occurs over time, involving a change \\nover time with discernible breaking points or points of transition or \\npassage from one stage to another (Glaser, 1978). Y\\nou will read \\naccounts of GT methodology as requiring identification of basic social \\nprocesses (Bryant & Lasky, 2007; Starks & Trinidad, 2007), but the basic \\nsocial process is just one theoretical code that may be appropriate to \\nthe theoretical integration of a grounded theory. “All BSPs [basic social \\nprocesses] are core variables, but not all core variables are BSPs” \\n(Glaser, 1978, p. 96). As one example, the seminal work of Glaser and \\nStrauss (1965a, 1968) employed a typology of awareness contexts and \\na trajectory of the dying process in hospitalized terminal care.\\nThe emergent nature of theoretical coding is achieved through \\nhand sorting of accumulated theoretical (analytic) memos:\\nFacilitating the emergence of relevant theoretical codes requires \\nclose attention to the ideas memoed, submersion at the concep­\\ntual level, a balance of logic and creativity, openness to the \\nunexpected, and confidence in following what emerges regard­\\nless of how counter-rational it may seem to extant theoretical \\nperspectives. (Holton, 2007a, p. 284)\\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n108\\nAs such, theoretical coding reminds us that the essential principle of \\nearned relevance in GT is not confined to conceptual emergence but is equally \\nimportant in theoretical integration through the emergence of relevant \\ntheoretical codes.\\nPerhaps You Are Wondering . . . \\nIs axial coding the same as theoretical coding?\\nWhile theoretical coding is not mentioned per se in Glaser and \\nStrauss (1967), the coding paradigm of conditions, contexts, actions, \\ninteractions, and consequences that underlie Strauss and Corbin’s \\n(1990, 1998) axial coding approach is implicit in the Awareness study \\n(Glaser & Strauss, 1965a), in which Glaser and Strauss explain how, \\nearly on, they “systematically worked out the concepts (and types) of \\ndeath expectations and awareness contexts, and the paradigm for the \\nstudy of awareness contexts . . . [and how] . . . a concern with death \\nexpectations and awareness guided the preliminary data collection; \\nthe systematic formulation of these concepts and of the paradigm \\nguid[ing] further data collection and the ensuing analysis” (p. 287). \\nStrauss (with Corbin) would go on to incorporate the use of this \\ncoding paradigm—the conditional matrix—as an essential step in \\ntheir process, referring to it as axial coding.\\nGlaser (1978) would extend the scope of theoretical coding to \\ninclude a wide range of potential models for theory integration by \\nelaborating on 18 different coding families (pp. 74–82) and insisted \\nthat the appropriate model(s) must emerge from theoretical sorting. \\nGlaser (1992) challenged Strauss and Corbin’s (1990) insistence on \\nthe use of axial coding as forcing a preconceived theoretical code, \\nundermining the emergent potential of GT and reducing it to \\nconceptual description. Later, he would further elaborate and extend \\nthe range of theoretical coding possibilities (Glaser, 2005). To sum up, \\naxial coding relies solely on one theoretical code (paradigm) applied \\nto the data for theoretical integration, whereas classic GT remains \\nopen to the emergence (through theoretical sorting) of a wide range \\nof potential theoretical codes around which to integrate and model \\nan emerging grounded theory.\\n\\n\\nChapter 7    Shaping the Theory\\n109\\nThis insistence on the emergence of theoretical codes through hand \\nsorting of memos is a hallmark of classic GT not espoused by those \\nwho follow other approaches. For example, Charmaz (2014) advocates \\ntheoretical codes for clarifying and sharpening analysis but implies \\nthat the basic social process is the theoretical code underlying all \\ngrounded theories. While the basic social process is certainly a domi­\\nnant theoretical code in grounded theories, such preconception effec­\\ntively precludes the integrative potential for other emergent theoretical \\ncodes. At the same time, it is worth noting that Charmaz (2014) appears \\nambivalent regarding the need for a core category, preferring the more \\nambiguous notion of “focused coding.” This is perhaps consistent with \\nher constructivist stance and its notion of multiple social realities. It is \\nnot, however, consistent with classic GT.\\nGlaser (1998) encourages the grounded theorist to develop a reper­\\ntoire of as many theoretical codes as possible as a means of offering \\ntheories that transcend the somewhat predictable offerings of extant \\ntheories. Skill in drawing upon such a repertoire also offers some \\nassurance against slipping into what Glaser (2005) terms pet or grabby \\ntheoretical codes. Here the theorist becomes so enthralled with the \\ndiscovery of an enticing theoretical code—one with imagery that \\ngrabs—that he or she begins to see it everywhere! As with substantive \\ncoding, theoretical codes must earn their relevance.\\nTheoretical Sorting\\nHand sorting of memos provides theoretical order and integration of \\nideas and requires the theorist to theoretically discriminate as to where \\neach idea fits in the emerging theory. The goal is to find the emergent fit \\nof all ideas so that everything fits somewhere with parsimony and scope \\nand with no relevant concepts omitted. The physical act of hand sorting \\nmemos further facilitates the preconscious processing of matured ideas \\nand guides the organization and integration of the overall theory. Many \\nresearchers use sticky notes to help with their theoretical sorting. Piko \\n(2014) describes how she used sticky notes on a large window to help her \\nvisualize the emergent integration of her grounded theory. Over a period \\nof weeks she sorted, compared, and resorted her notes, moving them \\naround on the window and making additional notes until she achieved \\nfull integration. Some colleagues provided us with photographs of the \\ntheoretical sorting of one of their grounded theories (see Figure 7.1).\\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n110\\nFigure 7.1    \\u0007\\nTheoretical Sorting (Sautrey & Mourmant, 2015. Research \\nin progress)\\n\\n\\nChapter 7    Shaping the Theory\\n111\\nOne Student’s Experience\\nTo begin, I printed out my memos, sat on the floor, and put them in a \\npile in front of me.\\nI placed like with like and grouped together memos with apparent \\nlinks.  \\nAs I picked each new memo off the pile, I compared it to the \\nothers, asking, “Where does this fit?” and wrote further memos \\nexplicating these relationships and theoretical connections. A lot of my \\nmemos contained more than one idea relating to different concepts, \\nand so I used scissors to cut them into relevant sections, sometimes \\ninto individual paragraphs or even sentences. This increased the amount \\nof paper I had to manage. \\n As I sorted memos, I wrote further memos \\nabout the relationships between ideas and integrated these into my sort. \\nBy the time my final theory was written, my memo bank had expanded \\nto over 400 memos. Although other researchers have been able to use \\nCAQDAS during this process, I struggled to see how this messy three-\\ndimensional (3D) process could translate onto a two-dimensional (2D) \\ncomputer screen. \\nAs I considered the spread of papers in front of me and \\nshifted the paper about, I was excited by the potential for my ideas to \\ncome together in any number of ways. \\nThis preliminary sort also enabled \\nme to weed out those immature memos that did not relate significantly \\nto the core category. I retained these in a separate file, meaning that I \\ncould go back to retrieve them if their relevance later emerged.\\nFinally, after 2 months of sorting and trying to find the best theoretical \\ncode, I settled on a basic social process (BSP). The BSP is the most \\ncommon theoretical code in CGT and structures the categories into \\ndistinct stages. (For further reading on the BSP, see Glaser & Holton, \\n2005.) As I tried to write up my theory, however, I became more and \\nmore confused. Seeing the lack of integration on paper confirmed my \\nfear that I had been guilty of forcing an inappropriate theoretical code \\non my data. In my mental fatigue and confusion, I had resorted to settling \\non a BSP, despite it not having best fit. I knew that I still had a pile of \\nmemos that I had not been able to integrate into my chosen structure, \\nbut had chosen them because I was fearful that the best theoretical \\ncode would never emerge. Buoyed by Glaser’s advice that often “wrong \\ntracks lead to right ways,” I began sorting again. I explored different \\ntheoretical codes by rereading Theoretical Sensitivity and Doing Grounded \\nTheory. I looked again at the theoretical coding families that I had found \\ntoo difficult to understand and looked for examples of them in other \\nCGT studies. Interestingly, most of the theories I read were examples \\n(Continued)\\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n112\\nSome theorists find that diagramming helps them to visualize \\npotential ways of theoretically modeling their emerging theories. \\n­\\nGlaser (2014) quotes one experienced grounded theorist:\\nI draw figures of the different options of theoretical codes to see \\nhow my concepts relate to each other. Some of my drawings may \\nnot be so grounded all the time, but it helps me to trigger my \\ncreativity so when I go back to my memos and write more memos \\non memos I see how everything finally fits together. (pp. 16–17)\\nAt the same time, however, Glaser (1998) cautions against using \\nsuch models as a short-form approach to presenting a grounded theory. \\nof BSPs, which is perhaps part of the reason behind my subconscious \\nforcing of a BSP onto my data. Glaser has warned against the popularity \\nof the BSP, emphasizing that not all core categories will be a staged \\nprocess occurring over time.  \\nWhile I had not intended to go looking for \\na process per se, as a novice researcher, I was perhaps swayed towards \\nthe simplicity and ‘grab’ of a process. Having increased my theoretical \\nsensitivity, I was able to see new possibilities. In particular, Glaser’s \\nbinary theoretical code resonated with my participants’ main concern: \\nbeing person driven in a service-driven organization. Glaser provided \\nthe following definition:\\n“This coding has to do with compliance to institutional or \\n­\\nnormal roles—binary retreat; or with non-compliance with role \\n­\\nrequirements—binary deconstruction. Thus, for example, when a \\nprofessional meets with a client, to what degree do they interact \\nin their roles—binary retreat; or drop the roles and just be their \\nhuman selves—binary deconstruction” (Glaser, 2005a, p. 22).\\nIt was exciting to read this definition for the first time, and it \\nsparked an intense period of sorting and a flood of memos as I realized \\nthat the binary theoretical code had significant fit and relevance to \\nmy data. By prioritizing client realities, practitioners deconstructed \\nservice ideals, making service delivery individualized and driven by \\nthe person. Conversely, the equilibrium point could be shifted in the \\nopposite direction as practitioners reinstated service ideals over \\nclient realities, retreating within their roles and service boundaries. \\n\\nI thus conceptualized Revisioning as being dual directional, whereby \\npractitioners made adaptations to their practice in order to go one way \\nor another. (Breckenridge, 2013, pp. 14–15)\\n(Continued)\\n\\n\\nChapter 7    Shaping the Theory\\n113\\nHe views a diagram as a cop-out, leaving the hard work of elaborating \\nthe theory to the reader who may not grasp its rich explanatory \\npower:\\nA diagram is but a precursor to a write up of [the theory]. Thus, \\nit is an aide to comprehending the meanings of the written \\ntheory. It is not a theory in and of itself . . . The meaning has to \\nbe written (p. 169)\\nAs an illustration of this, the diagrams presented in Walsh and col­\\nleagues (2013) started to be drawn in 2008. However, it took close to \\n\\n5 years for the theory to fully emerge and be elaborated in a manner \\nthat could be understood by, and useful to, other people. When doing \\nGT, one of the authors of the present book uses diagrams as memos, \\nand these diagrams keep being modified as new data are collected. She \\nfinds that they improve her theoretical sensitivity: They help in identi­\\nfying theoretical codes and the emergence of the theory. This is illus­\\ntrated in Appendix E.\\nIt is quite possible that in sorting, the theorist may see the possibil­\\nity of more than one theoretical code that might work to organize the \\nemerging theory. It is the theorist’s responsibility to select the code that \\nbest fits to integrate the relationship between the core category and \\nrelated concepts. One theoretical code always dominates, although \\nsub-core relationships may require additional theoretical codes to fully \\nexplain the complexity inherent in some grounded theories.\\nTheoretical Coding\\nWhile the core category of a grounded theory is always some type of the­\\noretical code, some theories may require more than one theoretical code \\nfor full integration. Holton (2006) illustrates her theory of rehumanizing \\nknowledge as a basic social process with three stages (finding and liken­\\ning, igniting passions, and mutual engagement) and a number of amplifying \\ncausal loops creating the energy that motivates and sustains the rehuman­\\nizing process.\\nAnalytic Rules\\nGlaser (1978) has suggested that failure to hand sort memos results in \\na thin, linear, and less than fully integrated theory. Sorting also leads to \\nmore memo writing as the theorist begins to see and record the relation­\\nships among concepts. This helps to further raise the conceptual level of \\n\\n\\nIgniting\\nPassions\\nFinding\\nLikening\\nTrust\\nAltruistic\\nAtmosphering\\nConnectedness\\nBonding\\nCommitment\\nExperimentation\\nLearning\\nEnergy\\nChallenge\\nConfidence\\nCreativity\\nMutual\\nEngagement\\nFigure 7.2    \\u0007\\nTheoretical Coding of the Rehumanizing Process (Source: Holton, 2006)\\n114\\n\\n\\nChapter 7    Shaping the Theory\\n115\\nthe theory and reduces the potential for the analyst to slip into descrip­\\ntion. Glaser (1978) has set out a number of analytic rules regarding the \\nprocesses of sorting, carrying forward, and integrating a grounded \\n­\\ntheory. The following is a summary of Glaser’s (1978, 2014) analytic rules:\\n\\t\\n1.\\t Starting to sort—Start anywhere. Sorting will force its own \\nbeginning, middle, and end for writing. The important thing \\nis to start. Trying conceptually to locate the first memos will \\nforce the analyst to start reasoning out the integration. Once \\nstarted, the analyst soon learns where ideas are likely to inte­\\ngrate best, and sorting becomes generative and fun.\\n\\t\\n2.\\t Core variable—Sort all other categories and properties only as \\nthey relate to the core category or BSP. This rule forces focus, \\nselectivity, and delimiting of the analysis. Theoretical coding \\nhelps the analyst decide on and figure out the meaning of the \\nrelation of a concept to the core variable. This theoretical code \\nshould be written and sorted into the appropriate pile with \\nthe substantive code.\\n\\t\\n3.\\t Promotion or demotion of core variables—This helps the GT \\nanalyst to explain some variation of a problem as it is pro­\\ncessed, thus achieving some theoretical coverage. The goal is \\nneither to cover all possible theoretical possibilities nor to \\nexplain all variation.\\n\\t\\n4.\\t Memoing—Once sorting on the core variable begins, the con­\\nstant comparisons are likely to generate many new ideas, espe­\\ncially on theoretical codes for integrating the theory. Stop sorting \\nand start memoing! Then, sort the memos into the integration.\\n\\t\\n5.\\t Carry forward—Analyst carries forward to subsequent sorts \\n(pages when writing) the use of each concept from the point of \\nits introduction into the theory. Requires only the use of the \\nconcept, not all illustration. The concept is illustrated only when \\nit is first introduced (to develop the imagery of its meaning).\\n\\t\\n6.\\t Integrative fit—All ideas must fit in somewhere in the outline, \\nor the integration must be changed or modified. This rule is \\nessential—if the analyst ignores this step, he will break out of \\nthe theory too soon, and necessary ideas and relations will not \\nbe used. This rule is based on the assumption that the social \\nworld is integrated and the job of the grounded theorist is to \\ndiscover it. If he cannot find the integration, he must resort and \\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n116\\nreintegrate his concepts to fit better. The analyst moves back \\nand forth between outline and ideas as he sorts, forcing under­\\nlying patterns, integrations, and multivariate relations between \\nthe concepts. This step is intensely generative, yielding many \\ntheoretical coding memos to be resorted into the outline.\\n\\t\\n7.\\t Sorting levels—Conceptual ordering of sorting is on at least \\ntwo levels, for example, the core category, then subcore cate­\\ngories and conceptual properties or dimensions. Or, for a \\nmonograph or a book, the ordering would be chapters and \\nthen sections in each chapter.\\n\\t\\n8.\\t Idea problems—Sorting forces the analyst to introduce an \\nidea in one place and then establish its carry forward when it \\nis necessary to use it again in other relations. When in doubt \\nabout a place to sort an idea, put it in that part of the outline \\nwhere the first possibility of its use occurs, with a note to \\nscrutinize it and pass it forward to the next possible place.\\n\\t\\n9.\\t Cutting off rules—Theoretical completeness implies theoreti­\\ncal coverage as far as the study can take the analyst. He must \\nexplain with the fewest possible concepts and with the great­\\nest possible scope as much variation as possible in the behav­\\nior and problem under study, sufficiently explaining the main \\nconcern and its resolution with concepts that fit, work, have \\nrelevance, and are saturated.\\n\\t\\n10.\\t Mechanics of sorting—Cut up memos to facilitate ease of sort­\\ning ideas. Keep notes as you are sorting, and sketch out mod­\\nels of potential theoretical codes for theory integration. Stop \\nsorting to memo. Even before intentional sorting begins, keep \\nfiles for memos arranged for some initial integration. Move \\nquickly when sorting. Avoid overthinking about any one sort \\nas the sort will change over time as the appropriate theoretical \\ncoding emerges.\\n\\t\\n11.\\t Theoretical pacing—Have a flexible but regular schedule with \\nperiods of uninterrupted sorting through memos to help in \\nbalancing the buildup of ideas preconsciously with the writing \\nof memos that capture preconscious processing. At the \\n­\\ntheoretical coding stage, the sorting and emergence of ideas is \\nat a deeper level as the theorist begins to see the emerging \\ntheory as an integrative whole. (Glaser, 1978, pp. 120–127; \\n2014, pp. 108–122)\\n\\n\\nChapter 7    Shaping the Theory\\n117\\nTable 7.1 summarizes our discussion of theoretical sampling, \\ntheoretical saturation, and theoretical coding as essential techniques \\nfor shaping the overall elaboration and integration of a grounded \\ntheory.\\nOne very effective way to test the integration and overall shape of \\na grounded theory is to draft a working paper. Start with a very brief \\nexplanation of the context and the main concern and how the core cat­\\negory works to process, manage, or resolve that issue or concern. Hand \\nsorting of memos leads to seeing how best to conceptually organize \\nthis initial write-up and assists in writing the theory both conceptually \\nand parsimoniously without reverting to detailed description.\\nTheoretical \\nsampling\\nTheory-driven process of data collection based on further \\ndeveloping the emerging theory\\nPursuing different slices of data for theoretical \\nelaboration\\nIterative process of cycling between existing data and \\ncollecting new data as required to pursue and develop \\nemerging ideas\\nTheoretical \\nsaturation\\nIntense conceptual property and dimension elaboration\\nConstant comparison of conceptual indicators until \\nadditional comparisons yield no further theoretical \\nspecification or elaboration\\nSaturation bounded by empirical limits of the data, \\nintegration, and density of the theory and researcher’s \\ntheoretical sensitivity\\nTheoretical \\ncoding\\nEmergent latent pattern recognition\\nCore category of a grounded theory is always some type \\nof theoretical code (e.g., process, range, continuum, \\ntypology, etc.)\\nConceptualizes how substantive concepts may relate to \\neach other as hypotheses\\nOverall integration of the emergent theory using \\nprocesses, frameworks, models, etc.\\nEarned relevance (i.e., modeling is emergent through \\nhand sorting of conceptual memos)\\nOften more than one theoretical code needed \\nTable 7.1    \\u0007\\nTechniques for Shaping a Grounded Theory\\n\\n\\nCLASSIC GROUNDED THEORY IN PRACTICE\\n118\\nIN SUMMARY\\nIn this chapter, we have explored the role of theoretical sampling, theo­\\nretical saturation, and theoretical coding in shaping an emerging \\ngrounded theory. We have discussed the techniques of theoretical sam­\\npling and theoretical saturation to build emergent concepts and catego­\\nries and the importance of hand sorting of memos for emergent \\nintegration of the final theory through theoretical coding. We take up \\nthe issues regarding writing grounded theory in Chapters 8 and 9.\\nTEST YOUR KNOWLEDGE\\nIn the following questions one or several answers may be correct.\\n1.\\t Theoretical sampling\\na.\\t Is the sampling of different existing theories or theoretical \\nframework\\nb.\\t Is deciding what data to collect next based on the theory that \\nis emerging from already collected data\\nc.\\t Is the sampling of only one type of data\\nd.\\t May include any type of data\\ne.\\t None of the above\\n2.\\t Theoretical saturation\\na.\\t Refers to intense development of conceptual properties\\nb.\\t Is the comparison of extant theories to an emergent grounded \\ntheory\\nc.\\t Continues until no new properties or dimensions of concepts \\nemerge from additional data\\nd.\\t Is achieved through theoretically sampling diverse compari­\\nson groups and data\\ne.\\t None of the above\\n3.\\t Theoretical coding\\na.\\t Is done when the grounded theory has emerged\\nb.\\t Is done to help a core category emerge from the data\\nc.\\t Is done after theoretical sampling has helped with saturating \\nthe core category\\nd.\\t None of the above\\n4.\\t Memo writing\\na.\\t Helps record the relationships among concepts\\nb.\\t Helps raise the conceptual level of the theory\\n\\n\\nChapter 7    Shaping the Theory\\n119\\nc.\\t Helps avoid falling into simple description\\nd.\\t None of the above\\n5.\\t Theoretical sorting\\na.\\t Is the process of sorting conceptual memos\\nb.\\t Is in aid of discovering relationships between the core cate­\\ngory and related concepts\\nc.\\t Can be done by hand or by computer\\nd.\\t Can be a tedious process requiring several attempts at resort­\\ning of memos\\ne.\\t Draws on the researcher’s creativity and theoretical ­\\nsensitivity\\nf.\\t None of the above\\n6.\\t Analytic rules\\na.\\t Help to raise the conceptual level of a grounded theory\\nb.\\t Guide the process of theoretical sorting\\nc.\\t Foster integrative fit of the emerging theory\\nd.\\t Provide a step-by-step guide to doing classic GT\\ne.\\t None of the above\\nFURTHER READING\\nGibson, B., & Hartman, J. (2014). Rediscovering skills for theoretical sampling. \\nIn Rediscovering grounded theory (Chapter 7). London: Sage.\\nGlaser, B. G. (1978). Analytic rules. In Theoretical sensitivity: Advances in the meth­\\nodology of grounded theory (pp. 120–127). Mill Valley, CA: Sociology Press.\\nGlaser, B. G. (1978). Theoretical coding. In Theoretical sensitivity: Advances in the \\nmethodology of grounded theory (pp. 72–82). Mill Valley, CA: Sociology Press.\\nGlaser, B. G. (1978). Theoretical sampling. In Theoretical sensitivity: Advances in \\nthe methodology of grounded theory (Chapter 3). Mill Valley, CA: Sociology \\nPress.\\nGlaser, B. G. (2005). The grounded theory perspective III: Theoretical coding. Mill \\nValley, CA: Sociology Press.\\nGlaser, B. G. (2014). Analytic rules. In Memoing: A vital grounded theory procedure \\n(pp. 108–122). Mill Valley, CA: Sociology Press.\\nGlaser, B. G., & Strauss, A. L. (1967). Theoretical sampling. In The discovery of \\ngrounded theory: Strategies for qualitative research (Chapter 3). New York: \\nAldine de Gruyter.\\nENDNOTE\\n1.\\t E-mail correspondence with Barney Glaser, March 9, 2007.\\n\\n\\n120\\nAfter studying this chapter, you will:\\n• understand why and how writing is, in itself, an essential part of the \\nclassic GT methodological protocol\\n• appreciate and recognize readiness to sort and write moments\\n• understand possible barriers that can block your writing and be aware \\nof some solutions to remove these barriers\\nW\\ne saw in a previous chapter the iterative nature of data \\nanalysis toward the emergence of a grounded theory. \\nWriting in GT is part of this iterative process and begins \\nwith the first data collected, as you start writing memos to help the \\nanalysis and conceptualization. Memoing was covered in Chapter 6. \\nIn this chapter, we are concerned with writing up the components of \\nyour emerging theory and how they are integrated together. When \\ndoing GT, writing and thinking are embedded and intertwined: \\n“Writing is the practical expression of thinking  .  .  .   writing is a \\nfundamental act of enquiry as valuable, if not more so, as generating \\nor collecting, reading and coding data” (Birks & Mills, 2012, p. 126).\\nCharmaz (2014) describes writing and rewriting GT as part of the \\ndiscovery process:\\n8\\nWriting as an Important Part \\nof Doing Classic GT\\n\\n\\nChapter 8    Writing as an Important Part of Doing Classic GT\\n121\\nYou’ll gain further insights and create more ideas about \\nyour data while you’re writing. You’ll see clearer connec­\\ntions between ­\\ncategories and draw implications from them. \\nThus writing and rewriting become crucial phases of the \\nanalytic process. Writing demands more than mere report­\\ning. Through writing and rewriting drafts, you can bring out \\nimplicit arguments, provide their context, make links with \\nextant literatures, critically examine categories, present your \\nanalysis, and provide data that support your analytic argu­\\nments. Each successive draft grows more theoretical and \\ncomprehensive. (p. 289)\\n\\t\\nTHE PROCESS LEADING TO\\t\\nl\\nTHE READINESS TO WRITE MOMENT\\nWriting in itself is considered by some researchers as a method of \\ninquiry (Richardson, 2000). This perspective is particularly well suited \\nto a GT approach as writing up is part of classic GT. Writing is not \\nsomething to do after the research has been completed. “The write-up \\nis a vital part of the method . . . Its timing is sequential, its doing not \\noptional” (Glaser, 2012, p. 12) because it helps conceptualization: \\n“Writing and thinking are inextricably intertwined in the construction \\nof a grounded theory” (Birks & Mills, 2012, p. 130).\\nAfter data have been collected and constantly analyzed, concepts \\nhave been generated, and memos have been written, the grounded theo­\\nrist often reaches a state of ideational overload. At this stage, researchers \\nare often facing a huge pile of memos, signaling a readiness to sort \\nmemos. As discussed in Chapter 7, the physical act of hand sorting \\nmemos helps the researcher preconsciously process this overload, thereby \\nguiding the organization and integration of the emerging theory.\\nOnce the core category of your grounded theory has emerged, \\nmemo writing accelerates as theoretical sampling progresses to elabo­\\nrate the core and related concepts and initial efforts are undertaken to \\ntheoretically sort memos for emergent patterns (theoretical coding). \\nWhen theoretical completeness appears to be achieved, memos are \\nagain sorted, leading to a “readiness to write” (Glaser, 2012, p. 4) \\nmoment that will prompt the researcher to write the first draft of the \\ntheory. The readiness to write moment is built into classic GT and \\nshould not be bypassed; if it is bypassed, it may be definitively lost. It \\nis an essential moment to allow the relationship between the researcher \\nand her data to evolve (Glaser, 1978).\\n\\n\\nWRITING AND EVALUATING CLASSIC GROUNDED THEORY\\n122\\nGlaser (2012) describes in great detail memo sorting as a crucial \\nepisode in the classic GT research process. To sort memos, researchers \\nshould pick up any memo, anywhere in the pile, and place it on a table, \\non the floor, or anywhere big enough to accommodate a large number \\nof small pieces of paper (Figures 8.1  and  8.2). They then pick up \\nanother memo and compare it to the first one; they either place it on, \\nor next to, the first one or place it in a new pile while writing another \\nshort memo about how this memo relates to the first one.\\nThis new memo is then placed between the two piles (perhaps \\nsuggesting a potential theoretical code). This process continues until \\nall initial memos are sorted into different piles with new memos \\nlinking all piles. Then the researcher writes the theory that he sees \\nunfolding from his data on the table. If there are any gaps or clarifi­\\ncations needed, the researcher goes back to the field to theoretically \\nsample. Once sorting is completed, the memos are merged and inte­\\ngrated into a readable text with good flow. Thus, there is no precon­\\nceived outline when you write as part of the ­\\nclassic GT process.\\nThe process leading to the readiness to write moment is summarized \\nin Figure 8.3.\\nWhen sorting memos, the researcher does not put a memo down idly. \\nIf it does not fit somewhere, it is put aside in a “left out” pile as nonfitting \\nor nonrelevant; usually this pile is small. If a memo can fit in several \\nplaces, it should be placed where it fits best to help the write-up of the \\nFigure 8.1    \\u0007\\nTheoretical/Memo Sorting on a Table (Source: Walsh, \\n2014a)\\n\\n\\nChapter 8    Writing as an Important Part of Doing Classic GT\\n123\\nFigure 8.2    \\u0007\\nSorting Memos on the Floor (Source: Sautrey & Mourmant, \\n2015. Research in progress)\\nemergent theory. Sorting helps generate the theory and writing it up. \\nWhen sorting memos, Glaser (1998, 2005a) advocates talking and writing \\nsubstantively but thinking theoretically. The idea here is that the researcher \\nstays close enough to the substantive area to ensure that the emergent \\ntheory remains relevant to those in the substantive area and that in orga­\\nnizing and writing up the theory, it does not become so conceptually \\nabstract that it no longer works to explain the main concern and its resolu­\\ntion. On the other hand, the researcher must resist the temptation to fall \\ninto copious description that inevitably bogs down the theory, holding it \\nhostage to one specific setting (i.e., specific people, time, and place), and \\nstealing the explanatory power of its broader general implications.\\n\\n\\nWRITING AND EVALUATING CLASSIC GROUNDED THEORY\\n124\\nFigure 8.3    \\u0007\\nThe Process Leading to the Readiness to Write Moment\\n1. Data\\ncollection\\n2. Concept\\ngeneration\\n3. Write\\nmemos\\n4. Information\\noverload\\n5. Readiness\\nto sort\\nmemos\\n6. Memo\\nsorting into\\npiles\\n7. Further memo\\nabout piles of memos\\nand their relationships\\nTheoretical\\nintegration:\\nReadiness\\nto write\\nOnce the researcher has completed analysis of empirical data, and after \\nthe core category has emerged, he can turn to the literature of his rele­\\nvant field as more data to be constantly compared in elaborating and \\nsaturating emergent concepts, developing further their properties and \\ndimensions (for an illustration, see Appendix C). Glaser (1978) describes \\nthe literature as a “prominent class of outside comparisons” (p. 51). Of \\ncourse, in doing so, the researcher must determine whether the litera­\\nture is empirically grounded. Even “conjectured” theory can yield \\ninteresting ideas, but its groundedness will need to be empirically con­\\nfirmed through further theoretical sampling to merit a place in the final \\nwrite-up of a grounded theory. Another approach to addressing the \\nliterature is to compare it with your emergent theory, discovering and \\nl\\nl\\nAPPROACHING THE LITERATURE IN A GT STUDY\\n\\n\\nChapter 8    Writing as an Important Part of Doing Classic GT\\n125\\nconsidering what aligns, what modifies, and what challenges extant \\ntheory. In so doing, the researcher begins to identify how the emergent \\ngrounded theory supports, extends, corrects, and contributes to extant \\nknowledge. As we saw in a previous chapter, the literature from other \\nfields of research may also be used to help make sense of collected data, \\nbuild new theory, and, more generally, write a compelling account of \\nthe emerging theory.\\n\\t\\nWRITING GT WITH MIXED\\t\\nl\\nQUANTITATIVE AND QUALITATIVE DATA\\nWriting as part of the GT process is particularly delicate and essential \\nwhen you use mixed quantitative and qualitative data. A researcher \\nmay use quantitative and qualitative data and analyze them as one set \\nin an exploratory GT stance with the help of those quantitative and/or \\nqualitative techniques that the researcher thinks will best serve the pur­\\npose of her research (e.g., Schall, 1983; Walsh, 2014a). In some instances, \\nneither qualitative nor quantitative data/methods are sufficient to fully \\ndevelop your theory; rather, all are necessary (Figure 8.4). The different \\nquantitative and qualitative data, methods, and techniques are mixed \\nand supplement each other within a single project. For instance, and as \\nhighlighted in Walsh (2014c), the researcher may need to qualify quan­\\ntitative data (e.g., make sense of the results of the factor analysis of \\nquantitative data with the help of collected qualitative data) or quantify \\nqualitative data as a compromise between numerical precision and \\n­\\nnarrative complexity (Sandelowski et al., 2009).\\nFigure 8.4    \\u0007\\nUsing Mixed Data to Write a Grounded Theory\\nPatterns\\nConcepts—\\nPropositions\\nQualitative\\nData\\nQuantitative\\nData\\n\\n\\nWRITING AND EVALUATING CLASSIC GROUNDED THEORY\\n126\\nWhen using mixed data, it is even more essential to write the \\nemerging grounded theory as it unfolds, when data becomes daunting \\nand memo sorting becomes compelling, so as not miss the readiness to \\nwrite moment. Attempting to later capture back elements that were \\nbypassed and not written will be extremely difficult because of the \\nembeddedness of the qualitative and quantitative analyses.\\nThe Consequences of Bypassing the Readiness to Write Moment: \\nTestimony by Isabelle Walsh on How the Publication of Walsh \\n(2014a; see Appendix C) Was Unnecessarily Delayed\\nThe grounded theory proposed in \\nWalsh (2014a) had to be written and rewrit­\\nten a number of times because the readiness to write moment was bypassed. \\nEven though, when sorting the various memos, I felt compelled to write what \\nwas emerging from my data, I just did not know how to write what I saw in my \\ndata: at the time, there were very few examples of mixed GT studies.\\nWhen I first wrote this mixed GT study (and submitted it to a top-tier \\ninformation systems journal), I first presented all quantitative results, then \\nall qualitative results, then attempted to integrate them all. But this was not \\nhow the theory had emerged. I was very lucky to have reviewers who were \\nvery patient and very open to grounded theory. They made me go back to \\nmy data and account faithfully to what had emerged, when and how. Not all \\nreviewers are as patient and open as these reviewers were. Guided by the \\nreviewers’ queries and criticisms, I had to work very hard to recapture the \\ntrue emergence of the theory.\\nEventually, I found that the easiest way to write mixed GT was not to try and \\nisolate quantitative and qualitative data/results that were embedded. Instead, I \\nfound that it was best to relate my stories as they unfolded, as impossible as it \\nmay appear at first. If quantitative and qualitative data/results were embedded, \\nthen I wrote “as was.” However, I wrote quantitatively, that is, with acronyms \\nfor variables/categories/concepts and numbers, when I used quantitative data, \\nand I wrote qualitatively, with fully developed names for variables/categories/\\nconcepts and verbatim, when I used qualitative data. What I wrote was then \\neasy for me, the reviewers, and (I hope) the readers to read, follow, and under­\\nstand. However, the whole process took me years and some very indulgent and \\npatient reviewers (I have not always been that lucky! Other reviewers might \\nhave rejected my work during the first round), whereas, if I had done the writ­\\ning as the theory emerged and when the readiness to write moment was at hand, \\nthe whole process would have been easier, less risky, and much faster.\\n\\n\\nChapter 8    Writing as an Important Part of Doing Classic GT\\n127\\n\\t\\nTHE FIRST DRAFT\\t\\nl\\nWriting as part of the GT process must be done as simply as possible \\nto highlight patterns as they emerge and to help move toward concep­\\ntualization and increased abstraction. Once the memos are theoreti­\\ncally sorted, the researcher is usually eager to write up the theory that \\nhe sees in the data to produce a first rough draft of his conceptual \\nideas. This first draft, or working paper, is “a ‘write up’ of piles of \\nideas [the piles of memos] from theoretical sorting” (Glaser, 1978, \\n\\np. 128). It is “writing concept to concept relations integrated into a \\nconceptual theory” (Glaser, 2012, p. 111) and certainly not like writing \\nfiction. Perfection in terms of style, grammar, and spelling is not \\nrequired for this first draft, which is not written to please a supervisor, \\na committee, an editor, or reviewers. The purpose of this first draft is \\nonly to put the theory on paper. In turn, this first draft becomes a \\nmemo in itself.\\nIn writing up your GT, it is important to stay on a conceptual level. \\nOne very effective way to do this is to write in the present tense; that is, \\ntalk about your theory as if it is going on now, for it surely is . . . some­\\nwhere! Writing in the past tense (as research findings) tends to drift into \\ndescription, producing an account of what happened during data collec­\\ntion rather than a conceptual explanation of the discovered latent pat­\\ntern of social behavior. Employing “illustration dosage” (Glaser, 1998, \\npp. 198–199) by adding brief data excerpts to illustrate key concepts \\nhelps control conceptual density, improving readability as well as \\nenhancing understanding. Too many lengthy quotes, however, can bog \\ndown your theory, interrupting the conceptual flow and making it dif­\\nficult to read.\\nAs part of an academic process, first drafts may eventually be sub­\\nmitted to international conferences in order to present one’s ideas \\nwithin one’s research field. Even though Glaser often advises not to \\nvoice one’s ideas and not to present them for other people’s judgment \\nuntil the theory has fully emerged, it may be useful and important for \\nyoung scholars in search of legitimacy to offer a working paper for \\npeers’ judgment. In our experience, the peer review process mostly \\nhelps strengthen results and identify weaknesses as long as the \\nresearcher takes the comments as provisional and stays true to what is \\nin the data. When reviews are positive, they provide great encourage­\\nment to complete a research project, which might otherwise have been \\nput aside.\\n\\n\\nWRITING AND EVALUATING CLASSIC GROUNDED THEORY\\n128\\nAnother Type of Memoing\\nOne of the authors of this book regularly submits her first drafts to inter­\\nnational conferences. She also encourages her doctoral students to do the \\nsame.\\nFor instance, the grounded theory of “IT culture creep” was published \\nin 2010 (Walsh et al., 2010). However, the first draft was published in 2008 \\nat the Americas Conference in Information Systems (Walsh & Kefi, 2008). \\nComments from peers obtained during the conference helped finalize the \\ntheory of IT culture creep and eventually publish it in a top tier research \\noutlet.\\nAnother example is the grounded theory of the “translated strategic \\nalignment model” that was published in a top tier research outlet in 2013 \\n(Walsh et al., 2013). The first draft of this theory was published in 2010 \\nat the Americas Conference in Information Systems (Walsh et al., 2010), \\nbefore categories were fully saturated.\\nThese publications at international conferences may be considered \\nas a type of memoing and will eventually help the final write-up of your \\ntheory.\\nProposing first or intermediate drafts of their grounded theories to \\ninternational conferences for peers’ perusal and informed comments \\nwill help prepare researchers for what we consider scholars’ ultimate \\npurpose: writing toward publishing in a recognized and well-ranked \\nresearch journal of their field. This issue is addressed at length in the \\nnext chapter.\\nl\\nl\\nWRITING BLOCKS\\nWriting as part of the GT process may be impeded in many ways for \\nmany reasons, and in some extreme cases, blocked. In his 2012 mono­\\ngraph about the act of writing grounded theory, Glaser investigates in \\ndepth the different elements that may block writing. These writing \\nblocks may be personal. For instance, the researcher may suffer from \\nperformance anxiety, lack of experience, poor reading ability that \\ninduces a lack of knowledge of existing literature, or an inability to see \\npatterns and conceptualize. Writing may also be blocked by method­\\nological issues. For instance, conflicts between GT and other approaches \\nresearchers might wish to use in combination. Finally, writing blocks \\nmay be induced by circumstances that are contextual or external to the \\n\\n\\nChapter 8    Writing as an Important Part of Doing Classic GT\\n129\\nresearcher. For instance, supervisors, committee members, or universi­\\nties may hinder writing through insisting on some prescribed writing \\nformats. Difficult life circumstances (health, family) may also hinder \\nwriting.\\nPersonal Blocks\\nSome examples of personal blocks are provided in Figure 8.5 and \\ndiscussed in this section. The fear of writing needs to be acknowledged \\nas legitimate for most inexperienced researchers; it will usually dimin­\\nish with time and practice. It is mostly linked to performance anxiety, \\nmore especially when a researcher lacks confidence, has no writing \\nexperience of any sort, or when the format prescribed to the researcher \\n(e.g., by his supervisor) is too constraining.\\nNovice grounded theorists may be overwhelmed by their data, \\nwhich could feed not one theory but multiple theories, leading to per­\\npetual outgrowing (Glaser, 2012, p. 112). This can be worsened when \\nresearchers “fall in love” with their data and search for perfection or \\nwhen too many redundant descriptive elements are blurring the over­\\nall theory. If writing is done as memos are sorted, then perpetual \\n­\\noutgrowing can be minimized. It will also probably help alleviate what \\nFigure 8.5    \\u0007\\nOvercoming Some Personal Writing Blocks\\nWRITING\\nWrite as\\nsorted\\nCreative\\nautonomy\\nMemo\\nsorting\\nExperience\\nPerpetual\\noutgrowing\\nWorrisome\\naccuracy\\nConfusion\\nFear of writing\\nFalling in love\\nwith your data\\nRedundancy\\nNo writing\\nexperience\\nPrescribed\\nformat\\nPerformance\\nanxiety\\nLack of\\nconfidence\\nPremature\\nwriting\\nInability to see\\npatterns\\nSupervisor\\nvalidation \\nConcept\\nvalidity\\nField\\njargon\\nMay be solved by\\nMay be solved by\\nMay be solved by\\nMay be solved by\\n\\n\\nWRITING AND EVALUATING CLASSIC GROUNDED THEORY\\n130\\nwe named “postpartum blues” (!) when one doctoral student, whom \\none author of this book helped supervise, seemed to fall into depres­\\nsion once her dissertation document had been submitted to her com­\\nmittee. If parting with your theory is done a little at a time, the final \\nbreak will be less difficult to accept.\\nWhat Glaser (2012) names “worrisome accuracy” (p. 119) is another \\npossible writing block induced by superfluous descriptive details, the \\nneed to conform to field jargon, or the desire to obtain a supervisor’s \\nvalidation. Fighting for and obtaining creative autonomy will solve \\nthis issue. Even though writing is an intrinsic part of classic GT, \\nattempting to write up one’s theory before memos are sorted into piles \\nwill lead to confusion and may well tempt the researcher into precon­\\nceiving the theory. It has to be remembered that writing as part of the \\nGT process is a “write-up of sorted memos” (Glaser, 2012, p. 120) as \\nthey provide conceptual meaning and will help fight confusion.\\nMethodological Blocks\\nSometimes, writing may be hindered by some methodological \\nblocks. For example, the researcher may feel compelled to use some \\nmethods or techniques that violate GT principles. These methods or \\ntechniques have then to be revised and adapted. We have found that \\nany methodological blocks may be addressed by going back to the \\nseminal texts, Glaser and Strauss (1967) and Glaser (1978), and keeping \\nclearly in mind GT’s main foundational pillars as highlighted in \\nChapter 3: emergence, theoretical sampling, and constant comparative \\nanalysis. We have found that if this is done, while remaining in an \\nexploratory stance that is an essential aspect of a GT approach, GT \\nmethodological protocol may be applied with any data, collected \\nthrough a variety of methods, and analyzed with the help of many data \\nanalysis techniques. In using quantitative data in particular, we have \\nfound that many methods and techniques may be adapted to fit GT’s \\nmain foundational pillars (see boxed text).\\nUsing Quantitative Techniques in a GT Stance\\nFactor analysis of quantitative data may be used in an exploratory fashion. \\nIts purpose is “to examine the underlying patterns or relationships for a \\nlarge number of variables and to determine whether the information can \\nbe condensed or summarized in a smaller set of factors or components” \\n\\n\\nChapter 8    Writing as an Important Part of Doing Classic GT\\n131\\nExternal/Contextual Blocks\\nContext may also hinder writing. This is when the “outside world” \\nimposes unwarranted constraining demands on the researcher. Every \\nresearcher will have her own ideal physical writing preferences, for \\nexample, being completely isolated from anybody else, listening to \\nmusic while writing, and so on. However, researchers will often have \\nto deal with nonideal contextual situations imposed by family life \\nobligations or the demands of a dissertation supervisor, for example, \\nthe requirement to review the literature ex ante, which goes against \\n(Hair, Black, Babin, Anderson, & Tatham, 2006, p. 101). This quantitative \\ntechnique may help with the selective and theoretical coding of quanti­\\ntative data by sorting the data “into the appropriate pile” (Glaser, 2004, \\n\\np. 191). \\nWhen applying this technique with the help of some software, make \\nsure not to inform the software with the number of factors you might \\nexpect (based on previous qualitative phases and results); rather, let the \\nslices of quantitative data inform the emerging number of factors.\\nCluster analysis is another statistical technique that we have found use­\\nful when doing GT with quantitative data. It is a statistical data analytic \\nmethod viewed mainly as an exploratory technique (Hair & Black, 1998) \\nand whose main purpose is “to group objects based on the character­\\nistics they possess” (Hair, Anderson, Tatham, & Black, 1995, p. 423). High \\nwithin-cluster homogeneity and between-cluster heterogeneity is the aim. \\nCluster analysis is very close to the GT comparative method. It searches \\nfor nonobvious patterns and relationships between important sets of data. \\nHowever, the way one uses cluster analysis (or any other multidimensional \\nscaling instrument) will, of course, determine whether theory is allowed to \\nemerge from the data or whether theory is forced upon those data.\\nEven techniques like structural equation modeling, traditionally con­\\nsidered as a confirmatory technique, may be adapted to fit a GT stance. \\nFor instance, when using mixed qualitative and quantitative data, you may \\nchoose partial least squares (PLS) path modeling to investigate the quanti­\\ntative data. This approach is generally viewed more as an exploratory, soft \\nmodeling approach than as a confirmatory one (Gefen, Rigdon, & Straub, \\n2011; Hair, Sarstedt, Ringle, & Mena, 2012; Ringle, Sarstedt, & Straub, 2012; \\nVinzi, Trinchera, & Amato, 2010). All possible paths between the various \\nconstructs under scrutiny may be investigated before deciding which paths \\nare the most probable when considering quantitative and qualitative data \\ntogether as one set.\\n\\n\\nWRITING AND EVALUATING CLASSIC GROUNDED THEORY\\n132\\nclassic GT’s main pillars or meeting impossible deadlines. This will \\ncreate a block that can be alleviated (as proposed by Lowe, cited by \\nGlaser, 2012) by engaging momentarily in other, purely menial, tasks \\nand then eventually coming back to the problematic contextual issue \\nafter having taken time to ponder over it.\\nThe most serious external or contextual writing block may some­\\ntimes stem from your PhD supervisor or your dissertation committee \\nwho may not be experienced grounded theorists and may try to \\nimpose their own pet approaches, which may violate some of GT’s \\nfoundational pillars. Beyond all precious advice that one may find in \\nvarious publications (see, e.g., Glaser, 2012), we have found that the \\nbest way to get around possible issues with supervisors or committee \\nmembers is to have your work recognized by your peers, through pub­\\nlication in top tier international conferences and/or peer reviewed \\njournals before you present your finished work to them. Writing for \\npublication is addressed in the next chapter.\\nIN SUMMARY\\nWriting in itself is an intrinsic part of GT. If ideational overload is over­\\nlooked and the readiness to write moment is bypassed, the completion \\nof your work will be delayed. Some personal and contextual blocks to \\nwriting may be overcome through memo sorting and simply writing \\nthings as they emerge and as you see them through collected data, on \\nthe way to building up your creative autonomy and developing your \\nGT writing experience.\\nTEST YOUR KNOWLEDGE\\nIn the following questions one or several answers may be correct.\\n1.\\t When doing GT\\na.\\t You have to write memos all the time as you analyze your \\ndata and ideas emerge\\nb.\\t You have to write your memos properly so that anybody can \\nunderstand them\\nc.\\t You have to wait to be ready to write your theory before you \\nstart memoing\\nd.\\t None of the above\\n\\n\\nChapter 8    Writing as an Important Part of Doing Classic GT\\n133\\n2.\\t When you are overtaken by information/ideational overload, \\nyou should\\na.\\t Stop doing GT\\nb.\\t Change your codes\\nc.\\t Start memo sorting\\nd.\\t None of the above\\n3.\\t The readiness to write moment\\na.\\t Comes after memo sorting\\nb.\\t Is when you are ready to write a first draft of some theoretical \\nstatement\\nc.\\t May be bypassed\\nd.\\t None of the above\\n4.\\t Drafts of your theory\\na.\\t Must include the relationships between concepts that \\nemerged from your data\\nb.\\t Involve writing up the memos that you have sorted into piles\\nc.\\t Eventually become memos until another, more densified, \\ntheory emerges\\nd.\\t None of the above\\nFURTHER READING\\nGlaser, B. G. (1978). Theoretical sensitivity: Advances in the methodology of grounded \\ntheory (Chapter 5). Mill Valley, CA: Sociology Press.\\nGlaser, B. G. (2012). Stop, write: Writing grounded theory. Mill Valley, CA: \\nSociology Press.\\nSchall, M. (1983). A communication-rules approach to organizational culture. \\nAdministrative Science Quarterly, 28(4), 557–581.\\nWalsh, I. (2014). A strategic path to study IT use through users’ IT culture and \\nIT needs: A mixed-method grounded theory. Journal of Strategic Information \\nSystems, 23, 145–173.\\nENDNOTE\\n1.\\t All articles in FQS are licensed under a Creative Commons Attribution \\n4.0 International License. Authors hold the copyright and retain \\n­\\npublishing rights without restrictions.\\n\\n\\n134\\n9\\nWriting Classic GT for \\nPublication\\nAfter studying this chapter, you will:\\n• understand that you can rarely publish a grounded theory as it actually \\nemerged from your data\\n• accept reviewers’ opinions and address their concerns even though you \\nmight not always agree with them\\n• have some general guidelines to help you get your work read, accepted, \\nand published\\nW\\nhether or not you have read the previous chapters, having \\narrived at this stage of the book probably means that you are \\nconvinced that using GT as a research paradigm, a method-\\nology, or an analysis framework is important and significant for your \\nresearch. However, writing as part of the classic GT process, that is, of \\n“doing” GT, and writing classic grounded theory for an audience, to \\nshare and disseminate your ideas, are two rather different processes.\\nl\\nl\\nDOING VERSUS WRITING\\nWriting—the way we write—matters “theoretically and practically. . . . \\nHow should we write our research? Who is our audience? . . . What are \\n\\n\\nChapter 9    Writing Classic GT for Publication\\n135\\nour goals? . . . What readers will we write for? . . . How should we write \\nso that our writing matters?” (Richardson, 1990, p. 9). These questions \\nhighlight many implicit issues, which can be broadly classified as \\n­\\nrhetorical, ethical, and methodological.\\nRhetorical Issues: The Way We Write\\nThe first issue with writing grounded theory for an audience is the \\nfact that no written words can fully describe the iterative GT process, \\nthe incessant iterations between data and conceptualization, or the \\nfinal conceptual leap that illuminates a grounded theory. If one \\nattempts to describe fully in writing, and in truth, the way their GT \\nresearch was conducted, it is most probable that few people would \\nread or understand it. When you write a classic grounded theory, it is \\ntherefore essential to give sufficient information to guarantee the \\nreader that all classic GT guidelines have been followed, but, at the \\nsame time, you must also organize your work in a way that respects \\nthe conventions of the intended audience.\\nThere are many different ways to write and account for your \\nresearch depending on your chosen outlet. When you write your dis­\\nsertation, you have to first please your thesis supervisor, then your \\ncommittee members and external examiners. When you write an arti­\\ncle to be published in an academic outlet, you have to please the \\nreviewers and the associate and senior editors. When you write a \\nmonograph, you mostly please yourself. Finally, when you write \\na methodological book like this book, your purpose will be fulfilled if \\nyour readers find it easier to apply the method(s) you describe after \\nreading your book.\\nWriting options and writing outlets are numerous. Even though \\nLyytinen, Baskerville, Iivari, and Te’eni (2007) have pointed to differ­\\nences in publication culture whereby European scholars have tended in \\npast years to favor book publications, most academics are currently \\nhired, promoted, and tenured on the basis of publications in so-called \\ntop-tier journals. As academics, we depend essentially on promotion \\nand tenure committees and on other committees or public bodies that \\nallocate resources to academic units and individual researchers. These \\nvarious committees will judge your work based mostly on your publi­\\ncations in top tier journals (Takeda, 2011).\\nFurthermore, if, while doing your PhD, you publish your classic \\ngrounded theory in a top-tier journal, or (as revision processes of such \\njournals are very lengthy) you at least avoid the dreaded “desk reject” at \\nthe outset and engage your work in the revision process of such a journal, \\n\\n\\nWRITING AND EVALUATING CLASSIC GROUNDED THEORY\\n136\\nit is unlikely that your thesis supervisor will object or question the use of \\nGT or the way you write your research, whatever way you wish to do so. \\nWhen you are a young researcher, your peers’ positive judgment is a very \\nconvincing argument that is difficult to contradict, even by an established \\nprofessor. Therefore, we will concentrate in this chapter on the type of \\nwriting required to publish in those so-called top-tier outlets.\\nDifferent journals emphasize different topics, methodologies, or \\nboth. To find out about a journal’s preferences, you can investigate \\ntheir mission statement and submission guidelines, which are available \\non journal websites. To publish in any research journal, your work has \\nto be presented in a way that meets the writing conventions of that \\njournal and the respective editors’ and reviewers’ expectations. There­\\nfore, you have to find ways to reconcile these conventions with classic \\nGT precepts.\\nIf you choose to publish in a research journal (rather than, for \\ninstance, in a monograph or a book), another issue that you face when \\nwriting classic grounded theory is that journal articles “have become \\nincreasingly narrow in scope and focused on a little bit of knowledge—\\nas if knowledge ‘really’ were a bin of bits” (Richardson, 1990 p. 16) and \\nwere only progressing incrementally. However, if you have success­\\nfully applied classic GT, you should have produced a theory that has \\nthe potential to be in rupture with existing literature, hence difficult to \\nreduce to incremental progress and a single journal article. One way to \\ncounteract this drawback is to rationalize, as much as possible, the way \\nyou write and fit in, as much as possible, the prescribed formats, with \\nestablished rhetorical devices that may be found in most social sciences \\n(de Vaujany, Walsh, & Mitev, 2011).\\nDifferent Writing Formats\\nThe editor in chief of a top-tier research journal, whom we know well, once \\nconfirmed to us that research articles are traditionally more suitable for \\nincremental discoveries. This came up when, new to the game of publish­\\ning, one of the authors of this book asked for his advice on publishing a \\nconceptual piece of work that would create some upheavals if published. \\nThe editor in chief had read and liked this piece of work. He strongly \\nadvised us to publish it as a book because the intellectual leap that it pro­\\nposed was too broad to fit in a research article and would involve a length \\nof manuscript that would be difficult to adhere to any research journal’s \\n­\\nwriting ­\\nconventions.  \\nAs we did not wish to go through the book format, \\n\\n\\nChapter 9    Writing Classic GT for Publication\\n137\\nFinally, and as highlighted by Fetters and Freshwater (2015), when \\naiming to publish, “effective authors compose their articles for the \\nunderstanding of the reader, and not their own egos” or seeking “to \\nimpress others” (p. 204). These authors propose that we consider every \\nword and sentence we write and ask ourselves the following questions: \\n“Will this make sense for the reader? Will this help the reader under­\\nstand what I want to convey?” (p. 204).\\nEthical Issues: Adjusting Your Work to Fit \\nEstablished Conventions\\nAcademic papers are encoded with academic conventions (e.g., the \\nmain current paradigmatic positions of the research field covered by a \\ngiven journal) and writing conventions (e.g., length, focus, method­\\nological stance) fitting traditional standards that readers expect will be \\nmet (Richardson, 1990). To help you identify these conventions and \\nverify that they are compatible with whom you are and with your work, \\nRichardson (2000) implicitly provided some valuable guidelines even \\nthough prescription was certainly not her purpose as she was solely \\naiming to identify different writing formats in qualitative research. \\nHowever, using some of her ideas out of context helped us to highlight \\nsome interesting propositions that can be applied while writing classic \\ngrounded theory geared toward publication in a top-tier outlet.\\nWe would propose for you to choose an article published in a top tier \\noutlet of your own field, an outlet where you would like to publish. \\nChoose an article that is fairly recent (published during the past 3 years) \\nbecause cyclic changes in editorial boards may change some of the applied \\nwriting conventions (de Vaujany et al., 2011). As the chosen article has \\nbeen published in the chosen outlet, it exemplifies in some ways this out­\\nlet’s writing conventions and will guide you toward framing your work \\nwithin these conventions. This article should be one that you admire, one \\nthat you believe makes important contributions to the field, and one \\nwhose philosophical assumptions are globally in tune with yours.\\nProvided reviewers do not subsequently ask you to forfeit some of \\nyour own philosophical assumptions, you then might be able to fit \\nyour work within the conventions of the given journal. That being said, \\nthe original article was divided into several articles, to which we added \\nempirical data. The original piece of work thus “gave birth” to a number of \\njournal articles, most of which have now been published.\\n\\n\\nWRITING AND EVALUATING CLASSIC GROUNDED THEORY\\n138\\nand whatever the writing conventions and philosophical assumptions \\nof the editorial board of a journal, our experience is that if your work \\nis considered innovative and important for your field, and if your \\n­\\nwriting is clear enough to allow the reader to understand your contri­\\nbution, you will eventually be published in a top-tier journal—perhaps \\nnot in the journal that you originally aimed for (you may have to inves­\\ntigate other top-tier journals with broader-minded editors in chief), but \\nyou will be published.\\nLearning From Experience\\nIt took me many years to be able to publish my first GT article in a top \\ntier journal. Before I actually succeeded in publishing it, I had to submit \\nit twice.  \\nAfter the initial submission in the first journal, I was asked to \\nreview and resubmit while removing the methodological GT approach \\nas well as the core category of my grounded theory! Although there \\nwas no doubt that my work was not convincing enough and that other \\nelements of these initial reviews helped me improve it, I did not review \\nand resubmit to the same journal. I felt that, if I did so, I would be lying \\nabout my research and that I would somehow forfeit my identity as a \\nresearcher. Instead, I improved my work based on some other elements \\nof the initial reviews that I felt highlighted some serious flaws in my work, \\nbut I submitted to another journal, which incidentally was higher ranked \\nand globally considered as “better” than the first one. Interestingly, during \\nthe reviewing process of this second submission, and after I had addressed \\nall revisions required by the reviewers, the associate editor advised me, \\nfor the last round of review, to bring in a coauthor who could guide me \\nto write within the conventions of the journal. This did not jeopardize my \\nphilosophical assumptions, nor did it entail disguising my research. It only \\nmeant that my writing format was not the right format for the chosen \\noutlet. I brought in an experienced coauthor who helped me understand \\nthe associate ­\\neditor’s and the journal’s writing expectations, and the article \\nwas eventually published.\\n—Isabelle Walsh\\nMethodological Issues: Writing to Be Read and Published\\nEven though doing GT is never a linear process, the writing format \\nof GT is rarely otherwise if one wishes to be read, easily understood, and \\n\\n\\nChapter 9    Writing Classic GT for Publication\\n139\\npublished in a top-tier journal. As hypothetical-deductive research has \\nleft an enduring imprint in most of the so-called top tier outlets, and \\nalthough sections found in published empirical articles are sometimes \\nnamed differently and sometimes merged, the basic standard and \\nexpected structure of a journal article that relates empirical research is, \\nglobally, fairly linear and includes six sections. Each of these sections as \\noutlined in the following box aims to answer particular questions.\\nThe Standard and Expected Structure of a Journal Article and \\nQuestions to Be Answered\\n1. Introduction\\n• Why should the reader bother to read your article? (What is the \\nissue you are investigating or addressing? Why is it important?)\\n• What are the major contributions of your work?\\n• How is your work organized?\\n2. Literature Review\\n• What does the literature tell us about the object of your \\nresearch?\\n• What is the gap that is being addressed?\\n3. Methodology\\n• What are the data, data collection methods, and data analysis \\ntechniques used in your research?\\n4. Results\\n• What have you found?\\n5. Discussion\\n• How do your findings fit in with existing literature?\\n• What are the limitations and contributions of your work?\\n6. Conclusion\\n• How has the research issue, highlighted in the introduction, been \\naddressed?\\n\\n\\nWRITING AND EVALUATING CLASSIC GROUNDED THEORY\\n140\\nOn one hand, this expected structure, in itself, may raise a num­\\nber of issues in relation to a classic GT stance. On the other hand, it \\nneed not raise any issues if we keep in mind that doing research and \\nwriting to publish are different processes, even though they are linked, \\nand writing research should reflect, at least in some manner and to \\nsome extent, the way the research has been conducted.\\nWhen you do classic GT, you basically start with a general area of \\ninterest and methodological guidelines that you will apply in congru­\\nence with your philosophical positioning, but you do not even know \\nwhich data are going to be collected or how you are going to collect \\nthem. You start by listening and observing in the field. It is only after \\nyou have identified the main concern and after a core category has \\nemerged to explain this main concern that you turn to the literature. \\nBut this does not mean this is the way you should actually structure \\nyour story for publication. It also does not mean that you have to lie \\nor ­\\npretend you did what you did not do. For instance, when you \\nwrite the introduction, in which traditionally the reader expects to see \\nyour research question, the fact that you comply with this does not \\nmean that the focus of the study was decided at the outset of the \\nstudy through preconceived notions. You can, in the introduction, \\nindicate that you began the study with an area of interest and phrase \\nthe research question in an open manner (e.g., What is the experience \\nof knowledge workers regarding work-related informal networks?). \\nEven when placing your literature review after the introduction and \\nbefore the results, you can clearly indicate that it was done ex post but \\nis presented ex ante in order to help the reader understand your \\nwork. This confirms that, as per classic GT precepts, the literature was \\nheld in abeyance until the core category of your grounded theory \\nemerged.\\nBeing a scholar means that you read a lot. Doing GT does not \\nmean that you stop doing so. Sometimes, an existing framework \\ndeveloped in a different field of research (or a different domain in your \\nfield) and discovered by chance through your readings helps you \\nunderstand and code your data, not because you tried to apply it and \\nforced your data into this framework but, rather, the opposite; that is, \\nthis theoretical framework was not preconceived but rather emerged \\nfrom other literature (which is also data as such) and helped you make \\nsense of some of your data (see the use of translation theory, a theory \\nfrom the field of sociology, in Walsh, 2010, or the use of values, needs, \\nand motivation theories, from the psychological field of research, in \\nWalsh et al., 2010).\\n\\n\\nChapter 9    Writing Classic GT for Publication\\n141\\nTo illustrate the difference between doing and writing GT, we give \\nthe example of an article published in a top tier journal of the \\ninformation systems field of research (Walsh et al., 2013), which was \\nfirst published online in 2013 and won a best paper award in May 2014. \\nTable 9.1 presents a summary of how the research was conducted and \\nhow it was presented for publication.\\nThe Way the Research Was Conducted\\nThe Way the Journal Article Was \\nWritten\\nPhases\\nDescription\\nSections\\nDescription\\n1\\nOpen coding\\nConstant \\ncomparison\\nIdentification of \\nthe main \\nconcern\\n1\\nIntroduction with \\ntraditional research \\nquestion as \\nformulated in Phase 6 \\nof the research \\n(Introduction)\\n2\\nTheoretical \\nsensitivity to \\ntranslation \\ntheory\\n2\\nTheoretical anchorage \\n(Literature review)\\n3\\nSelective \\ncoding\\nConstant \\ncomparison\\nEmergence of \\nthe core \\ncategory\\n3\\nResearch settings, \\nsamplings, data \\ncollection, coding \\n(Methodology)\\n4\\nLiterature \\nreview\\n4\\nDescription and \\nconceptualization \\n(Results)\\n5\\nTheoretical \\nsaturation\\n5\\nProposed theory and \\ndiscussion with \\nrespect to existing \\nliterature (Discussion)\\n6\\nTheoretical \\ncoding\\nIntegration and \\nelaboration of \\nthe grounded \\ntheory\\nArticulation of \\ntraditional \\nresearch \\nquestion\\n6\\nSummary of what was \\naccomplished and \\nanswer to the research \\nquestion (Conclusion)\\nTable 9.1    \\u0007\\nConducting a GT Research Project Versus Writing an \\nArticle for a Research Journal\\n\\n\\nWRITING AND EVALUATING CLASSIC GROUNDED THEORY\\n142\\nThe actual chronology of this research, as befitting classic GT, was \\nidentification of the main concern during many years of practical expe­\\nrience, consultancy missions, and preliminary research investigations \\nin three firms in 2008 and 2009. This main concern was the failure of \\ninformation technology (IT) projects in firms. It must be highlighted \\nthat this was not our (the researchers’) main concern, as the consultancy \\nmissions conducted in the three firms (and later theoretically sampled \\nfor our research) were not concerned with IT projects as such but with \\nchange management. The main concern identified was, indeed, the \\nparticipants’ main concern.\\nLater, while preparing a class in 2008 about organizational theories, \\nwe (by chance) discovered translation theory and actor network theory \\n(ANT; Callon, 1986; Latour, 1987). Theoretical sensitivity to this framework \\nallowed us to move on from open coding to relate various categories, \\nthus highlighting the core category (which had not been highlighted \\nbefore as such in the literature). The core category, Aligning stakeholders’ \\nneeds, emerged in 2009, after which we started selective coding leading to \\ntheoretical coding. The literature review was conducted from 2010 to 2012 \\nby a supervised doctoral student, who co-signed the final published \\narticle. All categories in our theory were fully saturated in 2012. The \\nresearch question was formulated at the end of 2012 while finalizing the \\nwriting of the article before submission to a journal. If one looks at the \\nway the article was written and published after three rounds of revision \\nand substantial modifications required by reviewers, it reads as follows:\\nIntroduction: Up front, the introduction includes an argument \\non the fact that our area of interest (the alignment between \\ninformation systems and business strategy) is an important \\nissue for both managers and researchers (it is considered as \\nessential in firms to drive toward competitive advantage), the \\nlimitations of the existing literature (the concept of strategic \\nalignment has mostly been studied through Henderson and \\nVenkatraman’s 1993 strategic alignment model, and this static \\nmodel relies on assumptions that are no longer valid), and the \\nresearch question (How do organizations achieve strategic \\nalignment in practice?).\\nLiterature Review: The second part of the article, titled “Theoreti­\\ncal Anchoring and Analytical Framework,” describes the model \\nthat predominates in the literature and why it is challenged. Then \\nANT is introduced as well as the important concept of translation, \\nwhich helped us relate, make sense of, and understand our \\nempirical data.\\n\\n\\nChapter 9    Writing Classic GT for Publication\\n143\\nMethodology: This part describes research settings and sam­\\nplings, data collection and coding, and our coding categories.\\nResults: The results include a descriptive part (the emerging \\nlandscape in the three firms investigated in depth) and our \\nconceptualization of seven dynamic needs alignments to \\nmaximize the chances of success of an IT project.\\nDiscussion: The theory that we propose (the translated strate­\\ngic alignment model [TSAM], which is a processual, non-\\nfunctionalist model) is discussed in this section together with \\nits contributions and how it addresses previous models’ short­\\ncomings, hence the gap filled in the literature, as well as the \\nmanagerial (practical) implications.\\nConclusion: The conclusion summarizes the work accom­\\nplished and how the research question was addressed in the \\narticle.\\nWe were familiar with the journal where the article was submitted \\nand knew that several members of the editorial board were grounded \\ntheorists. Despite this familiarity, we were still expected to deliver a \\ncertain standard format that could be easily read and understood by \\nany researcher or practitioner, whether or not they knew about GT. In \\ntruth, and as far as we have experimented, no research process that \\nleads to theorizing in rupture with existing literature is linear. How­\\never, top-tier journals expect our writing to be so. To complicate issues \\nfurther, they also expect a compelling story and significant results. An \\narticle, beautifully fitting the conventions and standards of a given \\nresearch field and journal but which does not have a compelling story \\nto tell, nor interesting results, stands little chance of being published in \\na top-tier journal.\\n\\t\\nWRITING CLASSIC GT USING MIXED DATA\\t\\nl\\nAnother major and delicate issue in writing classic GT is aiming to \\ndo so while using both qualitative and quantitative data and tech­\\nniques. Data, methods, and techniques have become, wrongly \\n(Walsh, 2014c), implicitly embedded in paradigms: quantitative data \\nand techniques in positivist hypothetical-deductive research that \\nuses a mathematical type of writing and corresponding symbols; \\nqualitative data and techniques in interpretive research that uses a \\nmore narrative style of writing. However, research is currently \\n\\n\\nWRITING AND EVALUATING CLASSIC GROUNDED THEORY\\n144\\nmoving beyond these extreme caricatures, and a solution has to be \\nfound to write mixed design research while retaining sense-making \\nfor the reader. Writing a compelling qualitative GT study toward \\npublishing it in a top tier journal is already difficult enough. Doing \\nso for a mixed GT study is an even greater challenge; it may, how­\\never, help greatly in the production of formal theories of great value \\nbecause these theories are often in rupture with existing literature \\n(Walsh, 2014c).\\nBeyond remaining within the established conventions of top tier \\noutlets and rationalizing the presentation of our work, we found \\nthrough trial and error that the easiest way to write the results of a \\n“mixed design GT” study is, in the results section of the work, to stick \\nas close as possible to the way the theory unfolded. Walsh (2014a) \\nmixed method grounded theory was written several times trying dif­\\nferent writing strategies before it was accepted in a top-tier journal. It \\nwas first submitted to what is considered a well-established and “tra­\\nditional” quantitative positivist journal. To have it be considered for \\npublication by such a journal, the qualitative data slices and the GT \\nstance were “obliterated.” It was, of course, rejected, but we were very \\nlucky to have a senior editor who went to the trouble of pointing out \\nweaknesses. Unsurprisingly, the weaknesses were positioned where \\nwe had obliterated qualitative data.\\nThe work was then submitted in full to a journal where we had \\npreviously published qualitative GT. When we submitted the article, \\nwe warned the editor in chief that the design was unusual and required \\nreviewers to be familiar with classic GT and at ease with both qualita­\\ntive and quantitative data and techniques. Here again, we were very \\nfortunate, as one of the reviewers challenged us to write our results as \\nclose as possible to the way our theory had unfolded. When we first \\nsubmitted it, we had tried to describe the results obtained from the \\nquantitative data collected and the techniques applied, then the quali­\\ntative ones, and finally the theory that had unfolded. But this did not \\nwork because both qualitative and quantitative data, methods, tech­\\nniques, and analyses were embedded. Next we tried mixing them since \\nthey had truly been embedded in the reality of our research journey, \\nbut we lost the reader because our explanations were too complicated. \\nFinally, in the third round of revision, we used a subterfuge: We listed \\nour theoretical propositions and explained how we had reached each \\nof them using mixed data, ­\\nmethods, and techniques. To be easily \\nunderstood by the reader, we talked and wrote quanti (using in the text, \\nnumbers, acronyms to ­\\ndesignate the variables, and mathematical \\n\\n\\nChapter 9    Writing Classic GT for Publication\\n145\\nsymbols) when we were describing quantitative data and techniques, \\nand we talked and wrote quali (using in the text nothing but words and \\nfull extended names for the variables) when we were describing quali­\\ntative data and techniques. In other words, we used established \\n­\\nphilosophical caricatures to make our work understood. Then, the fact \\nthat we used both qualitative and quantitative data and techniques to \\ndevelop some ­\\npropositions did not hinder understanding.\\nThis article (Walsh, 2014a) was the follow-up to a previous qualita­\\ntive grounded theory that had already been published (Walsh et al., \\n2010), so with the primary data set collected for this second work, we \\nused as secondary data what had been collected and used in the previ­\\nous study. But we could not say so as, otherwise, reviewers could have \\nidentified us and this would have jeopardized the double blind review­\\ning process. We resolved this issue by using our previous work as data \\n(obtained from existing literature). For further information on method­\\nological aspects of this article, the reader can refer to Appendix C and \\nan example of the mixed writing described earlier. The introduction of \\nelements extracted from the literature as data is provided in the follow­\\ning boxed text.\\nUsing the Literature as One of the Sources of Data\\nProposition 2: Individual IT culture has a positive influence on contextual \\nIT needs.\\nWalsh et al. (2010) found that the users who were highly IT\\n-accultur­\\nated also had high levels of contextual IT needs. When we investigated \\nthe path IITC à CONITNEE, we found β = 0.400, significance: p < 0.001. \\nHowever, this path was investigated with quantitative data obtained within \\nan academic context and a highly-computerized organization. Some ele­\\nments resulting from our qualitative data set alerted us. For instance, \\ninterviewee P4 is a young assistant professor with a couple of previous \\nteaching experiences and who has started a new job in “X” school; she \\ngrew up in a computerized home environment with a father who is an \\nIT engineer. She told us: I certainly could not do my job in ‘X’ school without \\nIT. This was not the case in my previous institution, although I personally feel I \\nneed IT to do my job as a professor [ . . . ] for instance, it’s the first time I have \\nbeen asked to use PowerPoint slides to teach. In my old school, nobody ever \\nused PowerPoint. Thus, this professor did not perceive the same contextual \\n(Continued)\\n\\n\\nWRITING AND EVALUATING CLASSIC GROUNDED THEORY\\n146\\nl\\nl\\nWRITING CONVENTIONS\\nIT needs when she taught in an old-fashioned, IT\\n-conservative university \\nas when she teaches in a forward-looking, highly-computerized university. \\nThis more generally suggests that the same person with a given level of \\nIT acculturation may perceive different contextual IT needs, depending on \\nthe organization they work in.\\nIn the wording of the items for the construct of contextual IT needs . . . it \\nis quite clear that we specify the organization that we refer to. \\nTherefore, in \\nthe present study, the influence of the organization is included in the con­\\nstruct of contextual IT needs and explains why this path was quantitatively \\nsupported; it might not be so in a different organization and proposition 2 \\nmight not extend to other substantive areas and other contexts. (Walsh, \\n2014a, p. 155)\\nWriting, like power, is a “sociohistorical construction” (Richardson, \\n1990, p. 51), and we inscribe values in our writing. Therefore, we can­\\nnot improvise if we want our research to be published in the top jour­\\nnals of our research fields. It is a long learning process. “Styles of \\nwriting . . . are not fixed or neutral but reflect the historically shifting \\ndomination of particular schools or paradigms” (Richardson, 1990, \\np.  16). For instance, when we started publishing our GT works, we \\nfound that GT appeared to be more easily accepted in European jour­\\nnals of our research field even though these journals were also well \\nrecognized in North America. This guided our choice of relevant out­\\nlets for our work. This was even more the case when we were trying to \\npublish mixed design GT that used mixed qualitative and quantitative \\ndata. This might be explained by the fact that Paul Lazarsfeld’s works, \\nto which GT is anchored, were, at the time, extremely well accepted in \\nEurope, where they have led to an important school of thought in \\n­\\nstatistics, led by Benzécri (1973).\\nThe situation is currently1 evolving, and North American journals \\nare being made more and more aware of what GT may bring to \\nresearch, and its use has been increasing exponentially during the past \\ntwo decades (Figure 9.1). Some fields, for instance, medicine and nurs­\\ning, have widely recognized the added value of GT (Figure 9.2). Other \\n(Continued)\\n\\n\\nChapter 9    Writing Classic GT for Publication\\n147\\nMedicine (40.1%)\\nOther (11.7%)\\nUndefined (2.3%)\\nEngineering (3.5%)\\nArts and Humanities (3.9%)\\nBusiness, Management (8.3%)\\nHealth Professions (6.4%)\\nComputer Science (8.7%)\\nPsychology (13.4%)\\nNursing (22.4%)\\nSocial Sciences (30.8%)\\n1500\\n1250\\n1000\\n750\\nDocuments\\n500\\n250\\n0\\n1974\\n1980\\n1986\\n1991\\n1996\\n2001\\n2006\\n2011\\nFigure 9.1    \\u0007\\nNumber of GT Works Published per Annum (All Research \\nFields Included)\\nFigure 9.2    \\u0007\\nGT Works Published by Subject Area\\nfields, such as the management/business field (Figure 9.2), are still \\nlagging behind, although a significant increase in the number of GT \\npapers published during the past decade can be witnessed even within \\nthese fields (Figure 9.3).\\n\\n\\nWRITING AND EVALUATING CLASSIC GROUNDED THEORY\\n148\\nl\\nl\\nSOME GUIDELINES FOR PUBLISHING \\nCLASSIC GT IN TOP TIER RESEARCH JOURNALS\\n1991\\n140\\n120\\n100\\n80\\n60\\nDocuments\\n40\\n20\\n0\\n1997\\n2002\\n2007\\n2012\\nFigure 9.3    \\u0007\\nGT Works Published in the Business Research Field\\nIf you are an experienced researcher who has already published a num­\\nber of works, you probably know your research field and its writing \\nconventions fairly well. Then, you might try to publish your GT work in \\nany research outlet and “fight” for it to be recognized even though it \\nmight not be congruent with the historical writing conventions of the \\noutlet that you have chosen. The “fight” might be long and arduous. If, \\nhowever, you are a novice researcher, you might not be prepared to take \\nthis chance. Here we provide some guidelines for novice researchers.\\nPresent your ongoing works in conferences and to colleagues \\nwithin and outside your institution, within and outside the GT research \\ncommunity, and as often as you can, before submitting to a journal. \\nEncourage and accept criticism. This will help you improve your work \\nand maximize your chances of being published.\\nWhen you think you are ready to submit to a research outlet, study \\nthe top tier journals of your field thoroughly, as well as their writing con­\\nventions and the philosophical assumptions of the editorial board. For \\ninstance, if you pick a journal that has always published 100% quantitative \\npositivist research, it is completely unlikely that a qualitative interpretive \\ngrounded theory will be accepted in this journal. On the contrary, if you \\npick a journal renowned for its qualitative or mixed methods research, \\n\\n\\n\\nChapter 9    Writing Classic GT for Publication\\n149\\nyou might have a chance proposing an interpretive qualitative grounded \\n­\\ntheory or a critical realist mixed design grounded theory, respectively.\\nOnce you have identified the journal(s), select articles that are \\nmeaningful and interesting to you and use them as models of the \\nexpected writing conventions. This does not mean that you reproduce \\nthem but, rather, that you use them as models of the type of writing \\nthat is expected and commonly accepted in the chosen outlet(s). \\nRewrite your work to fit the identified writing conventions as long as \\nthis does not cause you to forfeit some of your own basic philosophical \\nassumptions or “disguise” your research beyond what you deem \\nacceptable. If this is the case, choose another outlet.\\nIf you are lucky enough not to receive a “desk reject,” your article will \\nbe sent to reviewers. Be patient and humble when you read the review­\\ners’ comments. All reviewers cannot be wrong. Reviewers are usually \\nchosen within a population of researchers who have already published in \\nthe outlet that you have chosen. These reviewers are usually specifically \\nappointed to review your work because of their competence in your \\nresearch area and sometimes for their methodological competence. \\nHence, if they question some aspects of your work, there is a fair chance \\nthat the readers of the outlet you have chosen to publish in will as well.\\nKeep in mind that reviewing is a thankless and unpaid task. Even \\nthough we have, on occasions (though extremely rarely), encountered \\nsome reviewers who did not understand our work, the other reviewers \\nappointed at the same time usually allowed the associate editor to \\nminimize the impact of the “bad” reviewer and our work was eventu­\\nally published. Do not be discouraged if and when your work is \\nrejected. This is part of the learning process, and we all learn much \\nmore through being rejected than accepted. As long as your work is \\nreviewed, even if it is eventually rejected, you can draw some valuable \\nlessons from this process and keep improving your output.\\nNot Quite Ready . . .\\nThe paper we gave as an example earlier in this chapter (Walsh et al., 2013), \\nwhose results were eventually voted by a scientific board as most important \\nfor both the research and practitioner communities, was ­\\npreviously rejected \\nby another top-tier journal.  \\nAt the time of this first submission and subse­\\nquent rejection, it just was not ready for publication and the research flow \\nwas still obviously difficult to follow. However, through this first ­\\nrejection, \\nand the detailed reviews that accompanied the rejection, we learned a lot \\nand were able to readjust our writing format to better suit a top-tier outlet.\\n\\n\\nWRITING AND EVALUATING CLASSIC GROUNDED THEORY\\n150\\nBe humble and do not hesitate to call in, as a second author, some­\\none with experience and know-how in a domain where your own \\ncompetence is lacking. If you choose this person correctly, what you \\nmight learn in the process of coauthorship will be extremely valuable \\nand might open some doors that you did not even know existed.\\nIN SUMMARY\\nDoing GT and writing GT with the goal of publishing in a research jour­\\nnal involve different processes. It is often necessary to organize the way \\nwe write our grounded theories to fit, at least approximately, the writ­\\ning conventions of the outlets in which we choose to publish. All \\nreviewers cannot be wrong, and we should remain humble. If one of our \\nworks submitted for publication is rejected, and if we want our works \\nto be read and understood, we should take the opportunity to learn \\nfrom, rather than refuse, the judgment of our peers.\\nTEST YOUR KNOWLEDGE\\nIn the following questions one or several answers may be correct.\\n1.\\t In order to publish a grounded theory in a research outlet\\na.\\t You should forfeit your philosophical assumptions\\nb.\\t You should unconditionally fit in the research outlet’s ­\\nwriting \\nconventions\\nc.\\t You do not have to show that you have followed all of GT’s \\nfoundational guidelines\\nd.\\t None of the above\\n\\t\\n2.\\t Writing GT for an audience\\na.\\t Has to be done through the GT process\\nb.\\t Has to faithfully describe the full GT process\\nc.\\t Has to be done in iterative cycles\\nd.\\t None of the above\\n\\t\\n3.\\t Writing conventions\\na.\\t Are the same for all research outlets\\nb.\\t Are all clearly stated in writing by research outlets\\nc.\\t May be identified partially through the publications of \\n\\neach outlet\\nd.\\t None of the above\\n\\n\\nChapter 9    Writing Classic GT for Publication\\n151\\n\\t\\n4.\\t Writing for an audience\\na.\\t Is always a linear process\\nb.\\t Accounts for findings in a mostly linear fashion\\nc.\\t Is usually helped by regular exposure of results to one’s peers\\nd.\\t None of the above\\nFURTHER READING\\nRichardson, L. (Ed.). (1990). Writing strategies: Reaching diverse audiences \\n\\n(Vol. 21). Newbury Park, CA: Sage.\\nRichardson, L. (2000) Writing: A method of inquiry. In N. K. Denzin & Y. S. \\nLincoln (Eds.), Handbook of qualitative research (2nd ed., pp. 923–948), \\nThousand Oaks, CA: Sage.\\nENDNOTE\\n1.\\t The source of the data used in this section is the Scopus database, \\nlaunched in 2004 by Elsevier. On its website, it claims to be “the largest \\nabstract and citation database of research literature and select results \\nfrom the web . . . [covering] 27 million abstracts, 230 million references \\nand 200 million web pages.” In this database, we investigated works \\nthat include the term grounded theory in the title, abstract, or keywords.\\n\\n\\n152\\n1 0\\nEvaluating Classic Grounded \\nTheory\\nAfter studying this chapter, you will:\\n• understand the criteria for evaluating classic grounded theory\\n• recognize the quality of a classic grounded theory by applying these \\ncriteria\\nD\\nifferent approaches to research necessitate different criteria for \\nevaluating the quality of the research outcomes. Establishing \\nsuch criteria is a somewhat contentious aspect of the shifting \\nnature of research paradigms.\\nl\\nl\\nTHE CONSEQUENCES OF THE PARADIGM WAR\\nPatton (2002) refers to the traditional dichotomy in research approaches \\nas competing inquiry paradigms: “using quantitative and experimental \\nmethods to generate and test hypothetical-deductive generalizations \\nand using qualitative and naturalistic approaches to inductively and \\nholistically understand human experience and constructed meanings in \\ncontext-specific settings” (p. 69). He acknowledges a range of qualitative \\nresearch approaches with alternative criteria and concludes that issues \\n\\n\\nChapter 10    Evaluating Classic Grounded Theory\\n153\\nof quality and credibility in qualitative research relate to three concerns: \\nrigorous methods, credibility of the researcher, and philosophical belief \\nin the value of qualitative inquiry (p. 584).\\nVenkatesh, Brown, and Bala (2013) propose an integrative \\nframework for mixed methods inference quality and provide several \\nexamples of validity in quantitative and qualitative research. For \\nquantitative research, they propose to check on design validity (i.e., \\ninternal and external validity), measurement validity (i.e., reliability \\nand construct validity), and inferential validity (i.e., statistical \\nconclusion validity); for qualitative research, they propose to check on \\ndesign validity (i.e., descriptive validity, transferability, and credibility), \\nanalytical validity (i.e., theoretical validity, dependability, consistency, \\nand plausibility), and inferential validity (i.e., interpretive validity and \\nconfirmability). However, their work still appears somewhat biased by \\nthe paradigm war that we described in Chapter 2; also it cannot be \\napplied “as is” to evaluate classic grounded theories. For instance, \\nexternal validity, that is, “whether the cause-effect relationship holds \\nover variation in persons, settings” (p. 13), does not hold for a classic \\ngrounded theory. External validity will depend on the researcher’s \\nphilosophical stance and if the investigated theory is a simple \\nsubstantive theory bounded in time, space, and context; a general \\nsubstantive theory that extends to a substantive area; or a conceptually \\nformalized theory that extends to many different substantive areas.\\nAs highlighted in Chapter 2 and discussed by Walsh (2014c) and \\nWalsh and colleagues (2015b), we must be careful not to enroll in \\nparadigmatic caricatures. We must not automatically assimilate the use \\nof quantitative data with hypothetical-deductive research or the use of \\nqualitative data with interpretive research. Guba (1990) refers to a \\nparadigm dialogue, noting a plethora of alternative approaches to \\nresearch that have emerged over the past decades to challenge \\nconventional positivism: “United in what they oppose, they are \\nnevertheless divided, sometimes sharply, on what they espouse” (p. 9). \\nTo engage further in this dialogue is beyond the scope of this text, but \\ngiven the extensive use of qualitative data in GT, novice researchers \\ncan find themselves attempting to justify their work in relation to a \\nmenu of criteria. As for the use of quantitative data, or mixed data, with \\nan exploratory stance, it is still so rare in many fields that researchers, \\neven experienced ones, who adopt this path may also find themselves \\nin need of justification for their work to be understood and validated. \\nTherefore, here, we focus our attention on exploring criteria for \\nevaluating the quality of a classic grounded theory with any type of \\ndata, qualitative and/or quantitative, and from any philosophical \\n\\n\\nWRITING AND EVALUATING CLASSIC GROUNDED THEORY\\n154\\nstance that a researcher may choose (although such a stance should be \\nclearly stated early on in a GT work to enlighten the reader and make \\nthe work transparent).\\nl\\nl\\nCRITERIA FOR EVALUATING GROUNDED THEORY\\nFirst, let us consider some general criteria for evaluating the quality of \\ntheory-building research. Bacharach (1989) contends that “the goal of \\ntheory is to diminish the complexity of the empirical world on the basis \\nof explanations and predictions” (p. 513). Eisenhardt (1989) reminds us \\nthat strong theory-building research should result in new, even frame-\\nbreaking, insights and that a good theory “should display enough \\nevidence for each construct to allow readers to make their own \\nassessment of the fit with theory . . . a strong theory-building study has \\na good, although not necessarily perfect, fit with the data” (p. 548). While \\nnoting that theory is always provisional, Bacharach (1989) cautions,\\nTo dangle criteria above the head of a theorist like the Sword \\nof Damocles may stifle creativity. . . . However, during the \\nearly stages of theory building, there may be a fine line \\nbetween satisfying the criteria of the internal logic of theory \\nand achieving a creative contribution. A good theorist walks \\nthis line carefully. (p. 513)\\nWhetten (1989) contends that “sensitivity to the competing virtues \\nof parsimony and comprehensiveness is the hallmark of a good \\ntheorist” (p. 490).\\nWhile appreciating the perspectives of these experienced \\nmethodologists, it is important to remember that GT’s distinct research \\nparadigm sits outside the “classically opposed orthodoxies” (Patton, 2002, \\np. 69). Our consideration should, therefore, begin by recalling the nature \\nand purpose of GT: to systematically generate theory from data as an integrated \\nset of concepts and as hypotheses about the interrelationships among these \\nconcepts that focus a main issue or concern in the area under study and explain \\nhow that issue or concern is processed or resolved. Glaser (1998, 2005a) refers \\nto GT as asymptotic and emphasizes its propositional nature; it never \\nfully reaches truth but the more data analyzed, the more plausible the \\ngenerated theory. As an exploratory approach, GT’s credibility cannot be \\njudged on the basis of established knowledge. Its relevance comes from \\ndata, not extant literature. “The important question is the usefulness of the \\ntheory that has been generated” (Baker, Wuest, & Stern, 1992, p. 1359).\\n\\n\\nChapter 10    Evaluating Classic Grounded Theory\\n155\\nO’Connor, Netting, and Thomas (2008) note the intent of grounded \\ntheory is emergent understanding, not testable theory, and suggest \\n“the easy advice for those facing an IRB [Institutional Research Board] \\nwould be to demonstrate the clearest standards available, those \\nsuggested by its original developers” (p. 36). Glaser and Strauss (1971) \\noffered four specific criteria for evaluating grounded theories: “theory \\nthat fits the real world, works in predictions and explanations, is \\nrelevant to the people concerned, and . . . is readily modifiable” \\n\\n(p. 176). Charmaz (2006, 2014) proposes the criteria of credibility, \\noriginality, resonance, and usefulness.\\nGlaser (1992, pp. 116–117) reiterates the four criteria of fit, \\nworkability, relevance, and modifiability and later elaborates, as \\nfollows, to define these terms and propose related questions that may \\nbe asked to verify the quality of a classic grounded theory:\\n1.\\t Fit is another word for validity. Does the concept adequately \\nexpress the pattern in the data which it purports to conceptual­\\nize? Fit is continually sharpened by constant comparisons.\\n2.\\t Workability means do the concepts and the way they are related \\ninto hypotheses sufficiently account for how the main concern of \\nparticipants is resolved?\\n3.\\t Relevance makes the research important, because it deals with the \\nmain concerns of the participants involved. To study something \\nthat interests no one really or just a few academics or funders is \\nprobably to focus on non-relevance or even trivia for the partici­\\npants. Relevance, like good concepts, evokes instant grab.\\n4.\\t Modifiability is very significant. The theory is not being verified \\nas in verification studies, and thus never right or wrong . . . it \\njust gets modified by new data to compare it to . . . New data \\nnever provides a disproof, just an analytic challenge. (Glaser, \\n1998, pp. 18–19)\\nIn exploring these criteria further, we might draw some comparison \\nof fit with notions of validity. As Glaser and Strauss (1967) state,\\nthe theory should provide clear enough categories and \\nhypotheses so that crucial ones can be verified in present and \\nfuture research; they must be clear enough to be readily \\noperationalized in quantitative studies when these are \\nappropriate . . . that the categories must be readily (not forcibly) \\napplicable to and indicated by the data under study. (p. 3)\\n\\n\\nWRITING AND EVALUATING CLASSIC GROUNDED THEORY\\n156\\nWe might compare GT’s criterion of workability with external \\nvalidity and transferability (i.e., the theory is sufficiently abstract to \\ntranscend specific empirical incidents for theoretical generalizability). \\nIndeed, replication is built into the research process through theoretical \\nsampling and constant comparisons (Glaser & Strauss, 1965a, p. 290).\\nGT’s criterion of relevance speaks to the practical value or utility of \\nthe theory (Patton, 2002, p. 588). Its relevance, credibility, and \\nconfirmability are, of course, affirmed through its empirical grounding \\nin data and the “conceptual grab” (Glaser, 1978, p. 100) of the theory. \\nConcepts convey credibility when they are both “analytic—sufficiently \\ngeneralized to designate the properties of concrete entities (not the \\nconcrete entities themselves)” and “sensitizing—they yield a \\n‘meaningful’ picture—with apt illustrations” so that the reader can \\n“feel that vicariously he has been in the field” (Glaser & Strauss, 1965a, \\np. 291). We might align criteria of resonance and usefulness (Charmaz, \\n2006, 2014) here as well. Bacharach (1989) suggests that “a theory is \\nuseful when it can both explain and predict” (p. 501). Glaser (1963) \\nsuggests that plausibility is sufficient in developing theory; that “to \\ndevelop, not test, a model, it is sufficient to explore plausible relations \\nbetween variables, and not necessary to build a strong case on hard \\nfact” (p. 381). Glaser and Strauss (1965a) suggest that any grounded \\ntheory can be “applied and adjusted to many situations with sufficient \\nexactitude to guide thinking, understanding and research” (p. 292).\\nGT’s emphasis on the modifiability of a good grounded theory is \\nperhaps unique in expressly acknowledging its propositional nature as \\ndistinct from the verificational nature of hypothetical-deductive theory \\nbuilding and from the detailed descriptive nature of many qualitative \\nresearch works. While contending that the primary goal of a theory is \\n“to answer the questions of how, when, and why” (Bacharach, 1989, p. \\n498), Bacharach insists that a theory that is not testable “no matter how \\nprofound or aesthetically pleasing it may be, it is not a theory” (p. 512). \\nIndeed, while insisting that the goal of GT is not hypothesis testing, \\nGlaser and Strauss (1967) maintain that “evidence and testing never \\ndestroys a theory (of any generality), they only modify it. A theory’s \\nonly replacement is a better theory” (p. 28).\\nTo the criteria of fit, workability, relevance, and modifiability we \\nwould echo Glaser and Strauss (1967) in noting the criteria of “logical \\nconsistency, clarity, parsimony, density, scope [and] integration” (p. 3) \\nas indicators of a good grounded theory. Density refers to levels of \\nconceptualization, development of theoretical properties, and overall \\nconceptual integration (Glaser & Strauss, 1971, p. 190). Bacharach \\n\\n\\nChapter 10    Evaluating Classic Grounded Theory\\n157\\n(1989) notes that “the purpose of theoretical statements is twofold: to \\norganize (parsimoniously) and to communicate (clearly)” (p. 496). \\nIndeed, Glaser and Strauss (1967) acknowledge the importance of the \\n“parsimony of variables” (p. 110) through GT’s delimiting the \\nemergent theory to a core category and related concepts. Eisenhardt \\n(1989) reiterates this call for parsimony, cautioning that in studies \\nwith a “staggering volume of rich data, there is a temptation to build \\ntheory which tries to capture everything. The result can be theory \\nwhich is rich in detail, but lacks the simplicity of overall perspective” \\n(p. 547). Bacharach (1989) defines scope as “the range of the \\nphenomena encompassed by the theory” (p. 509), suggesting that “for \\nadequate scope . . . the constructs must, in turn, sufficiently, although \\nparsimoniously, tap the domain of the phenomenon in question” \\n\\n(pp. 506–507).\\nIN SUMMARY\\nIn this chapter, we have explored the distinct criteria for evaluating a \\nclassic grounded theory that go beyond the accepted criteria for both \\nhypothetical-deductive and qualitative research. Classic GT’s criteria \\nof fit, workability, relevance, modifiability, and parsimony facilitate the \\ngeneration of theory with grab, or what Charmaz (2006, 2014) proposes \\nas originality. Of course, like all efforts at theory generation, the quality \\nof a classic grounded theory will vary according to the theorist’s under­\\nstanding and thoroughness in applying GT principles and techniques, \\nthe theorist’s ability to stay open to discovery in the data, the incisive­\\nness of analysis, and the clarity and elegance with which the theorist \\nwrites up his or her theory.\\nTEST YOUR KNOWLEDGE\\nIn the following questions one or several answers may be correct.\\n1.\\t Criteria for evaluating classic grounded theories\\na.\\t Include objectivity, generalizability, and reliability\\nb.\\t Are similar to criteria for interpretive research\\nc.\\t Emphasize modifiability and parsimony\\nd.\\t Include fit, workability, and relevance\\ne.\\t None of the above\\n\\n\\nWRITING AND EVALUATING CLASSIC GROUNDED THEORY\\n158\\n2.\\t In a good classic GT\\na.\\t The concepts express patterns in the data\\nb.\\t Emerging concepts and their relationships account for how \\nthe main concern is resolved\\nc.\\t The concepts have grab\\nd.\\t The theory may be modified by new data\\ne.\\t None of the above\\n3.\\t In a good classic GT\\na.\\t Replication is built into the research process\\nb.\\t The theory does not need to have practical value\\nc.\\t It is sufficient to explore plausible relations between ­\\nvariables\\nd.\\t It is necessary to build a strong case on hard facts\\ne.\\t None of the above\\nFURTHER READING\\nGlaser, B. G. (1978). Theoretical sensitivity: Advances in the methodology of grounded \\ntheory (Chapter 1). Mill Valley, CA: Sociology Press.\\nGlaser, B. G. (2003). The grounded theory perspective II: Description’s remodeling of \\ngrounded theory methodology (Chapter 9). Mill Valley, CA: Sociology Press.\\nGlaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies \\nfor qualitative research (Chapter 9). New York: Aldine de Gruyter.\\n\\n\\nConclusion\\nM\\nany authors have, over the years, written methodological \\nbooks about GT. In so doing, they have often reinterpreted \\nand “remodeled” GT (Glaser, 2003) from their own philo­\\nsophical perspectives. In this book, our purpose was to revisit classic \\nGT as a full package while remaining as true as possible to its founders’ \\nthoughts. We have reminded the reader of the roots of GT (Chapter 1) \\nand the ultimate purpose of GT, that is, discovering theories no matter \\nwhat the researcher’s philosophical assumptions may be (Chapter 2). \\nWe have highlighted GT’s foundational pillars (Chapter 3), clarified \\ncommon sources of confusion related to GT (Chapter 4), helped the \\nreader to find and use all available data (Chapter 5), analyze the data \\n(Chapter 6), shape the emerging theory (Chapter 7), and write as both \\nan intrinsic part of doing GT (Chapter 8) and also toward publishing \\nin top tier research outlets (Chapter 9). Finally, we have offered \\n\\nthe reader advice on evaluating the quality of grounded theories \\n(Chapter 10).\\nThroughout this book, we hope to have shown that GT is much \\nmore than a qualitative research methodology befitting only an inter­\\npretive philosophical stance, even though it may accommodate such \\na stance. We have aimed to provide practical guidance while staying \\ntrue to the classic GT practice. However, it must be highlighted that, \\nas our own philosophical stance is critical realist, it is most probable \\nthat, unwittingly we may have, at least partially, remodeled GT that \\nis supposed to be philosophically neutral. In the same way, the \\nframeworks proposed by Strauss and Corbin (1990) or Charmaz \\n(2014) are epistemologically and methodologically biased interpreta­\\ntions and applications of GT. We do not call for a reconsideration of \\nthe intrinsic worth of these works, which might suit the philosophi­\\ncal assumptions of some researchers. However, grounded theory \\nbeing a grounded theory itself as often stated by Glaser, it has to be \\nhighlighted that these frameworks are at a lesser level of conceptual \\n159\\n\\n\\nCLASSIC GROUNDED THEORY\\n160\\nformalization than that of classic GT (Glaser, 1978; Glaser & Strauss, \\n1967), a fact that is often misunderstood and which Glaser has been \\ndefending for many years.\\nTo give a few illustrative, non-limitative examples, Strauss and \\nCorbin (1990) and Charmaz (2009, 2014) propose frameworks that use \\nmostly qualitative data—the former with a symbolic interactionist \\nphilosophical background, the latter with a constructivist philosophical \\nbackground. Glaser (2008) uses quantitative data with a post-positivist \\nbackground. Walsh (2014c) uses mixed qualitative and quantitative \\ndata with a critical realist background. All these frameworks have \\nbeen empirically developed through doing and supervising various \\nresearch projects with emergent substantive GT research designs \\n(GTDs) and have probably and unwittingly remodeled, at least to \\nsome extent, classic GT, whereas Holton (2006, 2007b) uses qualitative \\ndata and develops a substantive GT design in her works, from a \\ncritical realist perspective in line with classic GT, without remodeling \\nit to become a different framework. This highlights the possibility of \\nviewing classic GT as a meta-theory of inductive research design \\n(Walsh, 2014c).\\nClassic GT\\n(Glaser, 1978; \\nGlaser and Strauss, 1967)\\nMixed data\\nMeta-Theory\\nof Research\\nDesign\\nGT\\nFrameworks\\nSubstantive\\nGT Designs\\n(GTDs)\\n“All is data”\\nQualitative data\\nStrauss and\\nCorbin’s (1990)\\nFramework\\nCharmaz’s\\n(2009)\\nFramework \\nQuantitative\\ndata\\nGlaser’s (2008)\\nFramework\\nGTD\\nGTD\\nGTD\\nLarsen’s\\n(2003)\\nGTD\\nWalsh et al.’s\\n(2010) Walsh and\\nGettler-Summa’s\\n(2010) GTD\\nGTD\\nWalsh's (2014c)\\nMixed Typological\\nFramework\\nEpistemological\\nBackground\\nSymbolic\\nInteractionism\\nConstructivism\\nPost-\\nPositivism\\nCritical\\nRealism\\nConceptual Formalization of GT\\nHolton (2006,\\n2007b) GTD \\nGTD\\nGTD\\nGTD\\nGTD\\nGTD\\nGTD\\nFigure 11.1.    \\u0007\\nClassic GT as a Meta-Theory of Inductive Research \\nDesign (Adapted from Walsh, 2014c)\\n\\n\\n\\t\\nWHAT IS GT, AND WHY THE CONTINUAL REMODELING?\\t\\nl\\nThis naturally leads to the question as to what classic GT truly is; a \\nquestion that has recently been openly asked (see Walsh et al., 2015a, \\n2015b, and the related special issue of Organizational Research Meth­\\nods). As we saw, many different answers to this question may be \\nproposed. In various fields of research, GT has been described as a \\ntechnique (e.g., Lawrence & Tar, 2013), a method (e.g., Amsteus, 2014), \\na methodology (e.g., Manuj & Pohlen, 2012), or a paradigm (e.g., \\nHolton, in Walsh et al., 2015b; Levina, in Walsh et al., 2015b; Rodrí­\\nguez-Martin et al., 2013). Based on Glaser (1978), Fernandez (in \\nWalsh et al., 2015b) describes GT “as a basic social process in which \\nresearchers engage with their data and participants in creating the­\\nory.” Bailyn (in Walsh et al., 2015b) sees GT as a “perspective on data \\nand on what one can learn from data.” Finally, Walsh (2014c; Walsh \\net al., 2015b) proposes GT as a meta-theory of exploratory research \\ndesign.\\nIs GT a method, a technique, a methodology, a framework, a para­\\ndigm, a social process, a perspective, or a meta-theory of research \\ndesign?\\nGT is probably all of these at the same time. As such, and \\ndepending on the researcher’s perspective, it will keep being remodeled. \\nHowever, if we go back to the four possible different definitions of the \\nword paradigm highlighted in the preface of this book, Morgan (2007) \\nshowed that all are a “shared belief system that influences the kinds of \\nknowledge researchers seek and how they interpret the evidence they \\ncollect” (p. 50). Therefore, it may certainly be stated that, because of its \\n“transcendent” quality (Glaser, 2004), classic GT is a research paradigm \\nbefitting the discovery of new theories from any philosophical \\nperspective.\\nConclusion \\n161\\n\\n\\nAppendix A\\nDoing Qualitative Research Does Not Mean You Are \\nDoing Classic GT\\nNote the precise, planned-in-advance sample; the attention to complete \\ndata capture through use of a preformulated interview guide, taping, \\nand transcribing; the abundance of codes generated from line-by-line \\ncoding of the transcripts; the logical and thematic organization of the \\nresultant codes; and the descriptive (past tense) reporting of findings.\\nExcerpt From the Study Design\\nI selected eight individuals from the community for personal \\ninterviews. Those selected included four males and four females; six \\nhad participated in the focus group at the May reunion while two—\\none male and one female—had been unable to attend. Interviews \\nwere semistructured in that I used a set list of questions but \\nencouraged the interviewees to take the conversation where they \\nwished. Each interview lasted from 60 to 90 minutes and was audio-\\ntaped and transcribed. In addition to line-by-line coding the eight \\ninterview transcripts, I also completed coding of the focus group \\ntranscript. This initial coding was free form and highly intuitive and \\nyielded 155 codes.\\nAfter reviewing my coding, some duplication was evident. I \\nreviewed the codes and grouped them to remove duplications, \\nreducing the list to 75 coded concepts. I then identified three \\noverarching categories that I labeled Aspects of Community, Aspects of \\nPractice, and Aspects of Participation. I explored the correlation of the \\nvarious codes to the three overarching categories and assigned \\n\\nl\\nl\\nQUALITATIVE DESCRIPTION DATA COLLECTION AND ANALYSIS1\\n162\\n\\n\\nDoing Qualitative Research Does Not Mean You Are Doing Classic GT\\n163\\n43 codes to the Aspects of Community category, 38 to the Aspects of \\nPractice category, and 24 to the Aspects of Participation category.\\nBeginning with the interviews, I focused my attention to the most \\nfrequent codes in each category. I selected those coded to at least 12 of \\nthe 18 interviews to represent a reasonable distribution of individual \\nmember perceptions within the data. Based on this criterion, I was able \\nto narrow the selection of codes from 75 to 42: 26 correlating to Aspects \\nof Community, 10 correlating to Aspects of Practice, and 6 correlating to \\nAspects of Participation.\\nOf the 26 concepts correlating to Aspects of Community, I designated \\nfive as subcategories—Community Identity, Bunch of Emotions, \\nConnectedness, Commitment, and Power of this Community. Under these \\nfive subcategories I thematically distributed the remaining 21 concepts. \\nSimilarly, under the category Aspects of Practice, I designated three \\nconcepts as subcategories—Allowed Me to Grow, Sharpening the Saw, and \\nPlay—and thematically distributed the remaining seven concepts \\namong these three. Under Aspects of Participation, I designated one \\nconcept as a subcategory—Participation in Community—and assigned \\nthe remaining five concepts to this subcategory.\\nExcerpt From the Study Findings\\nMembers referred to the reassurance they felt in being part of a tight-\\nknit family “able to get together, tell old war stories, create new \\nexperiences, make new friends” (A.M., Interview, October 2001) and “a \\nplace where egos and agendas don’t seem to get in the way of genuine \\ndialogue and caring for one another” (L.G., Interview, October 2001). \\nImages emerged of good times, celebrations, vivid memories, and \\npowerful connections. Shared stories and memories reinforced the \\ncloseness. As in a family, members described the sense of belonging as \\nimportant: “knowing that I can wander into the group 20 years from \\nnow and be accepted as one of the gang. I’ve found a community where \\nI believe I am liked as much for my loony-ness as for my creativity . . . I \\nguess that for me the community, the fun, and the belonging is \\nparamount” (J.M., Interview, October 2001).\\nOur findings suggest that community identity is forged through \\nshared experiences: “I know that my experience in the community has \\nshaped how I will live the second half of my life. I want to stay \\nconnected to others who have that same mid-life grounding point” \\n(M.S., Virtual Interview, October 2001). “It was a slog at times. . . . I was \\ngrateful that I have been through some tough times in my military \\ncareer, because that gave me the ability to see the slog as something \\n\\n\\nCLASSIC GROUNDED THEORY\\n164\\nthat bound us all together. We got through it because of the fun after \\nand during class as well as by having the stamina to suffer the late \\nnights, the lack of sleep in the residencies and the expectations on us to \\nproduce” (J.M., Interview, October 2001).\\nl\\nl\\nCLASSIC GT DATA COLLECTION AND ANALYSIS2\\nNote the tentative and emergent approach to sampling, the use of field \\nnotes to capture data, the use of constant comparison and theoretical \\nsampling to progress theorizing, the emergence of a core category to \\nexplain resolution of the main concern, and the use of theoretical \\nsaturation and theoretical coding to fully develop the emergent \\nexplanation. Note also that the theory is written conceptually and in \\nthe present tense.\\nExcerpt From the Study Design\\nI had initially projected my fieldwork to involve approximately \\n50 participants. In accordance with GT methodology however, \\nsample size could not be definitively established in advance of the \\nfieldwork but rather emerged as the study progressed. In summary, \\n61 individuals participated in the study: 27 through personal \\ninterviews and 34 individuals in focus group sessions. Interviews \\nlasted from 60 to 90 minutes each. The focus groups occurred during \\ntwo retreats, each lasting three days. These events were not organized \\nspecifically for my research efforts but were events to which I was \\ninvited by individuals who, because they had already participated in \\ninterviews, were aware of my research and felt that I might gain \\nfurther data for my study.\\nField notes were written up directly during and after each \\ninterview or focus group session. Consistent with the principles of \\nclassic GT development, field notes were collected and analyzed \\naccording to the constant comparison method of joint coding and \\nanalysis with subsequent data collection directed to theoretical \\nsampling as concepts emerged. In open coding of my data, I first \\ncompared incidents to incidents in the data, and I would ask, “What \\ncategory (concept) does this incident indicate?” “What property of \\nwhat category does this incident indicate?” and “What is the \\nparticipant’s main concern?” (Glaser, 1998, p. 140). These questions \\nenabled me to transcend the descriptive detail of the data and abstract \\nincidents within my data as indicators of the latent pattern of social \\nbehavior that would eventually emerge as a conceptual theory.\\n\\n\\nDoing Qualitative Research Does Not Mean You Are Doing Classic GT\\n165\\nAmong the substantive codes that emerged were changing workplace \\ncontext, coping with change and humanizing workplace. The important \\nthing is that each concept has earned relevance in relation to the theory, \\nits relevance theoretically sampled for and sufficiently validated and \\nits properties and dimensions identified though constant comparison \\nand interchangeable indicators to theoretical saturation. As my analysis \\nprogressed, rehumanizing emerged as a potential core category. It \\nappeared to account for much of the variation around knowledge \\nworker concerns with the changing knowledge workplace and resultant \\ndehumanization. Accordingly, rehumanizing became the focus of further \\nselective data collection and coding efforts.\\nSubsequent data collection and coding were thereby further \\ndelimited to that which was relevant to the emergent conceptual \\nframework (Glaser & Holton, 2004, para. 56–57). Selective data \\ncollection and analysis continued until the core and related categories \\nwere sufficiently saturated and further coding of new incidents did not \\nyield any new properties. It was time to cease collecting and coding \\ndata and to turn to conceptual integration and articulation of the \\nemerging grounded theory.\\nExcerpt From the Emergent Theory\\nThe instability of many workplace environments results in a loss \\n\\nof autonomy and even identity for many knowledge workers. \\n\\nThe concept of rehumanizing explains how knowledge workers \\nresolve their concerns with the dehumanizing impact of a changing \\nknowledge workplace, that is, how they restore the human dimension \\nin their work relationships and working environments. Rehumanizing \\ngives meaning to their work while sustaining energy and commitment. \\nRehumanizing is characterized by authenticity, depth and meaning, \\nrecognition and respect, safety and healing and kindred sharing. The \\nbureaucratic relationships in many knowledge workplaces lack \\nauthenticity. They are power-laden transactions. Whether the power \\ncan be attributed to position, influence or alliance, the tensional shifts \\ncan leave individuals feeling uneasy, anxious, and without a base or \\ngrounding. The need to play roles in organizations and to assume \\ncorporate identities also leads to inauthentic voices, disconnecting \\nindividuals from what they really feel. Identity and purpose in work \\nare eroded as “everyone plays the script.” With the increasing \\nintensification and specialization of knowledge, workers feel siloed. \\nThe time available for more general interactions is reduced. Removing \\nopportunities for broad interpersonal contact further dehumanizes the \\n\\n\\nCLASSIC GROUNDED THEORY\\n166\\nwork environment and reduces the potential for authentic engagement. \\nMany highly skilled knowledge workers risk vulnerability in moving \\nup in the organization to take on managerial roles that progressively \\ndistance them from their areas of specialized expertise. There is a risk \\nof vulnerability, as well, in emotional disclosures in the workplace. \\nIndividuals frequently feel they have to restrict emotional displays or \\nrun the risk of being seen as weak, a vulnerability that can restrict \\ncareer progression. Emotions are held in check and authenticity \\ncompromised.\\nENDNOTES\\n1.\\t Holton (2001). Unpublished case study.\\n2.\\t From Holton (2006).\\n\\n\\nAppendix B\\nClassic GT Using Qualitative Data\\nT\\nhe following is based on Holton (2006)1 and elaborates her \\napplication of classic GT methodology, in particular the process \\nof concept development, to her generation of a grounded theory \\nof rehumanizing knowledge work through fluctuating support networks in \\nthe knowledge workplace.\\n\\t\\nTHE DATA\\t\\nl\\nMy data consisted of field notes from personal interviews, casual \\nconversations with knowledge professionals, and observation of \\nvarious focus group gatherings. Participants were drawn from the \\npublic and private sectors and a variety of professional fields under the \\ngeneral rubric of knowledge work. In summary, 61 individuals \\nparticipated in the study: 27 through personal interviews (13 face-to-\\nface interviews, 12 virtual interviews via e-mail, and 2 telephone \\ninterviews) and 34 individuals in focus group sessions. In addition, I \\nencountered opportunities in the field to augment the interview and \\nfocus group data with participant observations of informal networks as \\nwell as document analysis. Data were analyzed using the full \\ncomplement of procedures that comprise classic grounded theory \\nmethodology. In accordance with classic GT, concepts emerged through \\nthe process of constant comparison of interchangeable empirically \\ngrounded indicators.\\n167\\n\\n\\nCLASSIC GROUNDED THEORY\\n168\\nMy thesis explains the basic social structural process of fluctuating \\nsupport networks through which knowledge professionals self-organize \\nto overcome dehumanized work environments as well as the social \\nstructural conditions that precipitate their formation, namely, the \\nchanging knowledge workplace, coping with change, resistance to change, \\ndehumanization, and rehumanizing. Such networks operate outside the \\nformal organization. They are self-emerging, self-organizing, and self-\\nsustaining. Participation is voluntary and intuitive. The core category \\nis rehumanizing, a three-stage basic social psychological process through \\nwhich knowledge professionals engage in finding and likening, igniting \\nof passions, and mutual engagement, giving meaning to their work while \\nsustaining their energy and commitment. Rehumanizing restores \\nauthenticity, depth, and meaning to their work; it offers them recognition \\nand respect, safety and healing, and kindred sharing as they cope with the \\nturbulence that inevitably accompanies persistent and unpredictable \\norganizational change. In the process, they are reenergized, their passion \\nfor work is reignited, stimulating creativity, the desire to embrace \\nchallenge, experimentation, and innovation.\\nl\\nl\\nTHE EMERGENT THEORY\\nl\\nl\\nSTAYING OPEN\\nEarly in my research process, I had to learn the hard way that my \\nresearch interest was not the main concern of the knowledge \\nprofessionals that I was interviewing. Like many novice researchers \\nwishing to do GT, I initiated data collection focused on a preconceived \\nprofessional interest. I developed an interview protocol and diligently \\ntaped and transcribed the first five interviews. As I sat transcribing, I \\nwas struck not only by how much time the transcribing was taking but \\nalso by how dull the interviews were. While my interviewees were \\ncompletely respectful of my research and dutifully tried to answer each \\nquestion posed, it was clear to me that they had little genuine interest \\nin my topic. We were all doing our best, but I found myself despairing \\nand anxious as to how any of this was going to produce a theory! My \\nearly efforts at open coding of these data, heavily influenced by \\nqualitative data analysis procedures, had yielded 155 substantive \\ncodes—data overwhelm!\\nI went back to further study GT methodology, particularly the work \\nof Barney Glaser (1978, 1998), and realized that my approach was \\nprivileging my professional interest and not allowing the main concerns \\n\\n\\n    Classic GT Using Qualitative Data\\n169\\nof my participants to surface. So I set aside those first five interviews and \\nstarted over to collect new data without an interview protocol and without \\ntaping. My first two “new” interviews were with managers in a public \\nsector organization. Rather than launching into a long preamble about my \\nresearch interests, I simply asked them to tell me about their jobs. I listened \\nfor what my participants wanted to tell me. I followed up with the \\nquestion, “What concerns you most in your role?” That’s all it took to open \\nthe “flood gates.” In each case, the interviewees started voicing concerns \\nabout organizational change and how the nature of change was impacting \\ntheir ability to manage the motivation and productivity of their staff. I \\njotted down key words and phrases during each conversation (e.g., \\nfragmented knowledge pockets, structures eroding, people marginalized, \\nhow to free people up, opening people up to possibility).\\nImmediately after each meeting, I rushed to a nearby coffee shop \\nand started jotting down more detailed field notes using those key \\nwords and phrases to remind me of what had emerged in our \\nconversation. What these managers wanted to talk about wasn’t what \\nI had set out to research, but I trusted the process. I coded my field \\nnotes and memoed. I struggled to let go of my preconceived research \\ninterest and simply to trust what was emerging from the data, but as I \\ncontinued collecting and coding data, this main concern (changing \\nworkplace context) emerged over and over.2\\n\\t\\nOPEN CODING\\t\\nl\\nWith my more open field notes, and using the constant comparison \\nprocess, open coding led to a much reduced list of 57 substantive codes. \\nIt is worth noting that this latter list of codes is substantially smaller and \\nmore conceptual than that produced in my earlier efforts. This illustrates \\nnot only the potential for overwhelming the novice researcher with \\ndescriptive data capture but also the very real possibility of losing the \\ntrue conceptual power of classic GT to transcend the descriptive level \\n(Glaser, 1978, pp. 49–52; 1998, pp. 23–26).\\n\\t\\nCONSTANT COMPARISON\\t\\nl\\nThe purpose of constant comparison is to see if the data support and \\ncontinue to support emerging categories. At the same time, the process \\nfurther builds and substantiates the emerging categories by defining \\ntheir properties and dimensions. When comparing incidents to incidents \\nin the data, I would ask, “What category (concept) does this incident \\n\\n\\nCLASSIC GROUNDED THEORY\\n170\\nindicate?” “What property of what category does this incident indicate?” \\n“What is the participant’s main concern?” (Glaser, 1998, p. 140). These \\nquestions enabled me to transcend the descriptive detail of the data and \\nabstract the incidents within as indicators of latent patterns of social \\nbehavior that would eventually emerge as a conceptual theory.\\nEach concept is developed through constant comparison of \\ninterchangeable (empirical) indicators in the data. The number of \\nindicators per category is not as significant as the requirement to \\nsample sufficiently to achieve theoretical saturation. The important \\nthing is that each concept has earned its relevance in relation to the \\ntheory—relevance that is theoretically sampled for and sufficiently \\nvalidated (theoretically saturated).\\nExample of the Constant Comparative Method\\n1. Comparing Incident to Incident\\nIn coding my interviews, I noted several references like these:\\n. . .  \\na place where she found \\n“like minds,” common passion, and where there \\nwas a building of trust . . .  \\nauthentic connection . . .  \\nThrough her subsequent \\nparticipation, she is “getting to really know others. . . . Building connections”\\n“ . . . way of hearing quiet voices in organizations” . . . authentic voices in \\norganizations rather than corporate voices\\n. . . corporate relationships are not authentic, they are transactional/\\npower related . . . Where do we find the experience of being ourselves in \\norganizations? Where do we learn to be human?\\n. . . having a chance to “be heard,” to share with others who understand, \\nwho they can trust to help and to hold confidences\\n. . . the corporate model that most of us work in now squeezes out our \\nhumanity. We develop machine relationships—even odd corporate voices \\n. . . manner of speaking . . . a “dead sound” where our real personality has \\nbeen excluded, as has emotion and feeling\\nC. and R. bring that wonderful warmth and spontaneity to the \\ncommunity—nothing can be dull when they’re present—lots of depth and \\nexperience there—looking forward to getting to know both much better\\nBy comparing incident to incident, I developed the concept of authenticity.\\n2. Comparing Concept to More Incidents\\nAs I continued to code my data, I began comparing new incidents in the data \\nwith my concept of authenticity. I wrote memos to conceptually elaborate \\nmy ideas about the concept. Memos such as the following:\\n\\n\\n    Classic GT Using Qualitative Data\\n171\\n(Memo 1) The instability characteristic of current organizational \\nenvironments results in a loss of autonomy and even identity for many \\nknowledge workers. Power shifts leave individuals feeling “at sea”—uneasy, \\nanxious, without a base or grounding.\\n(Memo 2) The prevalence of the “machine” metaphor in organizations \\nhas dehumanized work environments. Many workers actually begin to view \\nthemselves as “cogs” in a wheel and begin to respond accordingly with \\nautomatic, mindless behaviors that further disconnect them from their \\nwork. As one interviewee said, “It appears that the corporate model that \\nmost of us work in now squeezes out our humanity. We develop machine \\nrelationships—even odd corporate voices . . . a manner of speaking a ‘dead \\nsound’ where our real personality has been excluded as has emotion and \\nfeeling.” Another interviewee commented, “This machine world is causing us \\nto become ill and depressed . . . We act in this impersonal and unreal way in \\nour whole lives. We even act like this to ourselves and no longer have a real \\nrelationship with ourselves. How can we learn and experience being human \\nagain? \\nWhat is the essence of being human? It is surely to hear our real voice.”\\n(Memo 3) The need to play roles in organizations—to assume corporate \\nidentities leads to inauthentic voices. This creates disconnect between what \\nindividuals really feel and what they feel they must voice. It erodes identity and \\nconnection with purpose in work. Individuals put on armor to protect and \\nto project. After a while nothing feels real—people may long for a return to \\nconnecting with what they value/what they are passionate about. \\nThey long for \\na place to be real, to feel safe to express their authentic selves, to explore, to \\nplay, to risk . . . a place where they can relax, release tensions, open up.\\n(Memo 4) Fluctuating networks offer a respite from time crunched, \\ntransactional, efficiency-driven work environments where there is little \\ntolerance or time for the curiosity, exploration, and fun. What is being \\nsacrificed? Not only the potential for creativity and innovation but the very \\nhumanity-centered nature of vocation—the joy in creating and problem \\nsolving through collaborative engagement. The Xerox technicians story \\n(Orr, 1990) . . . On their own time, they found a way to integrate the social \\nand technical aspects of their vocation. They raised the solving of work-\\nrelated problems to an art. \\nValuable, tacit knowledge was transferred almost \\neffortlessly among the technicians in an open environment where knowledge \\nsharing was the power—building individual reputations (“war stories”) while \\nenhancing organizational performance in service to clients. The exercise \\nraised the identity of \\n“master technicians,” provided a process for recognizing \\nand sharing expertise, bringing new technicians into a culture of sharing and \\nlearning—a culture of mutual benefit—an inclusive meritocracy.\\n(Continued)\\n\\n\\nCLASSIC GROUNDED THEORY\\n172\\nl\\nl\\nCONCEPTUAL MEMOING\\nl\\nl\\nCORE CATEGORY EMERGENCE\\nl\\nl\\nTHEORETICAL SAMPLING\\nIn classic GT, conceptual elaboration and integration are facilitated \\nthrough an extensive and systematic process of memoing that parallels \\nthe data analysis process. Memos are theoretical notes about the data \\nand the conceptual connections between categories written up during \\nthe coding and analysis process as the ideation of the emerging \\ntheory’s conceptualization. During the course of the constant \\ncomparison process, I wrote more than 400 memos capturing the \\ndevelopment of my theory.\\nI started collecting data in 2001. Rehumanizing emerged as a potential \\ncore category in September 2003. As a concept, it appeared to account \\nfor much of the variation around knowledge professionals’ concerns \\nwith the changing knowledge workplace and resultant dehumanization.\\nAt this point, rehumanizing became the focus of theoretical sampling. \\nSubsequent data collection and coding was thereby delimited to that \\nwhich was relevant to the emergent core and related concepts, \\nfacilitating their conceptual elaboration through intense property \\ndevelopment.\\nI theoretically sampled for authenticity to further elaborate the concept \\nin terms of its properties. For example, openness and honesty, trust, sharing, \\nbeing heard, safety, power, balance, and so on.\\n3. Comparing Concept to Concept\\nI began to compare authenticity with other concepts that were emerging, \\nwriting memos about the relationships to develop hypotheses that linked/\\nintegrated these concepts. Eventually, through hand sorting of memos, \\nauthenticity would emerge as a property of my core category, rehumanizing.\\n(Continued)\\n\\n\\n    Classic GT Using Qualitative Data\\n173\\nSelective data collection and analysis continued on 38 substantive \\ncodes until the core and related categories were sufficiently saturated \\nand further coding of new incidents did not yield any new properties. \\nThe research process had achieved theoretical saturation of the core \\nconcept—rehumanizing—and its related categories through constant \\ncomparison and interchangeability of indicators (Glaser, 2001, \\n\\npp. 191–193). Further coding and constant comparison yielded no new \\nconceptual ideation. It was time to cease collecting and coding data \\nand turn to conceptual integration and articulation of the emerging \\ngrounded theory.\\n\\t\\nSELECTIVE CODING AND THEORETICAL SATURATION\\t\\nl\\n\\t\\nTHE CONCEPTUAL ELABORATION OF IGNITING PASSIONS\\t\\nl\\nTo illustrate the process of conceptual development, I offer here a \\ndetailed account of the emergence and theoretical saturation of one \\nconcept, Igniting Passions, the central stage in the basic social \\npsychological process of rehumanizing.3 This same process was applied \\nto each concept that emerged as relevant in relation to the core category \\nof rehumanizing.\\nIgniting Passions is the catalytic middle stage in the basic social \\npsychological process of rehumanizing, facilitating the symbiotic \\nrelationship, continuous amplification, and interdependent functioning \\nof rehumanizing’s subcore processes of finding and likening and mutual \\nengagement. The dynamic capacity of igniting passions sustains the \\noverall rehumanizing process by continuously generating confidence, \\nenergy, commitment, and bonding among network members. A total of 62 \\nspecific indicators of the concept of igniting passions appear in data \\ncollected from 15 interviews and two focus groups conducted between \\nOctober 2001 and January 2004. A total of 22 memos were written \\nbetween November 2002 and April 2004 elaborating the concept of \\nigniting passions. The memos range in length from several lines to \\nseveral pages.\\nAs each incident can be an indicator of more than one concept, \\nseveral of the memos correspond to more than one category or concept. \\nTherefore, to facilitate hand sorting the memos (the final stage in \\nconceptual integration of the theory), the corresponding categories are \\nprinted in bold type across the top of the memos. Each memo appears \\nonly once in the final theoretical integration of the theory, but its \\n\\n\\nCLASSIC GROUNDED THEORY\\n174\\nappropriate placement was not known until all concepts had been fully \\nsaturated. I was ready to begin hand sorting of the memos for emergent \\ntheoretical codes.\\nThe category passion appeared in open coding of data collected as \\nearly as October 2001. Early memos capture the dimensions of feedback \\nand recognition and learning as being positively correlated with igniting \\npassions for network engagement.\\nConceptual Memos\\nRecognition Needs: Feedback and recognition fuel alignment of individual \\nand organizational passions\\nReciprocating Passions: Learning and knowledge sharing . . . Being selfish in \\nterms of learning . . . Someone commented . . . \\n \\n“When we are (selfish), we learn \\nto give to others—greedy to share . . .” “vessels” for fulfilling greedy learning/\\nsharing needs—reciprocal fulfillment of passions for earning and sharing. \\n\\nThe accidental learning presented is a “smorgasbord” that rekindles the \\ndesire to learn—it also restores the learner’s control and builds autonomy \\nself-select, sample, take what’s desired and needed. Restless spirits sated by \\ncontinuous learning opportunities—integrated learning and practice\\nPower \\nin \\ndiversity: \\n \\n \\nDiversity \\nof \\npersonalities, \\ntalents, \\nperspectives, \\nexperience—\\ncreates a bubbling energy and creative potential. \\nThe range of contributions . . . \\n\\nis immense—energy, passion, strength, vulnerability, commitment, opening up, \\nconnecting, authenticity, tenacity, validation, modeling, confirming, facilitating . . . \\n\\nThis heady emotional/transformational release. . . . This may be applicable to \\nnetworks and communities like _____ because of the nature of their domain/\\npractice leadership, learning, organizational culture. . . . It will be interesting to \\nsee whether any of this feeling shows up in their networks.\\nAdditional dimensions of igniting passions emerged as data collection \\nand analysis continued with igniting passions linked to learning as well \\nas to vocation and to enhancing one’s professional skills.\\nConceptual Memos\\nThe Passion of Vocation: Networks as keeping personal and professional \\npassions from being eroded, depleted in the hectic, humdrum of daily \\norganizational operations. “Our job is our work . . . our practice is our passion.” \\nDistinguishing between “practice” and “work”—between “vocation” and “job.”\\n\\n\\n    Classic GT Using Qualitative Data\\n175\\nPassionate Learning:  “really start to learn when they find a passion for a \\nsubject and then make a real connection to another learners and real time \\npractitioners.” Individual passion for learning is stimulated and reinforced \\nin community. Further data collection and analysis reveal the dimension of \\nalignment and suggest that networks can assist in enabling individuals to \\nalign their personal purpose with their work. As such, passions are ignited \\nthrough purposeful engagement.\\nPassion and Purpose: Networks may be a venue/vessel for individuals \\nto be able to bring their passion back into their organizational roles—\\nexperiences of participation speak of their reconnecting with previous \\n“dream jobs,” core values, interests, and passions—this desire to reconnect \\npassion and purpose may be a motivation for seeking/participating in \\nnetworks/communities of practice.\\nSustaining Passion: Networks as a way to find the people with a passion \\nfor organizational work . . . this is a simple principle of self-organization \\nat work—allocating resources to follow the passion. She talks of the \\norganizational challenge of getting past “assigned work” to matching \\nassignments with passions–may not be worth the effort of trying to \\norchestrate this (i.e. manage the process)—that it may be better to let \\nthings take their course and be satisfied with “60%.” . . .\\nLetting the system self-organize . . . organic . . . “if no one has passion, \\nwe should not be doing it!” . . . organizations can use networks to \\nmonitor “passions” and set strategic directions accordingly . . . The \\nhuman dimension—passion ignites passion—creates understanding and \\nengagement—networks as a forum to mobilize passion into action . . . \\nbringing diverse talents, skills together—mobilized by shared passion—\\nvision—system sorts itself/self organizes to achieve action—progress the \\nwork—enhance the practice.\\nEngagement: Concept of engagement coming up a great deal lately in \\na lot of the literature/press on workplace issues. Article in the Globe and \\nMail this week (March 5) by Barbara Moses titled  \\n“Embrace the Challenge: \\nBe Engaged” on this issue noting importance of both satisfaction and \\nengagement as key determinants of employee retention in the current \\nenvironment. Another article in the Globe and Mail (same day, by Tara \\nWohlberg) on spirit in the workplace. One of the great advantages of the \\n“knowledge age” is the seemingly endless potential for combining uniquely \\npersonal talents and passions into an unlimited array of new careers/roles. \\nAt the same time, we are suffering a \\n“hangover” from the old mental models \\nof the Industrial era which focus on efficiency, productivity, and competition \\n(“push” factors/stressors).\\n\\n\\nCLASSIC GROUNDED THEORY\\n176\\nA longer memo begins to integrate emergent conceptual ideas.\\nConceptual Memos\\nIgniting Passions: “. . . . passion. This is when you are really working with \\npeople—not just a “file” that you are responsible for.” In work involving \\n“the human dynamic” it is essential to come together face to face—there \\nis a cost benefit to doing so if it engages the passion of organizational \\nmembers—we speculate about whether perhaps organizations should \\nallocate resources on the basis of matching passions! . . .\\n“Try to develop a passion meter” . . . networks as a way to find the \\npeople with a passion for organizational work . . . this is a simple principle \\nof self-organization at work—allocating resources to follow the passion.\\nOrganizational challenge of getting past “assigned work” to matching \\nassignments with passions—may not be worth the effort of trying to \\norchestrate this (i.e., manage the process)—that it may be better to let \\nthings take their course and be satisfied with “60%” . . . Letting the system \\nself-organize . . . organic . . . “if no one has passion, we should not be doing it!” \\n. . . organizations can use networks to monitor “passions” and set strategic \\ndirections accordingly . . . The human dimension—passion ignites passion—\\ncreates understanding and engagement—networks as a forum to mobilize \\npassion into action . . . bringing diverse talents, skills together—mobilized \\nby shared passion vision—system sorts itself/self organizes to achieve \\naction—progress the work—enhance the practice.\\nCore Group: These were the individuals who really had a passion . . . \\nto go beyond the technical. They were “kindred spirits”—passionate. They \\nare still friends even though most are no longer working in the training \\ndepartments. . . . The task is done, but the community remains. They still \\nhave a passion for the work (learning) and are still personally connected. \\nThey stay in touch regularly—still goes to them when she \\n“needs boosting.” \\nThere is one or two whom she uses as “sounding boards” and as sources \\nof information and contacts.\\nBuilding Community as Central Value: Like minds, common passions, \\nrespect for diversity, trust—authentic connection—reciprocating—building \\ncapacity, strength, common good.\\nPower in Diversity: Diversity of personalities, talents, perspectives, \\nexperience—creates a bubbling energy and creative potential. The range \\nof contributions is immense—energy, passion, strength, vulnerability, \\ncommitment, opening up, connecting, authenticity, tenacity, validation, \\nmodeling, confirming, facilitating . . . Practice is “deep,” heartfelt, engaging, \\n\\n\\n    Classic GT Using Qualitative Data\\n177\\ntransformational for individual and the group as a whole. . . . The fractal \\neffect of networks—just imagine the potential to be able to release this \\nwithin organizations! \\nThis heady emotional/transformational release . . . \\nThis \\nmay be applicable to networks like these because of the nature of their \\ndomain/practice leadership, learning, org culture . . . It will be interesting to \\nsee whether any of this feeling shows up in networks in other domains—\\ntechnology networks, etc. . . . Need to sample for this . . .\\nThe Passion of Vocation: Networks as keeping personal and professional \\npassions from being eroded, depleted in the hectic, humdrum of daily \\norganizational operations. “Our job is our work . . . our practice is our passion.” \\nDistinguishing between “practice” and “work”—between “vocation” and job.”\\nPassionate Learning: “really start to learn when they find a passion for a \\nsubject and then make a real connection to other learners and real time \\npractitioners.” Individual passion for learning stimulated and reinforced in \\ncommunity.\\nReciprocating Passions: Learning and knowledge sharing . . . Being selfish in \\nterms of learning . . . Someone commented . . .”  \\nWhen we are (selfish), we \\nlearn to give to others—greedy to share.”\\nNetworks as a “vessel” for fulfilling greedy learning/sharing needs \\nreciprocal fulfillment of passions for learning and sharing. The accidental \\nlearning presented in Networks is a “smorgasbord” that rekindles the \\ndesire to learn—it also restores the learner’s control and builds autonomy \\nself-select, sample, take what’s desired and needed. Restless spirits sated \\n\\nby continuous learning opportunities—integrated learning and practice.\\nPoised to Learn: Individuals must be self-motivated to learn—have an \\nurge, desire to learn—be “greedy” to learn. Someone commented . . .” \\n“When you’re ready to learn, a teacher will appear . . .” Networks brings \\neager, “poised to learn” together with ready teachers.\\nPassion and Purpose: Networks may be a venue/vessel for individuals \\nto be able to bring their passion back into their organizational roles. \\nExperiences of network participation speak of their reconnecting with \\nprevious “dream jobs,” core values, interests, and passions—this desire \\nto reconnect passion and purpose may be a motivation for seeking/\\nparticipating in networks.\\nTime Challenge: Struggle to find time to participate—juggling priorities—\\ntime vs. needs—geographic distance adds further challenge.\\nA conceptual memo written January 5, 2004, goes back to earlier \\ndata for further analysis.\\n\\n\\nCLASSIC GROUNDED THEORY\\n178\\nConceptual Memo\\nPassion, Resistance, and Bonding:  Rereading field notes from interview with \\nA . . . , noted the many references to passion, in particular, the connection \\nbetween passion and bonding of network members. Appears that the \\ncommon passion that brings network members together—part of the \\nlikening that creates a network—is also the “glue” that bonds network \\nmembers. She goes on to describe the “passionate few” as bonding due to \\nthe resistance they encounter from the formal system—“the resistance \\nserves as a way to separate out those who really have a passion to keep \\nworking” . . . So . . . passion creates likening, resistance creates bonding and \\nreinforces passions. . . . a cyclic process that sustains member engagement in \\nfluctuating networks.\\nFurther data and memoing suggest that passion may be a key \\nvariable in work performance with passion being positively correlated \\nto performance and disconnected passion blocking performance.\\nConceptual Memo\\nDisconnected Passion: Individuals who exhibit resistance to change or appear \\nto be risk averse can become labeled in organizations as “non-performers.” \\nThey may be shunned by others—particularly by managers. This response \\nfurther isolates the individual and contributes to lower performance—a self-\\nfulfilling prophecy.  \\nWhat may lie at the heart of this superficial assessment of \\nnon-performance is a disengagement of the individual from their passion or \\npurpose in work. \\nThis disengagement may be a defense or coping mechanism \\nin response to increasing dehumanization from changing workplace context \\nor a botched or poorly executed change effort. Individuals are \\n“having trouble \\ntrying to reconnect.” They are “disillusioned about the organization and \\ntrying to cope with the change” around them. \\nThey’ve become \\n“disconnected \\nfrom their work.” If efforts are made to listen past the disengagement and \\nnon-performance, one can often find that there was a time when these \\nindividuals “had a real hunger for their work.” What they require is a chance \\nto reconnect with their passion, to feel depth and meaning in their work.\\nThey “need to be whole” to have their “eyes light up” again, to feel the \\nenergy of a new challenge and new learning. This isn’t happening in many \\norganizations because “managers are not listening.” The result of their \\ninattention is the veritable blocking of employee capability and desire in \\ntheir work. The lack of opportunity leaves individuals feeling powerless.\\n\\n\\n    Classic GT Using Qualitative Data\\n179\\nIn February 2004, I moved from open coding of data to selective \\ncoding for the core category (rehumanizing) and related concepts. I went \\nback to data that I had collected earlier in my research. Although these \\ndata had been collected using an interview protocol (early in my \\nresearch and before I fully understood the importance of staying open) \\nI did use it—all is data—to selectively code for indicators of rehumanizing \\nand related categories—including Igniting Passions. My analysis of \\nthese earlier data (collected between December 2002 and November \\n2003) yielded the following conceptual ideation.\\nConceptual Memo\\nIgniting Passions: Maturity of network members enhances purposeful \\nengagement and facilitates their kindred sharing. Maturity may reduce \\nopportunism and transactional interactions. Mature individuals may \\nappreciate the value of authentic connections and kindred sharing more \\nthan those with less experience in the work world. Reference to second \\ncareers may also be reference to more passionate engagement in the focus \\nof the network—where their hearts are—their passions engaged more so \\nthan in their “day jobs” in the formal organization.\\nContextual dimensions of network engagement energy, fun, \\nachievement, tension emerged from further data analysis as positively \\ncorrelated with passion.\\nConceptual Memos\\nDepth and Meaning in Igniting Passions: There’s no pretense—no \\nshowiness—in network interactions. Members are there for a purpose. \\nThere’s depth and meaning in their participation—“problem solving, sorting \\nout . . . let’s have some discussion on this.” Real purpose and value (meaning) \\nto those engaged and at the same time, it’s fun. They enjoy the action, the \\ninevitable debates. It stimulates the environment. It enables members to \\n“eyeball people and get stuff sorted out” so there’s real value—progress is \\nmade in work.\\nIgniting Passions: Pressured environment creates an energy that ignites \\npassion. Unpredictability creates energy—fast pace creates excitement and \\nstimulates networking. \\nWhile there’s anxiety created in this environment, it’s\\n(Continued)\\n\\n\\nCLASSIC GROUNDED THEORY\\n180\\nan enjoyable anxiety—it’s edgy, stimulating. . . . “What excites me is the pace of \\nit. I like that, I like the energy it gives me and the energy it creates so I like it fast \\nnot slow. . . . I think it’s very important. It feels like a life! Ahh . . . sometimes it’s too \\nmuch for sure but I’ll take any day of too much than too little.”\\nChallenge emerged as another dimension of igniting passions.\\nConceptual Memo\\nIgniting Passions: Passions are ignited by challenge—the “against all odds” \\nsyndrome—finding mutual commitment to a goal that others consider \\nimpossible or crazy. Setting themselves apart from the “masses,” the \\nordinary—taking on a challenge and making it work—high achievement \\norientation—success is sweeter when shared. Believing in the impossible \\nand then making it happen. There’s a charge in being challenged and being \\ncreative in solving an issue, a problem that ignites passionate engagement \\nwithin a network—draws members in.\\nLikening facilitates the challenge and achievement dimensions of \\nigniting passions.\\nConceptual Memo\\nIgniting Passions/Likening/Sustaining: “Newness” is part of likening—the \\nchance to meet others who not only share a sense of energy but who \\nare also open to what others have to offer—the chance to “show and \\ntell” interests, share knowledge and skills—creates an energy and ignites \\npassions—can feel out of control at times but in a good way—skills and \\nabilities are stimulated by joint problem solving and collaborative creativity \\n(see Bennis book on creative collaboration . . . ). Moving past challenge to \\nachievement also very important. Novelty likens but achievement sustains.\\nWhile igniting passions creates bonds, it can also create insularity.\\n(Continued)\\n\\n\\n    Classic GT Using Qualitative Data\\n181\\nConceptual Memo\\nIgniting Passions/ Sustaining: Passions are not always positive—they can also \\ninvolve spirited outbursts of anger. This is particularly the case when the \\ncore group of a network have developed such a close group identity that it \\ncompromises their relationship with others in the external environment—\\ninsularity leading to intolerance—impacts upon ability of the network to \\nfunction within the larger external environment of the formal organization. \\nInteractions become personalized and highly emotional—core becomes \\nsegregated—trust erodes and threatens sustainability . . . network members \\nmay limit/reduce their participation if they feel it jeopardizes their position \\nwithin the formal organization—cannot risk the consequences.\\nChallenge and the risk associated with it emerged again as a \\ndimension of igniting passions.\\nConceptual Memo\\nIgniting Passions: Challenge and the chance to achieve ignite passions. The \\nhigher the stakes, the greater the passion. That feeling of being on the \\nedge—risking it to overcome and achieve holds a special kind of magic—a \\npull that can be very empowering. The fear of failure mixed with the high \\nof success.\\nPassion ignites through an initial spark when individuals connect \\nand liken.\\nConceptual Memo\\nFinding/Igniting Passions: A network forms and fluctuates around an initial \\nspark—this may be a person who takes the lead in identifying and \\nconnecting people of “like minds.” Once like minds find each other, they \\nquickly self-identify and self-organize. Their mutual passion creates the \\nenergy for perpetuating interactions and sustained mutual engagement.\\n\\n\\nCLASSIC GROUNDED THEORY\\n182\\nKindred sharing facilitates igniting passions and sparks creativity, \\nlearning, and growth.\\nConceptual Memo\\nKindred Sharing/Creativity/Igniting Passions/Learning: Kindred sharing involves \\nventing—the chance to “blow off steam”—within a safe and empathetic \\nenvironment. But kindred sharing moves past venting to strategizing—a \\nproactive and empowering response to dealing with organizational issues—\\nproblems and frustrations. It’s exploring ways to overcome the hurdle \\nand move forward. It engages collective minds to addressing issues—the \\ncollective response not only enables a better response but it also builds \\nmomentum and support for moving forward. There’s arguing in kindred \\nsharing—those outside the network might see this as negative but it’s \\npart of sparking the creativity and problem-solving power of the collective \\nmindset in a network. It’s part of the fluctuating nature of the network. \\nEnergy builds through fluctuating emotions—passions ignite and creativity \\nis sparked. The outcome is learning and growth.\\nIgniting passions wards off boredom and stimulates the desire for \\nnetwork participation.\\nConceptual Memo\\nIgniting Passions: Boredom mitigation drives networking—igniting passions \\naccelerates network engagement and participation. Work environments \\nthat exude boredom challenge K.W. to seek avenues to sustain energy and \\nward off dehumanization.\\nAs the concept reaches theoretical saturation, a further memo \\nattempts to integrate the conceptual ideas that have emerged, organizing \\nthem into conditions, dimensions, and consequences of igniting passions.\\nConceptual Memo\\nIgniting Passions: “. . . . passion. This is when you are really working with \\npeople—not just a ‘file’ that you are responsible for.” In work involving “the \\nhuman dynamic,” it is essential to come together face to face—there is a \\n\\n\\n    Classic GT Using Qualitative Data\\n183\\ncost benefit to doing so if it engages the passion of organizational members—\\nwe speculate about whether perhaps organizations should allocate resources \\non the basis of matching passions! . . .\\n“Try to develop a passion meter.” Informal networks as a way to find the \\npeople with a passion for organizational work . . . this is a simple principle \\nof self-organization at work—allocating resources to follow the passion. \\nOrganizational challenge of getting past “assigned work” to matching \\nassignments with passions—may not be worth the effort of trying to \\norchestrate this (i.e., manage the process)—that it may be better to let \\nthings take their course and be satisfied with “60%” . . . Letting the system \\nself-organize . . . organic “if no one has passion, we should not be doing it!” \\nOrganizations can use networks to monitor “passions” and set strategic \\ndirections accordingly. The human dimension—passion ignites passion—\\ncreates understanding and engagement—Informal networks as a forum to \\nmobilize passion into action . . . bringing diverse talents, skills together—\\nmobilized by shared passion—vision—system sorts itself/self organizes to \\nachieve action—progress the work—enhance the practice.\\nConditions: Energized Environment: Pressured environment creates an \\nenergy that ignites passion. Unpredictability creates energy—fast pace \\ncreates excitement and stimulates networking.  \\nWhile there’s anxiety created \\nin this environment, it’s an enjoyable anxiety—it’s edgy, stimulating. . . .” \\n \\n“What \\nexcites me is the pace of it. I like that, I like the energy it gives me and the \\nenergy it creates so I like it fast not slow. . . . I think it’s very important. It feels \\nlike a life! Ahh . . . sometimes it’s too much for sure but I’ll take any day of \\ntoo much than too little.” There’s no pretense—no showiness—in network \\ninteractions. Members are there for a purpose. There’s depth and meaning \\nin their participation—“problem solving, sorting out. . . . .let’s have some \\ndiscussion on this. . . .” Real purpose and value (meaning) to those engaged \\nand at the same time, it’s fun. They enjoy the action, the inevitable debates. It \\nstimulates the environment. It enables members to “eyeball people and get \\nstuff sorted out” so there’s real value—progress is made in work.\\nInitial Spark: \\nA network forms and fluctuates around an initial spark—this \\nmay be a person who takes the lead in identifying and connecting people \\nof “like minds.” Once like minds find each other, they quickly self-identify \\nand self-organize. Their mutual passion creates the energy for perpetuating \\ninteractions and sustained mutual engagement.\\nBoredom Mitigation: Boredom mitigation drives networking—igniting \\npassions accelerates network engagement and participation. Work \\nenvironments that exude boredom challenge K.W. to seek avenues to \\nsustain energy and ward off dehumanization.\\n(Continued)\\n\\n\\nCLASSIC GROUNDED THEORY\\n184\\nDimensions: Recognition: Passion is positively correlated to recognition. \\nFeedback and recognition fuel alignment of individual and organizational \\npassions.\\nPersonal Alignment: Informal networks may be a venue/vessel for \\nindividuals to be able to bring their passion back into their organizational \\nroles. Experiences of network participation speak of their reconnecting \\nwith previous “dream jobs,” core values, interests, and passions—this \\ndesire to reconnect passion and purpose may be a motivation for seeking/\\nparticipating in networks. Maturity of network members enhances \\npurposeful engagement and facilitates their kindred sharing. Maturity may \\nreduce opportunism and transactional interactions. Mature individuals may \\nappreciate the value of authentic connections and kindred sharing more \\nthan those with less experience in the work world. Reference to second \\ncareers may also be reference to more passionate engagement in the focus \\nof the network—where their hearts are—their passions engaged more so \\nthan in their “day jobs” in the formal organization.\\nPurposeful Engagement: Informal networks as keeping personal and \\nprofessional passions from being eroded, depleted in the hectic, humdrum \\nof daily organizational operations. . . . “Our job is our work . . . our practice \\nis our passion.” Distinguishing between “practice” and “work”—between \\n“vocation” and “job” . . . “. . . passion.  \\nThis is when you are really working with \\npeople—not just a ‘file’ that you are responsible for.” In work involving “the \\nhuman dynamic,” it is essential to come together face to face—there is a cost \\nbenefit to doing so if it engages the passion of organizational members—we \\nspeculate about whether perhaps organizations should allocate resources on \\nthe basis of matching passions! . . .  \\n“Try to develop a passion meter.”\\nNetworks as a way to find the people with a passion for organizational \\nwork . . . this is a simple principle of self-organization at work—allocating \\nresources to follow the passion. Organizational challenge of getting past \\n“assigned work” to matching assignments with passions—may not be worth \\nthe effort of trying to orchestrate this (i.e., manage the process)—that it \\nmay be better to let things take their course and be satisfied with “60%” \\n. . . Letting the system self-organize . . . organic—“if no one has passion, \\nwe should not be doing it!” Organizations can use networks to monitor \\n“passions” and set strategic directions accordingly. \\nThe human dimension—\\npassion ignites passion—creates understanding and engagement—networks \\nas a forum to mobilize passion into action . . . bringing diverse talents, skills \\ntogether—mobilized by shared passion—vision—system sorts itself/self \\norganizes to achieve action—progress the work—enhance the practice.\\n(Continued)\\n\\n\\n    Classic GT Using Qualitative Data\\n185\\nChallenge: Passion is positively correlated to challenge. Passions are \\nignited by challenge—the “against all odds” syndrome—finding mutual \\ncommitment to a goal that others consider impossible or crazy. Setting \\nthemselves apart from the “masses,” the ordinary—taking on a challenge \\nand making it work—high achievement orientation—success is sweeter \\nwhen shared. Believing in the impossible and then making it happen.\\nChallenge and the chance to achieve ignite passions. The higher the \\nstakes, the greater the passion. That feeling of being on the edge—risking it \\nto overcome and achieve holds a special kind of magic—a pull that can be \\nvery empowering. The fear of failure mixed with the high of success. There’s \\na charge in being challenged and being creative in solving an issue, a problem \\nthat ignites passionate engagement within a network—draws members in.\\nLearning: “really start to learn when they find a passion for a subject and \\nthen make a real connection to other learners and real time practitioners.” \\nIndividual passion for learning is stimulated and reinforced in community. \\nLearning and knowledge sharing. . . . Being selfish in terms of learning . . . \\nSomeone commented . . .  \\n“When we are (selfish), we learn to give to others—\\ngreedy to share . . .” Networks as a “vessel” for fulfilling greedy learning/sharing \\nneeds—reciprocal fulfillment of passions for learning and sharing. \\nThe accidental \\nlearning presented in networks is a “smorgasbord” that rekindles the desire to \\nlearn—it also restores the learner’s control and builds autonomy self-select, \\nsample, take what’s desired and needed. Restless spirits sated by continuous \\nlearning opportunities—integrated learning and practice. Individuals must \\nbe motivated to learn—have an urge, desire to learn—be “greedy” to learn. \\nSomeone commented . . .  \\n“When you’re ready to learn, a teacher will appear . . . \\nnetworks bring the eager, poised to learn together with ready teachers.”\\nConsequences: Likening:  \\nPassion is positively correlated to likening. \\n“Newness” \\nis part of likening—the chance to meet others who not only share a sense \\nof energy but who are also open to what others have to offer—the chance \\nto “show and tell” interests, share knowledge and skills—creates an energy \\nand ignites passions—can feel out of control at times but in a good way—\\nskills and abilities are stimulated by joint problem solving and collaborative \\ncreativity (see Bennis book on creative collaboration . . .). Moving past challenge \\nto achievement also very important. Novelty likens but achievement sustains.\\nRereading field notes from interview with \\nA. . . . , noted the many references \\nto passion, in particular, the connection between passion and bonding of \\nnetwork members. Appears that the common passion that brings network \\nmembers together—part of the likening that creates a network—is also the \\n“glue” that bonds network members. So . . . passion creates likening.\\n(Continued)\\n\\n\\nCLASSIC GROUNDED THEORY\\n186\\nBonding: She goes on to describe the “passionate few” as bonding due \\nto the resistance they encounter from the formal system—“the resistance \\nserves as a way to separate out those who really have a passion to keep \\nworking” . . . Resistance creates bonding and reinforces passions . . . a cyclic \\nprocess that sustains member engagement in fluctuating networks. Bonding \\nreinforces passions but excessive bonding creates insularity.\\nInsularity is negatively correlated to network sustainability. Passions are \\nnot always positive—they can also involve spirited outbursts of anger. This is \\nparticularly the case when the core group of a network have developed such a \\nclose group identity that it compromises their relationship with others in the \\nexternal environment—insularity leading to intolerance—impacts upon ability \\nof the network to function within the larger external environment of the \\nformal organization interactions become personalized and highly emotional—\\ncore becomes segregated, trust erodes and threatens sustainability . . . network \\nmembers may limit/reduce their participation if they feel it jeopardizes their \\nposition within the formal organization—cannot risk the consequences.\\nPerformance: Passion is positively correlated to performance. Individuals who \\nexhibit resistance to change or appear to be risk averse can become labeled \\nin organizations as “non-performers.” They may be shunned by others—\\nparticularly by managers. This response further isolates the individual and \\ncontributes to lower performance—a self-fulfilling prophecy. What may lie at \\nthe heart of this superficial assessment of non-performance is a disengagement \\nof the individual from their passion or purpose in work. \\nThis disengagement may \\nbe a defense or coping mechanism in response to increasing dehumanization \\nfrom a changing workplace context or a botched or poorly executed \\nchange effort. Individuals are “having trouble trying to reconnect.” They are \\n“disillusioned about the organization and trying to cope with the change” \\naround them. They’ve become “disconnected from their work.” If efforts are \\nmade to listen past the disengagement and non-performance, one can often \\nfind that there was a time when these individuals “had a real hunger for their \\nwork.” What they require is a chance to reconnect with their passion, to feel \\ndepth and meaning in their work. They “need to be whole,” to have their “eyes \\nlight up” again, to feel the energy of a new challenge and new learning. This \\nisn’t happening in many organizations because “managers are not listening.” \\nThe \\nresult of their inattention is the veritable blocking of employee capability and \\ndesire in their work. \\nThe lack of opportunity leaves individuals feeling powerless.\\nCreativity: Arguing is part of igniting passions and sparking the creativity \\nand problem-solving power of the collective mindset in a network. It’s part \\nof the fluctuating nature of the network. Energy builds through fluctuating \\nemotions—passions ignite and creativity is sparked.\\n(Continued)\\n\\n\\n    Classic GT Using Qualitative Data\\n187\\nTheoretical sampling of data collected earlier in the study (October \\nand November, 2001) provided additional indicators of several \\ndimensions of igniting passions that had emerged previously—fun, \\nchallenge, learning, energy, confidence, and bonding.\\nConceptual Memo\\nIgniting Passions: There’s a strong desire to continue to network once \\nindividual passions have been ignited. Passions are fueled by the desire to \\ncontinue to experience the energy and synergy that result from mutual \\nengagement—to work and learn and laugh together. There’s a strong sense of \\nfun, of pushing the envelope. The desire to continue to move the network \\nforward creates its own sense of excitement and fuels a passionate belief \\nin the ability to make a difference.\\nConceptual elaboration concludes when the relationships among \\nindividually elaborated concepts emerge through the identification \\nand use of relevant theoretical codes to achieve an integrated theoretical \\nframework for the overall theory. The theoretical codes that emerged in \\nthis study include basic social process (both psychological and \\nstructural processes) and amplifying causal looping. Igniting passions \\nearned a pivotal position within the basic social psychological process \\nof rehumanizing, the core category of fluctuating support networks, itself \\na basic social structural process.\\n\\t\\nCONCEPTUAL INTEGRATION AND WRITING UP\\t\\nl\\nOnce I had achieved theoretical saturation of my core concept and \\nrelated categories, I proceeded to review and hand sort memos regarding \\nthe core and its properties and dimensions. As I began to sort and look \\nfor relationships between the various concepts, theoretical codes began \\nto emerge as an abstract modeling of the latent structural patterns that \\ncould integrate and explain the emerging theory. The first indication of \\nemergent theoretical codes was memoed in December 2003.\\nMy study may be viewed as that of a basic social process (BSP)—coping \\nwith change (in this case, the substantive area is the knowledge workplace). \\n(Continued)\\n\\n\\nCLASSIC GROUNDED THEORY\\n188\\nAs the overarching BSP\\n, coping with change includes both a basic social \\nstructural process (BSSP)—fluctuating networks of professional concern4—\\nand a basic social psychological process (BSPP)—rehumanizing—which is \\nalso the core category, integrating and explaining the BSPP of rehumanizing \\nthrough the BSSP of fluctuating networks of professional concern. \\nRehumanizing can be viewed as a structural condition affecting the nature of \\nfluctuating networks of professional concern. These networks have always \\nbeen there in the workplace as they are inherent to social organization \\ngenerally, but today’s increasingly compressed and dehumanized work \\nenvironments (changing workplace context) have brought the need for \\nrehumanizing to the fore as a means of addressing the main concern of \\nthose involved—coping with change (thereby magnifying the BSPP of \\nrehumanizing as a structural condition of the BSSP of fluctuating networks). \\nAs such, the BSSP of fluctuating networks of professional concern has \\ntaken on the properties of the BSPP of rehumanizing, including authenticity, \\ndepth/meaning, respect, safety, healing . . . As a preliminary suggestion, the \\nstages in the BSPP of rehumanizing may be finding, likening, igniting passions, \\nkindred sharing, experimenting, bonding, and sustaining. Some of these may \\nbe combined as research progresses; new ones may be identified . . . the \\nstructural process (of fluctuating networks) is of significance because it \\nexplains the organization of behavior (as emergent informal organization) \\nto address the main concern of the participants—coping with change \\nwithin the workplace—through a BSPP (rehumanizing) as antidote to the \\ndehumanizing impact of traditional formal organizational structures. This is \\nstarting to feel \\n“right” for me—things are fitting into place and I can now see \\nan overall conceptual framework around which to begin building the theory.\\nWhile continuing to consider basic social process as an appropriate \\ntheoretical code through which to integrate my emerging theory, I \\nremained open to the emergence of other theoretical codes as I \\ncontinued to hand sort and integrate memos. A final integration of the \\ntheory occurred in March 2004 with the emergence of the theoretical \\ncode of amplifying causal looping. The final sorting and integration \\nfollowed, generating a theoretical outline (conceptual framework) for \\nthe full articulation of the theory.\\n(Continued)\\n\\n\\n    Classic GT Using Qualitative Data\\n189\\nENDNOTES\\n1.\\t It is important to remember that this is the work of a novice conducting \\nher first grounded theory study. It is, as such, an exemplar but is by no \\nmeans offered as a marker of excellence in doing classic GT.\\n2.\\t As so often happens, there was a connection back to my initial research \\ninterest in communities of practice, but my grounded theory explains \\nmuch more about what is really going on as managers and other knowl­\\nedge workers voluntarily seek out and engage in work-related commu­\\nnities and networks. My theory explains that the motivation for their \\nparticipation is not simply learning and practice enhancement. It is \\nmuch more about their efforts to overcome the dehumanization they \\nexperience in their work and their workplace context in the midst of \\npersistent and unpredictable change.\\n3.\\t My grounded theory is rather complex in that it is a “theory within a \\ntheory”—the overall theory of fluctuating support networks (a basic social \\nstructural process) with its core category of rehumanizing (a basic social \\npsychological process) of which igniting passions is the central stage in \\nthe process (the core category of the rehumanizing process).\\n4.\\t I later renamed this concept fluctuating support networks to more effec­\\ntively reflect their substantive role in providing support to network \\nmembers.\\n\\n\\nAppendix C\\nDoing Classic GT Using Mixed Qualitative and \\nQuantitative Data\\nT\\nhis appendix is based on excerpts from Walsh (2014a) and \\nelaborates on her application of classic GT while using mixed \\nqualitative and quantitative data and techniques. The full article \\n“A Strategic Path to Study IT Use Through Users’ IT Culture and IT \\nNeeds: A Mixed-Method Grounded Theory” may be found in the \\nJournal of Strategic Information Systems, 23(2), pp. 146–173. Copyright \\n2014. (Permission granted by Elsevier, License No. 3674860603800, \\n\\nJuly 23, 2015). Examples of the mixed qualitative and quantitative data \\napplied in this work may be found in Appendix D.\\nl\\nl\\n1. A MIXED-METHOD GROUNDED THEORY DESIGN\\nIn this section, we detail the research design that was not set at the \\nbeginning of our study; instead, this emerged as our research was \\nconducted, and it was guided by the emerging theory.\\n1.1 The GT Framework\\nIf one does not take into account any paradigmatic consideration, \\ntwo main distinct variants of the GT framework may be found in the \\nliterature. One of these variants is favored by Glaser and the other by \\nStrauss (Melia, 1996; Urquhart et al., 2009), with main disagreements \\ninvolving coding issues and the role of extant literature. Our stance is \\nclosest to Glaser’s but, beyond any difference of opinion between \\n190\\n\\n\\n    Doing Classic GT Using Mixed Qualitative and Quantitative Data\\n191\\nGlaser and Strauss over the years, and if one goes back to their original \\n(1967) common work, Urquhart et al. (2009) showed that four main \\ncharacteristics should be found in a GT study: (1) Emergence of both \\nresearch design and outcome must be rooted in data and not \\npreconceived/imposed on data; (2) constant comparative analysis of \\nthe data must take place (i.e., data are constantly compared to \\npreviously collected and analyzed data, looking for similarities and \\ndifferences to help toward conceptualization and theorization); ­\\n(3) there \\nmust be theoretical sampling (i.e., sampling is directed by the emerging \\ntheory and continues until the saturation of concepts, categories, \\nproperties, and relationships is reached); and (4) the ultimate aim of GT \\nmust be theory building, substantive or formal theory (i.e., discovering \\nand constructing, rather than verifying. theory is the end purpose of a \\nGT study).\\n1.2 A Mixed-Method Approach\\nIn his 2008 work about quantitative GT, Glaser uses Lazarsfeld’s \\nwork and his techniques of quantitative analysis for discovery. \\nLazarsfeld “insisted on the combining of qualitative and quantitative \\nanalysis” (Glaser, 2008, p. 6), leading to a mixed-method approach. \\n­\\nA  mixed-method design includes different quantitative and/or \\nqualitative methods, which combine with and supplement each other \\nwithin a single project. In our work, we did not use qualitative data \\nand methods to draw hypotheses that we verified through quantitative \\ndata and methods; this would be what we understand to be a multi-\\nmethod approach (Morse, 2003, pp. 189–208). In our research, neither \\nqualitative nor quantitative data/methods were sufficient in themselves \\nto theorize; all were necessary. This article combines qualitative and \\nquantitative data and research methods in a mixed-method GT \\napproach.\\nWe used both secondary data (from the existing literature) and \\nprimary data (collected for the present study). When we collected \\nprimary data (quantitative and/or qualitative), we did so as we needed \\nthem to move the emerging theory forward and also to allow for \\ntriangulation. Qualitative data were collected when we needed rich \\ndescription of the emerging relationships between concepts. \\nQuantitative data were collected when we needed to move back from \\ndetails and obtain a synthetic perspective. All data (qualitative and \\nquantitative) were analyzed as one set.\\n\\n\\nCLASSIC GROUNDED THEORY\\n192\\n1.3 Study Context, Sample, and Data Collection\\nThe software that we more specifically investigated is an open \\nsource e-learning software platform, Moodle (Modular Object-Oriented \\nDynamic Learning Environment), which may be used by both \\nprofessors and students in third-level education and training contexts. \\nWe investigated the use of this platform in a European business school, \\nwhere this software is mostly used as an exchange platform for \\nprofessors and students. Professors can (if they wish) upload their class \\nmaterials. Their students have direct and unlimited access to this \\nplatform and can download what they need. The platform also allows \\nstudents to submit assignments, and grading may be done online by \\nthe professors. It also provides facilities for instant messaging, and \\nonline news, announcements, and timetable details. The use of the \\nplatform is not institutionally mandatory, but its use is presented as a \\ngreat help for both students and professors in fulfilling their tasks; it is, \\ntherefore, strongly recommended by administrators for both professors \\nand students. “Well, it’s on Moodle, you can find it for yourself” is a \\nrecurring answer from some administrative staff when either students \\nor professors ask questions. Most full-time professors use it at least to \\nprovide course materials, and students may feel compelled to use it to \\ngain access to these materials, although they could also obtain them \\nfrom other students if they do not want to use the platform. Some \\nprofessors might, however, insist that students upload their homework \\non the platform in order to obtain grades.\\nOur sampling was guided by the emerging theory. The slices of \\nquantitative data were collected through surveys administered at the \\nend of a school year (June 2012). Therefore, the participants in these \\nsurveys had had the opportunity to directly access the investigated \\nexchange platform for a significant amount of time (at least two \\nsemesters) and had been offered voluntary training sessions at the \\nbeginning of each semester.\\nThe qualitative slices of data were collected through interviews \\nconducted between September 2011 and September 2012 with both \\nstudents and professors. As we were part of the institution investigated, \\nwe also used participant observation. We used Walsh et al.’s (2010) \\ntypological work to guide us in our theoretical sampling. We identified \\nusers that fit some of the ideal types highlighted by Walsh et al. (2010). \\nMore than the seven interviewees we report on were involved; we \\nretained for the present study and report on only those users whose \\ninformation technology (IT) culture profile we could identify with \\nsufficient precision and who were different in terms of age, nationality, \\nand background to allow for different perspectives. Professors who \\n\\n\\n    Doing Classic GT Using Mixed Qualitative and Quantitative Data\\n193\\nwere interviewed were colleagues that we worked with, sharing \\nclasses, chores, or research projects (but not the present one) with them. \\nStudents who were interviewed were those with whom we had an \\nabove-average number of exchanges as we supervised their master’s \\ndegree dissertations. Interviewees were treated as case studies rather \\nthan as simple interviewees. For the first slice of qualitative data, \\nseveral extensive interviews were conducted with each interviewee \\nbetween September 2011 and May 2012; notes were taken and memos \\nwritten up.\\nTwo surveys (one for professors and one for students)—very \\nsimilar but with some questions adapted to the different end-users—\\nwere put online, and e-mails were sent to faculty and students to invite \\nthem to respond in May and June 2012. E-mails were sent to 375 faculty \\nmembers (85 full-time professors and 290 part-time professors) and \\n1,076 students, and 84 professors and 198 students answered the \\nsurveys (the response rates were 22.4% and 18.4%, respectively). All \\nreceived answers were used, and no outliers were removed. We \\nverified our results with early and late respondents’ data with no \\nsignificant difference. This indicates that nonresponse bias should not \\nbe a problem in the quantitative data collected for the present study \\n(Armstrong & Overton, 1977).\\nQuantitative data were analyzed with the help of partial least \\nsquares (PLS) structural equation modeling (SEM), applied in an \\nexploratory manner. The transcripts of the qualitative data were hand-\\ncoded a number of times and sorted out on a table (in parallel with our \\nquantitative investigations. After we had started to explore our \\nquantitative slices of data, together with the qualitative data already \\ncollected, we conducted further interviews in July and September 2012 \\nwith the seven participants theoretically sampled for the present study, \\nin order to verify our interpretation of the qualitative and quantitative \\ndata already collected and to expose this to their perspectives. These \\nlast interviews with each interviewee were recorded and relevant \\nelements transcribed. During these interviews, and only then, did we \\nbriefly describe to interviewees details of the research that was being \\nconducted. We also introduced them to such terms as IT acculturation \\nor IT culture and the distinction we drew between the two terms IT \\nutilization and IT use, which follows Walsh’s (2010) definition of these \\ntwo terms.1 Had we not done so, and as these terms were part of our \\nown vocabulary, we risked biasing their answers through the way we \\nasked our questions. Also, because some respondents were of different \\nnationalities and spoke different native languages, some of the \\ninterviews were conducted in English and others in French. In order to \\n\\n\\nCLASSIC GROUNDED THEORY\\n194\\ntranslate adequately important words such as usage and utilization from \\none language to the other, the meaning of the terms used had to be \\nfully understood by respondents and our interpretations/translations \\nverified. Therefore, we chose to inform interviewees about the meanings \\nwe attributed to these terms in order to avoid misinterpretation on our \\npart of the qualitative data collected. Further notes were made and \\nmemos were again written up immediately after these last interviews \\ntook place as, in most instances, these final conversations continued after \\nthe interviews were recorded. Qualitative data were first open-coded on \\nconcepts (e.g., IITC, utilization, global IT needs, contextual IT needs, and \\nsituational IT needs) and then theoretically coded in terms of the \\nrelationships between these concepts (e.g., individual IT culture à \\nglobal IT needs, position à contextual IT needs, situational IT needs à \\nutilization). Secondary data (from the literature), primary data \\n(collected for the present study), quantitative data (collected through \\nQuantitative\\ndata collected\\nfrom surveys\\nconducted with\\nstudents and\\nprofessors (282\\nrespondents)\\nQualitative\\ndata collected\\nfrom participant\\nobservation\\nand interviews\\nwith seven\\nparticipants\\nSecondary Data \\n(Deci & Ryan, 1985; Ryan & Deci, 2000;\\nVallerand, 1997, 2007; Walsh, 2010;\\nWalsh et al., 2010)\\nPrimary Data\\nConstant\\nComparative\\nAnalysis\\nEmerging Theory\\nEmerging Theory\\nEmerging Theory\\nEmerging Theory\\nConstant\\nComparative\\nAnalysis\\nFigure C.1    Using All Our Data as One Set1\\n\\n\\n    Doing Classic GT Using Mixed Qualitative and Quantitative Data\\n195\\nsurveys), and qualitative data (collected through interviews and \\nparticipant observation) were analyzed together, and, in a critical \\nrealist stance, we opted for the relationships between concepts that \\nappeared to best fit our whole data set, while considering at the same \\ntime broader perspectives reaching beyond our substantive area, in \\norder to start driving toward formal GT.\\nIn order not to confuse the reader, and although many different \\npaths were investigated that are not reported here due to space \\nallocation, we present in the results section only the final models that \\nemerged from and were congruent with our whole data set. As \\nquantitative and qualitative investigations and analyses were done \\nsimultaneously, while taking into account all data as one set, and \\nconstantly comparing and analyzing all data as they were collected \\n(see Figure C.1), we justify our models with previous works from the \\nliterature, quotes from interviews, quantitative reports, or all three of \\nthese. Using mixed methods allowed us to avoid the “Texas sharpshooter \\napproach”2 (Biemann, 2012), a bias that plagues many quantitative \\nstudies and that results mostly from their authors’ quest for publication \\n(Biemann, 2012).\\nThroughout the present research, we openly remain in an \\nexploratory stance and iterate between all data. For instance, some \\nrelationships were investigated and confirmed through quantitative \\nmethods with our substantive quantitative data set, but we neither \\nretained nor reported on them as they did not make sense in our \\ncomplete data set. Conversely, some other relationships that were not \\ncompletely confirmed through our quantitative data are discussed, \\nbecause, based on information obtained through our qualitative data \\nset, they could be important for further research in other substantive \\nareas with different, and less specific, targeted sampling.\\n1.4 Quantitative Measures\\nAll research variables were quantitatively modeled as reflective \\nusing multi-item Likert-type scales (ranging from 1 = “not true at all” \\nto 7 = “completely true”), except for the variables POSITION, IITC, and \\nUTILIZATION.\\nReflective measures for the constructs global IT needs (GLOBITNEE), \\ncontextual IT needs (CONITNEE), and situational IT needs (SITITNEE) \\nwere developed specifically for the present study. To ensure content \\nvalidity, we used qualitative data from the interviews that were \\nconducted, as well as Walsh et al.’s (2010) qualitative GT work and its \\ndetailed online coding appendices. Reflective measures for ease of use \\n(EOU) and usefulness (U) were adapted from Davis (1989). Benbasat \\n\\n\\nCLASSIC GROUNDED THEORY\\n196\\nand Barki (2007) stressed the “need to make sure usefulness is measured \\nbeyond perceptions where possible” (p. 216). The perceptual aspect of \\nusefulness was minimized, as we studied usefulness and ease of use as \\nassessed by users after they had had time to be trained and to test and \\naccess the investigated software for a significant period of time.\\nThe control variable POSITION was dummy-coded (Professor = 1 \\nand Student = 2).\\nFor the IITC construct, we used as a starting point the measure \\nproposed by Walsh (2009) and verified by Von Stetten, Wild, and \\nChrennikow (2011), which we aimed at improving. We modeled IITC \\nas a first-order reflective, second-order formative construct and \\nfollowed the guidelines provided by Diamantopoulos, Riefler, and \\nRoth (2008). The first-order latent reflective variables we started with \\nwere accomplishment needs satisfied through IT usage (ACCNEE), \\naffiliation needs satisfied through IT usage (AFFNEE), primary needs \\nsatisfied through IT usage (PRIMNEE), power needs satisfied through \\nIT usage (POWNEE), intrinsic motivation to know IT (INTMOTKNO), \\nextrinsic motivation to use IT through identified regulation \\n(EXTMOTID), and extrinsic motivation to use IT through external \\nregulation (EXTMOTEX). The items of this last construct were worded \\nnegatively and had to be reverse-coded to obtain the new variable \\nREV_EXTMOTEX. However, Von Stetten et al. (2011) found that this \\nsub-construct, proposed by Walsh (2009), did not have a significant \\ninfluence on IITC. We verified and confirmed this point in our \\npreliminary investigations and did not include this sub-construct in the \\nfinal measurement model of IITC, as we found theoretical reason not to \\ndo so: If one considers the self-determination continuum proposed by \\nGagné and Deci (2005), all sub-constructs of IITC are on the positive \\nend of this continuum except for the questioned variable EXTMOTEX, \\nwhich rather appears to capture some elements of amotivation \\n(Pelletier et al., 1997). Even recoding and inverting this variable does \\nnot solve this theoretical issue, as it has not been proven that \\namotivation is the reciprocal function of motivation. This is why we \\nchose not to include this sub-construct in the IITC measurement model. \\nMeasures of the other dimensions were adapted to improve reliability, \\nand content validity was ensured through the study of various scales \\nin the psychosociological field of research, which we adapted for our \\npurpose.3\\nConcerning the assessment of use, there is a lack of consensus in \\nthe literature on how to measure system usage/utilization, and there is \\na vast array of measures for it, as highlighted by Burton-Jones and \\nStraub (2006). Wu and Du (2012) categorized the existing measures \\n\\n\\n    Doing Classic GT Using Mixed Qualitative and Quantitative Data\\n197\\nfound in the literature as actual usage, reported usage, and measured \\nusage. Actual usage (e.g., the measures used by Devaraj et al., 2008, or \\nVenkatesh et al., 2002) involves objectively collected measures such as \\ncomputer logs. The other two categories (reported and measured \\nusage) imply subjective self-reported measures. Reported usage (e.g., \\nthe measures used by Adams et al., 1992; Keill et al., 1995; Taylor & \\nTodd, 1995) involves measures of duration and frequency that aim at \\nobjectivity even though these measures are reported by the users \\nthemselves: for instance, the number of hours spent on a system per \\nday. Assessed usage (e.g., the measures used by Davis, 1989; Igbaria & \\nParasuraman, 1989; Karahanna, Agarwal, & Angst, 2006) involves \\nordinal measures of intensity and extent, the ordinal scale being the \\nmain element of differentiation of this type of assessment. Over \\n\\nthe years, other elements of differentiation that involve the user’s \\nsubjective cognitive abilities were added. As a result of their \\ninvestigations, Wu and Du (2012) propose the use of rich measures of \\nassessed usage through multiple dimensions that mix different \\nmeasures. Originally, we had wished to assess IT utilization as defined \\nby Walsh (2010), that is, “the actual, objectively assessed, use of an IT” \\n(p. 152). As actual measures extracted from computerized reports were \\nvetoed by our faculty, we opted for self-reported ordinal measures of \\nutilization. We included in our measure of utilization the number of \\nfunctionalities used that assessed the extent of utilization (UT_FUNCT) \\nand the frequency of utilization that assessed the intensity of utilization \\n(UT_FREQ). We also wanted to assess the type of utilization, that is, \\nwhether users’ perception of the platform led them to a self-determined \\nor constrained utilization (variable UT_TYPE). For instance, if users go \\non the platform “as little as possible” or only when they feel “it is \\nabsolutely necessary,” their utilization might be considered closer to \\nmandatory than voluntary, and with differing degrees of self-\\ndetermination. We therefore specifically developed for the present \\nstudy a three-dimensional formative measure of utilization that \\nincluded the dimensions UT_TYPE, UT_FREQ, and UT_FUNCT. The \\nconstruct UTILIZATION was specified as formative, since changes in \\nany of the three indicators would lead to changes in utilization but, \\nconversely, a change in utilization does not imply a change in each of \\nthe three indicators.\\nAll measures were pretested with 25 students in a postgraduate \\nclass and with a selected sample of 20 faculty members, in order to \\nobtain comments and feedback before the pilot test was finally \\nconducted. After the pretest, the wording of some items was readjusted, \\nand one item was deleted from three of the reflective scales (EXTMOTID, \\n\\n\\nCLASSIC GROUNDED THEORY\\n198\\nGLOBITNEE, and CONITNEE) in order to improve reliability; these \\nthree scales were, thus, reduced from three to two items each. As these \\nscales were for reflective constructs and we aimed at a parsimonious \\ninstrument, this was deemed acceptable.\\nThe measurement models, as well as the relationships between all \\nvariables, were analyzed using a PLS approach, and bootstrapping as a \\nresampling technique (500 random samples), to generate t-statistics \\n(Chin, 1998). We used the SMARTPLS software (Ringle, Wende, & Will, \\n2005). PLS analysis was preferred because it does not require data with \\na normal distribution (Fornell & Cha, 1994), and it supports both \\nreflective and formative constructs (Gefen, Rigdon, & Straub, 2000). It \\nwas also preferred because our research objectives were exploratory \\n(Gefen et al., 2011; Hair et al., 2012; Ringle et al., 2012). PLS analysis is \\nparticularly suitable in situations of high complexity, with low theoretical \\ninformation (Jöreskog & Wold, 1982): We had an important number of \\nvariables and little theoretical backup on the relationships we wanted to \\ninvestigate. Another advantage of the PLS approach is its ability to work \\nwith a small sample size (Ringle et al., 2012; Urbach & Ahlemann, 2010).\\nAs advised by Glaser (1978, 2008), we let our data “speak” and guide \\nthe emerging theory. The propositions highlighted could not have been \\ndeveloped if all data had not been interpreted as one set, and if both \\nqualitative and quantitative methods and analyses had not been \\napplied in combination.\\nOur stance is critical realist (Bhaskar, 1979, 1989, 1998, 2002). \\nAlthough a debate about critical realist precepts is beyond the scope of \\nthis book, before we present our models and propositions, it is essential \\nto clarify what causality is for a critical realist. The notion of causality \\nas a “generative mechanism” is a core and defining feature of critical \\nrealism (Bhaskar, 2002). Generative mechanisms are best understood as \\n“tendencies,” as their activation is highly context-dependent (Bhaskar, \\n2002). In contrast with the Humean vision of causality (“A causes B”), \\ncommonly accepted in information systems (IS) traditional quantitative \\npositivist circles, a generative mechanism can be reformulated as \\n­\\n“A  generates B in context C” (Cartwright, 2003; Smith, 2010). For a \\ncritical realist, causality is thus a process of how causal powers are \\nactualized in some particular context: a process in which the generative \\nmechanisms of that context (C) shape (modulate, dampen, etc.) \\nl\\nl\\n2. THE EMERGING RESEARCH MODELS\\n\\n\\n    Doing Classic GT Using Mixed Qualitative and Quantitative Data\\n199\\n­\\nthe  particular outcomes. For instance, building on Tsang and Kwan \\n(1999)’s example, a car will drive adequately if it has four wheels and \\nan engine but only if somebody also turns on the ignition and a nail \\ndoes not puncture one of its tires. Therefore, the arrows in the diagrams \\nproposed in this section do not illustrate causality as traditionally \\nunderstood; they illustrate the activation of causal powers as revealed \\nin the substantive area (the use of the Moodle platform by students and \\nprofessors) and the context (a European business school) that were \\ninvestigated. The paths that are reported in the models are those that \\nbest fitted our set of data (qualitative and quantitative) within the \\nsubstantive area and the context that were being investigated. Where \\npossible, however, we used a reflexive approach and extended our \\nresults in order for them to be applied in further research in other \\nsubstantive areas and contexts. When we understood the collected data \\nas sufficient to demonstrate that a proposition would apply whatever \\nthe substantive area and context investigated, we added the word \\nmostly in the wording of the corresponding proposition. Mostly, \\nhowever, is used only as a heuristic device and should not be taken at \\nface value. It means that, based on the data we collected, the proposition \\nappears to extend to other substantive areas and other contexts beyond \\nthose researched for the present study; this would, of course, need to \\nbe verified in further research. When mostly is not used in the wording \\nof a proposition, it indicates that the corresponding proposition may \\nhold true only within our substantive area and investigated context.\\nAs our main concern was to investigate the path between users’ \\nindividual IT culture and their utilization of the e-learning platform, \\nand how the various constructs related to their perceived IT needs \\n(global, contextual and situational) fit in this path, we first describe in \\nthis section the new path to IT use that we propose through these new \\nconcepts. We then embed one of our newly defined variables in the \\nTAM nomological framework to position our work with respect to past \\nliterature.\\nQualitative and quantitative analyses were embedded, and our \\nresults emerged through constant iterations between quantitative and \\nqualitative data as they were collected and through the analysis of all \\nour data as one set. Therefore, for each model, we present our results \\nin a logical sequence that relates our findings as closely as possible to \\nhow we arrived at them.\\nFor each proposition, we detail elements from the literature (if \\navailable) and/or qualitative clues that led us to it. To avoid unnecessary \\nlength, we cite only when needed some verbatim that we consider as \\nmost illustrative for our arguments. We also report on the quantitative \\n\\n\\nCLASSIC GROUNDED THEORY\\n200\\nresults that were obtained as we explored the various quantitative \\npaths between the newly defined variables. We report on these through \\n(1) the standardized coefficients β of the investigated paths, which \\nindicate the strength of the relationships between each pair of variables, \\nand have values between 0 and 1; (2) the significance of these paths, \\nobtained through the bootstrapping procedure and expressed by the \\nprobability p that the hypothesis underlying this path might not be \\nverified; and (3) the R2 values for each variable, which inform us how \\nmuch of its variance is explained by its antecedent(s).\\nWith the help of our whole data set, we found that individual IT \\nculture mostly positively influences the global IT needs perceived by \\nusers (Proposition 1). In our substantive area and investigated context, \\nindividual IT culture also happened to positively influence the contextual \\nIT needs perceived by professors and students (Proposition 2). Both \\nglobal IT needs and contextual IT needs were found to mostly influence \\nthe situational IT needs perceived by users for a given IT proposed in \\ntheir work context and supposed, by deciding instances, to be of some \\nhelp in fulfilling some of the users’ tasks (Propositions 3 and 4). These \\nperceived situational IT needs in turn mostly significantly explain the \\nuse of this IT (Proposition 5). The position held by an individual within \\nan organization will mostly influence his or her contextual and \\nsituational IT needs (Propositions 6a and 6b); in our substantive area \\nand investigated context, it also moderated the effect of the perceived \\nsituational IT needs on the actual utilization of the investigated IT \\n(Proposition 6c).\\nENDNOTES\\n1.\\t “We define IT-utilization as the actual, objectively assessed, use of an IT, \\nand IT-usage as a socially constructed . . . phenomenon.” (Walsh, 2010, p. 8)\\n2.\\t “The fabled ‘Texas sharpshooter’ fires a shotgun at a barn and then \\npaints the target around the most significant cluster of bullet holes in the \\nwall. Accordingly, the Texas sharpshooter fallacy describes a false con­\\nclusion that occurs whenever ex post explanations are presented to inter­\\npret a random cluster in some data” (Biemann, 2012, p. 2).\\n3.\\t The investigated scales were the determination scale (Sheldon, Ryan, & \\nReis, 1996), basic psychological needs satisfaction scales (currently being \\nresearched by a team of researchers, including Deci and Ryan), global \\nmotivation scale (Guay, Mageau, & Vallerand, 2003), and the fundamen­\\ntal needs satisfaction scale in the context of sports (Gillet, Rosnet, & \\nVallerand, 2008).\\n\\n\\nAppendix D\\nSome Examples of Theoretical Coding Using Mixed \\nQualitative and Quantitative Data\\nA\\nlthough referring to Walsh (2014a), which is an empirical article, \\nthis appendix is extracted and adapted from Walsh (2014c), \\nwhich is a methodological article. The examples provided refer \\nto the same research project as Appendix C and provides some \\nexamples of theoretical coding using mixed qualitative and quantitative \\ndata in an embedded mixed GT design. This appendix also illustrates \\nhow, once the core category of a grounded theory has emerged, the \\nliterature may be used as secondary data.\\n201\\n\\n\\n202\\nTheoretical \\ncodes\\nGeneralizability\\nConstant comparative analysis of all data as one set\\nSecondary data (from previous phases and \\nfrom the literature)\\nPrimary qualitative data\\nPrimary \\nquantitative data\\nProposition 1: \\nIndividual \\ninformation \\ntechnology \\n(IT) culture \\nhas a positive \\ninfluence \\non the \\nindividual’s \\nglobal IT \\nneeds.\\nSubstantively \\nformalized\\nWe found, during previous phases of \\nour research, three main attitudinal \\nuser groups (proactive, passive, and \\nrefusal). These groups include \\ndifferent user profiles, with differing \\ndegrees of IT acculturation. We also \\nfound that the most IT-acculturated \\nusers are proactive. They are perceived \\nby managers as having a facilitating \\ninfluence during the implementation \\nof new IT in organizations. They are \\nalso those users with high levels of \\nglobal IT needs, whereas passive users \\nhave lower levels of global IT needs \\nand refusal users have none.\\n“For me, there is a clear link between an \\nindividual’s level of IT acculturation \\nand the need they perceive for IT in \\ntheir life.” (CEO of an IT firm)\\nIITC à \\nGLOBITNEE: \\nβ = 0.598, \\np < 0.001, \\nR² for \\nGLOBITNEE \\n= 35.8%\\nProposition 2: \\nIndividual IT \\nculture has a \\npositive \\ninfluence on \\ncontextual IT \\nneeds.\\nBounded by \\ncontext\\nWe found during previous phases of \\nour research that the users who were \\nhighly IT-acculturated also had high \\nlevels of contextual IT needs.\\n“I certainly could not do my job in X \\nschool without IT. This was not the case \\nin my previous institution, although I \\npersonally feel I need IT to do my job as a \\nprofessor . . . For instance, it’s the first \\ntime I have been asked to use PowerPoint \\nslides to teach. In my old school, nobody \\never used PowerPoint.” (Young assistant \\nprofessor with a couple of previous \\nteaching experiences and who has started \\na new job in X school; she grew up in a \\ncomputerized home environment with a \\nfather who is an IT engineer)\\nIITC à \\nCONITNEE: \\nβ = 0.400, \\np < 0.001\\n\\n\\n203\\nProposition 3: \\nGlobal IT \\nneeds \\ninfluence \\nsituational IT \\nneeds, \\npositively or \\nnegatively.\\nSubstantively \\nformalized\\nVallerand (1997, 2001) showed that \\nmotivation at the situational level \\nresults from the effects of both global \\nmotivation and “social factors at the \\nappropriate level of generality.” \\n(Vallerand, 1997, p. 275)\\n“I have fairly important IT needs; for \\nexample, I have several computers, and \\nI always need to be connected through \\nthe Web. When I travel abroad, I remain \\nconnected even if it costs me a lot of \\nmoney. . . . You evaluate the platform \\nand how it helps you fulfill needs \\nlinked with your responsibilities. I \\nfound that Moodle helped only with \\nuploading the files that I wanted the \\nstudents to have, and I found other \\nfunctionalities inadequate, so I use the \\nplatform only minimally.” (Young \\ninformations systems [IS] assistant \\nprofessor)\\nPOSITION à \\nSITITNEE:  \\nβ = –0.138,  \\np < 0.001\\nProposition 4: \\nContextual IT \\nneeds have a \\npositive \\ninfluence on \\nsituational IT \\nneeds.\\nBounded by \\ncontext\\nN/A\\n“The link between the need for IT that I \\nperceived as related to the school where \\nI did my Executive Master of Business \\nAdministration degree, and the need for \\nthe specific tool that was proposed to \\nhelp us in our tasks as students is \\nsignificant. . . . It was made clear to us, \\nfrom the beginning of the course that \\nwe were supposed to be IT-proficient.” \\n(Mature EMBA student)\\nCONITNEE à \\nSITITNEE: \\nβ = 0.280,  \\np < 0.001\\n(Continued)\\n\\n\\n204\\nTheoretical \\ncodes\\nGeneralizability\\nConstant comparative analysis of all data as one set\\nSecondary data (from previous phases and \\nfrom the literature)\\nPrimary qualitative data\\nPrimary \\nquantitative data\\nProposition 5: \\nSituational IT \\nneeds, related \\nto some \\nspecific \\nsystem or \\nsoftware, \\nhave a \\npositive \\ninfluence on \\nthe use of this \\nspecific \\nsystem or \\nsoftware.\\nSubstantively \\nformalized\\nIf users perceive needs for some \\nspecific IT in order to fulfill given \\ntasks, they will be driven to fulfill \\nthese needs (Maslow, 1954) and hence \\nuse this specific IT.\\nN/A\\nSITITNEE à \\nUTILIZATION: \\nβ = 0.676, \\np > 0.001\\nProposition 6a: \\nThe position \\nheld by an \\nindividual \\nwithin a given \\norganization \\naffects the \\nindividual’s \\ncontextual IT \\nneeds.\\nSubstantively \\nformalized\\nPost et al. (1999) show that \\norganization size and users’ academic \\neducation influence the level of users’ \\ncontextual IT needs.\\n“I have done several summer jobs \\nduring my studies—from stable work \\nwith horses, through factory work, to \\nhelping candidates in local elections; \\nobviously, I did not have the same \\nneeds for IT in all these jobs. It is \\nprobably in my current teaching \\nposition that I need IT the most.” \\n(Young assistant professor)\\nPOSITION à \\n\\nCONITNEE: \\nβ = 0.277, \\np > 0.001, \\nR² for \\nCONITNEE \\nexplained by \\nIITC and \\nPOSITION = \\n27.7%\\n(Continued)\\n\\n\\n205\\nProposition 6b: \\nThe position \\nheld by an \\nindividual \\nwithin an \\norganization \\naffects his or \\nher situational \\nIT needs.\\nSubstantively \\nformalized\\nN/A\\nN/A\\nPOSITION à \\nSITITNEE:  \\nβ = 0.454,  \\np < 0.001, R² for \\nSITITNEE \\nexplained by \\nCONITNEE, \\nGLOBITNEE, \\nand POSITION \\n= 35.8%\\nProposition 6c: \\nPosition has a \\nmoderating \\ninfluence \\nbetween \\nsituational IT \\nneeds and \\nutilization.\\nBounded by \\ncontext\\nN/A\\n“I teach in several schools and I am not \\nvery much into IT. If I were to get used \\nto all the different IT tools in the \\ndifferent schools where I teach, it would \\ntake me hours that I don’t have. Unless I \\nam ‘officially’ obliged to utilize a \\nplatform, I will not do it even if I need it \\n. . . even if it would make my exchange \\nwith the students somehow easier.” \\n(Professor who shares his time between \\nconsulting and teaching in business \\nschools, and who feels, as he says so \\nhimself, “constrained to utilize IT.”)\\nQuantitative \\nanalyses with \\nand without the \\nmoderating \\neffect\\n\\n\\nAppendix E\\nExamples of Diagrams as Precursors of a Theory\\nA\\ns we saw in Chapter 7, a diagram is not a theory. However, it \\nmay help the researcher’s theoretical sensitivity, thus nurturing \\nthe emergence of a theory, especially if the theory is complex \\nand multiple theoretical codes are emerging from the data.\\nTwo of the following diagrams are extracted from Walsh, Renaud, \\nand Kalika (2013); two are unpublished and were intermediary memos.\\nUsers’\\nIT Needs\\nTasks’ \\nIT Needs\\nMana-\\ngerial\\nIT Needs\\nHardware\\nNeeds\\nSoftware\\nNeeds\\nAlignment 4\\nAlignment 5\\nAlignment 6\\nAlignment 7\\nAlignment 1\\nAlignment 3\\nAlignment 2\\n\\u0007\\nMemo 1:    Unpublished\\n206\\n\\n\\nExamples of Diagrams as Precursors of a Theory\\n207\\nMemo 2:    Extracted from Walsh, Renaud, and Kalika, 2013\\nUsers’\\nIT Needs\\nTasks’ \\nIT Needs\\nMana-\\ngerial\\nIT Needs\\nTop Managers\\nHardware\\nNeeds\\nSoftware\\nNeeds\\nAlignment 4\\nAlignment 5\\nAlignment 6\\nAlignment 7\\nAlignment 1\\nAlignment 3\\nAlignment 2\\nIT Utilization\\nIS  Strategy\\nand Infrastructure\\nUsers’ \\nIT Culture\\nIT Technicians\\nSoftware Analysts\\nUsers’ Tasks\\n Business  Strategy\\nand Infrastructure\\nJob Experts\\n\\n\\nCLASSIC GROUNDED THEORY\\n208\\nMemo 3:     Unpublished\\nUsers’\\nIT Needs\\nTasks’ \\nNeeds\\nMana-\\ngerial\\nNeeds\\nHardware\\nNeeds\\nSoftware\\nNeeds\\nAlignment 4\\nAlignment 5\\nAlignment 6\\nAlignment 7\\nAlignment 1\\nAlignment 3\\nIS Strategy and Structure\\nUsers’ IT Culture\\nIT Utilization\\nUsers’ Tasks\\nJob Experts\\nTop\\nManagers\\nBusiness Strategy\\nand Structure\\nIT\\nTechnicians\\nSoftware\\nAnalysts\\nTranslators\\nTranslators\\nTranslators\\nTranslators\\nTranslated\\nStrategic\\nAlignment\\nAlignment 2\\n\\n\\nExamples of Diagrams as Precursors of a Theory\\n209\\nUsers’\\nIT Needs\\nTasks’ \\nIT Needs\\nMana-\\ngerial\\nIT Needs\\nTop Managers\\nHardware\\nNeeds\\nSoftware\\nNeeds\\nAlignment 4\\nAlignment 5\\nAlignment 6\\nAlignment 7\\nAlignment 1\\nAlignment 3\\nAlignment 2\\nIT Utilization\\nIS Strategy\\nand Infrastructure\\nUsers’ \\nIT Culture\\nIT \\nTechnicians\\nSoftware\\nAnalysts\\nBusiness Strategy \\nand Infrastructure\\nJob Experts\\nTranslation\\nUsers’ Tasks\\nT\\nr\\na\\nn\\ns\\nl\\na\\nt\\no\\nr\\n(\\ns\\n)\\n \\n=\\n \\nP\\nr\\no\\nj\\ne\\nc\\nt\\n \\nM\\na\\nn\\na\\ng\\ne\\nr\\n—\\n \\nC\\no\\nn\\ns\\nu\\nl\\nt\\na\\nn\\nt\\ns\\n…\\nTranslation\\nTranslation\\nLevel III = Delegates\\nLevel II = Intermediaries\\nLevel I = Actants\\nTranslation\\nTranslation\\nAlignments investigated in traditional IS strategic alignment models\\nUser-centered alignments mostly ignored in the literature\\nMemo 4:    Extracted from Walsh, Renaud, and Kalika, 2013\\n\\n\\n210\\nGlossary\\nBasic Social Process: A generic theoretical construct explaining funda­\\nmental patterns in the organization of social behavior as it occurs over \\ntime, involving a change over time with discernible breaking points or \\npoints of transition or passage from one stage to another. Basic social \\nprocesses may be psychological (e.g., rehumanizing, becoming, etc.) or \\nstructural (network, bureaucracy, routine) or may emerge as a combi­\\nnation of the two (e.g., rehumanizing knowledge work through fluctu­\\nating support networks).\\nCategory: The abstraction of individual concepts as a latent pattern.\\nClassic grounded theory: The systematic generation of theory from \\ndata as an integrated set of concepts and as hypotheses about the inter­\\nrelationships among these concepts that focus on a main issue or con­\\ncern in the area under study and explain how that issue or concern is \\nprocessed or resolved.\\nCode: The label attached to an incident during the substantive coding \\nstage of analysis. Codes are often descriptive, particularly in the initial \\nstages of data analysis.\\nConcept: The resultant higher level abstraction of a conceptual idea as \\nemerging through the cumulative comparison of codes.\\nConstant comparison: A foundational pillar of classic GT in which \\ndata are analyzed as they are collected and constantly compared with \\npreviously collected data for interchangeable indicators of emerging \\nconcepts. Comparison first of indicators to indicators, then indicators \\nto emerging concepts and finally concepts to concepts for emerging \\ntheoretical integration. Constant comparison is conducted in tandem \\n\\n\\nGlossary\\n211\\nwith theoretical sampling as a means of guiding the direction of further \\ndata sampling.\\nCore category: The category (i.e., theoretical model) that appears to \\nexplain how the main concern in the area under study is processed, \\nmanaged, or resolved by its accounting for much of the variation in the \\nway the main concern is addressed.\\nData: The empirical basis for theory building. In GT, “all is data” mean­\\ning that data are information that can be counted or expressed numeri­\\ncally (quantitative) or in the form of words (qualitative).\\nDimension: A measurable component of a concept or category. Dimen­\\nsions are not interchangeable; they are complementary to define a \\nconcept or category.\\nEmergence: A result of GT’s open and exploratory stance in relation to \\napproaching the research field. A foundational pillar of classic GT, \\nemergence requires the researcher to remain open to discovering a \\nmain concern motivating the interest or action within the area under \\nstudy.\\nExplication de texte: A technique in French literary analysis describing \\nthe close reading of text, line by line, to ascertain precisely what the \\nauthor is saying without reinterpreting or implying meaning. In GT, expli­\\ncation de texte aids in generating concepts that closely fit what is going on \\nin the substantive area by simply naming the concept as an abstraction.\\nField notes: Notes recorded by the researcher while in the research \\n\\nsetting that serve as in the moment reminders of incidents that may \\nindicate potential concepts.\\nFormal grounded theory: A grounded theory developed at a higher \\nlevel of empirical generalization and/or conceptual abstraction by \\nextending the general implications of the core category of a substantive \\ngrounded theory through subsequent theoretical sampling across a \\nwide range of settings and/or disciplines.\\nFramework: The general set of guidelines proposed for a specific \\napproach to research, for example, action research, case-study research, \\nclassic grounded theory.\\n\\n\\nCLASSIC GROUNDED THEORY\\n212\\nIncidents: Indicators of phenomena or experiences as observed or \\narticulated in data.\\nInterchangeability of indicators: Empirical incidents in data that are \\ninterchangeable in indicating a concept (i.e., indicating the same con­\\ncept) but that may also bring out different properties or values of the \\nconceptual idea to enhance theory elaboration.\\nMain concern: The issue that emerges across the coded data as the \\nprime motivator, interest, or problem in the research setting.\\nMemoing: The core ideational processing of theoretical ideas as they \\nemerge through coding and constant comparative analysis, memoing \\nis a continual process that helps to raise the data to a conceptual level \\nand develop the properties of each category.\\nMethodology: The specific combination of research methods and tech­\\nniques used in a research project.\\nMethods: The ways in which data are collected, for example, inter­\\nviews, observation, filming, or surveys.\\nOpen coding: The initial coding stage in classic GT in which the ana­\\nlyst codes each incident in the data for as many concepts as possibly \\nrelevant. Open coding proceeds (in tandem with memoing) until the \\nanalyst perceives a pattern emerging that may indicate a potential core \\ncategory.\\nParadigm: A worldview, epistemological stance, or system of shared \\nbeliefs held by a community of research practice.\\nPreconceptions: Assumptions embedded within a researcher’s world­\\nview or professional discipline that are taken for granted and assumed \\nto be of relevance.\\nPreconscious processing: Creative insights and theoretical realizations \\nthat occur seemingly serendipitously outside the analyst’s conscious \\nefforts to analyze the data.\\nProperty: A latent characteristic of a concept or category (e.g., \\nauthenticity, trust, healing) that cannot be directly and objectively \\nassessed.\\n\\n\\nGlossary\\n213\\nReadiness to sort: The moment of ideational overload following data \\nanalysis when the analyst feels a need to hand sort memos for \\n­\\ntheoretical realization.\\nReadiness to write: The moment when theoretical completeness \\nappears to be achieved and memos are sorted, prompting the analyst \\nto write the first draft of the theory.\\nSecondary data: Data that have been collected for some other \\npurpose.\\nSelective coding: Subsequent to the emergence of the core category, \\nfurther data collection and analysis is delimited to achieving theoreti­\\ncal saturation of the core category and related concepts.\\nSubstantive coding: The process of conceptualizing the empirical data \\nin which the theory is grounded. Incidents in the empirical data are \\ncoded for indicators of concepts from which a grounded theory is then \\ngenerated. Substantive coding consists of both the initial open coding \\nof data as well as the selective coding of data once the core category has \\nemerged.\\nSubstantive grounded theory: A grounded theory developed within a \\nspecific setting and context and generalizable to a limited number of \\nsimilar settings or groups.\\nTechniques: The instruments used to help analyze and make sense of \\nthe collected data, for example, text analysis, elaboration analysis, clus­\\nter analysis, and structural equation modeling.\\nTheoretical coding: The final stage in classic GT theory genera­\\ntion, theoretical coding conceptualizes and models how the sub­\\nstantive codes may relate to each other as hypotheses to be \\nintegrated into the final theory. Theoretical coding is not selected \\nand imposed on the theory but emerges through earned relevance \\nas the best fitting theoretical integrator of the core category and \\nrelated concepts.\\nTheoretical sampling: A foundational pillar of classic GT, the process \\nthrough which empirical data are selected and collected while guided \\nby the emerging theory as opposed to some predefined sample or \\nunit.\\n\\n\\nCLASSIC GROUNDED THEORY\\n214\\nTheoretical saturation: When in coding and analyzing data, no new \\nproperties emerge but rather more indicators of the same properties \\ncontinually emerge.\\nTheoretical sorting: The hand sorting of memos to provide emergent \\ntheoretical order and integration of ideas as the theorist seeks to \\ntheoretically discriminate as to where each idea fits in the emerging \\ntheory.\\n\\n\\n215\\nReferences\\nAdams, D., Nelson, R., & Todd, P. (1992). Perceived usefulness, ease of use, and \\nusage of information technology: A replication. MIS Quarterly, 16(2), 227–247.\\nAlvesson, M., & Sandberg, J. (2014). Habitat and habitus: Boxed-in versus box-\\nbreaking research. Organization Studies, 35(7), 967–987.\\nAmsteus, M. N. (2014). The validity of divergent grounded theory method. \\nInternational Journal of Qualitative Methods, 13, 71–87.\\nArmstrong, J. S., & Overton, T. S. (1977). Estimating nonresponse bias in mail \\nsurveys. Journal of Marketing Research, 14(3), 396–402.\\nBacharach, S. B. (1989). Organizational theories: Some criteria for evaluation. \\nAcademy of Management Review, 14(4), 496–513.\\nBackman, K., & Kyngas, H. A. (1999). Challenges of the grounded theory \\napproach to a novice researcher. Nursing and Health Sciences, 1, 147–153.\\nBaker, C., Wuest, J., & Stern, P. (1992). Method slurring: The grounded theory/\\nphenomenology example. Journal of Advanced Nursing, 17, 1355–1360.\\nBaskerville, R., & Pries-Heje, J. (1999). Grounded action research: A method for \\nunderstanding IT in practice. Accounting, Management and Information \\nTechnologies, 9, 1–23.\\nBenbasat, I., & Barki, H. (2007). Quo vadis, TAM? Journal of the Association for \\nInformation Systems, 8(4), 211–218.\\nBenzécri, J.-P. (1973). La place de l’a priori. In Encyclopédia universalis (Vol. 17, \\nOrganum, pp. 11–24). Edinburgh, UK: Encyclopaedia Britannica.\\nBhaskar, R. (1979). The possibility of naturalism. New York: Harvester Press.\\nBhaskar, R. (1989). Reclaiming reality. London: Verso.\\nBhaskar, R. (1998). Philosophy and scientific realism. In M. Archer, R. Bhaskar, \\nA. Collier, T. Lawson, & A. Norrie (Eds.), Critical realism: Essential readings \\n(pp. 16–47). London: Routledge.\\nBhaskar, R. A. (2002). Reflections on meta-reality: A philosophy for the present. New \\nDelhi: Sage.\\nBiemann, T. (2012, August). What if we are Texas sharpshooters? A new look at publica­\\ntion bias. Paper presented at the Academy of Management Conference, Boston.\\nBigus, O. E., Hadden, S. C., & Glaser, B. G. (1994). The study of basic social \\nprocesses. In B. G. Glaser (Ed.), More grounded theory methodology: A reader \\n(pp. 38–64). Mill Valley, CA: Sociology Press.\\n\\n\\nCLASSIC GROUNDED THEORY\\n216\\nBirks, D. F., Fernandez, W., Levina, N., & Nasirin, S. (2013). Grounded theory \\nmethod in information systems research: Its nature, diversity and oppor­\\ntunities. European Journal of Information Systems, 22(1), 1–8.\\nBirks, M., & Mills, J. (2012). Grounded theory: A practical guide (2nd ed.). \\nThousand Oaks, CA: Sage.\\nBlumer, H. (1956). Sociological analysis and the “variable.” American Sociological \\nReview, 21(6), 683–690.\\nBreckenridge, J. (2009). Demystifying theoretical sampling in grounded theory \\nresearch. Grounded Theory Review, 8(2), 113–126.\\nBreckenridge, J. (2013). Doing classic grounded theory: The data analysis pro­\\ncess. In Sage Research Methods Cases [Online database]. Retrieved from \\nhttp://srmo.sagepub.com/view/methods-case-studies-2014/n269.xml\\nBringer, J. D., Johnston, L. H., & Brackenridge, C. H. (2004). Maximizing trans­\\nparency in a doctoral thesis: The complexities of writing about the use of \\nQSR*NVIVO within a grounded theory study. Qualitative Research, 4(2), \\n247–265.\\nBryant, A., & Charmaz, C. (Eds.). (2007). The SAGE handbook of grounded theory. \\nLondon: Sage.\\nBryant, J., & Lasky, B. (2007). A researcher’s tale: Dealing with epistemological \\ndivergence. Qualitative Research in Organizations and Management: An \\nInternational Journal, 2(3), 179–193.\\nBryman, A. (1998). Quantitative and qualitative research strategies in knowing \\nthe social world. In T. May & M. Williams (Eds.), Knowing the social world \\n(pp. 138–157). Buckingham, UK: Open University Press.\\nBurton-Jones, A., & Straub, D. (2006). Reconceptualizing system usage: \\n\\nAn approach and empirical test. Information Systems Research, 17(3), \\n228–246.\\nCallon, M. (1986). Some elements of a sociology of translation: Domestication of \\nthe scallops and the fishermen of St. Brieuc Bay. In J. Law (Ed.), Power, action \\nand belief: A new sociology of knowledge (pp. 196–223). London: Routledge & \\nKegan Paul.\\nCartwright, N. (2003). Causation: One word, many things. London: Centre for \\nPhilosophy of Natural and Social Science.\\nCharmaz, K. (1990). “Discovering” chronic illness: Using grounded theory. \\nSocial Science & Medicine, 30(11), 1161–1172.\\nCharmaz, K. (2006). Constructing grounded theory: A practical guide through quali­\\ntative analysis. London: Sage.\\nCharmaz, K. (2009). Shifting the grounds: Constructivist grounded theory \\nmethods. In J. M. Morse et al. (Eds.), Developing grounded theory: The second \\ngeneration (pp. 127–154). Walnut Creek, CA: Left Coast Press.\\nCharmaz, K. (2014). Constructing grounded theory (2nd ed.). London: Sage.\\nChin, W. W. (1998). The partial least squares approach to structural equation \\nmodeling. Modern methods for business research, 295(2), 295–336. \\nClarke, A. (Ed.). (2005). Situational analysis: Grounded theory after the postmodern \\nturn. Thousand Oaks, CA: Sage.\\n\\n\\nReferences\\n217\\nClarke, A. E., & Friese, C. (2007). Grounded theorizing using situational \\n­\\nanalysis. The SAGE handbook of grounded theory (pp. 363–397). London: Sage.\\nThe conceptualization of data [Blog post]. (2010, May 1). From the Mirror. \\nRetrieved from http://thespeculum.blogspot.ca/2010_05_01_archive.html\\nCorbin, J. M. (1991). Anselm Strauss: An intellectual biography. In D. R. Maines \\n(Ed.), Social organization and social process: Essays in honor of Anselm Strauss \\n(pp. 17–42). New York: Aldine de Gruyter.\\nCreswell, J. W. (2007). Qualitative inquiry and research design: Choosing among five \\napproaches. London: Sage.\\nCreswell, J. W. (2013). Qualitative inquiry and research design: Choosing among five \\napproaches (3rd ed.). London: Sage.\\nCreswell, J. W. (2014). Research design: Qualitative, quantitative, and mixed methods \\napproaches (4th ed.). Thousand Oaks, CA: Sage.\\nDavis, F. D. (1989). Perceived usefulness, perceived ease of use, and user accep­\\ntance of information technology. MIS Quarterly, 13(3), 318–339.\\nde Bono, E. (1968). New think: The use of lateral thinking in the generation of new \\nideas. New York: Avon Books.\\nde Vaujany, F.-X., Walsh, I., & Mitev, N. (2011). An historically grounded critical \\nanalysis of research articles in IS. European Journal of Information Systems, \\n20(4), 395–417.\\nDevaraj, S., Easley, R. F., & Crant, J. M. (2008). How does personality matter? \\nRelating the five-factor model to technology acceptance and use. Information \\nSystem Research, 19(1), 93–105.\\nDey, I. (1993). Qualitative data analysis: A user-friendly guide. New York: \\nRoutledge.\\nDey, I. (1999). Grounding grounded theory: Guidelines for qualitative inquiry. San \\nDiego, CA: Academic Press.\\nDiamantopoulos, A., Riefler, P., & Roth, K. P. (2008). Advancing formative mea­\\nsurement models. Journal of Business Research, 61(12), 1203–1218.\\nDraucker, C. B., Martsolf, D. S., Ross, R., & Rusk, T. B. (2007). Theoretical sam­\\npling and category development in grounded theory. Qualitative Health \\nResearch, 17(8), 1137–1148.\\nEisenhardt, K. M. (1989). Building theories from case study research. Academy \\nof Management Review, 14(4), 532–550.\\nEvermann, J., & Tate, M. (2011). Fitting covariance models for theory genera­\\ntion. Journal of the Association for Information Systems, 12(9), 632–661.\\nFetters, M., & Freshwater, D. (2015). Publishing a methodological mixed meth­\\nods research article. Journal of Mixed Methods Research, 9(3), 203–213.\\nFine, G. A., & Deegan, J. G. (1996). Three principles of serendipity: Insight, \\nchance, and discovery in qualitative research. International Journal of \\nQualitative Studies in Education, 9(4), 434–447.\\nFornell, C., & Cha, J. (1994). Advanced methods of marketing research. Cambridge, \\nMA: Blackwell Business.\\nGagné, M., & Deci, E. (2005). Self-determination theory and work motivation. \\nJournal of Organizational Behavior, 26(4), 331–362.\\n\\n\\nCLASSIC GROUNDED THEORY\\n218\\nGefen, D., Rigdon, E. E., Straub, D. W. (2011). Editors’ comment: An update and \\nextension to SEM guidelines for administrative and social science research. \\nMIS Quarterly, 35(2), iii–xiv.\\nGibson, B., & Hartman, J. (2014). Rediscovering grounded theory. London: Sage.\\nGillet, N., Rosnet, E., & Vallerand, R. J. (2008). Développement d’une échelle de \\nsatisfaction des besoins fondamentaux en context sportif. Revue Canadienne \\ndes Sciences du Comportement, 40(4), 230–237.\\nGiske, T., & Gjengedal, E. (2007). “Preparative waiting” and coping theory with \\npatients going through gastric diagnosis. Journal of Advanced Nursing, 57(1), \\n87–94.\\nGlaser, B. (2004). Remodeling grounded theory. Forum Qualitative Research, 5(2), \\nArt. 4.\\nGlaser, B. G. (1961). Some functions of recognition in a research organization \\n(Unpublished doctoral dissertation). Columbia University, New York.\\nGlaser, B. G. (1963). Attraction, autonomy, and reciprocity in the scientist-\\nsupervisor relationship. Administrative Science Quarterly, 8(3), 379–398.\\nGlaser, B. G. (1964a). Comparative failure in science. Science, 143(3610), 1012–1014.\\nGlaser, B. G. (1964b). Organizational scientists: Their professional careers. New \\nYork: Bobbs Merrill.\\nGlaser, B. G. (1965). The constant comparative method of qualitative analysis. \\nSocial Problems, 12, 436–445.\\nGlaser, B. G. (1966). [Review of the book Professional employees: A study of scientists \\nand engineers, by Kenneth Prandy]. American Journal of Sociology, 72(3), \\n\\n331–332.\\nGlaser, B. G. (1978). Theoretical sensitivity: Advances in the methodology of grounded \\ntheory. Mill Valley, CA: Sociology Press.\\nGlaser, B. G. (1991). In honor of Anselm Strauss: Collaboration. In D. R. Maines \\n(Ed.), Social organization and social process: Essays in honor of Anselm Strauss \\n(pp. 11–16). New York: Aldine de Gruyter.\\nGlaser, B. G. (1992). Basics of grounded theory analysis: Emergence vs. forcing. Mill \\nValley, CA: Sociology Press.\\nGlaser, B. G. (1995). A look at grounded theory: 1984-1994. In B. G. Glaser (Ed.), \\nGrounded theory 1984-1995 (Vol.  1, pp.  3–17). Mill Valley, CA: Sociology \\nPress.\\nGlaser, B. G. (1998). Doing grounded theory: Issues and discussions. Mill Valley, \\nCA: Sociology Press.\\nGlaser, B. G. (2001). The grounded theory perspective: Conceptualization contrasted \\nwith description. Mill Valley, CA: Sociology Press.\\nGlaser, B. G. (2002). Conceptualization: On theory and theorizing using \\ngrounded theory. International Journal of Qualitative Methods, 1(2), 23–38.\\nGlaser, B. G. (2003). The grounded theory perspective II: Description’s remodeling of \\ngrounded theory methodology. Mill Valley, CA: Sociology Press.\\nGlaser, B. G. (2005a). The grounded theory perspective III: Theoretical coding. Mill \\nValley, CA: Sociology Press.\\n\\n\\nReferences\\n219\\nGlaser, B. G. (2005b, August 23). The roots of grounded theory. Keynote address to \\nthe 3rd International Qualitative Research Convention, Johor Bahru, \\nMalaysia.\\nGlaser, B. G. (2007). Doing formal grounded theory. Mill Valley, CA: Sociology Press.\\nGlaser, B. G. (2008). Doing quantitative grounded theory. Mill Valley, CA: \\nSociology Press.\\nGlaser, B. G. (2009). Jargonizing: Using the grounded theory vocabulary. Mill Valley, \\nCA: Sociology Press.\\nGlaser, B. G. (2010). The future of grounded theory. Grounded Theory Review, 9(2), \\n1–14.\\nGlaser, B. G. (2011). Getting out of the data: Grounded theory conceptualization. Mill \\nValley, CA: Sociology Press.\\nGlaser, B. G. (2012). Stop, write: Writing grounded theory. Mill Valley, CA: \\nSociology Press.\\nGlaser, B. G. (2013). No preconceptions: The grounded theory dictum. Mill Valley, \\nCA: Sociology Press.\\nGlaser, B. G. (2014). Memoing: A vital grounded theory procedure. Mill Valley, CA: \\nSociology Press.\\nGlaser, B. G., & Holton, J. (2004). Remodeling grounded theory. Forum \\nQualitative Sozialforschung/Forum: Qualitative Social Research, 5(2), Art.  4. \\nRetrieved from http://www.qualitative-research.net/fqstexte/2-04/2-04\\nglaser-e.htm\\nGlaser, B. G., & Holton, J. (2005). Basic social processes. Grounded Theory Review, \\n4(3), 1–21.\\nGlaser, B. G., & Strauss, A. L. (1965a). Awareness of dying. Chicago: Aldine.\\nGlaser, B. G., & Strauss, A. L. (1965b). Discovery of substantive theory: A basic \\nstrategy underlying qualitative research. American Behavioral Scientist, 8(6), \\n5–12.\\nGlaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies \\nfor qualitative research. New York: Aldine de Gruyter.\\nGlaser, B. G., & Strauss, A. L. (1968). Time for dying. Mill Valley, CA: Sociology \\nPress.\\nGlaser, B. G., & Strauss, A. L. (1971). Status passage: A formal theory. Mill Valley, \\nCA: Sociology Press.\\nGoulding, C. (1999). Consumer research, interpretive paradigms and method­\\nological ambiguities. European Journal of Marketing, 33(9/10), 859–873.\\nGoulding, C. (2002). Grounded theory: A practical guide for management, business \\nand market researchers. London: Sage.\\nGregor, S. (2006). The nature of theory in information systems. MIS Quarterly, \\n30(3), 611–642.\\nGuay, F., Mageau, G., & Vallerand, R. (2003). On the hierarchical structure of \\nself-determined motivation: A test of top-down, bottom-up, reciprocal, \\nand horizontal effects. Personality and Social Psychology, 29(8), 991–1004.\\nGuba, E. G. (Ed.). (1990). The paradigm dialog. Newbury Park, CA: Sage.\\n\\n\\nCLASSIC GROUNDED THEORY\\n220\\nGuthrie, W. (2000). Keeping clients in line: A grounded theory explaining how vet­\\nerinary surgeons control their clients (Unpublished doctoral dissertation). \\nUniversity of Strathclyde, Glasgow, UK.\\nGynnild, A. (2006). Growing open: The transition from QDA to grounded the­\\nory. Grounded Theory Review, 6(1), 61–78.\\nHair, J., Anderson, R., Tatham, R., & Black, W. (1995). Multivariate data analysis \\nwith readings (4th ed.). New York: Macmillan.\\nHair, J., & Black, W. (1998). Cluster analysis. In J. H. Hair, R. E. Anderson, \\nR.  L.  Tatham, & W. C. Black (Eds.), Multivariate data analysis (5th ed., \\npp. 469–518). New York: Prentice Hall.\\nHair, J., Black, W., Babin, B., Anderson, R., & Tatham, R. (2006). Multivariate data \\nanalysis (6th ed.). London: Pearson Education International.\\nHair, J. F., Sarstedt, M., Ringle, C. M., Mena, J. A. (2012). An assessment of the \\nuse of partial least squares structural equation modeling in marketing \\nresearch. Journal of the Academy of Marketing Science, 40(3), 414–433.\\nHall, R. (2012). Mixed methods: In search of a paradigm. In T. Lê & Q. Lê (Eds.), \\nConducting research in a changing and challenging world (pp.  71–78). \\nHauppauge, NY: Nova Science.\\nHämäläinen, M. (2014). Renaissance entrepreneurship: A grounded theory of entre­\\npreneurial recycling (Doctoral dissertation). Aalto University, Helsinki, \\nFinland.\\nHernandez, C. A. (2010). Getting grounded: Using Glaserian grounded theory \\nto conduct nursing research. Canadian Journal of Nursing Research, 42(1), \\n151–163.\\nHiggins, A., Barker, P., & Begley, C. M. (2008). “Veiling sexualities”: A grounded \\ntheory of mental health nurses responses to issues of sexuality. Journal of \\nAdvanced Nursing, 62(3), 307–317.\\nHolton, J. A. (2006). Rehumanising knowledge work through fluctuating support \\nnetworks: A grounded theory study (Unpublished doctoral thesis). University \\nof Northampton, Northampton, UK.\\nHolton, J. A. (2007a). The coding process and its challenges. In A. Bryant & \\nC. Charmaz (Eds.), The SAGE handbook of grounded theory (pp. 265–289). \\nLondon: Sage.\\nHolton, J. A. (2007b). Rehumanising knowledge work through fluctuating sup­\\nport networks: A grounded theory study. Grounded Theory Review, 6(2), \\n23–46.\\nHolton, J. A. (2008). Grounded theory as a general research methodology. \\nGrounded Theory Review, 7(2), 67–93.\\nHolton, J. A. (2011). The autonomous creativity of Barney G. Glaser: Early influ­\\nences in the emergence of classic grounded theory methodology. In \\n\\nA. Gynnild & V. Martin (Eds.), Grounded theory: The philosophy, method and \\nwork of Barney Glaser (pp. 201–223). Boca Raton, FL: BrownWalker Press.\\nHolton, J. A., & Grandy, G. (2015). Voiced inner dialogue as relational reflection-on-\\naction: The case of middle managers in healthcare. Manuscript submitted for \\npublication.\\n\\n\\nReferences\\n221\\nIgbaria, M., & Parasuraman, S. (1989). A path analytic study of individual char­\\nacteristics, computer anxiety and attitudes toward microcomputers. \\nJournal of Management, 15(3), 373–388.\\nJohnson, R., & Onwuegbuzie, A. J. (2004). Mixed methods research: A research \\nparadigm whose time has come. Educational Researcher, 33(14), 14–26.\\nJones, R., & Kriflik, G. (2006). Subordinate expectations of leadership within a \\ncleaned-up bureaucracy: A grounded theory study. Journal of Organizational \\nChange Management, 19(2), 154–172.\\nJöreskog, K., & Wold, H. (1982). The ML and PLS technique for modeling with \\nlatent variables: Historical and comparative aspects. In K. G. Jöreskog & \\nH. Wold (Eds.), Systems under indirect observation (Part I, 1263–1270). \\nAmsterdam: North-Holland.\\nJørgensen, L. B., Dahl, R., Pedersen, P. U., & Lomborg, K. (2013). Four types of \\ncoping with COPD-induced breathlessness in daily living: A grounded \\ntheory study. Journal of Research in Nursing, 18(6), 520–541.\\nJunglas, I., Niehaves, B., Spiekermann, S., Stahl, B. C., Weitzel, T., Winter, R., & \\nBaskerville, R. (2011). The inflation of academic intellectual capital: The case for \\ndesign science research in Europe. European Journal of Information Systems, 20(1), \\n1–6.\\nKarahanna, E., Agarwal, R., & Angst, C. (2006). Reconceptualizing compatibil­\\nity beliefs in technology acceptance research. MIS Quarterly, 30(4), \\n\\n781–804.\\nKearney, M. H. (2007). From the submime to the meticulous: The continuing \\nevolution of grounded formal theory. In A. Bryant & C. Charmaz (Eds.), \\nThe SAGE handbook of grounded theory (pp. 127–150). London: Sage.\\nKeill, M., Beranek, P., & Konsynski, B. (1995). Usefulness and ease of use: Field \\nstudy evidence regarding task considerations. Decision Support Systems \\n13(1), 75–91.\\nKelle, U. (2005). “Emergence” vs. “forcing” of empirical data? A crucial prob­\\nlem of “grounded theory” reconsidered. Forum Qualitative Sozialforschung/\\nForum: Qualitative Social Research, 6(2), Art.  27. Retrieved from http://\\nnbn-resolving.de/urn:nbn:de:0114-fqs0502275\\nKelle, U. (2007). The development of categories: Different approaches in \\ngrounded theory. In The SAGE handbook of grounded theory (pp. 191–213). \\nLondon: Sage.\\nKlag, M., & Langley, A. (2013). Approaching the conceptual leap in qualitative \\nresearch. International Journal of Management Reviews, 15(2), 149–166.\\nLangley, A. (1999). Strategies for theorizing from process data. Academy of \\nManagement Review, 24(4), 691–710.\\nLatour, B. (1987). Science in action. Cambridge, MA: Harvard University Press.\\nLawrence, J., & Tar, U. (2013). The use of grounded theory technique as a prac­\\ntical tool for qualitative data collection and analysis. Electronic Journal of \\nBusiness Research Methods, 11(1), 29–40.\\nLazarsfeld, P\\n. F., & Henry, N. W. (1968). Latent structure analysis. Boston: Houghton \\nMifflin.\\n\\n\\nCLASSIC GROUNDED THEORY\\n222\\nLazarsfeld, P. F., & Thielens, W., Jr. (1958). The academic mind: Social scientists in \\na time of crisis. Glencoe, IL: Free Press.\\nLee, A. S., & Baskerville, R. L. (2003). Generalizing generalizability in informa­\\ntion systems research. Information Systems Research, 14(3), 221–243.\\nLee, A. S., & Hubona, G. S. (2009). A scientific basis for rigor in information \\nsystems research. MIS Quarterly, 33(2), 237–262.\\nLeonardi, P. M., & Bailey, D. E. (2008). Transformational technologies and the \\ncreation of new work practices: Making implicit knowledge explicit in \\ntask-based offshoring. MIS Quarterly, 32(2), 411–436.\\nLincoln, Y. S., & Guba, E. G. (Eds.). (2000). Naturalistic inquiry (2nd ed.) London: \\nSage.\\nLocke, K. (2001). Grounded theory in management research. London: Sage.\\nLocke, K. (2007). Rational control and irrational free-play: Dual-thinking modes \\nas necessary tension in grounded theorizing. In A. Bryant & C. Charmaz \\n(Eds.), The SAGE handbook of grounded theory (pp. 565–579). London: Sage.\\nLyytinen, K. (2009). Data matters in IS theory building. Journal of the Association \\nfor Information Systems, 10(10), 715–720.\\nLyytinen, K., Baskerville, R., Iivari, J., & Te’eni, D. (2007). Why the old world \\ncannot publish? Overcoming challenges in publishing high-impact IS \\nresearch. European Journal of Information Systems, 317–326.\\nManuj, I., & Pohlen, T. L. (2012). A reviewer’s guide to the grounded theory meth­\\nodology in logistics and supply chain management research. International \\nJournal of Physical Distribution & Logistics Management, 42(8/9), 784–803.\\nMaslow, A. H. (1954). Motivation and personality. New York: Harper.\\nMcCallin, A. (2006). Grappling with the literature in a grounded theory study. \\nGrounded Theory Review, 5(2/3), 11–27.\\nMelia, K. (1996). Rediscovering Glaser. Qualitative Health Research 6, 368–373.\\nMerton, R. K. (1967). On theoretical sociology. New York: Free Press.\\nMerton, R. K. (1968). Social theory and social structure. New York: Free Press.\\nMingers, J. (2001). Combining IS research methods: Towards a pluralist meth­\\nodology. Information Systems Research, 12(3), 240–259.\\nMingers, J. (2004). Realizing information systems: Critical realism as an under­\\npinning philosophy for information systems. Information and Organization, \\n14(2), 87–103.\\nMonette, D., Sullivan, T., & Delong, C. (2011). Applied social research: A tool for \\nthe human services. Belmont, CA: Brooks/Cole.\\nMorgan, D. (2007). Paradigms lost and pragmatism regained: Methodological \\nimplications of combining qualitative and quantitative methods. Journal of \\nMixed Methods Research, 1(1), 48–76.\\nMorse, J. M. (2003). Principles of mixed methods and multimethod research \\ndesign. In Tashakkori and Teddlie (Eds.), SAGE Handbook of mixed methods in \\nsocial & behavioural research. Thousand Oaks, CA: SAGE.\\nMorse, J. M., & Mitcham, C. (2008). Exploring qualitatively-derived concepts: \\nInductive–deductive pitfalls. International Journal of Qualitative Methods, \\n1(4), 28–35.\\n\\n\\nReferences\\n223\\nNathaniel, A. (2008). Eliciting spill: A methodological note. Grounded Theory \\nReview, 7(1), 61–65.\\nO’Connor, M. K., Netting, F. E., & Thomas, M. L. (2008). Grounded theory: \\nManaging the challenge for those facing Institutional Review Board oversight. \\nQualitative Inquiry, 14(1), 28–45.\\nOrr, J. E. (1990). Sharing knowledge, celebrating identity: Community memory in \\na service culture. In D. Middleton & D. Edwards (Eds.), Collective remembering \\n(pp. 169–189). London: Sage.\\nPartington, D. (2000). Building grounded theories of management action. \\nBritish Journal of Management, 11(2), 91–102.\\nPartington, D. (2002). Grounded theory. In D. Partington (Ed.), Essential skills \\nfor management research (pp. 136–157). London: Sage.\\nPatton, M. Q. (2002). Qualitative research and evaluation methods (3rd ed.). \\nThousand Oaks, CA: Sage.\\nPelletier, L., Tuson, K., & Haddad, N. (1997). Client motivation for therapy \\nscale: A measure of intrinsic motivation, extrinsic motivation and amotiva­\\ntion for therapy. Journal of Personality Assessment, 68(2), 414–435.\\nPergert, P. (2008). Façading in transcultural caring relationships: Healthcare staff and \\nforeign-born parents in childhood cancer care (Doctoral thesis). Karolinska \\nInstitutet, Stockholm. Retrieved from http://openarchive.ki.se/xmlui/\\nhandle/10616/37795\\nPiko, L. M. (2014). Discovering Glaser: My experience of doing grounded theory. \\nGrounded Theory Review, 13(2). Retrieved from http://groundedtheoryreview\\n.com/2014/12/19/discovering-glaser-my-experience-of-doing-grounded-\\ntheory/\\nPost, G., Kagan, A., & Keim, R. (1999). A structural equation evaluation of CASE \\ntools attributes. Journal of Management Information Systems, 15(4), 215–234.\\nRichardson, L. (Ed.). (1990). Writing strategies: Reaching diverse audiences. \\nNewbury Park, CA: Sage.\\nRichardson, L. (2000). Writing: A method of inquiry. In N. K. Denzin & Y. S. Lincoln \\n(Eds.), Handbook of qualitative research (2nd ed., pp. 923–948). Thousand Oaks, \\nCA: Sage.\\nRingle, C. M., Wende, S., & Will, S. (2005). SmartPLS 2.0 (M3) Beta [Software]. \\nRetrieved from http://www.smartpls.de\\nRingle, M., Sarstedt, M., & Straub, D. (2012). A critical look at the use of PLS-\\nSEM in MIS Quarterly. MIS Quarterly, 36(1), iii–xiv.\\nRoderick, C. (2009). Learning classic grounded theory: An account of the \\njourney and advice for new researchers. Grounded Theory Review, 8(2), \\n49–63.\\nRodríguez-Martín, B., Martínez-Andres, M., Cervera-Monteagudo, B., Notario-\\nPacheco, B., & Martínez-Vizcaíno, V. (2013). Perception of quality of care \\namong residents of public nursing-homes in Spain: A grounded theory \\nstudy. BMC Geriatrics, 13(1), 65.\\nRunkel, P. J., & Runkel, M. (1984). A guide to usage for writers and students in the \\nsocial sciences. Totowa, NJ: Rowman & Allanheld.\\n\\n\\nCLASSIC GROUNDED THEORY\\n224\\nSandelowski, M. (2000). Combining qualitative and quantitative sampling, \\ndata collection, and analysis techniques in mixed-method studies. Research \\nin Nursing & Health, 23(3), 246–255.\\nSandelowski, M., Voils, C., & Knafl, G. (2009). On quantitizing. Journal of Mixed \\nMethods Research, 3(3), 208–222.\\nSandgren, A. (2010). Deciphering unwritten rules: Patients, relatives and nurses in \\npalliative cancer care (Doctoral thesis). Linnaeus University, Kalmar, \\nSweden. Retrieved from http://www.diva-portal.org/smash/record.jsf?p\\nid=diva2%3A310790&dswid=8891\\nSandgren, A., Thulesius, H., Fridlund, B., & Petersson, K. (2006). Striving for \\nemotional survival. Qualitative Health Research, 16(1), 79–96.\\nSandgren, A., Thulesius, H., Petersson, K., & Fridlund, B. (2007). “Doing good \\ncare”—A study of palliative home nursing care. International Journal of \\nQualitative Studies on Health and Well-Being, 2(4), 227–235.\\nSautrey, J., & Mourmant, G. (2015, June 12). The practical modern artist. Paper \\npresented at Journée AIM, IT and Culture, Technologies de l’Information \\net de la Communication et Culture: enjeux et perspectives dans le domaine \\nde l’art, Lyon, France.\\nSchall, M. (1983). A communication-rules approach to organizational culture. \\nAdministrative Science Quarterly, 28(4), 557–581.\\nSchön D. (1983). The reflective practitioner: How professionals think in action. \\nLondon: Temple Smith.\\nScott, H. (2009). Data analysis: Getting conceptual. Grounded Theory Review, \\n8(2), 89–111.\\nShah, S. K., & Corley, K. G. (2006). Building better theory by bridging the quan­\\ntitative-qualitative divide. Journal of Management Studies, 43(8), 1821–1835.\\nSheldon, K., Ryan, R., & Reis, H. (1996). What makes for a good day? \\nCompetence and autonomy in the day and in the person. Personality and \\nSocial Psychology Bulletin, 22, 1270–1279.\\nSimmons, O. E. (2010). Is that a real theory or did you just make it up? Teaching \\nclassic grounded theory. Grounded Theory Review, 9(2), 15–38.\\nSmith, M. L. (2010). Testable theory development for small-N studies: Critical \\nrealism and middle-range theory. International Journal of Information \\nTechnology and Systems Approach, 3(1), 41–56.\\nStarks, H., & Trinidad, S. B. (2007). Choose your method: A comparison of phe­\\nnomenology, discourse analysis, and grounded theory. Qualitative Health \\nResearch, 17(10), 1372–1380.\\nStjernsward, S., & Ostman, M. (2008). Whose life am I living? Relatives living in the \\nshadow of depression. International Journal of Social Psychiatry, 54(4), 358–369.\\nStrauss, A. L. (1987). Qualitative analysis for social scientists. Cambridge, UK: \\nCambridge University Press.\\nStrauss, A. L., & Corbin, J. M. (1990). Basics of qualitative research: Grounded the­\\nory procedures and techniques. Newbury Park, CA: Sage.\\nStrauss, A. L., & Corbin, J. M. (1998). Basics of qualitative research: Techniques and \\nprocedures for developing grounded theory (2nd ed.). Thousand Oaks, CA: Sage.\\n\\n\\nReferences\\n225\\nSuddaby, R. (2006). What grounded theory is not. Academy of Management \\nJournal, 49(4), 633–642.\\nTakeda, H. (2011). Examining scholarly influence: A study in Hirsch metrics and social \\nnetwork analysis (Unpublished doctoral dissertation). Georgia State University, \\nAtlanta. Retrieved from http://scholarworks.gsu.edu/cis_diss/44/\\nTaylor, S., & Todd, P. (1995). Assessing IT usage: The role of prior experience. \\nMIS Quarterly, 19(4), 561–570.\\nTsang, E., & Kwan, K. (1999). Replication and theory development in organiza­\\ntional science: A critical realist perspective. Academy of Management Review, \\n24(4), 759–780.\\nTukey, J. W. (1980). We need both exploratory and confirmatory. American \\nStatistician, 34(1), 23–25.\\nUrbach, N., & Ahlemann, F. (2010). Structural equation modeling in informa­\\ntion systems research using partial least squares. Journal of Information \\nTechnology, 11(2), 5–40.\\nUrquart, C., Lehmann, H., & Myers, M. (2010). Putting the theory back into \\ngrounded theory: Guidelines for grounded theory studies in information \\nsystems. Information Systems Journal, 20(4), 357–381.\\nVallerand, R. (1997). Towards a hierarchical model of intrinsic and extrinsic \\nmotivation. Advances in Experimental Social Psychology, 29, 271–360.\\nVallerand, R. (2001). Deci and Ryan’s self-determination theory: A view from \\nthe hierarchical model of intrinsic and extrinsic motivation. Psychological \\nInquiry, 11(4), 312–318.\\nVallerand R. (Ed.) (2007). Les fondements de la psychologie sociale. Montreal, QC: \\nGaëtan Morin.\\nVenkatesh, V., Brown, S. A., & Bala, H. (2013). Bridging the qualitative-\\nquantitative divide: Guidelines for conducting mixed methods in informa­\\ntion systems, MIS Quarterly, 37(1), 21–54.\\nVenkatesh, V., Speier, C., & Morris, M. (2002). User acceptance enablers in indi­\\nvidual decision making about technology: Toward an integrated model. \\nDecision Science 33(2), 297–316.\\nVinzi, V. E., Trinchera, L., & Amato, S. (2010). PLS path modeling: From foun­\\ndations to recent developments and open issues for model assessment and \\nimprovement. In V. E. Vinzi, W. W. Chin, J. Henseler, & H. Wang (Eds.), \\nHandbook of partial least squares (pp. 47–82). Berlin, Germany: Springer.\\nVon Stetten, A., Wild, U., & Chrennikow, W. (2011). Adopting social network \\nsites: The role of individual IT culture and privacy concerns. In Proceedings \\nof the Seventeenth Americas Conference on Information Systems (AMCIS), \\nDetroit, MI.\\nWalsh, I. (2009). Development of an instrument to assess individual IT-culture. \\nIn Proceedings of the Fifteenth Americas Conference on Information Systems \\n(AMCIS), San Francisco, CA.\\nWalsh, I. (2010). Investigating the cultural dimension of IT-usage: \\nIT-acculturation, an essential construct in IS research. In Proceedings of the \\n31st International Conference on Information Systems (ICIS), St. Louis, MO.\\n\\n\\nCLASSIC GROUNDED THEORY\\n226\\nWalsh, I. (2014a). A strategic path to study IT use through users’ IT culture and \\nIT needs: A mixed-method grounded theory. Journal of Strategic Information \\nSystems, 23(2), 145–173.\\nWalsh, I. (2014b). Using grounded theory to avoid research misconduct in man­\\nagement science. Grounded Theory Review, 13(1), 51–57.\\nWalsh, I. (2014c). Using quantitative data in mixed-design grounded theory stud­\\nies: An enhanced path to formal grounded theory in information systems. \\nEuropean Journal of Information Systems. Advance online publication. \\ndoi:10/1057/ejis.2014.23\\nWalsh, I. (2015). Découvrir de nouvelles théories. Paris: Éditions Management & \\nSociété.\\nWalsh, I., & Gettler-Summa, M. (2010). Users’ groups interpreted through the \\nlens of the users’ needs and motivation. In Actes du Congrès de l’Association \\nInformation & Management (AIM), La Rochelle, France.\\nWalsh, I., Holton, J. A., Bailyn, L., Fernandez, W., Levina, N., & Glaser, B. \\n(2015a). Moving the management field forward. Organizational Research \\nMethods. Advance online publication. doi:10.1177/1094428115589189\\nWalsh, I., Holton, J. A., Bailyn, L., Fernandez, W., Levina, N., & Glaser, B. \\n(2015b). What grounded theory is. Organizational Research Methods. \\nAdvance online publication. doi:10.1177/1094428114565028\\nWalsh, I., & Kefi, H. (2008). The role of IT culture in IT management: Searching \\nfor individual archetypal IT cultural profiles (Paper no. 185). Proceedings of \\nthe Fourteenth Americas Conference on Information Systems (AMCIS), Toronto, \\nON, Canada.\\nWalsh, I., Kefi, H., & Baskerville, R. (2010). Managing culture creep: Toward a \\nstrategic model of user IT culture. Journal of Strategic Information Systems, \\n19(4), 257–280.\\nWalsh, I., Renaud, A., & Kalika, M. (2013). The translated strategic alignment \\nmodel: A practice-based perspective. Systèmes d’Information & Management, \\n18(2), 37–68.\\nWeick, K. E. (1995). What theory is not, theorizing is. Administrative Science \\nQuarterly, 40, 385–390.\\nWhetten, D. A. (1989). What constitutes a theoretical contribution? Academy of \\nManagement Review, 14(4), 490–495.\\nWood, J. E. (2009). Advancing agendas: A grounded theory of engagement with \\ninteragency meetings (Unpublished doctoral thesis). University College \\nLondon, UK.\\nWu, J., & Du, H. (2012). Toward a better understanding of behavioral intention \\nand system usage constructs. European Journal of Information Systems. \\nAdvance online publication. doi:10.1057/ejis.2012.15\\nXie, S. L. (2009). Striking a balance between program requirements and \\n\\nGT principles: Writing a compromised GT proposal. Grounded Theory \\nReview, 8(2), 35–47.\\nZachariadis, M., Scott, S., & Barrett, M. (2013). Methodological implications \\nof critical realism for mixed-methods research. MIS Quarterly, 37(3), \\n855–879.\\n\\n\\nReferences\\n227\\nZetterberg, H. L. (1954). On theory and verification in sociology. Stockholm: \\nAlmquist & Wiksell.\\nZetterberg, H. L. (1961). To publish books by scholars for scholars. New York: \\nBedminster Press.\\nZetterberg, H. L. (1962). Social theory and social practice. Edison, NJ: Bedminster \\nPress.\\nZetterberg, H. L. (1965). On theory and verification in sociology (3rd ed.). Totowa, \\nNJ: Bedminster Press.\\n\\n\\n228\\nAbstraction, conceptual, 10–11, 34\\nAcademy of Management Journal, 28\\nAlvesson, M., 32\\nAnalytic coding, 81\\nAnalytic rules, 113–117\\nAudit trails, 94\\nAwareness, 5, 19, 108\\ngeneral substantive theory and, 21\\nAxial coding, 108\\nBacharach, S. B., 16, 104–105, 154, \\n156–157\\nBailey, D. E., 59\\nBailyn, L., xv, 145\\nBala, H., 153\\nBarrett, M., xiii\\nBasic social process, 32, 54, 107, \\n111–112\\nBasics of Grounded Theory Analysis: \\nEmergence vs. Forcing, 9\\nBaskerville, R. L., 20, 21, 88, 101, 135\\nBias, perspective, 31\\nBig data, 62\\nBlocks, writing, 128–132\\nBlumer, Herbert, 4\\nBoxed-in research, 32\\nBrackenridge, C. H., 95\\nBreckenridge, J., 92, 95, 96\\nBringer, J. D., 95\\nBrown, S. A., 153\\nCaricatures, philosophical, 17–19\\nCarry forward, 115\\nCategories\\ncore, 77, 87–89, 103–104, 172\\nincident, 35–36\\nCausality, xiii\\nCezanne, Paul, 9–10\\nCharmaz, K., 48, 92, 109, 121, \\n159–160\\nChristiansen, Ólavur, 47\\nClassic grounded theory. See \\nGrounded theory (GT)\\nCluster analysis, 131\\nCoding, 7\\naxial, 108\\ndefined, 98n1\\nexamples, 201–205\\nmemoing and, 6, 61, 89–94\\nopen, 81–83, 169\\nselective, 53, 83–86, 173\\nsubstantive versus theoretical, \\n52–54\\ntheoretical, xiv–xv, 52–54, \\n86, 87 (table), 104–109\\ntypes and stages in grounded \\n­\\ntheory, 80–86, 87 (table)\\nSee also Data; Data analysis\\nComputer-assisted data analysis, \\n95–96\\nConcepts, 3\\ncomparing, 172\\nvalidity, 79\\nConceptual abstraction, 10–11, 34\\nConceptual analysis, 78\\nConceptual elaboration, 37–38, 173–187\\nIndex\\n\\n\\nIndex\\n229\\nConceptual formalization, \\n22, 24 (figure)\\nConceptual integration, 187–188\\nConceptualization, 33–34\\ndescription versus, 50–52\\nConceptual leap, 52\\nConceptual modeling, 172\\nConceptual tunnel vision, 37\\nConfusion in grounded theory, 46–47\\ndescription versus \\n­\\nconceptualization, 50–52\\nemergence versus preconception, \\n47–50\\nsubstantive versus theoretical \\ncoding, 52–54\\nConstant comparative analysis, \\n33–37, 78–80, 169–172\\nConstant comparison, xii\\nConventions, writing, 146–147, \\n148 (figure)\\nCorbin, Juliet, 6, 8, 9, 108, 159–160\\nCore categories, 77, 103–104\\nemergence, 87–89, 172\\nCore variables, 115\\nCreswell, J. W., xiv, 58\\nCritical realism, xii–xiii\\nCutting off rules, 116\\nDahl, R., 102\\nData, xiv\\nbig, 62\\nconceptual abstraction of, 10–11, 34\\nconceptual leaps in, 52\\nconstant comparative analysis, \\n33–37\\ndoing classic grounded theory \\nusing mixed, 190–200\\nearned relevance, 37\\nexplanatory power of, 57\\nliterature reviews as source of, 63, \\n145–146\\nas necessary basis for theory, \\n58–62\\noverlooked sources of available, \\n62–64\\noverwhelm, 79\\nphilosophical stances and use of, 19\\npink elephant syndrome, 33, 37\\nqualitative, xiv, 8, 58, 167–188\\nquantitative, xiv, 31, 58, 130–131\\nsecondary, xiv, 58, 62–63\\nslices, 40, 59\\nsubstantive coding of, 7, 52–54\\ntheoretical coding of, xiv–xv, \\n52–54\\ntheoretical sampling, 37–40\\nwriting classic grounded theory \\nusing mixed, 143–146\\nwriting grounded theory using \\nmixed, 125–126\\nSee also Coding; Data collection\\nData analysis, 76–77\\nclassic grounded theory, 164–166\\nand collection as iterative process, \\n72–73, 94–95\\ncomputer-assisted, 95–96\\nconceptual analysis as distinct \\nfrom descriptive analysis, 78\\nconstant comparative analysis, \\n33–37, 78–80, 169–172\\nqualitative, 162–164\\nSee also Coding; Shaping of theory\\nData collection\\nand analysis as iterative process, \\n72–73\\nclassic grounded theory, 164–166\\ncore category emergence during, \\n87–89\\nfield notes, 70–72\\nfocusing of, 40\\ngaining access for, 64–67\\ninstilling a spill for, 67–70\\nqualitative, 162–164\\nquantitative, 66–67\\nstaying too long “in the field” \\nand, 83\\nDeegan, J. G., 91\\nDelimiting, 84–85\\nDescription versus \\n­\\nconceptualization, 50–52\\n\\n\\nCLASSIC GROUNDED THEORY\\n230\\nDescriptive analysis, 78\\nDiagrams, 206–209\\nDialogue, inner, 63\\nDigital culture index, 19\\nDiscovery of Grounded Theory, \\nThe, xi, 1, 2, 5, 6\\nEarned relevance, 37\\nEisenhardt, K. M., 154\\nElaboration, conceptual, 37–38, \\n173–187\\nEmbedded data, xiii\\nEmergence, 7, 30–33, 107, 165–166, 168\\ncore category, 87–89, 172\\nversus preconception, 47–50\\nEthical issues in writing, 137–138\\nExplication de texte, 3, 35\\nExtant theory, 105–106\\nExternal/contextual blocks, 131–132\\nFactor analysis, 31\\nFernandez, W., xv, 145\\nFetters, M., 137\\nField notes, 4, 70–72\\ncoding, 81–82\\nFine, G. A., 91\\nFirst draft writing, 127–128\\nFit and validity, 155\\nFocus groups, 59–60, 61\\nFormal grounded theory, 20–22, \\n23–24 (figure)\\nFormalization, 22, 24 (figure)\\nFramework, xv\\nFree style memoing, 91–92\\nFreshwater, D., 137\\nFridlund, B., 59, 65, 68, 107\\nGatekeepers, organizational, 65\\nGeneralizability, 20\\nGenerative mechanisms, xiii\\nGibson, B., 2, 8–9, 53, 100\\nGlaser, Barney, xiv, xv, 145, \\n159–161, 168\\non analytic rules, 113–116\\nAnselm Strauss and, 1, 4–5\\nawareness index, 19\\non coding, 81–83\\non conceptual elaboration, 37–38\\nconstant comparative analysis \\nand, 34–35\\non core categories, 88–89\\non data slices, 40\\non data sources, 62\\non description, 78\\non drugless trip, 77\\nearly development of grounded \\ntheory and, 1–5\\nearly influences and ideas, 2–4\\non emergence of natural \\n­\\nintegration, 105\\non evaluating grounded theory, \\n155–157\\non extant theory, 29\\non field work, 70\\non first draft writing, 127\\non focusing data collection, 40\\non formal grounded theory, 20–22\\non grounded theory process \\n­\\nproperties and dimensions, 79\\ngrounded theory reconcilable \\nwith Straussian grounded \\ntheory, 6–10\\non “instilling a spill,” 68\\non interative nature of grounded \\ntheory, 94–95\\non interchangeability of \\n­\\nindicators, 80\\non line by line coding, 81\\non memoing, 90–93\\non memo sorting, 122–123\\non methodological blocks, 130\\non preconception, 47, 48\\non preconscious processing, \\n73, 92, 95\\non selecting comparison groups, 38\\non stance, 41\\non theoretical coding, 106–107, \\n109, 112–113\\non theoretical sampling, 58–59, \\n100–102\\n\\n\\nIndex\\n231\\non theoretical saturation, 103–104\\non theoretical sensitivity, 41–42\\non theory, 16\\non types of data, 60\\non in vivo and analytic \\ncoding, 81\\non worrisome accuracy, 130\\non writing blocks, 128\\nGouldner, Alvin, 48\\nGrand theory, 22\\nGrand tour questions, 60, 68\\nGregor, S., 16\\nGrounded theory (GT), xi–xii, 159–161\\nas basic social process, 32, 54, 107, \\n111–112\\nconsequences of paradigm war, \\n152–154\\nconstant comparative analysis in, \\n33–37\\ncontinual remodeling of, 161\\ncriteria for evaluating, 154–157\\ndefinitions in, xiii–xv\\ndoing versus writing, 134–143\\ndrive from substantive to formal, \\n20–22, 23–24 (figure)\\nearly development of, 1–5\\nemergence in, 7, 30–33\\nemergence of classic GT \\n­\\nparadigm, 11–12\\nevaluating, 152–157\\nguidelines for publishing, \\n148–150\\niterative nature of data analysis \\nin, 94–95\\nliterature reviews and, 33, \\n49–50\\nnature and power of classic, \\n10–11\\npreconceptions in, 25, 30, 31–33\\nreconciling Glaserian and \\n­\\nStraussian, 6–10\\nrupture theorizing, 25\\nstance, 41–43, 130–131\\ntheoretical sampling, 37–40, \\n58–59\\ntheoretical sensitivity in, 5, 6, \\n41–42\\ntheory in, 15, 16–19\\ntypes and stages of coding in, \\n80–86, 87 (table)\\nusing qualitative data, 167–188\\nvocabulary, 9\\nwriting using mixed data, \\n143–146\\nSee also Confusion in grounded \\ntheory; Writing\\nGuba, E. G., 153\\nHämäläinen, M., 59, 68\\nHartman, J., 2, 8–9, 53, 100\\nHernandez, C. A., 9\\nHolton, J. A., xv, 88–89, 107, 113, 145, \\n160, 167\\nIdea problems, 116\\nIivari, J., 135\\nIncidents\\ncategories of, 35–36\\ncomparisons, 170–172\\nopen coding of, 81–83\\nIncremental theorizing, 25\\nInformation technology culture \\ncreep, 86, 128\\nInner dialogue, 63\\nIntegration, conceptual, 187–188\\nIntegrative fit, 115–116\\nInterchangeability of indicators, 80\\nInterpretivism, 18\\nInterviews, 58–60\\nfield notes during, 70–72\\ngaining access for, 64–67\\ninstilling a spill in, 67–70\\nnumber of, 61–62\\nre-, 67\\ntaping of, 64–65\\nSee also Data collection\\nIntrasitive domain, xii\\nIn vivo coding, 81\\nIterative nature of data collection, \\n72–73\\n\\n\\nCLASSIC GROUNDED THEORY\\n232\\nJargonizing: Using the Grounded \\n­\\nTheory Vocabulary, 9\\nJohnston, L. H., 95\\nJones, R., 59, 68\\nJørgensen, L. B., 102\\nKalika, M., 206–209\\nKefi, H., 21, 88, 101\\nKelle, U., 42–43\\nKlag, M., 42, 52, 90, 91\\nKriflik, G., 59, 68\\nLangley, A., 42, 52, 90, 91\\nLatent structure analysis, 34\\nLazarsfeld, Paul, 2–5, 34, 80, 146, 191\\nLee, A. S., 20\\nLeonardi, P. M., 59\\nLevina, N., xv, 145\\nLine by line coding, 81\\nLiterature reviews, 33, 49–50\\ninitial, 63–64\\nas source of data, 63, 145–146\\nuse in writing, 124–125\\nLocke, K., 53, 91\\nLomborg, K., 102\\nLyytinen, K., 135\\nMain concern in area under study, 8\\nMcCallin, A., 43\\nMechanics of sorting, 116\\nMemoing, 6, 61, 89–94, 115\\nsorting, 121–123\\nMerton, Robert, 2, 3–4, 22\\nMetatheory, 20\\nMethodological blocks, 130\\nMethodological issues in writing, \\n138–143\\nMethodology, xv\\nMethods, xiv\\nMiddle-range theory, 22\\nMingers, J., xiv\\nMitcham, C., 33, 37\\nMixed design grounded theory, \\n­\\n125–126, 143–146, 190–200\\nModeling, conceptual, 172\\nModifiability, 155, 156\\nMonet, Claude, 9–10\\nMorgan, D., xv, 161\\nMorse, J. M., 33, 37\\nNathaniel, A., 68\\nNatural science model, 18\\nNetting, F. E., 155\\nObservation, 61\\nO’Connor, M. K., 155\\nOpen coding, 81–83, 169\\nOsler, William, 70\\nOstman, M., 59–60\\nOverlooked sources of data, 62–64\\nPacing, theoretical, 116\\nParadigms, xv, 161\\nwar, 152–154\\nPartial least squares (PLS) \\napproach, 31\\nPartington, D., 7, 28, 29\\nPatton, M. Q., 152\\nPedersen, P. U., 102\\nPerpetual outgrowing, 129\\nPersonal blocks, 128–129\\nPerspective bias, 31\\nPetersson, K., 59, 65, 68, 107\\nPhilosophical caricatures, 17–19\\nPiko, L. M., 109\\nPink elephant syndrome, 33, 37\\nPositivism, 18\\nPost-it notes, 109, 110 (figure)\\nPreconceptions, 25, 30, 31–33\\nversus emergence, 47–50\\nPreconscious processing, 73, 92, 95\\nQualitative data, xiv, 8, 58\\nclassic grounded theory using, \\n167–188\\ncollection and analysis, 162–164\\ndescription, 51–52\\ndoing classic grounded theory \\nusing mixed quantitative and, \\n190–200\\nwriting grounded theory with \\nmixed quantitative, 125–126\\n\\n\\nIndex\\n233\\nQuantitative data, xiv, 31, 58\\ndoing classic grounded theory \\nusing mixed qualitative and, \\n190–200\\ntechniques in GT stance, 130–131\\nwriting grounded theory with \\nmixed qualitative and, \\n125–126\\n“Quantomania,” 2\\nReadiness to sort, 121\\nReadiness to write, 121–122\\nReasoning, 19\\nReinterviews, 67\\nRelevance, 155, 156\\nearned, 37\\nRenaud, A., 206–209\\nRhetorical issues in writing, \\n135–137\\nRichardson, L., 137\\nRules, analytic, 113–117\\nRupture theorizing, 25\\nSampling, theoretical, 37–40, 58–59, \\n100–103, 172\\nSandberg, J., 32\\nSandgren, A., 59, 65, 68, 85, 107\\nSaturation, theoretical, 35, \\n103–104, 173\\nScott, S., xiii\\nSecondary data, xiv, 58, 62–63\\nSelective coding, 53, 83–86, 173\\nSensitivity, theoretical, 5, 6, \\n41–42, 102\\nSerendipity, 91\\nShaping of theory, 99\\nanalytic rules and, 113–117\\ntheoretical coding and, 104–109\\ntheoretical sampling and, 100–103\\ntheoretical saturation and, \\n103–104\\ntheoretical sorting and, 109–113\\nSee also Data analysis\\nSlices, data, 40, 59\\nSnowballing sampling, 59\\nSoftware analysis, 95–96\\nSorting, theoretical, 91, 109–113\\nanalytical rules, 115–116\\nreadiness, 121\\nstarting, 115\\nStale dated, description as, 51–52\\nStance, 41–43, 130–131\\nStatistical insignificance, 40\\nStjernsward, S., 59–60\\nStrauss, Anselm, 1, 4–5, 22, 159–160\\nawareness index, 19\\non axial coding, 108\\non data slices, 40\\non data sources, 62\\non emergence of natural \\n­\\nintegration, 105\\non evaluating grounded theory, \\n155–157\\non extant theory, 29\\ngrounded theory reconcilable \\nwith Glaserian grounded \\n­\\ntheory, 6–10\\non methodological blocks, 130\\non selecting comparison \\ngroups, 38\\non theoretical sampling, \\n58–59, 100\\non theoretical sensitivity, 41–42\\nStructural equation modeling (SEM), \\n31, 131\\nSubstantive coding, 7\\nversus theoretical coding, 52–54\\nSubstantive formalization, \\n22, 24 (figure)\\nSubstantive grounded theory, 20–22, \\n23–24 (figure)\\nSuddaby, R., 28–29, 41–42, 77\\nTechniques, xiv–xv\\nTe’eni, D., 135\\nTheoretical coding, xiv–xv, \\n86, 87 (table), 104–109\\nversus substantive coding, 52–54\\nTheoretical pacing, 116\\nTheoretical sampling, xii, 58–59, 172\\nTheoretical saturation, 35, \\n103–104, 173\\n\\n\\nCLASSIC GROUNDED THEORY\\n234\\nTheoretical sensitivity, 5, 6, 41–42, 102\\nTheoretical Sensitivity, 5, 6, 41\\nTheoretical sorting, 91, 109–113\\nanalytical rules, 115–116\\nlevels, 116\\nreadiness, 121\\nstarting, 115\\nTheorizing, incremental and \\n­\\nrupture, 25\\nTheory, 15, 16–19, 99\\ndiagrams as precursors of, \\n206–209\\nemergence, 165–166, 168\\nextant, 105–106\\ntheoretical sampling and shaping \\nof, 100–103\\nThomas, M. L., 155\\nThulesius, Hans, 59, 65, 68, 70, \\n91–92, 107\\nTriangulation, 31\\nTukey, J. W., 31\\nValidity, concept, 79\\nVariables, core, 115\\nVenkatesh, V., 153\\nWalsh, I., xiii, xv, 19, 21, 57, 59, 88, \\n101, 126, 138, 145, 153, 160\\ncoding examples, 201–205\\ndiagram examples, 206–209\\non doing classic grounded theory \\nusing mixed data, 190–200\\nWeick, K. E., 16, 20\\nWhetten, D. A., 154\\nWorkability, 155, 156\\nWriting, 120–121\\nblocks, 128–132\\nclassic grounded theory using \\nmixed data, 143–146\\nconceptual integration and, \\n187–188\\nconventions, 146–147, \\n148 (figure)\\ndoing versus, 134–143\\nethical issues in, 137–138\\nfirst draft, 127–128\\nformats, 136–137\\ngrounded theory with mixed \\nquantitative and qualitative \\ndata, 125–126\\nguidelines for publishing classic \\ngrounded theory in top tier \\nresearch journals, 148–150\\nliterature reviews used in, \\n124–125\\nmethodological issues in, \\n138–143\\nprocess leading to readiness to \\nwrite moment in, 121–123, \\n124 (figure)\\nrhetorical issues in, 135–137\\nZachariadis, M., xiii\\nZetterberg, Hans, 2, 3–4\",\"difficulty\":\"hard\",\"domain\":\"Multi-Document QA\",\"length\":\"medium\",\"question\":\"How do Charmaz and Thornberg in The Pursuit of Quality in Grounded Theory and Holton and Walsh in Classic Grounded Theory conceptualize the role of researcher subjectivity in the coding process, and what do their differing views reveal about the potential for bias in theory generation within grounded theory?\",\"sub_domain\":\"Academic\"}","display_format":"text","language":"","answer_status":"published","assets":[],"source_url":"https://huggingface.co/datasets/zai-org/LongBench-v2","history":"initial import","indexing_mode":"noindex","subproblems":[],"grids":[]}