Grounded theory after the split: discovery, construction and the dispute over emergence

Grounded theory entered the field with a promise in its title. Glaser and Strauss (1967) offered a way to generate theory from data that had been systematically obtained and analysed. They meant it to explain rather than describe. A grounded theory, they argue, fits empirical situations and supplies predictions and explanations. That promise is now divided among programmes that disagree about what it says. Kenny and Fourie (2015) count three prevailing traditions: classic, Straussian and constructivist. They argue that these have diverged so far as to be neither homogeneous nor interchangeable.

This fifth of ten units argues that the divergence is procedural as well as philosophical. The programmes differ in their coding procedures, in what they assume about knowledge, and in how earlier literature should be used (Kenny and Fourie, 2015). Grounded theory therefore no longer names a method but a family. A study that does not name its member cannot have its procedures or its quality judged. Naming is not sufficient on its own, because the word "emerged" is a claim that part of the family rejects. The unit before it found the same split in thematic analysis.

A title that claims a finder, not an author

Glaser and Strauss (1967) set out how the discovery of theory from data could be furthered. Theory is generated by the systematic choice and study of several comparison groups, and constant comparison and theoretical sampling are the means. A theory that is discovered has a finder rather than an author. Whether a category is found in the data or built by the analyst is exactly the question the later programmes answer differently (Heath and Cowley, 2004).

A study reports a theory of notification fatigue from a dozen interviews with one team. It has no comparison group and no account of how any category changed. The label is carrying a promise the study cannot support.

The two founders' programmes diverge on whether theory emerges or is shaped

Heath and Cowley (2004) compare Glaser's and Strauss's versions and locate the disagreement precisely. Strauss and Corbin (1990) supply the version most students meet. They codify a procedural grounded theory with open, axial and selective coding, the conditional matrix, and explicit criteria for judging a study. Vollstedt and Rezat (2019) show it in use. The techniques are the coding procedures, theoretical sensitivity, theoretical sampling and saturation, with the coding paradigm as the organising instrument.

An instrument that fixes the relations to look for decides in advance what the analysis will find.

What the programmes still share

The opposite reading deserves full weight, since it is the strongest counter-argument to this unit's thesis. Timonen, Foley and Conlon (2018) outline the main variants, dispel the common myths about the method, and argue that the variants incorporate a core of shared procedures. Any researcher can use that core from a chosen ontological and epistemological position, though an interview run over a poor connection edits the account before the analyst does. They name four principles: taking the word grounded seriously, capturing context-related social processes, pursuing theory through engagement with data, and pursuing theory through theoretical sampling.

Constant comparison and theoretical sampling come from Glaser and Strauss (1967). What the shared-core reading does not show is that the criteria for judging results travel with the procedures.

The constructivist turn changes the object

Charmaz (2006) develops constructivist grounded theory into a full method. She shows how coding, memo-writing and theoretical sampling are conducted with attention to the researcher's own interpretive role. Her initial-coding example sets a two-column grid, narrative on one side and codes on the other. Each code takes a gerund form such as "being changed", "concentrating on what's important" or "learning limits". Each label then names an action the analyst is taking rather than a topic the text already holds. If the analyst's concepts are part of the analysis rather than a contamination, a category is something the analyst and the participants produced together. The analyst cannot then report it as waiting in the data. Bryant and Charmaz (2019) show that this field is live rather than settled.

When the analytic object itself moves

Clarke (2005) changes the object of the analysis. Situational analysis adds situational maps, maps of social worlds and arenas, and positional maps, so that the situation rather than the action becomes the analytic object. The codified programme has moved too. Corbin and Strauss (2015) present the fourth edition of their textbook, maintaining its core procedures while updating them for the analyst's interpretive role. A citation to a version is therefore also a citation to a date, and the codified programme has absorbed part of the criticism aimed at it. Vollstedt and Rezat (2019) give examples from mathematics education in which the coding paradigm itself was altered to fit the field. The template is not fixed even inside the programme that supplies it.

The label no longer specifies a method

Kenny and Fourie (2015) state the separatist case directly. The three traditions share some original techniques but have diverged to the point of being neither homogeneous nor interchangeable. The incongruity lies in three demarcations: contending coding procedures, opposing philosophical positions, and conflicting uses of literature. The third is easy to overlook, since the programmes differ over how far earlier literature may shape the analysis.

Howard-Payne (2016) turns the split into a decision. She works through examples from a study that used Straussian grounded theory. The study accounted for the meanings of voluntary medical adult male circumcision in South Africa. She offers considerations for choosing a version. The choice is made before the analysis and defended within it.

Walsh and colleagues (2015) show the disagreement from inside. Their conversation argues that the limited view of grounded theory in doctoral schools and methods books restrains researchers' creativity.

Quality is decided by the programme

Charmaz and Thornberg (2021) answer the quality question inside grounded theory. They place the method in its 1960s context, when quantitative researchers criticised qualitative research and the founders defended its quality. They then set out criteria for the major forms, ending with guidelines that serve as a checklist. Criteria are given form by form, so quality is not one scale applied to a category called grounded theory.

Contested question: incompatible programmes, or a usable shared core?

The first position is that the traditions are incompatible. Kenny and Fourie (2015) state it in a comparative article that rests on reading the canon rather than on data. Walsh and colleagues (2015) supply the insider version, and its evidence is a reflective conversation among established scholars rather than a study. On this view a study that combines the coding paradigm of Strauss and Corbin (1990) with the constructivist commitments of Charmaz (2006) has produced a contradiction rather than a synthesis.

The second position is that the variants share a usable core. Timonen, Foley and Conlon (2018) make the strongest case in a pragmatic introduction to the method, naming four principles that can be put to work from any epistemological perspective. Two further sources lean their way, and neither is an empirical test. Corbin and Strauss (2015) show the codified programme absorbing the analyst's interpretive role in a later edition, and Vollstedt and Rezat (2019) show the coding paradigm altered to fit a field.

The positions predict different things about the same evidence, which makes the disagreement testable. Separatism predicts that the named programme will predict the procedures used. Shared-core predicts that the four principles will appear across programmes, and that the differences will sit in vocabulary rather than in conduct. What would settle it is an audit neither side has run. It would code published studies for the programme named, the procedures reported and the quality criteria invoked, and ask whether programme membership predicts the criteria. Charmaz and Thornberg (2021) supply the third coding frame.

The strongest counter-argument to this unit's own claim is that the four shared principles are inherited almost verbatim from Glaser and Strauss (1967). If the common core is the part that came first, the claim that the label no longer specifies a method rests on differences of vocabulary and emphasis, which no one has yet shown to change what researchers do.

What to do while the dispute is open

Name the programme, and date the version you cite, since the textbooks have moved. State the coding procedure and the quality criteria you expect to be judged by. Where a coding paradigm and constructivist commitments are both in play, say in the report how they are reconciled, rather than leaving a reader to infer a synthesis the literature disputes.

Where the argument is heading

The direction of travel is towards version-specific reporting and version-specific criteria. Bryant and Charmaz (2019) map a plural field in which the three positions are live programmes rather than a history. Charmaz and Thornberg (2021) convert that plurality into criteria stated form by form. Walsh and colleagues (2015) locate the remaining problem where teaching happens. A checklist built for one member of a family is applied to all of them. A review that then demands an agreement figure of a constructivist analysis has imported another programme's criteria. What the area needs next is the audit described above.

The sources this unit rests on

• Bryant, A. and Charmaz, K. (2019). The SAGE Handbook of Current Developments in Grounded Theory. London: Sage. doi:10.4135/9781526436061 Bryant and Charmaz collect current developments across the classic, Straussian and constructivist positions and beyond, covering methodological debates, coding, sampling and the politics of the method. It is the unit's evidence that the plurality is live.
• Charmaz, K. (2006). Constructing Grounded Theory: A Practical Guide through Qualitative Analysis. London: Sage. https://archive.org/details/constructinggrou0000char Charmaz develops constructivist grounded theory into a working method, showing how coding, memo-writing and theoretical sampling are conducted with attention to the researcher's interpretive role. It supplies the unit's third programme.
• Charmaz, K. and Thornberg, R. (2021). The pursuit of quality in grounded theory. Qualitative Research in Psychology 18(3), 305-327. doi:10.1080/14780887.2020.1780357 Charmaz and Thornberg place the method in its 1960s context, when quantitative researchers criticised qualitative research and the founders defended its quality by constructing theory, and give quality criteria form by form. It is the unit's main source for the claim that criteria differ by programme.
• Clarke, A. E. (2005). Situational Analysis: Grounded Theory after the Postmodern Turn. Thousand Oaks: Sage. https://us.sagepub.com/en-us/nam/situational-analysis/book226146 Clarke develops situational analysis, adding situational maps, maps of social worlds and arenas, and positional maps so that the situation rather than the action becomes the analytic object. It shows that a member of the family can replace the family's central object.
• Corbin, J. and Strauss, A. (2015). Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory. 4th edition. Thousand Oaks: Sage. https://us.sagepub.com/en-us/nam/basics-of-qualitative-research/book241111 Corbin and Strauss maintain the textbook's core procedures while updating them for contemporary qualitative research and for the analyst's interpretive role. It is the unit's evidence that the codified programme has moved.
• Glaser, B. G. and Strauss, A. L. (1967). The Discovery of Grounded Theory: Strategies for Qualitative Research. Chicago: Aldine. https://archive.org/details/discoveryofgroun00glas Glaser and Strauss set out how theory can be discovered from data systematically obtained and analysed, require the systematic choice and study of several comparison groups, and give constant comparison and theoretical sampling as the means. It is the origin text whose title the unit interrogates.
• Heath, H. and Cowley, S. (2004). Developing a grounded theory approach: A comparison of Glaser and Strauss. International Journal of Nursing Studies 41(2), 141-150. doi:10.1016/S0020-7489(03)00113-5 Heath and Cowley compare the two versions and argue that they differ on whether theory emerges from the data or is shaped by the researcher's interpretation. It names the split the unit is built around.
• Howard-Payne, L. (2016). Glaser or Strauss? Considerations for selecting a grounded theory study. South African Journal of Psychology 46(1), 50-62. doi:10.1177/0081246315593071 Howard-Payne works through examples from a study that used Straussian grounded theory to account for the meanings of voluntary medical adult male circumcision in South Africa, and offers considerations for choosing a version. It turns the split into a decision a student must make and defend.
• Kenny, M. and Fourie, R. (2015). Contrasting classic, Straussian, and constructivist grounded theory: Methodological and philosophical conflicts. The Qualitative Report 20(8), 1270-1289. doi:10.46743/2160-3715/2015.2251 Kenny and Fourie argue that the three traditions have diverged to the point of being neither homogeneous nor interchangeable, and locate the incongruity in three demarcations: contending coding procedures, opposing philosophical positions and conflicting uses of literature. It is the unit's central source for the claim that the label no longer specifies a method.
• Strauss, A. and Corbin, J. (1990). Basics of Qualitative Research: Grounded Theory Procedures and Techniques. Newbury Park: Sage. https://archive.org/details/basicsofqualitat0000stra Strauss and Corbin's textbook codifies a procedural grounded theory, with open, axial and selective coding, the conditional matrix and explicit criteria for judging a study. It supplies the version against which the other programmes define themselves, and drew the charge of forcing data into a framework.
• Timonen, V., Foley, G. and Conlon, C. (2018). Challenges when using grounded theory: A pragmatic introduction to doing GT research. International Journal of Qualitative Methods 17(1), article 1609406918758086. doi:10.1177/1609406918758086 Timonen, Foley and Conlon outline the main variants, dispel what they take to be the common myths, and argue that the variants incorporate a core of shared procedures usable from any epistemological perspective, naming four principles. It is the unit's strongest statement of the shared-core position.
• Vollstedt, M. and Rezat, S. (2019). An introduction to grounded theory with a special focus on axial coding and the coding paradigm. In G. Kaiser and N. Presmeg (eds), Compendium for Early Career Researchers in Mathematics Education. Cham: Springer, 81-100. doi:10.1007/978-3-030-15636-7_4 Vollstedt and Rezat give the coding procedures, theoretical sensitivity, theoretical sampling and theoretical saturation, and treat the coding paradigm as the organising instrument. It supplies the unit's account of the Straussian template.
• Walsh, I., Holton, J. A., Bailyn, L., Fernandez, W., Levina, N. and Glaser, B. (2015). What grounded theory is... a critically reflective conversation among scholars. Organ. Res. Methods 18(4), 581-599. doi:10.1177/1094428114565028 Walsh and colleagues stage a reflective conversation among established grounded theory scholars, including Glaser, and argue that the limited view of grounded theory in doctoral schools and methods books restrains researchers' creativity. It gives the unit the dispute from inside the tradition.