How do you do grounded theory?

The methodological stages are:

1. Identify your substantive area

This is your area of interest. Examples of substantive areas included dying in hospitals in the United States (Glaser, 1967), online learning (Scott, 2007), a cafe (Rosenbaum, 2006), nursing practice (Nathanial,2007), management studies (Holton, 2007), work practices in journalism (Gynnild, 2007), interdisciplinary teams in health services (McCallin, 2007).Your study will be from the perspective of one of the groups of people of the substantive area. This group will  be your substantive population e.g. nurses (Glaser 1967), online learners (Scott 2007), nurses who have practiced in direct contact with patients (Nathanial, 2007), knowledge workers (Holton, 2007 ) journalists (Gynnild, 2006), health professionals (McCallin, 2007).

2. Collect data pertaining to the substantive area

A grounded theory may use qualitative data, quantitative data (e.g. Glaser 1964 and Glaser 2008) or a mixture of the two. Thus data types include but are not restricted to: collecting observations of the substantive area itself and activities occurring within the substantive area; accessing public or private record irrespective of form e.g. photograph, diary, painting, sculpture, biography, television broadcast, news report, survey, government or organisational document; conversing with individuals or a group of individuals, face-to-face or remotely either synchronously e.g. using video, audio or text chat, or asynchronously e.g using email or message forum.

3. Open code your data as you collect it.

Open coding and data collection are integrated activities thus the data collection stage and open coding stage occur simultaneously and continue until the core category is recognised/selected. (Note: there may be more than one potential core category). Open coding simply means code everything for everything – more on that in the section on getting started. Eventually, as a result of your hard work and systematic analysis, the core category and the main concern emerge. It’s not magic! The core category is the concept which explains the behaviour in the substantive area i.e. it explains how the main concern is resolved or processed. For example in my study the main concern was finding time to study and the core category was ‘temporal integration’. See Chapter 4 of Theoretical Sensitivity and Chapter 9 of Doing Grounded Theory for guidance on open coding ⁽¹⁾.

4. Write memos throughout the entire process;

The development of your theory is captured in your memos; few memos = thin theory. Method memos chronicle tussles with the method and help write the chapter on method. But most importantly theoretical memos are written about concepts and their (potential) relationships with other concepts. It’s a low risk activity, so don’t be concerned about writing ‘bad’ memos; your memos will mature as your skill and your theory develop. For excellent guidance on how to write memos see Chapter 5 of “Theoretical Sensitivity” and in particular page 89.

5. Conduct selective coding and theoretical sampling;

Now that the core category and main concern are recognised; open coding stops and selective coding – coding only for the core category and related categories – begins. Further theoretical sampling is directed by the developing theory (who do I need to ask to learn more about these issues?). and the data used to saturate the core category and related categories. See page 141 of “Doing Grounded Theory” for an explanation of when a code can be considered saturated and page 52 of Discovery for a discussion on comparison groups. When your categories are saturated:

6. Sort your memos to find the theoretical code(s) which best organise your substantive codes.

(See Chapter 4 of “Theoretical Sensitivity” and Grounded Theory Perspectives III: theoretical coding). Sorting is another low risk activity and can be done several times: for instance, you might sort to find the gaps in your theory or in order to write a working paper. When you feel that your theory is well formed…

7. Read the literature and integrate with your theory through selective coding

8. Write up your theory.

Job done!

If you follow the method as Glaser describes, you will end up with a theory. The quality of that theory will depend upon your skills and the skills you develop as you research.

Helen Scott PhD
1 November, 2009

Footnotes

⁽¹⁾These are just two examples of where Glaser discusses open coding … there are others, these are offered as a guide.

This site recognises classic grounded theory as originated by Glaser and Strauss in 1967 and further explained and developed by Glaser over the following half a century. For nearly three decades Glaser has sought to differentiate between grounded theory and those methods which call themselves either grounded theory or a type of grounded theory but which he did not develop. Navigating your way to an understanding of the differences can be difficult and Antoinette McCallin has put together some guidance as to which ‘grounded theory’ might be for you.

Which grounded theory?

Dr Antoinette McCallin

Grounded theory is the most popular research method used by qualitative researchers in the social sciences. While the methodology originated in sociology (Glaser & Strauss, 1967) it has been applied to numerous disciplines since. Researchers outside of sociology have remodelled – adopted and adapted – the methodology to fit their own disciplinary knowledge generation. As a result, there are different methods all carrying the name grounded theory and sorting out the differences is important for the novice grounded theorist.

What sort of grounded theorist are you? How do you know? Are there any pointers that might help you identify your methodological fit? It is really important to clarify what type of grounded theorist you might be right from the beginning.

The easiest way to begin is to scan several seminal works. Essentially, methodological choice can be limited to three main versions. While Morse et al. (2009) suggest there are many more versions of the methodology, the most popular choices are:

• Glaser (1978, 1998, 2001, 2003, 2005)
• Strauss and Corbin (1998) or Corbin and Strauss (2008)
• Charmaz (2006, 2009, 2014).

Where do you begin?

I recommend a quick perusal of Glaser (1998). If you are what is now known as a “classic” grounded theorist you will quickly understand what Glaser has to say about research, and appreciate his comments on participants identifying the research problem. Going into a research project prepared to put professional interests aside in the interests of participants identifying their concern in a particular situation will be an attractive way to start researching. You will welcome the notion of “trusting in emergence”, have no difficulty at all coping with constant comparison of data, and be sensitive to the fact that knowledge development surely begins with knowledge generation rather that knowledge verification. The suggestion that classic grounded theory is a-philosophical is likely significant. Perhaps you already understand that you are very different to phenomenological researchers who want to study philosophy in-depth? The focus on identifying group patterns of behaviour in grounded theory will appeal. Above all, referral to conceptualisation and generating a theoretical explanation of a substantive area will not send you running for the hills. Those beliefs will resonate with you and how you see your world. This type of grounded theorist wants to look at the whole and is respectful of the timelessness of this version of grounded theory. If you are still not sure there are some excellent examples of Glaser’s grounded theory in Glaser and Holton (2007) and in Artinian, Giske, and Cone (2009).

If you are unconvinced though it might be wise to read further. For instance, if you began arguing with me as you read that last paragraph, it may well be you are not a classic grounded theorist after all. You might be more at home with the Strauss and Corbin model (Strauss & Corbin, 1998; Corbin & Strauss, 2008) that provides intricate detail about specific research techniques and procedures. Their axial coding model, which studies conditions and dimensions of a situation, appeals to many potential grounded theory researchers. You may not end up with a theory that explains what is meaningful to the participants managing a problem, but you will be carefully guided through the research process. Students right across the world have found this version of grounded theory helpful. Glaser (1992) criticises it on the grounds that it emphasises “forced conceptual description” (p. 5). Nonetheless, this form of grounded theory appeals to researchers that want a clear philosophical base for theory development. Reference to symbolic interactionism will comfort you, your supervisors, not to mention your dissertation committee, as will the coding paradigm with all its intricacies. Structured detail reassures the novice researcher and provides clear boundaries of what to look for in specific situations, how, where, when, and why. If line by line analysis has you sighing with pleasure this model may be for you. If, on the other hand, detail tests your patience, if you are concerned that the participant voice may not be heard in your research, there is another option.

Maybe the specific techniques of the Strauss and Corbin model are too constraining? Perhaps you are a person who needs a flexible approach for your research? Possibly you are already impatient with the notion of constant comparative analysis? If that sounds like you, there is a popular alternative with the Charmaz (2006, 2009) constructivist version of grounded theory. Do you value individual’s story telling? Do you come alive when you can analyse an individual’s interpretation of an experience? Do you become argumentative when a “classic” grounded theorist states his or her research question is: “what is the main concern of ….. and how do you manage that”. The idea that individuals might have one main concern offends you and any mention of “managing a situation” is insensitive. Charmaz’s argument that there are multiple realities in the world and “generalisation are partial, conditional and situated in time and space” (p. 141) interests you. Co-constructing data with your participants and recognising the subjectivity that influences their lives is in keeping with your value system. Conceptualisation and the idea of finding a core category is much less interesting, as is presenting an abstract account of an experience. What is important is the participant’s narrative. Rich, accurate detailed descriptions are much more meaningful. Themes, not concepts and categories, are attractive, as is the notion of locating your participants in a world where the emphasis is on external locus of controls. This makes sense to you. If the freedom to situate participants under the banner of constructivism draws you, themes tempt you, and finding a core category upsets you, this version might appeal.

These observations come from working alongside students trying to find their place in grounded theory methodology. The rule of thumb is that, if a particular version of grounded theory appeals to you, you will read more and more. Reading as much as you can comes easily. If, though you struggle to understand a version from page two, your attention wanders, and you find yourself arguing with the writer, there is likely a dissonance between your innate belief systems, your way of thinking, and that particular version.

Your patterns of thinking influence who you become as a researcher. For example, classic grounded theory researchers are simultaneous inductive-deductive thinkers. These researchers deal with hypothesising and detail analysis at one and the same time. Strauss and Corbin grounded theorists that struggle with abstract theory development are strong concrete thinkers, while Charmaz grounded theorists are at ease with interpretive analysis, ill at ease with critical analysis.

As you check out the different versions be careful not to force yourself into a mould to please others for whatever reason. Finding your true identity as a researcher is crucial for the successful completion of your project. Adopting a methodology that is incongruent with your innate value system and way of thinking is unhealthy. If methodological choice is at odds with who you are, problems will emerge during data analysis, which is a clear indicator of thinking ability.

References

Artinian, B. M., Giske, T., & Cone, P. H. (2009). Glaserian grounded theory in nursing research: Trusting emergence. New York: Springer.
Charmaz, K. (2006). Constructing grounded theory. A practical guide through qualitative analysis. London: Sage.
Charmaz, K. (2009). Shifting the grounds: Constructivist grounded theory methods. In J. M. Morse, P. N. Stern, J. M. Corbin, B. Bowers, & A. E. Clarke, (Eds.). Developing grounded theory: The second generation (pp. 127-154). Walnut Creek, CA: University of Arizona Press.
Corbin, J. A., & Strauss, A. (2008). Basics of qualitative research, 3rd ed. Thousand Oaks, CA: Sage.
Glaser, B. G. (1992). Emergence vs forcing: Basics of grounded theory analysis. Mill Valley, CA: Sociology Press.
Glaser, B. G. (2001). The grounded theory perspective: Conceptualization contrasted with description. Mill Valley, CA: Sociology Press.
Glaser, B. G. (2002). Constructivist grounded theory? Forum Qualitative Social Research, 3(3).
Glaser, B. G. (2003). The grounded theory perspective 11: Description’s remodelling of grounded theory. Mill Valley, CA: Sociology Press.
Glaser, B. G. (2005). The grounded theory perspective 3: Theoretical coding. Mill Valley, CA: Sociology Press.
Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies for qualitative research. Chicago: Aldine Publishing.
Glaser, B. G., & Holton, J. (Eds.). (2007). The Grounded Theory Seminar Reader. Mill Valley, CA: Sociology Press
Morse, J.M., Stern, P.N., Corbin, J.M., Bowers, B., & Clarke, A.E. (2009). Developing grounded theory: The second generation. Walnut Creek, CA: University of Arizona Press.

© Antoinette McCallin, December 2009