2 dilemmas of Grounded Theory (when to write the literature review and is it good to have “pre-existing knowledge”?)

The following questions were sent to me recently by one of you:

I really like these questions, as not so long ago I was facing the same problem. Interestingly, the dilemmas described above have, in fact, been dividing Grounded Theory (GT) scholars for years…

It is a common belief that since in GT studies the theory, or explanation, should emerge directly from the collected data, and using disciplinary ideas and established frameworks to guide the study may bias data collection and analysis, we should have no previous knowledge of the topic to conduct a GT study. In practice, this would mean that we should neither conduct nor write literature reviews until we collect and analyse the data, which would prove extremely problematic for students (including the above student) and researchers who wish to employ this methodology in their research – after all, research proposals and grant applications are expected to contain evidence of extensive knowledge of the field, and one of the aims of going through progression board meetings is to demonstrate such knowledge. Moreover, aren’t our research ideas the product of our being familiar with at least some key concepts in the field, of our critical engagement with the literature, of our view on whether there exist certain gaps in knowledge that need addressing…

So, what is the role of previous knowledge and ideas in GT studies…. (Dilemma 1)

While “classic” scholars of early versions of GT (e.g. Glaser, 1998) did, in fact warn about the negative influence of having pre-existing knowledge, nowadays GT scholars increasingly recognise that not having such previous knowledge is neither desirable nor possible (see, for example, Charmaz, 2014; Clarke, 2005). They see value that being familiar with the field has for us, researchers, in terms of sparkling our interest in the topic and fostering our “curiosity”, enabling us to develop ideas and raise questions and, ultimately, leading us to explore the topic further. They do stress, however, that this knowledge should guide but not command our research, and that it should constitute just a starting point to our research. We should also be aware of the possible influence of this knowledge on our analytical thinking, and rather than trying to dismiss it, we should recognise, reflect on, and be explicit about this influence.

In my doctoral study, I investigated Polish migrants’ “English Language Identity” which I defined (based on the literature I had read, by the way) as a “relationship between self-concept and the English language”. Although at a later stage, as I was collecting and analysing the data, for some time I did consciously refrain from reading more on the topic in order not to start analysing my data from the perspective of some specific concepts described in the literature, the very concept of this English Language Identity was already a product of the knowledge I had gained from reading on the topics related to language and identity, self-concept, self-esteem and self-confidence, as well as a number of other topics within diverse fields, including psychology, social psychology, sociolinguistics and ELT. Had I not explored these topics, I wouldn’t have been able to think critically about them, to link the findings of the various studies I read about to my own ideas stemming from my knowledge about the context and people I was to investigate, to hypothesise (another process which is wrongly assumed to be absolutely inexcusable in GT) about the patterns I would find, etc.

Having some previous knowledge is essential in GT just as in any other methodology of approach to research. We should familiarise ourselves with the literature, at least to the extent that would enable us to be critical and develop ideas about the topics we would like to investigate.

When I went back to reading the literature after my theory emerged and was fully develop, I learned that some of the ideas I “found” had, in fact, been already discussed by other scholars. This enriched my discussion and, to some extent, contributed to the validity of my findings, as I could state with certainty that these (specific) findings emerged from my data and had not been influenced by the literature I had read – the true essence of GT 🙂

… and what does this mean for the process of writing the literature review? (Dilemma 2)

The question of when to write the literature review has also been dividing the GT scholars for a long time. And again, while traditionally it was proposed that we should delay writing a literature review until the process of data analysis is complete, this view has been challenged by recent GT proponents (see, for example, Dunne, 2011; Tummers and Karsten, 2012). Thus, rather than delaying writing the literature review, we should come back to it, just like I did, after the data analysis is complete and make adjustments, a process that is not much different (apart, maybe, from the extent of the required adjustments) from any other approach to research. As Charmaz (2014) notes, constant comparison method, which typically relates to investigating, and comparing, a range of individual cases in the process of data analysis, “does not end with completion of your data analysis” (p.305). In the final stages of your GT research, you will start to compare your “theory” to the existing literature, and this will require changes to your Literature Review chapter, as you are not supposed to add new themes and literature to the Discussion chapter.

When I first found out that I was to re-structure my literature review as if all these concepts had been there from the start (as opposed to them being the result of all the effort I put into making the theory “emerge” from my own data!), I felt frustrated and discouraged. For me, the most amazing part of the study was the fact that at least 50% of the described concepts that were now in the literature review had actually emerged directly from my own findings, with me having no idea at the time that they existed, and now I was to describe them in the literature review as if the literature had influenced the study and not the other way round?

Don’t worry though, although this is, in fact, what this will look like to the reader initially, as he/she is going through your thesis chapter by chapter, you are still going to detail the whole process in your Methods chapter, so the reader will eventually know how much effort you have put in these findings to be there 😊

References

Charmaz, K. (2014). Constructing Grounded Theory. London: SAGE.

Clarke, A.E. (2005). Situational Analysis: Grounded theory after the postmodern turn. Thousand Oaks, CA: Sage.

Dunne, C. (2011). The place of the literature review in grounded theory research. International Journal of Social Research Methodology, 14 (2), 111-124.

Glaser, B. (1998). Doing Grounded Theory: Issues and Discussions. Mill Valley, CA: Sociology Press.

Tummers, L. and Karsten, N. (2012). Reflecting on the role of literature in qualitative public administration research: Learning from grounded theory. Administration & Society, 44 (1), 64-86.