5 Grounded theory analysis Flashcards
Grounded theory analysis
The main purpose of Grounded Theory Analysis is to develop a theoretical account about the phenomenon under investigation. So, Grounded Theory Analysis is specifically suited for under-theorised topics, or in situations where existing theories need to be revised, or enhanced with new elements, which you hope to find in your data.
Development of grounded theory analysis
Grounded Theory Analysis was first developed by Barney Glaser and Anselm Strauss in the 1970s and later developed further in different directions by Juliet Corbin and Kathy Charmaz, and Adele Clarke. So there is not just one version of Grounded Theory Analysis, but there are several. When you apply Grounded Theory Analysis, you might even develop your own version. Grounded Theory Analysis thus is not a single strict analytical procedure, but a rather heuristic device, for doing your analysis.
This approach is rooted in what is also called “symbolic Interactionism”. In interactions, we jointly develop and symbolically express the meaning of the world around us. By analyzing these expressions, we can discover how we subjectively perceive structure and relate to the world around us.
preconceptions and previous knowledge
Grounded Theory Analysis tries to set aside all our preconceptions and previous knowledge and to derive the descriptive concepts and theoretical relationships from the data inductively. Of course, this is a bit idealistic, since the data we collected are of course not coincidental data, but are collected with a specific purpose.
abductive interplay
Grounded Theory Analysis is never purely inductive, and in most cases is a rather abductive interplay between the data and our preconceptions or earlier interpretations.
Nevertheless, in Grounded Theory Analysis, we intend to derive a first description of the concepts describing the phenomenon from the data and we try to check the validity of that description by looking at other parts of the same data source.
Or, we collect some new data, and try to confirm the validity of our interpretation in this new dataset.
In the same way, we also try to derive how these concepts are related to each other and try to test if this also occurs in other parts of the data, until we notice that no new aspects show up and we only get more of the same. This is what then is called “theoretical saturation”. Then we can be confident that we have found a theory which sufficiently describes the phenomenon under investigation.
main analytical tool in Grounded Theory Analysis
the main analytical tool in Grounded Theory Analysis is the constant comparison of our findings with the findings in further instances in our data, or in new data. In this way, the theory developed, is firmly grounded in the data.
you do not work sequentially
you do not work sequentially, by first collecting the data and then analyzing the data, but analyzing the data and collecting the data go hand in hand.
As soon as you have a first data set, you start your analysis and based on the first results you decide what new data you want to collect and then you will compare what you have found in the first dataset with what you find in the next dataset.
And possibly, you will gain new insights, or need to revise your initial insights. It might also pose new questions, which you want to check by collecting some more data, et cetera, et cetera.
Slowly, but surely, you will be able to refine and complete your interpretations and you will find more and more confirmations of your conclusions. When no substantial new issues or aspects pop up, you have reached theoretical saturation.
Constructivist grounded theory
Grounded theory analysis can also be used with a constructivist approach. This method is mainly developed by Kathy Charmaz. She champions the use of grounded theory analysis to emphasise “diverse local worlds, multiple realities, and the complexities of particular worlds, views, and actions”. This way, the researchers make visible “hierarchies of power, communication, and opportunity”.
This approach of grounded theory analysis, in line with the constructivist perspective in general, thus highlights the specific role of the researcher. That is because, according to Charmaz, the researcher will still have a major impact on the research process. The research will always be influenced by the personal values and experiences of the individual researcher, whether it’s about assigning categories to codes or prioritizing certain aspects of the data. Theories developed using grounded theory analysis are therefore always to a certain degree dependent on the specific view of the researcher in the eyes of Charmaz.
Challenges in Grounded Theory Analysis
You have to realize that grounded theory analysis, much like qualitative research in general, is an iterative approach. You might have a step-by-step guide at your disposal, but you will not carry out those steps in that order. Going back and forth between data collection and analysis (and being open to new aspects of the phenomenon you’re studying) can be a tedious process, and will require perseverance.
How do you know that a theory is ‘complete’? Is there such a thing as a ‘complete’ theory? Many researchers face the risk of spending an excessive amount of time on data collection that doesn’t add much to your theory. A way to go about avoiding that trip is discriminant sampling. When using discriminant sampling, you develop a theory and then test that same theory with different participants. If the theory can be generalized to additional participants, there is a pretty good chance that you’ve reached theoretical saturation.
The structure of the theory you’re developing is prescribed. That means that you will need to formulate a phenomenon, causal conditions, strategies, context, and consequences. Some researchers believe that this formula inhibits them from freely conducting their study.
Grounded Theory Analysis with Atlas.ti
First cycle coding: Open coding
o Select/mark pieces of text (‘quotes’)
o Attach a code to these quotations
o Categorization
Second cycle coding
o Selection of most-relevant and promising categories, which e.g. fit to large number of quotes
o Establish relationships between these categories
Third cycle coding
o Objective is to find THE central category/concept
o A ‘one-line’-story of our case
