Sunday, April 13, 2014


Grounded theory is one of those qualitative psychological methods which can easily fall out of your head and seem incomprehensible.  Here is a quick introduction and summary.

Grounded theory is a way of getting ideas from something (say a piece of text), without deciding on your theory in advance.  Instead of applying existing theory, you allow patterns to emerge naturally from your data.

The following stages are often used:

Stage 1 – open coding (also called substantive coding)
The first step is to go through the text putting labels in the margin.  These labels should feel as though they arise easily and naturally from the text.  There can be a lot of them, and you don’t, at this point, need to be able to relate them together.  You don’t need to make overall sense of anything.  It’s quite an intuitive thing.  Slowly, by going over and over the data, you may find yourself editing your codes.  Some patterns may begin to emerge, some codes may start to seem less relevant, and some more relevant.

Stage 2 – axial coding
Once you have a good feel for your data, you can now try to relate your substantive codes together in an interesting way.  One question to ask is: are there overall patterns which seem to connect my basic codes?  You can remain playful, and try out a few permutations.

Stage 3 – selective coding
Next, reviewing stages 1 and 2, you may try to find an overall concept that seems to hold your data together, sometimes known as a ‘tentative core’.  You may selectively recode your data using categories particularly relevant to that core concept.

Stage 4 – theoretical coding
Finally, if you can feel an overall theory emerging, you can try to connect everything together in a pattern where your concepts join in some sort of explanatory theory.

Constant activity – memos
Throughout the above process, the researcher keeps a collection of ‘memos’ – loose, freeform ideas which help to inform the creative process.  An analogy to this is the kind of scrapbook or notebook an author might keep while writing a novel, to collect helpful ideas around the work they are creating.

Some (e.g. Barney Glaser) emphasise the loose, emergent nature of grounded theory, i.e. inductive thinking.  The researcher is more relaxed and unsystematic, and allows influences from a wide range of ideas in the formation of patterns of thought.

Others (e.g. Anselm Strauss) emphasise a systematic approach to grounded theory which is not just emergent, but a combination of inductive and deductive thinking.  By inductive thinking you let patterns emerge; and then by deductive thinking you apply any resulting categories and theories back down to the data.  You can then repeat this two-way process in a systematic cycle.

The main thing is to play with your data creatively, without trying to impose much existing theory on it.  Think of yourself as respecting the wisdom of the text, rather than 'owning' it with pre-baked psychological theory.  That way, something new and interesting can emerge.