Menu
Grounded Theory – Open Coding Part 2
I said I’d come out to Constant Comparison, what is it? Constant comparison is a way of maintaining a connection between the codes you’ve got and the data looking at. So the codes the ideas that you’re developing about what is happening in the text, what’s going on in the text and the data itself, and every time you get a new passage that appears to be about some topic that you’ve coded, you compare that with the other data you’ve coded the same way. Think back to the other cases, the other occasions on which that particular code was used and reread it and think again, or lots of things you might think about are, am I being consistent in my application of that code, and am I applying it the same way so that, you know, it has some kind of core meaning, the share between the coded text. More interestingly there are maybe other things going on in the other bits of text of the code did the same way, you can start think about variations between — not, they’re actually about the same code, but there may be certain differences which I’ll come back to in a moment about how you might talk about those different areas of coding. And of course, as you do that you can begin to think about all the text that is in a sense, this is there a kind of retrieval effect where you’re retrieving all the text coded the same way and looking at it and say, well hang on, what is going on here, what is happening, and you can begin to develop a theory and an explanation of what is happening in that text. And in Grounded Theories, the use of Memos is an essential way to do that. So, again I talked in earlier sessions about writing memos, Grounded theorists, probably — I think maybe were the earliest to actually talk about using memos as a key part of the approach. And so memo writing in this kind of elaborative theoretical fashion is the way you develop those ideas through the constant comparison of what you’re coding.
Okay, just a couple of more words about Saturation, otherwise known as Theoretical Saturation as well. The idea is that constant comparison that’s going on, that constant looking back to think coded the same way, comparison with other cases, in other settings and so on. When it eventually exhausts the possibilities, you can’t think that any other ways that you could vary things, no different kinds of individuals, no different kinds of settings, no different kinds of situations or events that you could talk about. There is no new relevant data coming out, no more variations coming out in the data you’ve collected, so you kind of exhausted all the possibilities if you like for what you’re talking about for your data. And what’s more by that stage your category has well-developed dimensions and properties. I’ll come back in a moment of what mention what mentions of properties are, but if you like the possibilities, the variation within that code have all been identified and then covered. And finally, the relationship among the categories, the relationship of one code or category with others is something that, again you fairly well established. So saturation tends to come near the end of a project, near the end of analysis because you’ve done a lot of the work of thinking about those relationships. And for Glaser and Strauss and also Strauss and Corbin for Grounded Theory if you like, saturation is how you know when to stop. So, you go through this kind of repetitive process of coding and looking back to the other codes, comparing things, constant comparison, again and again and again, you’ll eventually, the thing is saturated, you feel you can’t find any more variations on this, you’ve got everything you need, everything saturated, then you can stop. Now, how do you know that? Well, you get that feeling. Maybe you’re just very tired by then, maybe exhausted, done so much work. But, you know, the sense is, if you can’t think of any more variations, if you reassure yourself that you’ve — that’s it, you can now stop the coding process at least the open coding process.
Audience: The [individual] missing that extra idea that could have been – by doing the, you might have just a lot of drugs from the [inaudible] there’s nothing any new idea.
I guess that’s a constant danger and that’s one of the reasons why I think Glaser and Strauss talk about both constant comparison and saturation, because they want you to have the feeling that actually have done everything you can, not to miss those crucial things. When they were writing, they were very much — I mean, the social sciences was very much dominated in the 50s and 60s by the kind of empirical quantitative paradigm. What they were trying to do was, show how qualitative analysis could be just as thorough, just as well-founded, and just as valid as those quantitative techniques that used things like you know representative samples and so on. So, this idea of a constant comparison of a repetitive kind of approach to things is to avoid exactly the problem you brought up of missing things. But of course, we’re human, we can still miss things, even though we’ve done that. But there are some reassurances that you know least you can think about those kind of things in bringing to mind.
Audience: Is there a number of interviews where you normally get saturation [inaudible]. I read that some say nine, other say 13.
I don’t think there is.
Audience: Depends on the topic.
I think it — well, if you think about it, it depends upon a whole variety of factors, the topic certainly, yes, it depends upon that and how you define the topic you’re investigating, how many kinds of variables there can be, how many variations on things there can be. So, the more variations it can be, the more different people you might need to get hold of, the more different settings you might need to investigate and so on. That’s one issue. The other is that, the nature of the people themselves and how you approach what they’re doing. Again, you know, you could have a situation in which you’re asking fairly straightforward questions, which because you might expect a smaller sample to give you what you need, but if you go into a much more kind of discursive kind of approach or constructivist kind of approach, which for example, Charmaz would recommend, you might get a lot more variation just in simply the way people are talking about things. Okay, you need to investigate all those possibilities, so a larger sample might come from there as well. So, I don’t think there is a simple answer for that honestly. I remember some years ago a colleague was — many many years ago, a colleague was doing his PhD part-time and he was doing it by qualitative analysis and wanted to know how many people to interview and he went to Ann Oakley’s thesis it was then on housework I think it was, and asked how many she did and it turned out — this is a famous sociologist, Ann Oakley, and she did 50 interviews. This must have been the mid-60s she was doing her work. That’s quite unusual now, I mean, when you look at PhD thesis now, I talk to other colleagues about how many of their students are doing and so on, it’s much much lower, it’s in the 20s and 30s that kind of. So, I think over the years it’s got easier to do a PhD, you see.
Audience: [inaudible]
Sorry?
Audience: I have seen some [inaudible] the other day [inaudible].
14? Doing grounded theory with 14? Yes, that is interesting, I think that’s too few. There are some approaches, some, for example phonological approaches, where smaller numbers are a [perfect] acceptable, it is much more intensive, much more concerned with, you know, the experience of the individual and that’s fine, smaller numbers. But I think for a Grounded Theory approach you’d expect more than that. But, maybe less than 50. And it depends, again, yeah it could be a team of you work here, maybe you want to do, you know, a hundred interviews, I’ve certain come across the examples of that, over a hundred interviews and usually with team of people doing and not a PhD thesis.
Copyright Disclaimer
Under Title 17 U.S.C. Section 107, allowance is made for “fair use” for purposes such as criticism, comment, news reporting, teaching, scholarship, and research. Fair use is permitted by copyright statute that might otherwise be infringing.
