qualitative research
Types of qualitative research | Mike McDonald | March 21st, 2003
I have just read the Lazaraton article that accompanies Unit 2 of IIC and done Task 2.3. After reading Keith's comments, I think I understand the article in general, but I'm still not quite sure what qualitative research is. Figure 1 gives some general ideas on qualitative research traditions, but doesn't pin it down.
The text on pp. 460-461 seems to say that in the field of applied linguistics there are two major types of qualitative research: ethnographies (which include anthropological and educational ethnographies) and case studies (which include interaction analysis such as classroom interaction analysis). Is this correct? If so, where does written discourse analysis fit into the picture? Is that also a type of interaction analysis?
A further question: is all valid research either quantitative or qualitative, or are there types of research that are neither? Lazaraton seems to imply this on p. 461 when she says "it is also debatable whether the other orientations described in Cumming (1994) should be considered qualitative. Is the critical ethnography authored by Canagarajah (1993) to be viewed as an interpretive orientation, an ideological one, or both?" This suggests that there is a category of ideological research that is neither quantitative nor qualitative.
Mike McDonald
Types of qualitative research | Colin Graham | March 21st, 2003
Hi Mike,
By written discourse analysis I am assuming you mean the analysis of written discourse, rather than a discourse analysis that has been written! I would say, from a TDA angle, that you can't really talk broadly about discourse analysis as being either qualitative or quantitative until you identify the method of analysis you are undertaking.
In a given piece of text, for example, you could run it through a concordancer or text analyzer and count word tokens and frequency of occurrence, followed by a key word analysis to highlight unusual topic structures, for example. Stubbs does this in his analysis of the Little Red Riding Hood story. I would say that was definitely quantitative.
At the other extreme, you have the critical discourse analytical (CDA)models of people like Fairclough and Wodak who not only 'compress' their linguistic analysis of the text but also focus on selected features and introduce interdisciplinary triangulation of their views from the politic, economic or social sciences. They also adopt an overtly ideological stance, the stating of which is a (pre-) condition of the analysis being accepted as a valid CDA. Widdowson has criticized this ideology as being a weakness of CDA, whereas CD analysts see it as a strength because of its overtness... It seems to me that Widdowson likes to criticize everything that isn't part of his way of thinking!!
So, I think it may be easier to think in terms of the model of analysis as being largely quantitative or qualitative and the ideology as being an added dimension. Action research, for example, is often seen as a qualitative form of research but that doesn't mean that it doesn't or can't include some quantitative aspects. The important thing is probably to be clear about how you are going to validate your findings, and that can be more difficult with qualitative analyses, which can be viewed as (more) subjective and thus less reliable.
Colin
\(^_^)/
Banzai!
Types of qualitative research | Mike McDonald | March 21st, 2003
Colin,
From the viewpoint of Lazaraton's article, though, isn't that an instance of the use of quantitative methods within a fundamentally qualitative approach? I am still hazy on what separates quantitative research from qualitative research using quantitative methods, but I had the impression that if you were dealing with a single piece of text, even using statistical methods, you would be doing qualitative research. Certainly it would be difficult to generalize from a single piece of text, wouldn't it?
So, I think it may be easier to think in terms of the model of analysis as being largely quantitative or qualitative and the ideology as being an added dimension. Action research, for example, is often seen as a qualitative form of research but that doesn't mean that it doesn't or can't include some quantitative aspects. The important thing is probably to be clear about how you are going to validate your findings, and that can be more difficult with qualitative analyses, which can be viewed as (more) subjective and thus less reliable.
This is a good point (about validation), and one that I don't think the Lazaraton article really dealt with. On the whole I found the article rather frustrating, as it didn't define the territory very thoroughly; didn't draw up any clear categories of qualitative, quantitative, or other types of applied linguistic research; and didn't come to any convincing conclusions. At the same time, it did offer plenty of food for thought, which I am sure is why Keith included it.
Cheers,
Mike McDonald
Qualitative Research | Keith Richards | March 21st, 2003
Hi
In response to his question, I've sent Mike a very short section from Level 1 of my forthcoming book on qualitative inquiry in TESOL, but I didn't want to send it to the list because it would be an attachment and not relevant to many participants. However, if you are interested in a simple statement that includes a comment on the qualitative/quantitative issue, just send me an email and I'll send you the attachment. If you do, I hope you won't mind if I just hit the reply button rather than adding a message. Thanks.
All the best
Keith
Types of qualitative research | Mike McDonald | April 2nd, 2003
I have just read the Watson-Gegeo article that accompanies unit 3 of IIC, and found the following explanation:
"Qualitative research is an umbrella term for many kinds of research approaches and techniques, including ethnography, case studies, analytic induction, content analysis, semiotics, life histories, and certain types of computer and statistical approaches." (Kirk and Miller, 1986)
This seems to cover quite a lot of ground, but it is rather mystifying. Ethnography, case studies, and life histories I can more or less understand. I have a vague idea of what semiotics is. "Certain types" leaves me baffled.
And can anyone give me a definition of analytic induction or content analysis?
Mike McDonald
Types of qualitative research | Keith Richards | April 2nd, 2003
Hi
Mike's question about the Kirk and Miller quote is understandable since this covers a research tradition (ethnography), an approach that combines traditions (case studies), a procedure (analytic induction), a method (content analysis - perhaps my understanding is at fault here, but I understand this to be just what it says it is and pretty woolly at that), something which is sometimes regarded as a method, sometimes as an approach and sometimes as a tradition (life history; c.f. narrative, life story, biography), a field (semiotics) and a ragbag ('certain types of computer and statistical approaches'). At best it shows the very wide range of what is possible, at worst it's a hodge podge (and what is semiotics doing there?).
Analytic induction starts with a definition of a phenomenon to be explained and the formulation of a hypothesis that is then checked against the data. If any negative (deviant) cases are found, either the phenomenon is redefined in order to exclude the case or the hypothesis is reformulated and checked once more against the data. The aim is to repeat these steps until the researcher identifies a hypothesis that accounts for all cases. It's a pretty rigid procedure that I think has relatively limited appeal in its pure form but features commonly as part of broader strategies of analysis.
Hope that helps
Keith
with 
and 
