Methodspace - home of the Research Methods community

I'm wondering if it is possible to use both qualitative AND quantitative methods to generate theory?  ie. In the initial stage of the study, the researcher would use qualitative methods and based on the results, the researcher would then use quantitative methods to collect more data and THEN generate theory based on both the qualitative and quantitative data collected? 

I would especially be interested in techniques to analyzing quantitative data to generate them into theory.


Any advise, suggestions, or any discussions regarding the above would be welcome!

Views: 1441

Replies to This Discussion

Classic grounded theory is a general method for analyzing qualitative and/or quantitative data in such a way that explanatory theory emerges.
Just a small comment: It's an overstatement - even if understandable - to say that "theory emerges" from the data. It's better to use the statement expressed e.g. by Lyn Richard that theory is constructed by the reasearcher.
I think you'll find this relevant - looks at the epistemology surrounding the 'generation' of theory: http://www.qualitative-research.net/index.php/fqs/article/viewArtic...

"Since the 1960s it is one of the most crucial and widely accepted insights of epistemology and cognitive psychology that "there are and can be no sensations unimpregnated by expectations" (LAKATOS 1978, p.15) and that the construction of any theory, whether empirically grounded or not, cannot start ab ovo, but has to draw on already existing stocks of knowledge.
Hi Vincent,

Your question is too general, I think, to receive specific answers. Generally speaking different paradigms and approaches to research, namely, quant, qual, mixed, etc. have their own procedures to produce models and theories. For example, some researchers have used quant approach to produce theory-like or models, while others have used qual approaches and still others mixed ones. In quant approaches, factor analysis, regression, structural equation modelling and the like have been used to produce models and theories, and in qual approaches ethnographies, and grounded theory have been good candidates.

Had you made it clear what research question you're addressing and what type of data and how much of it you'd gathered, you'd have got better answers.
Best;
Mehdi
Hi Mehdi Riazi,

I'm very interested in learning more about generating theory/model from quantitative methods. I have only come across studies where the researcher would test theory using the quantitative data analysis as the ones you mentioned but none would build theory using quantitative methods alone. And hence my motive to use both quant and quali methods to build the theory because of the lack of theory building quantitative methods literature out there. Please, could you by any chance point me to any papers/literature regarding building theory using quantitative methods?
Hi Vincent,

A quick response to your query in the areas of Applied linguistics and education would be Canale and Swain (1981) on communicative competence and latest Gardner's studies on motivation and attitude toward language learning.
Hope this helps.
Best;
Mehdi
Hi Mehdi,

I have checked the sources that I have access to but the closest paper I could find is Canale and Swain (1980) Theoretical Bases of Communicative Approaches to Second Language Teaching and Testing and in the paper, they proposed a theoretical framework so I believe this is not the one that you recommended. I am also searching for Gardner's paper on language learning and hopefully will have access to this and will get back to you. Cheers
This isn't actually an answer to your question, but maybe tells something about the research process to explain social phenomena. Peter Abell wrote somewhere that explanation presupposes three phases:
- to study a case
- to compare more cases and
- finally to collect more (ie. "enough") cases to verify statistically/quantitatively the theoretical ideas generated in the first two phases.

Of course, this is only a "handbook answer" to your question about working with the empirical data. But in general, case studies and comparisons are for constructing theories and statistical/quantitative methods are primarily for testing the theoretical ideas after case studies and comparing cases. - In fact, some statistical textbooks even warn about "snooping the data" for doing something else than just testing theoretically based hypotheses set up at the beginning of the research.
Hi everyone.

Like Vincent, I am also looking for papers/publications that made use of quantitative AND qualitative methods to generate theory. Unfortunately, I was not able to find any yet. A "slight" difference is that in the first phase of the study the researcher would use a quantitative method (i.e., a survey questionnaire), and based on the results he researcher would then use a qualitative method (i.e., in-depth semi-structured interviews) to collect more data and THEN generate theory based on both the qualitative and quantitative data collected.

Again, any advice, suggestions, or any discussions regarding the above would be more than welcome.
Hi Nuno Goncalo,

Having read your question and linking it to Vincent's, I am just thinking if generating Theory is something we can decide for in advance and at the initial stages of a research project? In other words, can I as a researcher say my purpose in doing this research is to produce a theory? I'd like other members respond to this as well, but I can't see that with just a research project (something like the one you described: starting with a questionnaire and then performing some interviews) we can produce a theory. Aren't theories the outcome of several studies put together? To use Pentti's words "case studies and comparisons are for constructing theories". So, I'm really interested to see how other friends think about the issue of constructing theories and the reason for that. I mean the purpose of a research project is to contribute to the body of the knowledge of the field which might or might not contribute to theory development.
Best;
Mehdi
With classic grounded theory it is not so much a case of producing theories as using a systematic method to discover theories that explain patterns of behavior.
Hi Vincent,

I generated a theory from Qual guided by Grounded Theory and generalize them in a bigger population (Quan phase) with structural equation modelling. Therefore I used a sequantial exploratory mixed methods design, Qual -> Quan. Qual to inductively generate a theory (on a small sample) and quan to test and generalize the theory (on a bigger sample) to be accepted by scientific community. Hope I did not further confuse you :-)

RSS

Follow us:

© 2014   Created by SAGE Publications.

Badges  |  Report an Issue  |  Terms of Service