Written By: David Morgan, Professor, Department of Sociology, Portland State University
Long ago, in the mid 1980s, when I first began giving talks about focus groups, I had to spend a fair amount of time explaining what they were. By 1996 when I wrote an article for the Annual Review of Sociology, I reported that over 100 articles a year were being published that used focus groups. For 2010, an equivalent literature search shows over 2,000 articles for that year along.
Clearly methods are not static — in the past 25 years, focus groups have gone from being virtually unknown in the social sciences to being very widely used indeed. There is, however, a sense in which focus groups have been more static than I would like.
In particular, starting with the first articles that I wrote, I both called for and looked forward to future innovations in the field. In fact when Richard Kreuger and I and others first began using focus groups, almost all the sources we had were from marketing research. Consequently, we spent a fair amount of time adapting that work so that it made more sense in the context of social science research. One example would be reducing the size of the typical group, because our participants were typically much more engaged with the topic than people who were talking about products and advertising. Another was learning how to work with people who were already acquainted rather than strangers, because so many of our groups came from work places and other intact social settings.
Thus, our early use of focus groups required a fair amount of innovation, and we expected that trend to continue. Unfortunately, that hasn’t been the case. Instead, people have been satisfied to rely on a rather restricted range of research designs — essentially using what “authorities” such as Krueger and I said. The fact that we were giving good advice is evident from the number of people who have successfully followed that path. But the other side of the coin is that too few researchers have gone beyond that advice.
The idea of using a well-proven research design has obvious appeal. Consider the choices that students make as they work on their theses. Do you really want to try something new and risky for such an important project? And the same goes for those who are expecting to generate funded grants or peer-reviewed publications — do you really want to deviate from what your peers already value as “standard” research?
Yet my own experience has been that doing something new can be highly rewarded, and I think there is a viable strategy in my experience. In particular, I began by using techniques that were already very successful in marketing research, so there was every reason to believe that they could be adapted to what I wanted to do. So one suggestion is to start close to what you already know is likely to work. New ideas don’t have to be radical departures from what others are already doing.
Also, my first project was essentially a pilot study. I worked with one graduate student, with a “seed money” grant to pay 40 participants $25 apiece. So the lesson there is that it is entirely possible to start small. In particular, that one study produced both a substantive article and a methodological article, as well as a masters thesis for the student.
One reason that I’ve been thinking about these topics lately is that I’m working on a new method: dyadic interviews. The idea is to create something like a micro-focus group by having just two participants. Like a focus group, they respond to the questions by talking to each other, but the dynamic of a two-person conversation is quite different from a group discussion. Still, the previous lesson still holds: we are concentrating on an approach that is closely related to something that is already successful, and I’m starting small.
I am working with just one or two students at a time, and we are doing pieces of research that amount to pilot studies. This way we can build on what we learn, one step at a time. Last year we presented a symposium at the IIQM meetings in Vancouver, and we are now in the beginning stages of writing two papers.
I’m very excited about our results so far, but there is still going to be a process of “selling” dyadic interviews as a new way of doing things. We need to convince our colleagues to, once again, expand their notions of what we can do to collect interesting and valuable qualitative data. Based on my previous experience, however, I think that is only half the story.
Beyond promoting what we hope will be a successful method, I want to do more to encourage an innovative future for this new method. So, I plan to give out lots of advice about not just how to make dyadic interviews work, but also what kinds of things we could do next. What are possible steps that are both close enough the core of what we are doing and relatively small in terms of the resources they require? With any luck, we’ll have not just an innovative new method, but a method with a built in tradition of innovation.
Finally, I want return to my theme at the beginning by saying that again that research methods are dynamic rather than static. That doesn’t just happen, however. Instead, we all need to work together to make new ideas happen.