Are We Too Limited on Group Size? What About 2 or 3 Person “Mini-Groups”?

  

When I first became interested in focus groups, we were still receiving most of our advice about how to do focus groups from marketing researchers, where that advice often included groups as large as 9-12. Since then, the recommended size for social science focus groups has fallen to a range of 4 to 8. In other words, we have explored the upper sizes of the old range and found them too large for most of our purposes.

With regard to larger size groups, the most likely source of the difference between marketing focus groups and social science focus groups is the higher level of involvement that our participants have with the topic. Put simply, when the participants are engaged with a topic, they all have something to say, and larger groups can become disorganized. Thus, when a group of 10 people each have contributions on a question, this makes it hard to maintain an orderly group dynamic. This contrasts with the topics in marketing research where the participants may have low levels of involvement with a topic, which means that larger groups can generate a wider range of subjects for conversation.

 

If our reasons for limiting the upper size limit of our groups are reasonably clear, the reasons for limiting the lower end of the size range are less clear. The most likely reason why we don’t conduct groups of size 2 or 3 is that the group dynamics change, which can also mean that we need to change both our moderating style and the kinds of questions that we ask.

 

This is an example of my larger argument that we a “stuck in a rut.” Where is the kind of experimentation that would examine whether 2 and 3 person mini-groups are a useful format for doing social science research?

 

I will cover three points that make the case for mini-groups.

 

First, they maintain the fundamental characteristic of focus groups by providing interactive discussions. There is an undeniable group dynamic in 2 and 3 person groups, and this interaction is the distinctive feature of group interviewing. In particular, the participants can engage in the kind of “sharing and comparing” that is one of the major advantages of focus groups. This means that sets of 2 and 3 participants can build on each others’ contributions in a number of different ways, using the dynamics of a conversational setting to extend the data beyond what would routinely be available in a one-on-one interview.

 

Second, 2 and 3 person interviews borrow some of the advantages of one-on-one interviews by allowing us to hear more from each individual participant. In a 90 minute focus group with 6 people, we hear an average of 15 minutes per person. Moving down from 6 to 3 people obviously means that the time per person doubles, and a dyadic interviews give 45 minutes to each person. This means a considerable increase in the amount of depth and detail that we get from each participant.

 

Finally, using small groups would help solve one of the biggest current problems in focus groups, the difficulty in recruiting enough participants. Group interviews inherently require bringing people together at the same time, and this has become increasingly difficult over the 25 year history of focus groups in the social sciences. Working with smaller groups reduces the problems in finding enough participants to put together a group.

 

This advocacy for 2 and 3 person interviews does, however, create a series of questions, including: What kinds of topics are most appropriate for small group interviews? What level of involvement do participants need in order to sustain this kind of interview? How should we modify the kinds of questions that we use now? How should we modify the kinds of moderating styles that we use now?

 

Overall, 2 and 3 person interviews create both a number of opportunities and a number of unanswered questions, but my claim is that the value of the opportunities clearly justifies the work needed to make this method a reality.

This entry was posted in Uncategorized. Bookmark the permalink.