The Composition of the Group
The focus group is characterized by homogeneity but with sufficient variation
among participants to allow for contrasting opinions. By homogeneity,
we mean participants have something in common that you are interested
in, such as the following:
* An occupation. (You want to explore professional development needs of
practicing dentists in the state.)
* Past use of a program or service. (You want to evaluate an educational
program, comparing the experiences of people who completed the program
in the past year with groups of people who dropped out of the program.)
* Age. (You want to talk to teenagers about smoking. You conduct some
groups with 7th graders, some with 9th graders, and some with llth graders.)
v Gender. (You want to talk to men who have had false positives on prostate
screening tests to find out how this affected their quality of life.)
* Family characteristics. (You want to talk to women who have had babies
in the past year to get help in designing a program for new moms.)
There are at least two reasons we are concerned about homogeneity. One
is for analysis purposes. The other is the participants' comfort- the
degree to which sharing will be influenced by differences in participants'
characteristics.
Think about your analysis when considering group composition. If you want
to compare and contrast different types of people's views, you must separate
these different types of people into different groups. For example, it
is much easier to analyze differences between users and nonusers when
they are in separate groups than if they are mixed together. So if you
are interested in similarities and differences between or among different
types of people, separate them and conduct a series of groups with each
type of group.
Related to this, it is a fallacy to assume that any one individual can
"represent" his or her neighborhood, race, gender, or culture.
Each person speaks for himself or herself. When asked, however, these
individuals may attempt to offer insights about the opinions of the entire
group, and the degree to which they have similar opinions may vary greatly.
If you want to capture the opinions of a certain group of people, then
you'll need to conduct a sufficient number of focus groups with that particular
category of people. A focus group of diverse people is not sensitive enough
to pick up trends of subcategories of people.
The other reason (besides analysis) that we consider homogeneity is the
degree to which these factors will influence sharing within the group
discussion. We want people to feel comfortable to share. Some mixes of
participants do not work well because of limited understanding of other
lifestyles and situations. For example, care must be exercised in mixing
individuals from different life stages and styles unless the topic clearly
cuts across these life stages and styles. Recently, we were conducting
focus groups with women who were pregnant. We wanted their help in designing
a health and education program for new moms. Our only screens were that
they be pregnant and from the participating county. We had teenagers who
didn't want to be pregnant in the same groups with forty-year-old women
who were ecstatic about being pregnant for the first time. The young women
tended to be quiet and deferred to the others in the group. In hindsight,
it would have been better if we had held separate groups for the teenage
moms. You strive for a balance between having enough variation within
the group to get contrast and yet not so much variation that participants
are inhibited and defer to those whom they perceive to be more experienced,
knowledgeable, or better educated.
At times it is unwise to mix genders in focus groups. Men may have a tendency
to speak more frequently and with more authority when in groups with women-sometimes
called the "peacock effect"-and this can irritate women in the
group.
A related topic is the involvement of both husband and wife in the same
focus group discussion. There is a tendency for one spouse to remain silent
and defer to the talkative spouse. Even if the silent spouse disagrees,
it appears that he or she is reluctant to comment even when such comments
are solicited from the moderator. As a result, we have found that focus
groups of four married couples turn out to be a discussion of four people
with four rather silent partners.
|