Focus Group Composition in Heterogeneous Populations
Loading...
Date
2013-05
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
The Ohio State University
Abstract
Often, groups of people behave differently than individuals. Influential members in a group can cause other members in the group to make irrational decisions. Focus groups are used within marketing to see how a group will react to a new idea or product. Focus group researchers will often create separate focus groups consisting of people who are alike so that influential members are less likely to occur. However, influence occurring in the population due to interactions consisting of heterogeneous people may not show up in these focus groups due to their homogeneity. In this paper I present a model of a population with two types of reactions to a new policy, non-dissenter and dissenter. Dissenters are against the new policy, and feel very strongly about their opinion. Non-dissenters are not against the policy, but their opinion is weak. If the policy is enacted, members of the population interact on a one-on-one basis, sharing their opinion of the policy until the population has reached a consensus on the policy. Modeling shows that even a small proportion of dissenters can cause large shifts in the average opinion in the population overtime. I then present a model of a focus group where members within the focus group share their opinion to all other members. I find that focus groups that consist of members from the targeted population can be used to estimate the population consensus, and that these focus groups can be composed of randomly chosen members of the population. I then go on to show that homogenous focus groups, or focus groups consisting of all non-dissenters or all dissenters, create a bias if used in the estimation.
Description
Keywords
Bayesian Inference, Focus Groups, New Coke, Influence Activities, Group Behavior