Four Types of Group

What it is

The grouping comes from a paper in the Journal of Personality and Social Psychology by Lickel, David Hamilton, Grazyna Wieczorkowska, Amy Lewis, Steven Sherman and Neville Uhles. In the first study, participants in the United States rated forty groups on eight properties, among them size, duration and how alike the members were. They also rated each group's entitativity, the degree to which it is perceived as one coherent unit, and sorted the groups by similarity. The second study repeated the procedure in Poland.

Clustering and multidimensional scaling of the results produced the same four types in both countries. The authors labelled them intimacy groups, task groups, social categories and loose associations. A family sits among the intimacy groups and a work committee among the task groups; these are our illustrations of the labels rather than findings of the paper. Social categories are large and their members resemble one another, and loose associations are the groups that are easy to join and easy to leave.

A later review chapter by Donelson Forsyth and Jeni Burnette, citing this work, sums up that people more readily see a set of people as a group when the members have strong bonds, interact often and share clear boundaries.

In effect

In the third study, participants rated the properties of groups they belonged to themselves. The results replicated the earlier ratings and showed that people valued their membership most in the groups they perceived as high in entitativity.

The grouping sits close to the continuum that Colin Fisher calls groupiness in The Collective Edge, which runs from a social category through a crowd and a co-acting group to a real team. Fisher approaches the dimension from group performance and the 2000 paper from perception, so the two cover similar ground from different directions.

A later pair of experiments adds a condition. Rutchick, Hamilton and Sack found that whether entitativity judgements rest on similarity or on interaction depends on how a group is construed, which suggests that the position of a group on the entitativity scale is not fixed by the group alone.

What it does not say

The grouping describes how participants rated and sorted a set of forty groups. It does not claim that every group belongs to exactly one type, or that the four types exhaust the kinds of group. It is a statement about perception, about how groups are seen, and the paper reports no test of how groups perform. The examples above are ours. This page draws on the paper's abstract, so it does not report figures from inside the paper.

---