Agent-based models can test whether identity needs drive residential migration patterns

Evidence

Because of the difficulty of studying migratory patterns of populations in real time and the inability to experimentally manipulate features of real-world social contexts, agent-based modeling provides a very useful tool for studying the influence of the needs for inclusion and distinctiveness on the formation of groups in geographical space. Tests of ODT have not examined whether group members migrate to particular locations as means of satisfying their needs for inclusion and distinctiveness. Through agent-based modeling, we can create populations with simulated neighborhoods and vary the initial numbers of group members in various neighborhoods. We can then create agents with inclusionand distinctiveness-seeking rules and examine whether agents tend to settle in neighborhoods where there are a particular percentage of other in-group members in that neighborhood. More specifically, what one might expect are neighborhoods that are dominated by groups that each share a moderate proportion of the population (e.g., 35%) as opposed to highly heterogeneous neighborhoods with many groups with a small share of the population. — Roderick M. Kramer, Geoffrey J. Leonardelli, Robert W. Livingston, Social Cognition, Social Identity, and…, 846

Categorization

• SEED: Angles: ‘The 35-percent rule’ as a testable claim about why diverse neighborhoods self-segregate into moderate pluralities rather than full heterogeneity.