Comparing methods of social preference assessment in childhood

A central question in social cognitive development concerns the degree to which children prefer social ingroup members relative to social outgroup members. Forced-choice measures and continuous rating scales are often used to assess these preferences, but little work has examined the extent to which these two methods yield similar or divergent estimates. In Study 1, we used a within-subjects design to assess gender-, race-, and accent-based preferences in 5-6-year-old predominantly white children (N = 100) with both a forced-choice and a rating measure (on a 1-6 scale); replicating prior work, children expressed ingroup preference along all three dimensions regardless of how they were assessed. In Study 2, we replicated the discrepancy between forced-choice and rating in children's ingroup gender preferences in a more racially diverse sample (N = 55). In both studies, while responses on forced-choice and rating measures were correlated, estimates of ingroup preference were stronger in each domain when assessed with a forced-choice measure. We discuss the implications for researchers who wish to assess social group preferences.