AbstractThis study is one of a series of experiments designed to examine how sociostructural factors such as group numbers, power and status affect intergroup behaviour. Using a variant of Tajfel's ‘minimal group’ paradigm the present study investigated the intergroup behaviour of college students categorized as numerical minority, majority or ‘equal’ group members. The effects of salient (S) versus non‐salient (S̄) group categorizations were also examined. These manipulations yielded a 3 × 2 design matrix consisting of majority/equal/minority × salient (S)/non‐salient (S̄) group conditions. Unlike most previous studies using this paradigm, subjects' responses on Tajfel's point distribution matrices were supplemented with subjects' report of their own and outgroup's point distribution strategies. As expected, minimal group results were replicated in the ‘equal’ group (S̄) condition such that mere categorization into ingroup/outgroup was sufficient to foster intergroup discrimination. However salient (S) equal group members were more fair than discriminatory in their responses. Minorities (S/S̄) were generally less fair than equal groups, showed high levels of absolute ingroup favouritism (S̄) while simultaneously attempting to establish positive distinctiveness from majorities. Though majorities were generally fair (S/S̄), they also appeared to be more concerned than minorities about maintaining positive differentials between themselves and minorities. Although, majority (S/S̄) and equal group (S̄) members accurately reported their actual distribution strategies, minorities (S/S̄) and equal (S) group members were not as accurate in their self reports. Overall the present results are consistent with hypotheses derived from Social Identity Theory. But the results also show that sociostructural variables such as group numbers can have an important impac