Social psychological processes underlying intergroup behaviour should be placed in their sociostructural contexts. Previous experimental studies have usually created groups that do not differ on sociostructural dimensions such as group numbers, power, and status. The impact of these factors on intergroup behaviour constituted the major aim of the present research. Predictions were formulated assuming that individuals are motivated to achieve a positive social identity (Tajfel & Variants of the "minimal group" paradigm see Tajfel & Turner, 1979) were used, first, to replicate previous minimal group results, and second, to assess the independent and combined effects of power, status, and group as in previous studies, allocation matrices developed by Tajfel and his colleagues were the main dependent measures. Unlike previous studies, these were supplemented with extensive possession questionnaire items that included items on social identifications and various intergroup perceptions. First, the classic minimal group discrimination effect, replicated under conditions of equal group numbers, power, and status, was eliminated when group members had little or no power (study 2). Categorization per se was net sufficient for intergroup discrimination. Without usable power, minimal group members did net discriminate. However, categorization per se was sufficient in eliciting prejudice. Second, results of this research clearly shewed that increases in group status and group power led to concomitant increases in matrix discrimination. Membership in minority groups polarised these patterns of behaviour (and perceptions) more than membership in majority groups. Status was the best predictor of social identification (and related perceptions), while power best predicted actual behaviour. In conclusion, evidence from these studies indicated that group numbers, power, and status, independently and in combination, have a strong impact on intergroup behaviour and perceptions. / Thesis / Doctor of Philosophy (PhD)