This article examines students from ethnic minorities and majorities with regard to the relationships between their social and academic integration and their quality of learning. A total of 523 students at four universities completed a questionnaire: analyses of variance were used to examine mean differences, and structural equation modelling (lisrel) was used to analyse differences in relationships between integration and learning. After one year of study, minority students had performed less well and had obtained a lower number of credits, even though their approaches to learning had been no different. Neither had there been any differences in integration. Furthermore, analyses revealed a different set of relationships between integration and learning for minority and majority students. For majority students, the impact of formal academic integration was positively related to grades, credits and approaches to learning. However, in the group of minority students, the role of formal academic integration was inconclusive: negatively related to grades, but positively related to the deep approaches to learning. In the conclusions, these results are interpreted and some suggestions for future research are presented. (Contains 4 tables and 2 figures.)