Despite claims that diversity is negatively related to social cohesion, there is now extensive evidence that intergroup contact is associated with a reduction in prejudice and an improvement in various measures of intergroup relations. However, this conclusion is based primarily on the impact of positive contact, and negative contact has been overlooked until recently, since when it has been claimed to have stronger effects than positive contact. I first review results of some of my research group’s recent research, assessing the impact of positive contact in various forms: direct contact; extended contact; contact via social networks; and contextual effects of contact. I illustrate the policy impact of contact through a recent study of an ethnic merger between two previously segregated schools. I then consider the impact of negative contact, arguing that it is not very fruitful to ask which has the ‘greater’ impact, but, rather, that one should investigate how positive and negative contact interact. I illustrate current knowledge using a range of methods – experiments, social network analysis, cross-sectional and longitudinal survey analysis, and observational studies. I conclude that a broad approach to conceiving and measuring ‘contact’ is necessary, and that contact conceived in this manner urges caution regarding premature conclusions about the impact of diversity on our societies.