A field study was conducted in the Italian context to examine the longitudinal effects of contact on improved intergroup relations, and to test whether the effects were different for majority and minority group members. Furthermore, we examined the processes underlying contact effects. Participants were 68 Italian (majority) and 31 immigrant (minority) secondary school students, who completed a questionnaire at two time points. The results of regression analyses showed that, consistent with the contact hypothesis (Allport, 1954), quantity and quality of contact longitudinally improved outgroup evaluation and increased the attribution of positive stereotypes to the outgroup; the reverse paths were non-significant. Notably, whereas quantity of contact improved intergroup attitudes and stereotypes for both majority and minority participants, quality of contact had reliable effects only for the majority group. Intergroup anxiety and empathy mediated the longitudinal effects of quantity of contact for both Italians and immigrants; the cross-lagged effects of contact quality on criterion variables for the Italian group were mediated by intergroup empathy. The theoretical and practical implications of findings are discussed. Copyright © 2010 John Wiley & Sons, Ltd.