The importance of teamwork skills as part of employability has been widely acknowledged and accompanied by active research on successful cooperative learning. However, relatively few studies have focused on the effects of gender on students' group work, and only a limited number of empirical studies exist that examine students' group work process and performance through the results of self- and peer-assessment. This study examines the effects of gender on group work process and performance using the self- and peer-assessment results of 1001 students in British higher education formed into 192 groups. The analysis aggregates all measures on the group level in order to examine the overall group performance. Further, a simple regression model is used to capture the effects of group gender compositions. Results suggest that students in gender balanced groups display enhanced collaboration in group work processes. The enhanced collaboration could be associated with less social loafing behaviours and more equitable contributions to the group work. However, the results imply that this cooperative learning environment does not lead to higher student performance. Students' comments allow us to explore possible reasons for this finding. The results also indicate underperformance by all-male groups and reduced collaborative behaviours by solo males in male gender exception groups (i.e., groups consisting of one male student and other members being female). The results thus have implications for the composition of groups. The pedagogical implications of these findings are discussed.