Women are underrepresented in science, technology, engineering, and mathematics (STEM) disciplines. Evidence suggests the microclimate of the classroom is an important factor influencing female course grades and interest, which encourages retention of women in STEM fields. Here, we test whether the gender composition of small (8±9 person) learning groups impacts course performance, sense of social belonging, and intragroup peer evaluations of intellectual contributions. Across two undergraduate active learning courses in introductory biology, we manipulated the classroom microclimate by varying the gender ratios of learning groups, ranging from 0% female to 100% female. We found that as the percent of women in groups increased, so did overall course performance for all students, regardless of gender. Additionally, women assigned higher peer- evaluations in groups with more women than groups with less women. Our work demonstrates an added benefit of the retention of women in STEM: Increased performance for all, and positive peer perceptions for women.