Small to moderate effect sizes suggest that working together with a friend and simply having a friend were related significantly and positively both to cognitive and academic performance outcome

Do Friendships Afford Academic Benefits? A Meta-analytic Study. Kathryn R. Wentzel, Sophie Jablansky, Nicole R. Scalise. Educational Psychology Review, https://link.springer.com/article/10.1007/s10648-018-9447-5

Abstract: Using meta-analytic techniques, we examined systematically the evidence linking friendship to academically related outcomes, asking: To what extent is friendship related to academic performance and to academically related cognitive skills? Based on 22 studies that yielded 81 effect sizes and 28 independent samples, we examined relations between friendship and academically related cognitive skills (e.g., scientific reasoning, linguistic skills, spatial memory) and performance (e.g., academic grades, test scores). The role of friendship was defined in one of two ways: working with mutual friends on academic tasks and the experience of having friendships (as indicated by having at least one reciprocated friend or a number of friends). Small to moderate effect sizes suggest that working together with a friend and simply having a friend were related significantly and positively both to cognitive and performance outcomes. Student (sex, age, country of origin) and methodological (measurement, design) characteristics were not significant moderators of relations between friendship and academically related outcomes.