Students' relationships with peers is fundamental to their mental health. A 3-month study presented here investigated the nature and frequency of students' self-reported recess problems and the degree to which these were concomitant with two often-used measures of children's social competence: peer acceptance and mutual friendships. Seven specific student complaints were investigated, including three problems with peer conflict, three problems with social isolation, and one problem with play enjoyment. Results reveal that certain recess problems were occurring with surprising frequency. Children experienced the seven problems in 3% to 8% of their recesses, with not being allowed to join a group in play being the most frequent recess problem. Significant correlations were reported between recess self-reports and the size of children's friendship networks. Some, but not all, children with frequent recess problems were of low peer acceptance and/or had few identified friends. None of the seven recess problems differed markedly by grade, and few differences were noted by gender, suggesting that there are more similarities than differences in recess problems across age and gender. Correlations exist between the measures of social competence used in developmental research and children's complaints of recess problems. Contains 60 references. (RJM)