Abstract
There were three goals in this study: To determine the relation between boys’ temperament and rough‐and‐tumble (R&T) play; to determine the group composition of R&T, games, and other forms of reciprocal social interaction; and to determine the extent to which the vigor and flexibility components of R&T are related to social affiliation and social problem solving flexibility. Thirteen sociometrically defined popular boys, with a mean age of 114 months, were observed on their school playground during recess for 20 sessions. Additionally, they nominated peers they liked most and least, ordered peers in terms of dominance, and solved hypothetical social problems. Teachers completed temperament questionnaires. Results indicated that R&T groups were similar to other social groupings in terms of number and dominance symmetry. The flexibility dimension of R&T, but not vigor, was related to affiliation and social problem solving. Temperament was minimally associated with measures of behavior. Results are discussed in terms of the similarity of design features of R&T and social problem solving flexibility.