The purpose of the present study was to examine the predictive validit
y of several theoretical factors drawn from control, social learning,
and differential association theories for explaining adolescent sexual
behavior. The sample consisted of 1478 adolescents who attended ten p
rivate schools located in different areas of this country. When examin
ed together in multivariate regression equations, the theoretical fact
ors accounted for 47.2% and 42.7% of the respective variances in frequ
ency of sexual intercourse among males and females. The single factor
of differential association with peers accounted for most of the respe
ctive variances (males, 33.1%; females, 26.8%). The discussion address
ed the theoretical implications of the present findings for future res
earch.