A randomized, double-blinded, placebo-controlled cross-over clinical trial was used to determine the role of sex steroids on the development of aggressive behaviors in 35 boys and 14 girls. Depo-testosterone (to boys) or conjugated estrogens (to girls) was administered in 3-month blocks alternating with placebo at three dose levels approximating early, middle and late pubertal amounts. The Olweus Multifaceted Aggression Inventory was administered after each placebo and treatment period to ascertain the effect of sex steroids on self-reported aggressive behaviors. We employed a strict intent-to-treat analytical model. The data demonstrated significant hormone effects on physical aggressive behaviors and aggressive impulses, but not in verbal aggressive behaviors nor aggressive inhibitions in both boys and girls. These results are the first to causally relate the administration of physiological doses of sex steroids to changes in aggressive behaviors in adolescents.