Research has shown that living away from one's biological father is associated with a greater risk of adverse child and adolescent outcomes; yet, the role of the father-child relationship in understanding this association has not been directly investigated. This study uses data on biological fathers' relationships with their children from the 1979 National Longitudinal Survey of Youth (N=2,733) to assess whether father involvement mediates the relationship between family structure (i.e., father absence) and four measures of adolescent behavior. Differences in father involvement are shown to account for a sizeable fraction of the variance in outcomes by family structure. Father involvement does not affect boys and girls differently but is more beneficial when the father lives with the adolescent. (Contains 1 note.)