This study uses Children of the National Longitudinal Survey of Youth to test for evidence of a causal relationship between maternal alcohol, marijuana, and cocaine use, and children�s behavior problems. Ordinary least squares (OLS) results provide strong evidence that substance use is associated with behavior problems. However, OLS estimation fails to account for unobserved factors that may be correlated with substance use and child behavior. To account for this problem, mother�child and family fixed-effects models are tested. The results suggest that maternal illicit drug use is positively associated with children�s behavior problems, while alcohol use has a less consistent impact.