This article explores how different raters have different strengths and weaknesses. Parents know their children well, but have a narrower framework on which to base their judgments about behavior problems. Moreover, in some cases such as hyperactivity, they are prone to exaggerate differences between their children. Teachers have greater expertise with children, but there is the possibility of correlated errors when both members of a twin pair are rated by the same teacher, and increased error variance when co-twins are rated by different teachers. Additionally, subtle contextual differences may operate across raters. Despite these strengths and weaknesses, parent ratings, same-teacher ratings and for the first time, different-teacher ratings yielded similar conclusions regarding the etiology of behavior problems. This consistency of results across rater types, in the face of such different strengths and weaknesses, provides strong evidence for substantial heritability and negligible shared environmental influence on behavior problems in the early school years. The study described in this article examines sex differences in genetic and environmental effects on behavior problems in a sample of over 3,500 twin pairs. Given the inconsistency of previous results based on an extensive review of literature, the authors had no specific hypotheses concerning sex differences. However, the large sample, the focus on a single age, and the fact that parent and same- and different-teacher ratings of behavior problems were obtained on the same measure, allow for a more rigorous test of sex differences in the etiology of behavior problems.