but both genes and shared environment were important for teacher and parent ratings (Baker et al., 2008). Aggressive behavior in another sample of twins aged 7–12 years was found to be largely influenced by genetic (or familial) factors (76%–84%), as reported by parents. Data were collected from one or both parents; however, only mother-reported ratings were included in the analyses, as they accounted for the majority of the ratings collected (85.3%–90.1%). In contrast, when teacher ratings were used, aggressive behavior was found to be slightly less influenced by genetic (or familial) factors (42%–61%; Haberstick et al., 2006a). Significant shared environmental effects were not found for either parent or teacher ratings, and were therefore dropped from the models, suggesting that any shared environmental influences are likely to be included in the genetic component. Apart from rater bias, which may result in inflated twin similarity across the board when using single raters for multiple children, the varying patterns of genetic and environmental influence across methods of assessment could be the result of different raters reporting different aspects of the child's aggressive behavior. This could arise in part because individuals behave differently in different situations (e.g., school vs. home) or because some types
