Furthermore, most previous studies have relied upon adult-based behaviors, such as criminal convictions or antisocial personality disorder, or on retrospective accounts of adolescent behavior; in contrast, we incorporate prospective measures of externalizing behavior from multiple reporters and demonstrate that the effects of GxE are evident with respect to behavior spanning childhood through early adulthood. Another strength of our study is the inclusion of prospective reports of behavior from multiple reporters. Self-reported aggression and delinquency are far higher than maternal and teacher reports, a discrepancy that is more pronounced in the reports of covert delinquent behavior than in reports of overt aggression. Similarly, Collishaw and colleagues (2009), using a different scale of problematic behavior, also reported that youth self-reports of emotional and conduct difficulties were higher than parent or teacher reports. Our findings seem to be driven primarily by maternal reports; perhaps mothers are more aware of their child’s delinquent behavior than are teachers, to the degree that such behavior might be more frequent outside of school. That self-reports contribute little to our findings is interesting: perhaps adults are more inclined
