Peer-assessed outcomes were examined at the end of treatment (14 months after study entry) for 285 children (226 boys, 59 girls) with attention deficit hyperactivity disorder (ADHD) who were rated by their classmates (2,232 classmates total) using peer sociometric procedures. All children with ADHD were participants in the Multimodal Treatment Study of Children with ADHD (MTA). Treatment groups were compared using the orthogonal treatment contrasts that accounted for the largest amount of variance in prior MTA outcome analyses: Medication Management + Combined Treatment versus Behavior Therapy + Community Care; Medication Management versus Combined Treatment; Behavior Therapy versus Community Care. There was little evidence of superiority of any of the treatments for the peer-assessed outcomes studied, although the limited evidence that emerged favored treatments involving medication management. Post hoc analyses were used to examine whether any of the four treatment groups yielded normalized peer relationships relative to randomly selected-classmates. Results indicated that children from all groups remained significantly impaired in their peer relationships.Despite the widely accepted premise that children with attention deficit hyperactivity disorder (ADHD) have problematic peer relationships, the mechanisms underlying these peer difficulties remain poorly understood. Deficits in social skills knowledge, performance, and selfcontrol all have been postulated as important factors (for reviews, see Mrug, Hoza, & Gerdes, 2001;Wheeler & Carlson, 1994). Whereas earlier studies (e.g., Cunningham, Siegel, & Offord, 1985) suggested that inattentiveness may be a factor limiting opportunities for observational learning of social skills in children with ADHD, more recent work focuses on performance deficits (Barkley, 2000). Further, it is now widely accepted that a major contributing factor to the peer problems of children with ADHD is their largely unrestrained, overbearing interaction style, characterized by hyperactivity, aggression, bossiness, and other forms of controlling behavior that make them highly aversive to peers (see Mrug et al., 2001;Whalen & Henker, 1992, for reviews).Stimulant medication is useful for reducing rates of these negative and controlling behaviors (Cunningham et al., 1985) but typically yields few corresponding increases in positive behaviors (see Landau &Moore, 1991, andHenker, 1991, for reviews). Nonetheless, laboratory studies requiring normal peers to rate videotaped behavior of children with ADHD in both medicated and unmedicated states suggest that medication-related improvements in behavior are quite apparent to nondeviant peers even after short periods of observation (e.g., 15 min; Whalen, Henker, Castro, & Granger, 1987). Such findings suggest that medication may be an important part of any intervention for the peer problems of children with ADHD but that additional interventions are needed.Most psychological interventions for childhood peer problems are based on the social sk...