In a household community sample of 1,285, 9-17 years-olds with mental disorders who had received outpatient specialty mental health services in the past year were compared with youths with mental disorders who had not received those services to determine if samples drawn from clinical settings are representative of youths with mental disorders in the general population. Those who had used services were more impaired, less competent, more likely to have comorbid disorders, more likely to belong to non-Hispanic White relative to other ethnic groups, and less likely to be prepubertal girls. Their parents were more educated, but less satisfied with family life, engaged in less monitoring of their children, and more likely to have used mental health services themselves. These findings suggest the hypothesis that samples of youths with mental disorders drawn from outpatient clinical settings are not representative of all youths with mental disorders. If confirmed, this would indicate the importance of population-based samples for the study of psychopathology in youths.