Annual screenings of preschool children at kindergarten registration identified 158 children having high levels of aggressive, hyperactive, impulsive, and inattentive behavior. These “disruptive” children were randomly assigned to four treatment conditions lasting the kindergarten school year: no treatment, parent training only, full-day treatment classroom only, and the combination of parent training with the classroom treatment. Results showed that parent training produced no significant treatment effects, probably owing largely to poor attendance. The classroom treatment produced improvement in multiple domains: parent ratings of adaptive behavior, teacher ratings of attention, aggression, self-control, and social skills, as well as direct observations of externalizing behavior in the classroom. Neither treatment improved academic achievement skills or parent ratings of home behavior problems, nor were effects evident on any lab measures of attention, impulse control, or mother–child interactions. It is concluded that when parent training is offered at school registration to parents of disruptive children identified through a brief school registration screening, it may not be a useful approach to treating the home and community behavioral problems of such children. The kindergarten classroom intervention was far more effective in reducing the perceived behavioral problems and impaired social skills of these children. Even so, most treatment effects were specific to the school environment and did not affect achievement skills. These findings must be viewed as tentative until follow-up evaluations can be done to determine the long-term outcomes of these interventions.