Eight hyperactive children were treated with a behavioral intervention focusing on teacher and parent training over a period of 5 months. Three times, before therapy and after 3 weeks and 13 weeks of intervention, children received methylphenidate during 3-week probe periods. Each week in a probe they received either a placebo, .25 mg/kg, or .75 mg/kg methylphenidate. Classroom observation of on-task behavior suggested that effectiveness of the behavioral intervention was between that of the two dosages of medication before therapy. Both dosages resulted in higher levels of on-task behavior when administered after 13 weeks of behavioral intervention than when administered before therapy. Teacher rating data showed equivalent effects of therapy and the low dosage of methylphenidate alone but a stronger effect of the high dose alone; only the high dose resulted in improved behavior after 13 weeks of behavioral intervention. As a group, only when they received the high dose of methylphenidate after 13 weeks of behavioral intervention did children reach the level of appropriate behavior shown by nonhyperactive controls. However, this level was also reached by two children with the low dose and by one child without medication, and it was not reached by one child. The results suggest that the combination of psychostimulant medication and behavior therapy may be more effective in the short-term than either treatment alone for hyperactive children in school settings. In addition, parent ratings and clinic observation of parent-child interactions suggested that children had improved in the home setting, high-lighting the importance of behavioral parent training in the treatment of hyperactivity.