Background. Several pilot studies have evoked interest in robot-assisted therapy (RAT) in children with cerebral palsy (CP). Objective. To assess the effectiveness of RAT in children with CP through a single-blind randomized controlled trial. Patients and Methods. Sixteen children with CP were randomized into 2 groups. Eight children performed 5 conventional therapy sessions per week over 8 weeks (control group). Eight children completed 3 conventional therapy sessions and 2 robot-assisted sessions per week over 8 weeks (robotic group). For both groups, each therapy session lasted 45 minutes. Throughout each RAT session, the patient attempted to reach several targets consecutively with the REAPlan. The REAPlan is a distal effector robot that allows for displacements of the upper limb in the horizontal plane. A blinded assessment was performed before and after the intervention with respect to the International Classification of Functioning framework: body structure and function (upper limb kinematics, Box and Block test, Quality of Upper Extremity Skills Test, strength, and spasticity), activities (Abilhand-Kids, Pediatric Evaluation of Disability Inventory), and participation (Life Habits). Results. During each RAT session, patients performed 744 movements on average with the REAPlan. Among the variables assessed, the smoothness of movement (P < .01) and manual dexterity assessed by the Box and Block test (P = .04) improved significantly more in the robotic group than in the control group. Conclusions. This single-blind randomized controlled trial provides the first evidence that RAT is effective in children with CP. Future studies should investigate the long-term effects of this therapy.