The aim of this study was to evaluate the effects of a modified version of constraint-induced (CI) movement therapy on bimanual hand-use in children with hemiplegic cerebral palsy (CP; age range 18 mo to 4 y) and to make a comparison with conventional paediatric treatment. Twenty-one children (13 females, eight males) completed the CI therapy programme and 20 children (12 males, eight females) served as a control group. Children in the CI therapy group were expected to wear a restraint glove for 2 hours each day over a period of 2 months. The training was based on principles of motor learning used in play and in motivational settings. To evaluate the effect of treatment, the Assisting Hand Assessment (AHA) was used. Assessments took place on three occasions: at onset, after 2 months, and 6 months after the first assessment. A significant interaction was found between group and AHA measure (ANOVA, F(2,74) = 5.64, p = 0.005). The children who received CI therapy improved their ability to use their hemiplegic hand significantly more than the children in the control group after 2 months, i.e. after treatment. Effect size was high after treatment and remained medium at 6 months. As the treatment was tailored to each child's capacity and interests, little frustration was experienced by the children.