PURPOSE: To investigate the relative utility of Canadian Occupational Performance Measure (COPM) (adapted for children) and Goal Attainment Scaling (GAS) as outcome measures for paediatric rehabilitation. METHODS: A two-group pre-post design investigated the impact of a 3-month programme. Forty-one children with spastic hemiplegic cerebral palsy (mean 3.9 years; GMPM level 1; 21 boys, 10 girls) were randomized to occupational therapy only and occupational therapy plus one Botulinum Toxin A injection. The latter was considered a 'proven' intervention for the purpose of this instrumentation study. Intervention impact was investigated using GAS and COPM. Instrument sensitivity, convergent validity, goal/problem profiles and administration were evaluated. RESULTS: Both instruments were sensitive to within group change and detected significant between group change. Likert scale coding for GAS scores was more sensitive than the traditional weighted GAS or COPM. Different constructs were measured by each instrument. COPM was more time efficient in training, development and administration. CONCLUSION: Study aim, logistic and resource factors should guide the choice of COPM and/or GAS instruments as both are sensitive to change with a proven intervention and both evaluate different constructs.