The aim of this study was to determine the reliability and magnitude of error of three lower-limb clinical measures for children with cerebral palsy (CP): the Modified Ashworth Scale of Spasticity (MAS), passive range of movement (PROM) and the modified Tardieu scale (MTS). Six physiotherapists measured 11 females and seven males (mean age 6 years 4 months, SD 2 years 4 months; age range 2 years 4 months to 10 years) on two occasions using a repeated measures design, collecting all data over 6 days. The severity of CP spanned all five levels of the Gross Motor Function Classification System and all children demonstrated varying degrees of spasticity. Exclusion criteria included botulinum toxin injections, inhibitory plasters, and orthopaedic surgery within the 6 months before study entry. For PROM and the MTS interrater reliability was acceptable with an intraclass correlation coefficient of 0.7, but results for MAS were lower. Standard error of measurement for repeated measures of PROM and MTS was about five degrees, but 95% confidence interval ranges were considerably higher. Test–retest results varied widely, particularly for the MAS. These measurement tools should be used with caution when evaluating changes in young children with CP.