The activPAL (TM) activity monitor has potential for use in youth with Cerebral Palsy (CP) as it has demonstrated acceptable validity for the assessment of sedentary and physical activity in other populations. This study determined the validity of the activPAL (TM) activity monitor for the measurement of sitting, standing, walking time, transitions and step count for both legs in young people with hemiplegic and asymmetric diplegic CP. Seventeen participants with CP Gross Motor Function Classification System level I completed two video recorded test protocols that involved wearing an activPAL (TM) activity monitor on alternate legs. Agreement between observed video recorded data and activPAL (TM) activity monitor data was assessed using the Bland and Altman (BA) method and intraclass correlation coefficients (ICC 3,1). There was perfect agreement for transitions and high agreement for sitting (BA mean differences (MD): -1.8 and -1.8 s; ICCs: 0.49 and 0.95) standing (MD: 0.8 and 0.1 s; ICCs: 0.59 and 0.98) walking (MD: 1 and 1.1 s; ICCs: 0.99 and 0.94) timings and low agreement for step count (MD: 4.1 and 2.8 steps; ICCs: 0.96 and 0.95) for both legs. This study found clinically acceptable agreement with direct observation for all activPAL (TM) activity monitor functions, except for step count measurement with respect to the range of measurement values obtained for both legs in this study population.