The primary aim of this study was to assess the ability of the CSA accelerometer to measure physical activity in preschool children. A secondary aim was to examine inter-instrument differences and the effect of accelerometer placement on output. Eleven subjects (mean age = 4.0 years, SD = 0.4) wore the CSA-7164 for a 45-min preschool exercise class. They were observed throughout the class, and their engagement in activity was quantified using the Children’s Physical Activity Form (CPAF). The effect of accelerometer positioning (left vs. right hip) was assessed in 10 subjects over 2 days. CSA output during the class was highly correlated with the CPAF score (r = 0.87, p < .001), and rank order correlations between the 2 methods were also highly significant (r = 0.79, p < .01). Differences in CSA output between left and right hip reached statistical significance (paired t, p < .05), but these differences were small and probably of limited biological significance. The CSA appears to be an appropriate tool for assessment of physical activity in preschool children, but further studies on stability of activity as measured by CSA, as well as its validity, are urged.