PURPOSE: The primary aim of this study was to examine the accuracy of a hip (Evenson algorithm) and wrist-worn (Crouter algorithm) accelerometer in assessing time spent in different intensity categories in pre-adolescent girls during semi-structured dance classes using direct observation (D.O.) as the criterion measure. The secondary aim of this study was to examine the validity of a wrist-worn accelerometer for dichotomizing pre-adolescent girls as meeting or not meeting different preselected doses of moderate-to-vigorous PA compared to the hip-worn accelerometer. METHODS: Data were collected and analyzed on a total of 6 participants (age = 10.22 ± 2.38) for the primary aim. Additionally, data was collected and analyzed on a total of 20 participants (age = 8.6 ± 1.6) for the secondary aim. RESULTS: Compared to D.O., the wrist-worn accelerometer was inaccurate in measuring time spent in light PA, vigorous PA and MVPA. Additionally, the hip-worn accelerometer was inaccurate in measuring time spent in sedentary time, light PA, vigorous PA and total PA. Further, for the secondary aim, there was a significant difference between device location and meeting PA dosage for three days and five days. CONCLUSION: Traditional accelerometer algorithms rely on the activity count cut-point method which provides mixed to poor results of activity intensity classification regardless of wear location. Future research should move away from the activity count cut-point method and aim to develop algorithms that use more of the rich data available from the accelerometers’ acceleration signal.