A prospective study of the reliability and convergent validity of three physical activity measures in a field research trial.

This investigation sought to determine the test-retest reliability and convergent validity of three commonly used physical activity measures. The Stanford Physical Activity Recall, the Caltrac Activity Monitor, and a daily physical activity log were administered to 45 subjects over a 3-week period. The results indicated high levels of test-retest reliability for the Stanford and the daily log; low test-retest reliability was obtained for the Caltrac. Convergent validity was also high between the Stanford and daily log but was low for both of these measures with the Caltrac. These results suggest that the Stanford Physical Activity Recall may be the self-report measure of choice for large studies due to its high reliability and convergent validity with a more time-consuming daily log, as well as its cost- and time-efficiency. Based on the results, it is recommended that the Caltrac activity sensors not be used for field studies of physical activity until the logistical problems of their use in clinical trials are solved.