Sources of validity evidence needed with self-report measures of physical activity

Over the years, self-report measures of physical activity (PA) have been employed in applications for which their use was not supported by the validity evidence.

To address this concern this paper 1) provided an overview of the sources of validity evidence that can be assessed with self-report measures of PA, 2) discussed the validity evidence needed to support the use of self-report in certain applications, and 3) conducted a case review of the 7-day PA Recall (7-d PAR).

This paper discussed 5 sources of validity evidence, those based on: test content; response processes; behavioral stability; relations with other variables; and sensitivity to change. The evidence needed to use self-report measures of PA in epidemiological, surveillance, and intervention studies was presented. These concepts were applied to a case review of the 7-d PAR. The review highlighted the utility of the 7-d PAR to produce valid rankings. Initial support, albeit weaker, for using the 7-d PAR to detect relative change in PA behavior was found.

Overall, self-report measures can validly rank PA behavior but they cannot adequately quantify PA. There is a need to improve the accuracy of self-report measures of PA to provide unbiased estimates of PA.