A measurement model of youth physical activity using pedometer and self, parent, and peer reports

Because no single measure is able to accurately assess all types of physical activity (PA), some researchers advocate use of multiple methods and multiple informants to measure PA. However, little research has tested the validity of proxy reports of youth PA.

The current study determined whether peer, parent, and self-reports reflects a latent measure of youth PA. As a test of construct validity, pedometer data were included in the model and covaried with the second-order target youth PA factor to determine the relationship between the youth PA factor and an objective measure of PA.

Participants included 291 target youth (ages 10, 12, or 14 years), a peer, and a parent of each target child (N = 873). Each participant reported about target children's vigorous PA during the past 7 days, days of PA in a typical week, and PA compared to others the same age and sex. Pedometers recorded the average number of steps taken per day by target youth over 7 days.

Analyses indicated an acceptable fit of the model to the data, as all variables loaded significantly on their respective factors, and all factors had significant loadings on the higher-order target PA factor. A moderate correlation was observed between the higher-order youth PA factor and the pedometer measure.

Combining multiple reports of youth PA has the potential to yield a more comprehensive measure of youth PA but may not be practical for all studies.