Dyadic Analysis of a Self-report Physical Activity Measure for Adult-Youth Dyads

Abstract

Adult physical activity levels influence youth physical activity levels, but the nature of this relationship is still unknown. Most research focusing on this topic has been conducted with accelerometers, which are ideal since self-report physical activity measures can be biased. However, self-report measures for physical activity are useful to include in studies to gather information at low-cost. The purpose of this study was to further develop a self-report adult-youth dyad measure of physical activity. This study was conducted using secondary data analysis of the physical activity measures used in an intervention on behavioral nutrition (iCook 4-H). Participants were a sample of 214 adults (M = 39.0, SD = 8.0 years) and youth (M = 9.4, SD = 0.7 years) pairs. Accelerometer data was collected for a subset of youth (n = 122). There was dependency between the adult-youth physical activity data, and a dyadic confirmatory factor analysis model showed good fit to the data and achieved metric invariance, a measure to determine if the same construct was being measured in both youth and adults. Invariance was confirmed across matched versus unmatched sex pairs and some evidence of invariance with youth accelerometer data. Based on study findings, when using self-report measures of physical activity, researchers should measure both members of the adult-youth dyad to get more accurate measurements. Further validation of these findings is needed using an objective physical activity measure, like accelerometers, with all participants and more diverse samples.

Keywords: Adult-youth dyad; Dyadic confirmatory factor analysis; Physical activity measurement; Self-report.