Failure to generalize determinants of self-reported physical activity to a motion sensor

We hypothesized that the prediction of physical activity from psychological determinants would differ when free-living physical activity was assessed by electromechanical or self-report methods. We assessed outcome-expectancy values and perceived barriers for physical activity (OE_Values/P_Barriers) and physical self-efficacy in 44 young (24 yr ± 4.8) Caucasian men (N= 13) and women (N = 31). Two weeks later, subjects kept a daily diary and wore a Caltrac® motion sensor during all ambulation for seven consecutive days. Reliability (P₁) across days was high for MET-h of activity estimated by the diary (0.82) and for Caltrac® counts (0.87). Canonical correlation analysis yielded one significant linear combination (R_C = 0.55, P < 0.01; redundancy = 0.19 to 0.24) of the set of psychological variables [OE_Values/P_Barriers (β= 0.76) and physical self-efficacy (B = 0.33)] and the set of physical activity estimates [7-d diary (B = 0.89) and Caltrac® (β = 0.24)]. Multiple correlations indicated that the linear combination of OE_Values/P_Barriers and physical self-efficacy explained 26% of the variation in MET-h reported in the 7-d diary (P < 0.01) but was unrelated to Caltrac® counts (P> 0.05). The strongest bivariate model included the 7-d diary and OE_Values/P_Barriers (r = 0.51, P < 0.01). Caltrac® counts were weakly related to OE_Values/P_Barriers (r = 0.34, P < 0.05). A weak correlation between Caltrac® and the 7-d diary (r = 0.35, P < 0.01) was consistent with past studies. We conclude that the accuracy of predicting physical activity from psychological variables implicated as determinants of physical activity can depend on the method used to assess physical activity. The prediction of self-reported activity did not generalize to an objective motion sensor. Our findings support that the psychological determinants of physical activity will not be known until the dissociation of self-reported physical activity from concurrent objective monitoring is reconciled or understood.