This study evaluated the validity of the Previous Day Physical Activity Recall (PDPAR) self-report instrument in quantifying after-school physical activity behavior in fifth-grade children. Thirty-eight fifth-grade students (mean age, 10.8 ± 0.1; 52.6% female; 26.3% African American) from two urban elementary schools completed the PDPAR after wearing a CSA WAM 7164 accelerometer for a day. The mean within-subject correlation between self-reported MET level and total counts for each 30-min block was 0.57 (95% C.I., 0.51–0.62). Self-reported mean MET level during the after-school period and the number of 30-min blocks with activity rated at ≥ 6 METs were significantly correlated with the CSA outcome variables. Validity coefficients for these variables ranged from 0.35 to 0.43 (p < .05). Correlations between the number of 30-min blocks with activity rated at ≥ 3 METs and the CSA variables were positive but failed to reach statistical significance (r = 0.19–0.23). The PDPAR provides moderately valid estimates of relative participation in vigorous activity and mean MET level in fifth-grade children. Caution should be exercised when using the PDPAR to quantify moderate physical activity in preadolescent children.
S.G. Trost, B. McGraw, and R.R. Pate are with the Department of Exercise Science at the University of South Carolina, Columbia SC 29208. D.S. Ward is with the School of Public Health at the University of North Carolina at Chapel Hill, Chapel Hill, NC 27599.