Validation of the children's leisure activities study survey questionnaire for 12-year old South African children

Abstract

This study evaluated the reliability and construct validity of a modified Children's Leisure Activities Study Survey (CLASS) questionnaire for 12-year old South African children. Test-retest reliability of the questionnaire was examined in 108 (n=45 boys; n=63 girls) school children aged 12 years at 3 weeks interval. Construct validity was assessed by comparing data on the reported minutes spent in physical activities (PAs) obtained from the CLASS questionnaire responses with ten physical fitness (PF) parameters, assessed using the Eurofit test battery. Cronbach alpha's test of internal consistency, intra-class correlation coefficients (ICC) and the paired t-test were used to determine reliability. Validity was evaluated with the Spearman correlation coefficients. Substantial internal consistency and significant Intra-class correlations estimates for all intensities of PA and sedentary time (ST) were found (Cronbach's alpha ranged from 0.71 to 0.84, p < 0.05; ICC ranged from 0.73 to 0.95, p < 0.05). Non-significant differences were found in the means of test and retest measurements. BMI had moderate negative correlations with both moderate and total PA (r = -0.34, r = -0.36, respectively, p < 0.05). Aerobic fitness was significantly correlated with all intensities of PA and ST (r = 0.63, 0.57, 0.43 and -0.34, respectively; p < 0.05). ST was significantly and negatively associated with sit-ups (r = -0.38, p < 0.01) and shuttle run (r = -0.29, p < 0.05). Similarly, the results of the shuttle run (10x5 meter run), sit-ups, bent arm hang, and standing broad jump tests, had significant associations with vigorous PA. The CLASS questionnaire is a valid and reliable measure of PA in 12-year old South African children.