AN EXAMINATION OF ADOLESCENTS' PERCEPTIONS OF THE SCHOOL PHYSICAL ENVIRONMENT RELATED TO PHYSICAL ACTIVITY

The aim of this study was to examine the relationship between an objective and perceived measure of the physical environment related to physical activity within a school setting. A self-report questionnaire to assess the perceived physical environment was completed by 850 male and female students in four high schools in rural Alberta, Canada. Two researchers completed objective observations in the same four schools. The students were grouped as perceiving the school environment (a) accurately, (b)affording greater opportunities, or (c) providing fewer opportunities. Discriminant analysis revealed two significant functions which differentiated between the three groups based on variables associated with physical activity in youth (i.e., age, self-efficacy, peer network, family network, and teacher relationship) and on activity level. The first discriminant function's canonical correlations showed that teacher relationship (. 91) and activity level (. 42) were most predictive of individuals who perceived more opportunities than assessed in the objective measure. The second function's canonical correlations found that peer network (. 63), self-efficacy (. 41), and activity level (-.37) were the best at predicting individuals who perceived inaccurately. When examined by gender the discriminant analysis revealed only one significant function. For both females and males, this function's strongest predictors included teacher relationship and energy expenditure, while for males only, self-efficacy and family network were also highly correlated. The discriminant analysis's model equation correctly classified 68.0% of the cases, 16.6% more cases than would be predicted by chance alone. Overall, the small mean difference score (0.37) between the three groups directs researchers to use either objective or perceived measures of the physical environment, as both should provide similar results.