Predictors of Weekend Physical Activity in Adolescents

Wednesday, March 14, 2012
Poster Areas 1 and 2 (Foyer Outside Exhibit Hall C) (Convention Center)
Erin E. Centeio and Darla M. Castelli, The University of Texas–Austin, Austin, TX

Background/Purpose: Guided by the social cognitive theory, Active + Healthy = Forever Fit was a three-year physical education initiative resulting in the development of a student centered elective curriculum. As previously reported, enactment of this curriculum produced increased physical activity outside the school day (Centeio & Castelli, 2011). The purpose of this present study was to identify the predictors associated with weekend physical activity.

Method: At one Midwest school, 1224 students (Mage=16.71) enrolled in grades 9-12, volunteered to provide their Fitnessgram test scores, a belief survey (Motl et al., 2002), and physical activity recalls (Weston, Petosa, & Pate, 1997) for research. These measures were taken at the beginning, middle, and end of the school year. Further, demographic variables such as SES and ethnicity were secured.

Analysis/Results: Using a multiple regression analysis, student weekend physical activity was regressed on demographic, fitness, previous physical activity, and belief variables. This analysis revealed a significant model (p<.001) predicting students' weekend physical activity F(3,1224) = 36.905, R2 = .083. Previous physical activity (β=.261, p<.001), beliefs (β=.101, p<.001), and amount of screentime (β=-.074, p=.007) were contributors.

Conclusions: Knowing the predictors of physical activity among adolescents outside of the school day should guide physical education content. Specifically, lessons could be provided about screentime and building efficacy toward regular participation. Understanding student engagement in physical activity needs to be contextually and attitudinally directed. Further study about how teacher instruction modifies physical activity beliefs and experiences is warranted.