Perceived motor competence and cardiorespiratory fitness predict physical activity in children: Insights from an allometric longitudinal model

This study examined to what extent the factors of the Stodden et al (2008) conceptual model of motor development (i.e., Body Mass Index [BMI], actual and perceived motor competence [AMC and PMC, respectively], cardiorespiratory fitness [CF], and physical activity [PA]) might explain AMC and PA respectively, longitudinally, over a three-year period, using an allometric approach. A sample of 549 children (49.9% girls), aged 8 to 11 years at baseline undertook assessment of weight, height, PMC, CF, AMC and PA, once per year at the same time point. Using multilevel modelling for the longitudinal data, BMI, PMC, CF and PA were employed to predict AMC and subsequently, BMI, PMC, CF and AMC were employed to predict PA, thereby examining bidirectionality in the pathways suggested by the Stodden et al (2008) conceptual model of motor development. Results indicated that, for AMC, there was a strong relationship with body size (inverse BMI), but this effect of body size disappeared when CF was introduced into the model, with CF being the prime predictor of AMC longitudinally. For PA, PMC and AMC were significant. However, when CF was introduced into the model, the effect of AMC disappeared, leaving PMC and CF as significant predictors. The present study suggests, using an allometric and longitudinal approach that the trajectory of AMC over time is curvilinear in nature and CF is the prime predictor of AMC longitudinally, over PMC, weight status and PA.