Predictive power of leisure constraint-negotiation models within the leisure-time physical activity context: A partial least squares structural equation modeling approach

Although prediction is a stated goal of leisure constraints and constraint negotiation research, to date, empirical studies have solely focused on statistical explanation when comparing competing models. Our research, therefore, examined the predictive power of five leisure constraint-negotiation models within the context of leisure-time physical activity (LTPA). To do so, we utilized online survey data from 299 Japanese and 286 Euro-Canadian adults and employed partial least squares equation modeling's (PLS-SEM) predictive function. Our results indicated that the independence model-in which constraints, negotiation, and motivation directly and separately predict participation-demonstrated better predictive power than the four alternatives. This finding was consistent across strenuous, moderate, and mild levels of LTPA and between the two national groups. We discuss the theoretical, practical, and methodological implications of our findings, and call for future research and theory development based on both explanatory and predictive evidence.