In this study, we tested two alternative models for investigating the ability of dimensions of personal grit (consistency of interest and perseverance of effort) and autonomous motivation to predict the leisure-time physical activity (LTPA) over a one-year period of 256 youth (aged 11-15 years at the beginning of the study). Specifically, we compared a model in which dimensions of grit were assumed to predict LTPA through autonomous motivation with a model in which autonomous motivation was assumed to predict LTPA through dimensions of grit. We tested the two models using variance-based structural equation modelling with Warp PLS v7.0 software. Both models showed a good fit with the data and equally explained 41% of the variance in LTPA over this one-year time period. However, the model based on the total effect of autonomous motivation (β = 0.448, p < 0.001, ES = 0.229) on LTPA through the dimensions of grit was stronger than the model based on the total effect of perseverance of effort (β = 0.356, p < 0.001, ES = 0.126) on LTPA through autonomous motivation, and the total effect of consistency of interest was not significant (β = -0.029, p > 0.05, ES = 0.003). Our findings underscore the relatively greater importance of the interplay between autonomous motivation and perseverance of effort, versus consistency of interest, in predicting LTPA among youth over a one-year period.
Keywords: autonomous motivation; consistency of interest; leisure time physical activity; longitudinal effects; perseverance of effort.