Fig 1.
Hypothesized relations among constructs of the extended trans-contextual model.
Effects of past behavior on all other model constructs have been omitted for clarity.
Table 1.
Summary of hypothesized direct and indirect effects in the extended trans-contextual model.
Fig 2.
Parameter estimates from the Bayesian path analysis of the extended trans-contextual model for leisure-time physical activity including informative priors.
PE = Physical education context; LT = Leisure time context. Model parameters omitted for clarity: past physical activity behavior→perceived autonomy support, β = 0.390, 95% CI [0.116, 0.664], p = .003; past physical activity behavior→autonomous motivation (PE), β = 0.368, 95% CI [-0.076, 0.799], p = .051; past physical activity behavior→autonomous motivation (LT), β = 1.382, 95% CI [1.109, 1.654], p < .001; past physical activity behavior→attitude, β = 0.194, 95% CI [-0.109, 0.495], p = .106; past physical activity behavior→subjective norms, β = 0.337, 95% CI [-0.193, 0.867], p = .103; past physical activity behavior→perceived behavioral control, β = 0.593, 95% CI [0.249, 0.938], p < .001; past physical activity behavior→intention, β = 0.293, 95% CI [0.014, 0.574], p = .019; past physical activity behavior→habit, β = 0.799, 95% CI [0.397, 1.206], p < .001; past physical activity behavior→self-control, β = 0.153, 95% CI [-0.041, 0.348], p = .061; past physical activity behavior→physical activity behavior, β = 0.672, 95% CI [0.518, 0.827], p = .001.
Table 2.
Parameter estimates (β) with 95% credibility intervals for hypothesized effects from the Bayesian path analyses of the extended trans-contextual model for leisure-time physical activity.