A Four-Year Longitudinal Study of Relationships Among Actual and Perceived Motor Competence and Physical Activity in Spanish Primary School Children

The purpose of this study was: 1) by mediational analysis, to analyze the cross-sectional relationships between actual motor competence (AMC), perceived motor competence (PMC) and physical activity (PA) at four time points (one year apart), 2) by mediatorial analyses, to analyze the longitudinal relationship between these factors (each variable controlled by the previous year), 3) by a cross-lagged model, to analyze the longitudinal association between AMC, PMC and PA across the four years. A convenience sample of 111 Spanish students aged 8-10 years participated in this study. Analyses were carried out through structural equation models to test simple mediation, changes over time and cross-lagged relationships. Objective 1: Analysis supports PMC's mediating role between AMC and PA and also a bidirectional relationship between PMC and PA over the four years. Objective 2: All variables showed they were influenced by previous years, except for PA-T3 and PA-T4 when the model went from AMC to PA. The only bidirectional relationship that remained constant in all years was between PMC and PA. Objective 3: The cross-lagged panel analysis revealed significant autoregressive relationships for AMC, PMC and PA, as well as relationships between PMC and PA for the four years. Longitudinally, an interesting mediational path was found: AMC-T1 → PMC-T2 → PMC-T3 → (PMC-T4 and PA-T3) → PA-T4. The longitudinal association between AMC, PMC and PA, including the prospective meditational role of PMC, was confirmed across the four years with a particular consistency in the bidirectional relationship between PMC and PA.