برآوردن نیاز و انگیزش در تربیت بدنی مدار تغییر در فعالیت بدنی در زمان اوقات فراغت در اوایل نوجوانی را پیش بینی می کند

Abstract Objectives We examined (1) how psychological need fulfillment and motivation in physical education and leisure-time physical activity change during early adolescence, and (2) the degree to which need fulfillment and motivation predict trajectories of change in physical activity. Design Longitudinal survey.

Highlights • PA increased and identified and introjected regulation decreased on average. • Competence increased and autonomy and relatedness decreased on average. • Autonomy and relatedness positively predicted change in PA. • Intrinsic motivation and identified regulation positively predicted change in PA. • The longitudinal associations between PE experiences and PA behavior are reported.

Results Preliminary analyses Examination of the skewness, kurtosis, z-score statistics, and pairwise scatterplots indicated that the data were approximately normally distributed, linear, and there were no univariate outliers. However, a significant Mahalanobis distance value (Mahalanobis' distance > 22.56, p < .001) suggested one possible multivariate outlier. All subsequent analyses were conducted with and without this participant and their inclusion did not affect the interpretation of the findings so all participants were included in all analyses reported. The final sample size was N = 134 students, with n = 134, 110, 105, 109, 98, and 98 at each time point, respectively. Other than students who missed a time point, the only missing data were two missing reports of physical activity. This missing data is most likely due to students closing the online survey before submitting their responses to the physical activity items, as they were on the last page of the questionnaire. Although MLM is robust to missing data ( Raudenbush & Bryk, 2002), the final model was examined with and without cases that contained missing individual items, and with and without cases with two or fewer time points to examine whether the results in this study were affected by missing data. Interpretation of results remained the same in all analyses so all cases were retained. A series of ANOVAs were used to examine if students who completed all six time points were significantly different on study variables when compared to students who completed five or fewer time points. Students who completed all six time points reported significantly higher scores (p < .01) on intrinsic motivation, identified and introjected regulation at time zero, and introjected regulation at time two. Descriptive statistics, internal consistencies, and correlations for each wave of data collection are presented in Table 1. Boys had significantly higher levels of physical activity compared to girls at time three and five (p < .01). Internal consistency values were all > 0.70 except for introjected regulation at time 1 (α = .65). All correlations were in the expected directions except for the consistent positive correlation between introjected regulation and the autonomous motivations. Table 1. Descriptive statistics, internal consistencies, and correlations. 6th Grade 7th Grade Variable 1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 1. PA – 0.52* 0.17 0.40* 0.42* 0.39* 0.21* −0.14 – 0.42* 0.21* 0.25* 0.35* 0.37* 0.27* −0.04 2. Competence 0.53* – 0.31* 0.60* 0.56* 0.52* 0.38* −0.14 0.53* – 0.39* 0.38* 0.44* 0.37* 0.22* −0.25* 3. Autonomy 0.44* 0.30* – 0.60* 0.53* 0.47* 0.32* −0.32* 0.04 0.49* – 0.72* 0.67* 0.62* 0.36* −0.44* 4. Relatedness 0.51* 0.43* 0.58* – 0.72* 0.59* 0.51* −0.21* 0.18 0.37* 0.16* – 0.68* 0.68* 0.44* −0.24* 5. IM 0.48* 0.50* 0.54* 0.69* – 0.82* 0.53 −0.27* 0.16 0.35* 0.55* 0.66* – 0.90* 0.53* −0.28* 6. IDENT 0.38* 0.34* 0.46* 0.67* 0.83* – 0.64* −0.09 0.26* 0.33* 0.46* 0.60* 0.92* – 0.54* −0.22* 7. INTRO 0.25* 0.20* 0.23* 0.45* 0.52 0.59* – 0.20* 0.23* 0.28* 0.25* 0.51* 0.57* 0.62* – 0.24* 8. EXT −0.14 −0.14 −0.40* −0.12 −0.25* −0.08 0.28* – −0.04 −0.16 −0.49* −0.15 −0.21* −0.11 0.23* – Mfall 2.35 3.70 3.17 3.98 4.24 4.45 3.62 3.25 2.81 3.97 3.05 3.54 4.10 4.11 3.07 3.31 SDfall 1.00 1.48 1.30 1.42 1.66 1.49 1.48 1.73 0.93 1.53 1.43 1.69 1.90 1.87 1.64 2.20 Mspring 2.78 3.85 2.96 3.75 4.25 4.38 3.38 3.12 2.67 4.21 3.25 3.75 4.33 4.23 3.07 3.36 SDspring 0.93 1.48 1.38 1.60 1.78 1.70 1.66 1.87 0.95 1.45 1.48 1.68 1.86 1.78 1.74 2.07 αfall 0.86 0.77 0.76 0.95 0.90 0.85 0.65 0.81 0.90 0.82 0.83 0.96 0.93 0.92 0.73 0.93 αspring 0.85 0.77 0.81 0.96 0.90 0.89 0.79 0.85 0.90 0.83 0.85 0.96 0.94 0.90 0.78 0.91 Scale range 0–4 0–6 0–6 0–6 0–6 0–6 0–6 0–6 0–4 0–6 0–6 0–6 0–6 0–6 0–6 0–6 8th Grade Variable 1 2 3 4 5 6 7 8 1. PA – 0.43* 0.22* 0.25* 0.18 0.17 0.06 −0.20* 2. Competence 0.47* – 0.40* 0.37* 0.40* 0.36* 0.15 −0.27* 3. Autonomy 0.08 0.46* – 0.66* 0.58* 0.52* 0.25* −0.27* 4. Relatedness 0.18 0.55* 0.79* – 0.64* 0.58* 0.30* −0.20 5. IM 0.25* 0.53* 0.64* 0.69* – 0.90* 0.51* −0.13 6. IDENT 0.21* 0.44* 0.61* 0.63* 0.90* – 0.58* −0.03 7. INTRO 0.23* 0.37* 0.21* 0.33* 0.44* 0.53* – 0.33* 8. EXT −0.10 −0.33* −0.55* −0.40* −0.49* −0.38* 0.10 – Mfall 2.71 3.28 4.16 4.02 4.30 4.17 2.92 3.12 SDfall 0.87 1.53 1.51 1.59 1.68 1.65 1.70 1.99 Mspring 2.50 3.19 4.21 3.98 4.11 3.92 2.91 3.25 SDspring 0.91 1.56 1.34 1.56 1.66 1.75 1.79 1.89 αfall 0.91 0.90 0.87 0.97 0.92 0.90 0.80 0.91 αspring 0.90 0.88 0.79 0.96 0.91 0.91 0.82 0.85 Scale range 0–4 0–6 0–6 0–6 0–6 0–6 0–6 0–6 Note: Correlations for fall semesters are below the diagonal, spring semesters are above the diagonal. PA = leisure-time physical activity; IM = intrinsic motivation; IDENT = identified regulation; INTRO = introjected regulation; EXT = external regulation. *p < .05. Table options Multilevel modeling analyses Results of the unconditional model for physical activity showed significant residual and intercept variance parameters (p < .001). Fifty percent of the variance in physical activity was attributed to between-student differences, suggesting that there was substantial variance available to explain by adding between-student predictors in subsequent models. Between-student variation for each psychological need and motivation ranged from 44% to 65%. Results for the unconditional growth models are found in Table 2 and illustrated in Figure 1. Each variable demonstrated a significant average change over time, except intrinsic motivation and external regulation. There was a significant linear decrease in identified regulation and a significant linear increase in competence need fulfillment over time. Introjected regulation had a quadratic trajectory, with an initial decrease that plateaued during the spring semester of grade seven. The cubic trajectory of physical activity showed an initial increase that plateaued in grade seven and then resumed increasing in grade eight. The cubic trajectory of autonomy and relatedness need fulfillment initially decreased, transitioned to a positive trend in the fall of grade seven, and returned to negative trend in the fall of grade 8. Gender, age, and teacher were considered as potential covariates. However, these covariates did not predict the trajectories of change for each variable and did not affect the results of the study. Therefore, they were excluded from all subsequent models. Overall, the inclusion of significant linear and non-linear effects of time reduced the within-student variance of self-reported physical activity behavior by 9%, and motivation and psychological needs by 1%–7% as indicated by the pseudo-R2 statistic. Table 2. Linear, quadratic, and cubic trajectories of self-reported physical activity, psychological need fulfillment, and motivation over time. Fixed effects Random effects Variable Intercept Linear term Quadratic term Cubic term σ2 τ00 Physical activity 2.35** 0.95** −0.23** 0.02* 0.41** 0.45** Competence 3.71** 0.09** 0.75** 1.40** Autonomy 3.17** −0.93** 0.30** −0.03* 0.95** 1.04** Relatedness 3.98** −1.19** 0.34** −0.03* 1.10** 1.38** Intrinsic motivation 4.24** 1.39** 1.76** Identified regulation 4.45** −0.11** 1.41** 1.48** Introjected regulation 3.62** −0.44** 0.04* 1.45** 1.30** External regulation 3.26** 1.91** 1.83** Note: Models were re-specified when higher order terms were not significant. *p < .05. **p < .01. Table options Full-size image (31 K) Figure 1. Plots of significant unconditional growth curve models of physical activity, psychological need fulfillment, and motivation. (A) significant cubic trajectory of physical activity; (B) significant linear trajectory of competence; and cubic trajectory of autonomy and relatedness; (C) significant linear trajectory of identified regulation and significant quadratic trajectory of introjected regulation. Figure options Each psychological need was added as a within student predictor of physical activity behavior, and individual means of the psychological needs were examined as between-student predictors of the intercept and trajectory of physical activity behavior. The results for these models are reported in Table 3. Competence and relatedness were significant within-student predictors of physical activity. Students reported more physical activity behavior when they also reported more than their average perceptions of competence and relatedness. The mean levels of competence, autonomy, and relatedness were significant, positive, between-student predictors of the intercept for physical activity. Students with higher levels of psychological need fulfillment reported more physical activity behavior at the beginning of sixth grade. The linear, quadratic, and cubic effects of time were significant predictors of physical activity. In addition, the interactions between mean perceptions of autonomy and relatedness and the linear term of time were significant. These significant interactions were plotted in Figure 2. As indicated by the region of significance (p < .05), students who had relatively lower perceptions of autonomy, less than 0.85 on a six point scale (2.05 standard deviations below the mean), and relatedness, less than 3.41 on a six point scale (0.29 standard deviations below the mean), had greater average increases in physical activity over time. Specifically, autonomy positively predicted linear increases in physical activity over time for students whose autonomy was less than 0.85, while autonomy was not significantly associated with changes in physical activity over time for students whose autonomy was greater than or equal to 0.85. Similarly, relatedness positively predicted linear increases in physical activity over time for students whose relatedness was less than 3.41, while relatedness was not significantly associated with changes in physical activity over time for students whose relatedness was greater than or equal to 3.41. Compared to the unconditional growth model, competence, autonomy, and relatedness respectively explained 53%, 13%, and 23% of the between-student variance of self-reported physical activity, while the cross-level interactions of time with autonomy and relatedness predicted a small proportion of additional variance (0.51% and 0.42% respectively). Table 3. Psychological need fulfillment and motivation as predictors of self-reported physical activity behavior. Variable Competence Autonomy Relatedness Intrinsic motivation Identified regulation Introjected regulation External regulation γ SE γ SE γ SE γ SE γ SE γ SE γ SE Fixed effects Intercept 0.83** 0.16 1.41** 0.21 1.08** 0.21 1.17** 0.20 1.06** 0.22 1.75** 0.17 2.57** 0.16 Linear change 0.92** 0.25 1.08** 0.26 1.12** 0.26 1.08** 0.25 1.10** 0.25 0.96** 0.25 0.94** 0.25 Quadratic change −0.23** 0.08 −0.24** 0.08 −0.25** 0.08 −0.24** 0.08 −0.24** 0.08 −0.23** 0.08 −0.23** 0.08 Cubic change 0.02* 0.01 0.02* 0.01 0.02* 0.01 0.02* 0.01 0.02* 0.01 0.02* 0.01 0.02* 0.01 Time-varying predictor 0.07* 0.03 0.03 0.03 0.06* 0.03 0.07** 0.26 0.07** 0.02 0.03 0.02 −0.02 0.02 Time-invariant predictor 0.39** 0.04 0.30** 0.06 0.33** 0.06 0.28** 0.05 0.30** 0.05 0.19** 0.05 −0.07 0.04 Interaction with linear change −0.03** 0.01 −0.03** 0.01 −0.02* 0.01 −0.03** 0.01 Random effects Within-student residual 0.41** 0.03 0.41** 0.03 0.40** 0.02 0.40** 0.02 0.40** 0.02 0.41** 0.03 0.41** 0.03 Between-student residual 0.21** 0.04 0.39** 0.06 0.34** 0.05 0.35** 0.05 0.35** 0.05 0.40** 0.06 0.44** 0.07 Note: Models were re-specified when higher order terms were not significant. All interactions between the quadratic and cubic trajectory of time and each predictor were not significant. **p < 0.01, *p < 0.05. Table options Full-size image (54 K) Figure 2. Plots of significant interactions between the linear term of time and the between-student terms of (A) autonomy; (B) relatedness; (C) intrinsic motivation; and (D) identified regulation predicting self-reported physical activity. Low trajectories represent one standard deviation below the mean and high trajectories represent one standard deviation above the mean. Figure options Each motivation variable was added as a within student predictor of physical activity behavior, and individual means of each motivation variable were examined as between-student predictors of the intercept and trajectory of physical activity behavior. The results for the models are reported in Table 3 and illustrated in Figure 2. Intrinsic motivation and identified regulation were significant within-student predictors of physical activity. Students reported more physical activity behavior when they also reported more than their average perceptions of intrinsic motivation and identified regulation. Greater mean levels of intrinsic motivation, identified regulation, and introjected regulation predicted greater physical activity at the beginning of sixth grade, while external regulation was not a significant predictor. Only mean perceptions of intrinsic motivation and identified regulation had significant interactions with the linear term of time. The region of significance (p < .05) for average intrinsic motivation and identified regulation indicated that students who had relatively lower perceptions of intrinsic motivation, less than 3.59 on a six point scale (0.39 standard deviations below the mean), and identified regulation, less than 3.70 on a six point scale (0.36 standard deviations below the mean), reported greater increases in physical activity over time. Specifically, intrinsic motivation positively predicted linear increases in physical activity over time for students whose intrinsic motivation was less than 3.59, while intrinsic motivation was not significantly associated with changes in physical activity over time for students whose intrinsic motivation was greater than or equal to 3.59. Similarly, identified regulation positively predicted linear increases in physical activity over time for students whose identified regulation was less than 3.70, while identified regulation was not significantly associated with changes in physical activity over time for students whose identified regulation was greater than or equal to 3.70. Compared to the unconditional growth model, intrinsic motivation and identified regulation each explained 22% of the between-student variance in self-reported physical activity behavior while the cross-level interactions between time and intrinsic motivation and identified regulation accounted for a small percentage of additional variance (0.35% and 0.40% respectively).