استراتژی های خود ناتوان سازی در کلاس های تربیت بدنی: تاثیر نظریه ضمنی ماهیت انسانی و نظارت: جهت گیری هدف

Abstract Aim: To investigate the role of implicit theories of ability and achievement goals on self-handicapping strategies in physical education classes. Hypotheses: It was expected that incremental theories of ability would be negatively associated with self-handicapping strategies, whereas fixed theories of ability would enhance pupils’ self-reported use of such strategies. It was also hypothesised that low perceived competence would reinforce self-handicapping among pupils holding fixed theories of ability and an ego goal orientation. Method: A cross-sectional study of 9th graders in Norway (N=343; 166 boys and 177 girls) was conducted in which pupils responded to a questionnaire measuring sub-dimensions of fixed and incremental theories of ability, achievement goal orientations, perceived competence and self-handicapping in physical education. Results: Regression-based path analyses revealed that a fixed theory of ability had a direct positive effect on self-handicapping. The effects of an incremental implicit theory of ability on self-handicapping were negative and mediated by a task orientation. High perceived competence was found to buffer the aversive affect of holding a stable theory of ability on self-handicapping. Conclusion: The findings illustrate the importance of studying implicit motivational beliefs in physical education classes in order to provide an understanding of self-handicapping strategies among pupils.

3. Results Three sets of analyses were conducted. The first examined the bivariate relationships among the theories of ability, achievement goal orientations, perceived competence, and self-handicapping. In addition, gender differences with respect to these variables were assessed. In the second set of analyses, the hypothesised model shown in Fig. 1 was tested using path analysis. In the third set, moderated hierarchical multiple regression procedures were used to investigate the hypothesis that perceived competence moderates the effects of fixed theories of ability and an ego goal orientation on self-handicapping. 3.1. Descriptive analysis Means, standard deviations and correlations between variables are shown in Table 1. The mean values indicate that these pupils are very high on task orientation, hold a relatively strong incremental/learning conception of ability and perceive their ability in PE quite positively. The mean score of 2.17 for self-handicapping suggest that, in general, these pupils do not strongly identify themselves with being users of self-handicapping strategies. Table 1. Descriptive statistics (alphas, means, standard deviations) and zero-order correlation for theories of ability, achievement goals, perceived competence, and self-handicapping behaviour (N=343) Variable alpha M SD 1. 2. 3. 4. 5. 6. 7. 8. 1.Self-handicapping 0.76 2.17 0.99 – 0.15a 0.30c 0.16b −0.23c 0.03 −0.24c −0.51c 2. General 0.80 2.09 0.79 – 0.28c 0.44c −0.13a 0.21c −0.15b −0.12a 3. Stable 0.74 2.46 0.90 – 0.50c −0.27c 0.18c −0.23b −0.32c 4. Gift 0.84 2.78 1.03 – −0.13a 0.32c −0.16b −0.04 5.Learn/ incremental 0.74 4.11 0.57 – 0.06 0.34c 0.31c 6. Ego 0.91 3.41 1.05 – 0.02 0.25c 7. Task 0.70 4.64 0.45 – 0.22c 8. Perceived competence 0.78 3.85 0.87 – a p<0.05. b p<0.01. c p<0.001. Table options Generally, the expected pattern of results was found among theories of ability, achievement goal orientations and self-handicapping. That is, a learning/incremental theory of ability and a task orientation were both significantly negatively correlated with self-handicapping. The three dimensions of a fixed theory of ability (conceiving of ability as a natural gift, as stable and general) were all significantly and positively correlated with self-reported self-handicapping, whereas an ego goal orientation was not. Perceived ability was also significantly negatively associated with self-handicapping. Gender differences were revealed on the following constructs: boys were more likely to be ego oriented (M=3.80) than girls (M=3.04) (t=7.17, p<0.001); boys (M=2.96) were more likely to conceive of a theory of ability as a natural gift than girls (M=2.62) (t=3.14, p<0.001); boys (M=2.27) were more likely than girls (M=1.92) to report a theory of ability as general (t=4.05, p<0.001). Moreover, girls (M=2.27) reported more self-handicapping behaviour than boys (M=2.07) (t=−1.86, p<0.05). Given these differences by gender, it was controlled for in all subsequent analyses by entering gender into all regressions in the first step. 3.2. Path analyses Fig. 1 shows the hypothesised model used to assess the relations among theories of ability, achievement goals and self-handicapping behaviour. Path analyses were conducted to examine the direct and indirect relations among the variables. The path model was constructed in two steps. First, each of the achievement goal orientations (task and ego) was separately regressed on the theories of ability (Stable, Gift, General and Incremental/learning). Next, self-handicapping behaviour was regressed on the theories of ability and the two goal orientations. Gender was controlled for in all analyses. Significant paths of the fully estimated model are shown in Fig. 2. Standardised beta coefficients are reported and thus the path coefficients illustrate the strength of each explanatory variable controlling for all other explanatory variables in the model. Results are organised around the three sets of relations proposed in the model in Fig. 1: (A) influences of implicit theories of ability on achievement goal orientations and indirect (A–B) and direct (C) influences of theories of ability on self-handicapping. Path analysis of self-handicapping strategies in PE. Only significant paths of ... Fig. 2. Path analysis of self-handicapping strategies in PE. Only significant paths of the fully estimated model are shown. Standardised regression coefficients are presented. Figure options 3.2.1. Influences of implicit theories of ability on achievement goal orientations Task and ego goal orientations were separately regressed on the four dimensions of theories of ability. The incremental/learning dimension positively influenced a task goal orientation (beta=0.30, p<0.001), whereas the stable dimension negatively influenced a task goal orientation (beta=−0.13, p<0.05). As hypothesised, the gift and stable dimensions both positively predicted an ego goal orientation (beta = 0.22, p<0.001, and beta=0.13, p<0.05, respectively, see Fig. 2), indicating that pupils who believe in a conception of ability as a fixed and given capacity are more concerned with demonstrating how much they have of this ability relative to others. Unexpectedly, however, the Incremental/learning dimension of a theory of ability also positively, although modestly, influenced an ego goal orientation (beta=0.11, p<0.05). 3.2.2. Influences on self handicapping The full model explained 15% of the total variance in self-handicapping (see Fig. 2). When all variables were in the model, self-handicapping was explained by a task goal orientation but not by an ego orientation. A task orientation yielded a negative, inhibiting influence on self-handicapping (beta=−0.14, p<0.001). Thus, when pupils were highly task oriented, they were less likely to report self-handicapping behaviour in PE. Furthermore, a stable (beta=0.22, p<0.001) as well as a general implicit theory of ability (beta=0.14, p<0.001) both had facilitating effects on self-handicapping in PE. Thus, pupils who conceive of ability as stable and pupils who endorsed an implicit theory of ability as general across different activities in PE were more apt to report self-handicapping. Although not shown in Fig. 2, gender was a significant influence on self-handicapping (beta=0.16, p<0.001), indicating that girls reported self-handicapping more than did boys. 3.2.3. Achievement goals as mediators between theories of ability and self-handicapping. The next set of analyses was conducted to determine whether achievement goal orientations mediated the relationship between pupils’ theories of ability and self-handicapping. According to Baron and Kenny (1986), four conditions need to be established to determine mediation (see Fig. 1): (1) a significant relation exist between the independent variable and the mediating variable (Path A), (2) a significant relation exists between the mediating variable and the dependent variable (Path B), (3) a significant relation exist between the independent variable and the dependent variable (Path C), and (4) the relation between the independent variable and the dependent variable (Path C) is significantly reduced when the mediating variable is included in the regression equation. If the relation between the independent variable and the dependent variable is reduced to zero, then there is full mediation and if the relation between the independent variable and the dependent variable is attenuated but not reduced to zero, then partial mediation is said to exist (Baron & Kenny, 1986). The results of the path analysis results have already shown that for task goal orientation the two first conditions necessary for mediation have been established. Specifically, an incremental/learning theory of ability as well as a theory of ability as stable both explained a task goal orientation, which in turn explained self-handicapping. By contrast, for an ego goal orientation, only the first condition (Path A) has been established. Whereas a theory of ability as a gift, general and incremental/learning predicted an ego orientation, an ego orientation did not in turn predict self-handicapping. Thus, further mediation analyses were only performed with task goal orientation as a potential mediator. To establish the third and fourth conditions for mediation, a multiple regression strategy suggested by Judd and Kenny (1981) was used. Hierarchical multiple regression analyses were conducted with the theories of ability (stable and incremental/learning) in the first step and a task goal orientation entered in the second step. These analyses allowed for a determination of the relationships between these two theories of ability and self-handicapping behaviour independent of the task goal orientation (Condition 3). Further, it was possible to examine the degree to which the regression coefficients for the theories of ability were attenuated after the task goal orientation was entered into the regression (Condition 4). Table 2 presents the results of the hierarchical regression analyses. Given that there is no standard objective test to determine if a variable is fully or partly mediated, two criteria—the decrease in significance level and the decrease in the magnitude of the beta coefficient—are used together to examine mediation. Table 2. Summary of hierarchical regressions examining task goal orientation as mediator of theories of ability self-handicapping behaviour (N=343) a Predictor Betab Betac R2 Step 1: Theories of ability 0.10*** Learning/incremental −0.14** −0.10 Stable 0.25*** 0.23*** Step 2: Achievement goals 0.12*** Task goal orientation −0.13** a Note. Measures of gender were included in all regressions as statistical controls. **p<0.01. ***p<0.001. b Standardized regression coefficients without task goal orientation entered into the regression. c Standardized regression coefficients with task goal orientation entered into the regression. Table options In the first step of the regression analyses, when the two theories of ability variables were regressed on self-handicapping, both an incremental/learning theory of ability and a stable theory of ability were significant influences. When the task goal orientation was included in the analysis (step 2), the beta coefficient for the learning/incremental dimension dropped from −0.14, (p<0.01) to −0.10 and was no longer significant. This suggests that the preventive effect of an incremental/learning theory of ability on self-handicapping was mediated by a task goal orientation. The beta coefficient for a stable implicit theory of ability, however, dropped only marginally, from 0.25 (p<0.001) to 0.23 (p<0.001) and remained highly significant also in the presence of the task goal orientation (see Fig. 2). This suggests that holding an implicit theory of ability as stable had a significant positive direct effect on self-handicapping. 3.2.4. The moderating role of perceived competence on self-handicapping To test the potential moderating role of perceived competence in the relationship between fixed theories of ability, ego goal orientation and self-handicapping, cross-product terms between each of the predictors and perceived competence were computed. All variables were centred around their means before computing the cross-products in order to reduce multicollinearity (Aiken & West, 1991). The interaction terms were entered as a final step in path analysis of self-handicapping behaviour. The path analyses also comprising the interaction terms (i.e. the different goals and theories × perceived competence terms) revealed that only the Perceived Competence × Stable Theory of Ability interaction term was significant (beta = −0.37, p<0.05) and explained an additional 2% of the variance in self-handicapping (not shown in Fig. 2). The predicted values of self-handicapping for various values of a theory of ability as stable (−1SD, median and +1SD), and perceived competence (above and below the median) were graphed to aid in the interpretation of the interaction effect (see Baron & Kenny, 1986). As shown in Fig. 3, the Theory of Ability as Stable × Perceived Competence interaction indicates that as perceptions of competence decreased, holding a theory of ability as stable increased the likelihood that pupils would report self-handicapping. The interaction of perceived competence and a theory of ability as stable in the ... Fig. 3. The interaction of perceived competence and a theory of ability as stable in the reported use of self-handicapping strategies.