کمال گرایی خصلتی، حق تعیین سرنوشت و فرآیندهای خودارائه گری در رابطه با رفتار ورزشی

Abstract Objectives Motivational and self-presentational processes pervade all aspects of our lives including exercise behaviors. Furthermore, trait perfectionism has been shown to heighten self-presentational tendencies and energize achievement striving (Flett and Hewitt, 2002 and Hewitt et al., 2003). How maladaptive and adaptive perfectionism traits relate to these cognitive and behavioral processes specific to the exercise context remains to be determined. This study employed structural equation modeling to examine the associations between maladaptive and adaptive perfectionism, self-determination of exercise behavior, self-presentation in exercise, and exercise behavior.

Introduction Perfectionism has been conceptualized as a predisposing personality trait (Hewitt & Flett, 1991a) characterized by a tendency to set extremely high standards for performance and evaluate the self in a critical manner (Cox et al., 2002 and Flett and Hewitt, 2006). Traditionally, perfectionism has been studied from a clinical standpoint and viewed as a trait that is problematic and in need of adjustment (Bieling, Israeli, & Antony, 2004). This perspective has placed an emphasis on studying the negative correlates of perfectionism, including psychosocial adjustment problems and a vulnerability to distress (Flett and Hewitt, 2002, Frost et al., 1990 and Hewitt and Flett, 1991b). The examination of perfectionism in the exercise context has further emphasized the role of this construct in maladjustment. For example, previous research has indicated that perfectionism may be a critical antecedent of exercise dependence and other forms of compulsive and excessive exercise behavior (Coen and Ogles, 1993, Hagan and Hausenblas, 2003, Hall et al., 2007 and McLaren et al., 2001). However, recent conceptual advances have highlighted that perfectionism may comprise both maladaptive and adaptive dimensions (Bieling et al., 2004, Cox et al., 2002, Slade and Owens, 1998 and Stoeber and Otto, 2006). Despite the growing body of research acknowledging the maladaptive/adaptive dichotomy of perfectionism (Flett & Hewitt, 2006), little empirical effort has been devoted to the examination of the two broad dimensions in the exercise setting. The two higher-order constructs of maladaptive and adaptive perfectionism are derived out of the individual conceptualizations of Hewitt and Flett (1991a) and Frost and his colleagues (Frost et al., 1990). Maladaptive perfectionism which has also been labeled maladaptive evaluative concerns perfectionism (Frost, Heimberg, Holt, Mattia, & Neubauer, 1993) entails a socially prescribed tendency to evaluate oneself harshly, and to perceive that others require perfection from oneself (Hewitt & Flett, 1991a). Maladaptive perfectionism comprises the facets of socially prescribed perfectionism (SPP) from Hewitt and Flett’s (1991b) Multidimensional Perfectionism Scale (HF-MPS) as well as the subscales of concern over mistakes (COM), doubts about actions (DA), and parental perceptions (PP) from Frost et al. (1990) Multidimensional Perfectionism Scale (F-MPS). Maladaptive perfectionism has consistently shown a link with negative outcomes such as depression (Cox et al., 2002) psychological distress (Bieling et al., 2004 and Enns et al., 2001), and burnout (Zhang, Gan, & Cham, 2007). The concept of adaptive perfectionism, which has also been labeled positive strivings perfectionism (Frost et al., 1993), entails a self-oriented tendency to set high personal standards and achievement striving. Adaptive perfectionism incorporates the facets of self-oriented perfectionism (SOP) from the HF-MPS, high personal standards (HPS), and a need for organization and precision (O) from Frost et al. (1990) model. Research has shown positive relationships of adaptive perfectionism with positive affect related to performance (Bieling, Israeli, Smith, & Antony, 2003), life satisfaction (Bergman, Nyland, & Burns, 2007), and academic performance (Cox et al., 2002). However, some researchers hold strong doubts that perfectionism can be positive or adaptive (e.g., Flett & Hewitt, 2006). For instance, certain studies have failed to demonstrate significant associations between indicators of adaptive perfectionism and positive outcomes (Miquelon et al., 2005 and Mitchelson and Burns, 1998). Despite the body of research showing a consistent link between perfectionism and exercise behavior, few studies have sought to examine the psychological processes that underpin this relationship. A recent study examining adaptive and maladaptive dimensions of perfectionism in relation to University students’ exercise related cognitions and behaviors (Longbottom, Grove, & Dimmock, 2010), found that adaptive perfectionism was positively associated with a motivational profile that reflected self-efficacy, planning, and persistence in physical activity. In contrast, maladaptive perfectionism was associated with impeding and maladaptive motivation dimensions that reflected uncertainty about the conduct of exercise, fear or failure, and avoidance of physical activity (Longbottom et al., 2010). As such, the authors highlighted the need to examine how motivational orientations influence the relationship between perfectionism and exercise. The link between multidimensional perfectionism and motivation from a self-determination theory perspective Self-determination theory (SDT; Deci & Ryan, 2000) is one theoretical approach to human motivation that has received increasing attention in the exercise domain and has been successfully used in previous studies on perfectionism (Miquelon et al., 2005 and Van Yperen, 2006). SDT proposes that motivation varies in the extent to which it is autonomous (self-determined) or controlling (Deci and Ryan, 1985, Deci and Ryan, 2000 and Deci and Ryan, 2000). SDT proposes that three distinct forms of motivation exist, namely, intrinsic motivation, extrinsic motivation and amotivation,1 which represent different degrees of internalization of external values and goals – and thus differ in the degree to which they are self-determined or autonomous (Deci & Ryan, 1985). Extrinsic motivation is comprised of at least three main types of behavioral regulation: external, introjected, and identified2 (Deci & Ryan, 1985). External regulation reflects the least autonomy whereby the individual engages in a behavior to obtain external rewards or to avoid punishment (Deci & Ryan, 1985). In the exercise context, an example of external regulation would be exercising to obtain external recognition. When guided by introjected regulation, behaviors are only partially internalized. With introjection, action is undertaken in an attempt to avoid negative emotions or support conditional self-worth (e.g., exercising because of guilt or shame about not exercising, or to improve appearance on which self-worth is reliant). Identified regulation is a more autonomous form of extrinsic motivation whereby behavior is undertaken because certain outcomes of the activity are highly valued by the individual, although they may not enjoy the activity itself (e.g., exercising because one values the benefits associated with exercise such as improved health). Only when an individual engages in exercise for the inherent pleasure in the activity, is the behavior said to be fully self-determined or intrinsically motivated. In summary, these regulations lie on a continuum from lower to higher self-determination and reflect the extent of the internalization process (Deci & Ryan, 1985). The continuum conceptualization allows for the computation of the relative autonomy index (RAI; Ryan & Connell, 1989) that gauges the degree of self-determination in an individual’s behavioral regulation. Research investigating the applicability of the basic tenants of SDT within the exercise domain has shown that self-determined exercise motivation is a significant predictor of exercise adherence over time (Chatzisarantis, Hagger, Biddle, Smith, & Wang, 2003), psychological well-being (Wilson & Rodgers, 2007), and exercise behavior (Edmunds, Ntoumanis, & Duda, 2006b). In addition, Wilson and collaborators (Wilson, Rodgers, Blanchard, & Gessell, 2003) revealed that more self-determined forms of motivation were the strongest predictors of exercise behaviors, exercise attitudes, and physical fitness among participants in a 12-week exercise intervention. These findings provide support for the argument that self-determined motivation engenders the most positive behavioral and psychological consequences (Ryan & Deci, 2000) and is not associated with problematic exercise behaviors and cognitions (Robbins & Joseph, 1985). More controlling forms of motivation have been shown to correlate positively with strenuous exercise behavior (Edmunds et al., 2006a and Edmunds et al., 2006b) and weekly exercise behavior in individuals reporting symptoms of exercise dependence (Edmunds et al., 2006b). In addition to determining direct effects, the SDT approach has provided a comprehensive explanatory system that has chartered the antecedents and more importantly the processes that lead to exercise behavior (Hagger & Chatzisarantis, 2008). For instance, utilizing the SDT framework, Ingledew, Markland, and Sheppard (2004) expanded previous research linking the NEO-Five Factor model of personality (NEO-FFI; Costa & McCrae, 1992) and exercise participation. The study demonstrated that neuroticism was related to less self-determined exercise motivation, whereas a higher degree of self-determined motivation for exercise was related to openness and extraversion (Ingledew et al., 2004). The SDT framework may therefore prove useful in advancing our understanding of the role of perfectionism as an antecedent to exercise perceptions and behavior. According to the dual process model of Slade and Owens (1998) and initial theorizing by Hewitt and Flett (1991a), it has been suggested that adaptive perfectionism may be related to self-determined forms of motivation because it is characterized by a personal motivation to strive for perfection and a tendency to approach achievement striving with self-referenced criteria (Hewitt and Flett, 1991b and Mills and Blankstein, 2000). In contrast, maladaptive perfectionism may be associated with more controlling forms of motivation, such that achievements for maladaptive perfectionists are a means to win approval or to avoid blame and negative evaluations from others (Hewitt & Flett, 1991a). In accordance with SDT, Miquelon et al. (2005) showed that SPP enhanced non-self-determined academic motivation, which led to higher adjustment difficulties in the academic setting. In turn, SOP facilitated self-determined motivation which was linked to lower levels of psychological adjustment difficulties (Miquelon et al., 2005). In the sport domain, Gaudreau and Antl (2008) found that self-determined motivation in sport mediated the relationship between personal standards perfectionism and task-oriented coping, while non-self-determined motivation in sport was shown to mediate the relationship between evaluative concerns perfectionism and disengagement-oriented coping. Based on previous findings, it is feasible that broad dimensions of maladaptive and adaptive perfectionism may show a differential association to contextual self-determined motivation in the exercise domain. Integrating multidimensional perfectionism, motivation regulation, and self-presentation processes Whilst the incorporation of motivation constructs may be a step forward in explaining the interplay between perfectionism traits and exercise engagement, studies that assess expressive aspects of traits may also be relevant to the understanding of perfectionism in exercise. It has been argued that self-presentation or impression management represent a stylistic expression of the trait dimensions of perfectionism (Hewitt & Flett, 1991b). Self-presentation or impression management refers to processes by which people attempt to control the impressions others form of them (Leary & Kowalski, 1990). It consists of two discrete processes: impression motivation and impression construction (see Leary & Kowalski’s 1990 Two-Component Model). Impression motivation is a cognitive factor reflecting the desire or motivation to create a specific impression on others. In the exercise context, this definition corresponds to an individual’s motivation to present the impression that he or she is an exerciser (i.e., a fit, toned, and active person). Impression construction is a behavioral component that involves the tactics used to create the desired impression. For example, in the exercise setting, individuals may purposely wear particular clothing, exercise in the presence of others, or emphasize one’s athletic ability in order to convey the desired impression of an exerciser (Conroy et al., 2000). Hewitt et al. (2003) indicated that perfectionists’ compelling need for admiration and acceptance focuses attention on social situations and events that can be used to present the self in an ideal manner. Empirical research has confirmed that perfectionists, particularly those characterizing high maladaptive perfectionism, use self-presentational strategies as a way of managing social impressions (Hewitt et al., 1995 and Hewitt et al., 2003). Consequently, it is plausible that perfectionism may be associated with a strong desire to self-present as an exerciser, particularly in light of research indicating that portraying oneself as an exerciser can enhance one’s social appeal (Lindwall and Martin Ginis, 2006 and Martin et al., 2000). Although research has shown self-presentation to have a direct positive effect on exercise behavioral engagement (Gammage, Hall, & Martin Ginis, 2004), the role of self-presentation as a mediating factor has been neglected in the literature (Conroy et al., 2000). This strategy may offer new insight into the process by which personality traits, such as perfectionism, may be involved in energizing or compromising exercise engagement and may further the understanding of the social processes of exercise behavior. According to Leary and Kowalski (1990) impression management is influenced by factors such as an individual’s self-concept (e.g., the individual’s belief that she or he is fit and/or healthy), the belief that significant others value exercise lifestyles and exercisers, and demonstrating fitness and health conscious behaviors such as exercising regularly. Furthermore, items on the impression motivation scale are characterized by words and phrases such as value, important to me, and want. These items suggest that impression motivation involves the internalization of the positive exercise stereotype or the social image attached to being seen as fit, toned, or an exerciser and thus reflect a certain degree of self-determination. However, the extent to which such behaviors and cognitions are regulated in a self-determined fashion has not been empirically validated. Considering empirical research supporting a distinction between perfectionism traits and motivational orientations (Gaudreau and Antl, 2008 and Miquelon et al., 2005), the present study proposes that the extent to which impression management reflects self-determined motivation will depend upon the perfectionism trait in question. It can be argued that a sense of helplessness about exerting control over evaluative standards and a greater desire to please others and avoid punishment may heighten self-presentational concerns in individuals reporting high levels of maladaptive perfectionism (Hewitt & Flett, 1991b). In the exercise context, this may mean that less self-determined motivation may underpin impression motivation and taking steps to create the desired impression in those characterized by maladaptive perfectionism. Considering the precipitating nature of SOP (an indicator of adaptive perfectionism) in eating disorder behavior (Hewitt et al., 1995) and perfectionistic self-promotion (Hewitt et al., 2003), the role of adaptive perfectionism facets in fostering healthy exercise adaptations requires scrutiny. Alternatively, it can be argued that the predisposition of those characterizing this trait to adopt self-referenced criteria for achievement striving (Hewitt and Flett, 1991b and Mills and Blankstein, 2000) may promote self-determined reasons to impression manage. Whilst previous research has acknowledged the role played by self-determined motivation in adherence to physical activity behavior (Hagger, Chatzisarantis, & Biddle, 2002), recent research has also proposed that self-determined motivation in exercise may downplay social evaluations and alleviate concerns about one’s physique (Thøgerson-Ntoumani & Ntoumanis, 2006). The different forms of motivation within SDT have been described in the motivation literature as the “why” of actions (Wininger, 2007). In this light, the framework can provide clarification of the mediators or mechanisms by which predictor factors (e.g., personality traits) influence exercise behavior and psychological outcomes. Additionally, it provides information on how constructs can be targeted to create behavior change (Hagger & Chatzisarantis, 2008). Therefore, the assessment of the interplay between self-presentation, self-determined motivation, and adaptive perfectionism might prove useful in clarifying the conditions under which self-presentation in exercise leads to positive exercise pursuits and in the development of successful self-presentational interventions designed to increase exercise behavior. The present study Based on the above discussion, a structural model was proposed that examines the differential associations of two broad dimensions of dispositional perfectionism with contextual motivation, self-presentation processes, and exercise behavior. The proposed relationships displayed in Fig. 1 are hypothesized as follows: H1. Utilizing the Relative Autonomy Index (RAI) as a gauge of overall self-determination (Ingledew et al., 2004), adaptive perfectionism is expected to show a positive association with self-determined motivation for exercise; maladaptive perfectionism is expected to show a negative relationship with self-determined motivation. H2. Both adaptive perfectionism and maladaptive perfectionism will show positive associations with impression motivation and impression construction. H3. Impression motivation, impression construction, and the RAI are expected to show a positive association with aerobic exercise behavior. Hypothesized model of the relationships between adaptive and maladaptive ... Fig. 1. Hypothesized model of the relationships between adaptive and maladaptive perfectionism, self-determination in exercise, self-presentation processes and exercise behavior. Note. Regression coefficients and measurement error variances were specified for the motivation and self-presentation composite factors. SOP = Self-Oriented Perfectionism, O = Organization, HPS = High Personal Standards, SPP = Socially Prescribed Perfectionism, COM = Concern Over Mistakes, DA = Doubts About Actions, PP = Parental Perceptions, IM = Impression Motivation, IC = Impression Construction, RAI = Relative Autonomy Index, Exercise Behavior = Intensity (METs equivalence) × Duration (minutes per session) × Frequency (days per week).

Results Characteristics of the sample On average, participants exercised 3.26 days per week (SD = 1.71) for 47.83 min per session (SD = 21.41) for a total of 169.25 min per week (SD = 129.52). The average MET intensity for an exercise session across all participants was 6.57 METs (SD = 21.41), categorized as vigorous activity according to the Compendium of Physical Activities ( Ainsworth et al., 2000). The mean BMI of participants was 22.82 (SD = 3.79), which is within the healthy range for BMI classifications (18.5–25; Kuczmarski & Flegal, 2000). Preliminary analyses Descriptive statistics and correlations for all variables in the model are presented in Table 1. The internal reliabilities for the scales were typically satisfactory, but sub-optimal for the Doubts about Action subscale in the F-MPS and Introjected Regulation in the BREQ-2 (see Table 1). Confirmatory factor analysis (CFA) demonstrated that the two-dimensional model of perfectionism provided a better fit to the data when compared against a modified model with the Doubts about Action subscale removed (∆χ2 (1) = 83.869, p < .001). Inspection of other fit indices indicated acceptable model fit (χ2 = 46.109, df = 13, IFI = .95, CFI = .95, TLI = .92, RMSEA = .08). Overall, the results supported the two factor model of perfectionism specified in Cox et al. (2002), and the Doubts About Action subscale was therefore retained in subsequent factor analyses. Calculation of the RAI had the effect of addressing the low internal consistency for the Introjected regulation scale. Composite scores for the RAI and self-presentation constructs were modeled as single indicator latent variables by specifying the values of the regression coefficients and the measurement error variance associated with each composite variable ( Bollen, 1989). This process accounts for measurement error and thus reduces bias in the model when including composite scales. Table 1. Descriptive statistics and correlation coefficients between all variables in the study. Correlations M SD 1 2 3 4 5 6 7 8 9 Multidimensional Perfectionism 1. Self-Oriented Perfectionism (SOP) 23.68 6.13 1.00 2. Organization (O) 14.10 3.95 .40** 1.00 3. Personal Standards (PS) 17.52 4.23 .71** .40** 1.00 4. Socially Prescribed Perfectionism (SPP) 15.29 6.61 .43** −.01 .31** 1.00 5. Concern Over Mistakes (COM) 11.85 4.35 .47** .13* .40** .60** 1.00 6. Parental Perceptions (PP) 12.77 4.77 .29** −.04 .22** .66** .47** 1.00 7. Doubts About Action (DA) 7.62 2.48 .21** −.01 .13* .47** .43** .48** 1.00 8. Adaptive Perfectionism 55.30 11.79 .87** .72** .86** .30** .42** .19** .15* 1.00 9. Maladaptive Perfectionism 47.53 14.95 .44** .03 .33** .85** .78** .82** .74** .33** 1.00 Motivation Regulations for Exercise 10. Amotivation 1.46 .73 .02 −.07 −.13* .27** .20** .28** .25** −.08 .32** 11. External Regulation 1.93 .85 .11 −.13* −.01 .46** .40** .45** .43** .01 .54** 12. Introjected Regulation 2.92 1.03 .21** .12 .18** .18** .38** .17** .21** .22** .28** 13. Identified Regulation 3.80 .86 .11 .08 .21** −.11 .01 −.16** −.01 .18** −.11 14. Intrinsic Motivation 3.80 .87 .12 .19** .18** −.20** −.09 −.21** −.05 .21** −.18** 15. Relative Autonomy Index 7.85 6.01 .01 .15* .15* −.38** −.28** −.40** −.28** .14* −.43** Self-Presentation Processes 16. Impression Motivation 17.81 4.05 .31** .16* .28** .22** .22** .18** .13* .33** .22** 17. Impression Construction 8.72 4.61 .21** .07 .19** .35** .37** .25** .34** .20** .40** Exercise Behaviors 18. Total Exercise Behavior 909.54 820.43 .06 .11 .16* −.10 −.03 −.17** −.15** .13* −.15* 10 11 12 13 14 15 16 17 18 Motivation Regulations for Exercise 10. Amotivation 1.00 11. External Regulation .33** 1.00 12. Introjected Regulation −.18** .28** 1.00 13. Identified Regulation −.50** −.16* .49** 1.00 14. Intrinsic Motivation −.42** −.25** .23** .68** 1.00 15. Relative Autonomy Index −.75** −.61** .05 .73** .81** 1.00 Self-Presentation Processes 16. Impression Motivation −.28** .07 .36** .36** .27** .24** 1.00 17. Impression Construction .10 .16* .29** .17** .15** −.02 .37** 1.00 Exercise Behaviors 18. Total Exercise Behavior −.23** −.17** .33** .45** .36** .38** .27** .11 1.00 Note. Total Exercise Behavior = Frequency (days per week) × Duration (minutes per session) × Intensity (METs equivalence). *p < .05, **p < .001. Table options Assessing model fit The full structural model specified in Fig. 1 was subsequently tested to confirm the hypothesized relationships. The structural model provided a good fit to the data (χ2 = 122.876, df = 56, IFI = .94, CFI = .94, TLI = .92, RMSEA = .068). Inspection of the modification index was associated with the covariance between the error term for self-determined motivation and self-oriented perfectionism, suggesting that model fit would improve if the covariance between the two error terms was treated as a free parameter to be estimated. Post-hoc re-specification of models allowing measurement errors to correlate have been criticized because they may capitalize on random, sample-specific characteristics of the data ( Cole, Ciesla, & Steiger, 2007). The modification was therefore rejected and the model was accepted as the final model. Regarding the standardized coefficients of the model ( Fig. 2), the results show that all paths were in the expected direction as hypothesized. Final accepted path model (N = 254). All coefficients are standardized ... Fig. 2. Final accepted path model (N = 254). All coefficients are standardized coefficients. *p < .05, **p < .001 (two-tailed), ns = non-significant. Note. SOP = Self-Oriented Perfectionism, O = Organization, HPS = High Personal Standards, SPP = Socially Prescribed Perfectionism, COM = Concern Over Mistakes, DA = Doubts About Actions, PP = Parental Perceptions, IM = Impression Motivation, IC = Impression Construction, RAI = Relative Autonomy Index, Exercise Behavior = Intensity (METs equivalence) × Duration (minutes per session) × Frequency (days per week). Figure options Results of the tests of standardized direct, indirect, and total effects are presented in Table 2. As expected, maladaptive perfectionism had an inverse direct effect on the RAI (direct effect = −.833, p < .001) whereas, adaptive perfectionism showed a positive direct effect on the RAI (direct effect = .512, p < .001). Maladaptive perfectionism had a positive direct effect on both impression motivation (direct effect = .756, p < .001) and impression construction (direct effect = .932, p < .001). Adaptive perfectionism had a negative direct effect on impression construction (direct effect = −.284, p < .001) and a non-significant, negative direct effect on impression motivation (direct effect = −.08, p < .05). 5 The RAI partially mediated the significant indirect effect of adaptive perfectionism on impression construction (indirect effect = .284, p < .001) and fully mediated the relationship between adaptive perfectionism and impression motivation (indirect effect = .387, p < .001). 6 The negative relationship between maladaptive perfectionism and the RAI partially mediated the indirect effect of maladaptive perfectionism on impression motivation (indirect effect = −.432, p < .001) and impression construction (indirect effect = −.629, p < .001). The RAI had a positive direct effect on impression motivation (direct effect = .756, p < .001) and impression construction (direct effect = .555, p < .001). Table 2. Significance test of the standardized direct and indirect effects. Total Indirect effect Direct effect B B B SE Z Mediation model I: Direct paths from perfectionism to self-presentation processes Adaptive perfectionism Full-size image (3 K) impression motivation .307 .387 −.080 .147 −.701 Adaptive perfectionism Full-size image (3 K) impression construction .000 .284 −.284 .154 −2.399** Maladaptive perfectionism Full-size image (3 K) impression motivation .127 −.629 .756 .276 4.676*** Maladaptive perfectionism Full-size image (3 K) impression construction .440 −.462 .932 .297 5.411*** Partial mediation model II: Direct paths from motivation and self-presentation to exercise behavior RAI Full-size image (3 K) aerobic exercise behavior .744 .188 .556 .078 5.603*** Impression motivation Full-size image (3 K) exercise behavior .142 .000 .142 .072 1.975* Impression construction Full-size image (3 K) exercise behavior .149 .000 .149 .067 1.977* Partial mediation model II: Perfectionism to exercise behavior Adaptive perfectionism Full-size image (3 K) exercise behavior .327 .327 – – – Maladaptive perfectionism Full-size image (3 K) exercise behavior −.375 −.375 – – – *p < .05, **p < .01, ***p < .001. Table options In the prediction of aerobic exercise behavior, positive total effects were observed for the RAI (total effect = .744, p < .001) impression motivation (total effect = .142, p < .05), and impression construction (total effect = .149, p < .05). The indirect effect of the RAI on exercise behavior explained 25.27% of the total effect. The sequential pathways from the RAI to impression motivation and impression construction mediated the total effect of maladaptive perfectionism on aerobic exercise behavior (total effect = −.375, p < .001). 7 A positive total effect was found for adaptive perfectionism on aerobic exercise behavior via a positive association with the RAI (total effect = .327, p < .001). Altogether, the model explained 42% of the variance in exercise behavior.