athletes’ broad dimensions of dispositional perfectionism: … · 2017-10-11 · coping in the...

356

Journal of Sport & Exercise Psychology, 2008, 30, 356-382 © 2008 Human Kinetics, Inc.

The authors are with School of Psychology, University of Ottawa, Ottawa, ON, Canada.

Athletes’ Broad Dimensions of Dispositional Perfectionism:

Examining Changes in Life Satisfaction and the Mediating Role of Sport-Related

Motivation and Coping

Patrick Gaudreau and Sheilah AntlUniversity of Ottawa

This study examined the associations of dispositional perfectionism, contextual motivation, sport-related coping, goal attainment, and changes in life satisfaction during a sport competition. A sample of 186 athletes completed measures of dispositional perfectionism, contextual motivation, and life satisfaction at Time 1 (before a competition) as well as measures of coping, goal attainment, and life satisfaction at Time 2 (after a competition). Results of structural equation modeling supported a model in which self-determined and non-self-determined motivation partially mediated the relationships between different dimensions of perfection-ism and coping. It was also shown that disengagement-oriented coping mediated the negative relationship between evaluative concerns perfectionism and change in life satisfaction. In a similar way, goal attainment mediated the relationships of both task- and disengagement-oriented coping with change in life satisfaction. For the most part, these results are consistent with the motivational properties of evaluative concerns and personal standards perfectionism and with literature regarding coping and self-determination theory.

Keywords: personality, self-regulation, subjective well-being, intrinsic motivation, stress management

Perfectionism has been conceived as a personality disposition characterized by a tendency to strive for perfection and to evaluate the self in a critical manner (Cox, Enns, & Clara, 2002; Flett & Hewitt, 2006). This generally accepted definition implies that perfectionism can be divided into distinct dimensions. Recently, it has been proposed that two broad dimensions of perfectionism, namely personal standards and evaluative concerns perfectionism (e.g., Dunkley, Blankstein, Halsall, Williams, & Winkworth, 2000), can be derived out of the individual

Perfectionism, Motivation, and Coping 357

conceptualizations of Hewitt and Flett (1991) and Frost and his colleagues (Frost, Marten, Lahart, & Rosenblate, 1990). These dimensions, which provided the basis for this study, are consistent with the general definition of perfectionism representing the personal striving for perfection and the predisposition to evaluate oneself in a critical manner, respectively.

The first broad dimension, evaluative concerns perfectionism (ECP), entails a socially prescribed tendency to evaluate oneself harshly, to doubt one’s capacity to bring about desired outcomes, and to perceive that others require perfection from oneself (Dunkley & Blankstein, 2000). Evaluative concerns perfectionism, which has also been labeled perfectionistic concerns (Stoeber & Otto, 2006), incorporates the facets of socially prescribed perfectionism from Hewitt and Flett’s (1991) framework as well as the constructs of concerns over mistakes, doubts about actions, and parental criticism from Frost and colleagues’ (1990) conceptualization. There is now clear evidence in the general psychological literature that ECP is linked to negative outcomes such as depression (e.g., Cox et al., 2002; Dunkley, Blankstein, Masheb, & Grilo, 2006), negative affect (Bieling, Israeli, Smith, & Antony, 2003; Dunkley, Zuroff, & Blankstein, 2003), and psychosomatic symptoms (Dunkley & Blankstein, 2000). Among athletes, the ECP facets have been related to higher anxiety (Frost & Henderson, 1991; Hall, Kerr, & Matthews, 1998), anger (Vallance, Dunn, & Dunn, 2006), and burnout (Gould, Udry, Tuffey, & Loehr, 1996), as well as to lower confidence (Frost & Henderson, 1991; Hall et al., 1998) and self-esteem (Gotwals, Dunn, & Wayment, 2003).

The second broad dimension, personal standards perfectionism (PSP), involves a self-oriented tendency to set highly demanding standards and to strive for their attainment (Dunkley & Blankstein, 2000). Personal standards perfectionism, which has also been labeled perfectionistic strivings (Stoeber & Otto, 2006), incorporates the facets of self-oriented perfectionism from Hewitt and Flett’s (1991) model and the construct of high personal standards from Frost and colleagues’ (1990) frame-work. A literature review by Stoeber and Otto (2006) indicated that 10 out of 15 studies provided positive evidence for the healthy nature of PSP. Specifically, stud-ies have shown positive relationships of PSP facets with variables such as positive affect (Bieling et al., 2003), self-efficacy (Martin, Flett, Hewitt, Krames, & Szanto, 1996), academic achievement (Bieling et al., 2003; Cox et al., 2002), and goal progress (Powers, Koestner, & Topciu, 2005). Despite such empirical support, the healthy nature of PSP remains largely equivocal (e.g., Dunkley et al., 2006; Flett & Hewitt, 2006). For instance, the review of the literature for this article established that only 4 out of 17 published studies reported a significant positive bivariate cor-relation between self-oriented perfectionism and indicators of subjective well-being (Frost, Heimberg, Holt, Mattia, & Neubauer, 1993; Hewitt, Flett, & Mikail, 1995; Miquelon, Vallerand, Grouzet, & Cardinal, 2005; Schwartz et al., 2002).1 Given these inconsistent results, it is not surprising that different researchers consider PSP to be either a healthy (Stoeber & Otto, 2006), an unhealthy (Flett & Hewitt, 2005), or a neutral (Bieling, Israeli, & Antony, 2004) component of perfectionism.

Despite the mounting debate surrounding the healthy versus unhealthy nature of PSP (Dunkley et al., 2006; Flett & Hewitt, 2006), little empirical effort has been devoted to the mediating mechanisms through which the two broad dimensions of perfectionism could relate to distinct outcomes. Separate streams of research with college students have respectively indicated that motivational orientations (e.g.,

358 Gaudreau and Antl

Miquelon et al., 2005) and coping processes (e.g., Dunkley et al., 2003) medi-ate the relationship between perfectionism and emotional adjustment. As such, Miquelon and colleagues (2005) highlighted the need to examine how motivation influences coping strategies and how these two processes can combine to medi-ate the relationship between perfectionism and psychological adjustment. From a therapeutic standpoint, a more integrated approach is needed to identify change-able mechanisms on which sport psychologists could intervene to diminish the problems associated with ECP. Hence, this study proposed an integrative model in which motivational orientations and coping processes are conceived as media-tors through which the dimensions of perfectionism distinctly relate to athletes’ adjustment (see Figure 1).

From Perfectionism to Motivation

Self-determination theory (SDT; Deci & Ryan, 2002), which is gaining widespread acceptance as an integrative framework for the study of human motivation (Valler-and, 1997), has been successfully used in previous studies on both perfectionism (Van Yperen, 2006) and coping (Knee & Zuckerman, 1998). Self-determination theory categorizes the various reasons for which an activity is pursued into six types of motivation that can be regrouped in two broad dimensions of motivation (e.g., Knee & Zuckerman, 1998; Miquelon et al., 2005). Self-determined activities are performed for pleasure (i.e., intrinsic) because they are perceived as an integral part of the self (i.e., integrated), and because they are viewed as important for a person (i.e., identified). In contrast, non-self-determined activities are performed to avoid guilt and shame (i.e., introjected) and to obtain rewards or to avoid punishments (i.e., extrinsic). Non-self-determined motivation also includes amotivation, which represents a lack of valuing the activity.

Little attention has been allocated to the linkage between dispositional perfec-tionism and motivational processes within the self-determination theory. However, external pressure in the form of performance-contingent feedback and rewards (Deci & Ryan, 2002) has been pinpointed as an undermining factor of self-determined motivation. As such, the dual-process model of Slade and Owens (1998) and the work of Hewitt and Flett (1991) have proposed that self-oriented perfectionism, a facet of PSP, should foster self-determined forms of motivation given the tendency to approach success with self-referenced criteria and the personal motivation to strive for perfection. On the contrary, ECP should yield non-self-determined motivation owing to the tendency to avoid blame and negative evaluations from others and because of the perceived controlling environment.

In a recent study with college students, semipartial correlations indicated that self-oriented perfectionism—controlling for socially prescribed perfectionism—was positively associated with intrinsic motivation (Van Yperen, 2006). Socially prescribed perfectionism—controlling for self-oriented perfectionism—was positively associated with amotivation and with a variable aggregating extrinsic, identified, and introjected motivation. The conceptual meaning of this variable remains unclear because both identified and intrinsic motivations should be treated as self-determined types of motivation. Nonetheless, the results of a prospective study by Miquelon and her collaborators (2005) indicated that students with a high level of self-oriented perfectionism were showing higher self-determined

359

Fig

ure

1 —

Hyp

othe

size

d m

odel

of

the

rela

tions

hips

bet

wee

n di

spos

ition

al p

erfe

ctio

nism

, con

text

ual m

otiv

atio

n, s

ituat

iona

l cop

ing,

goa

l atta

in-

men

t, an

d lif

e sa

tisfa

ctio

n. +

, pos

itive

rel

atio

nshi

p; −

, neg

ativ

e re

latio

nshi

p.


motivation toward school. In contrast, socially prescribed perfectionism was positively associated with non-self-determined motivation. Despite these promising findings, it remains to be determined whether broad dimensions of dispositional perfectionism can predict contextual motivation in the sport domain.

Perfectionism, Motivation, and Coping

As shown in Figure 1, dispositional perfectionism is expected to correlate signifi-cantly with motivational orientations, which, in turn, should correlate significantly with coping processes. Coping represents the behavioral and cognitive efforts by which a person attempts to manage the demands of a stressful situation (Lazarus, 1999). Although athletes use a variety of coping strategies (Hoar, Kowalski, Gaud-reau, & Crocker, 2006), the coping construct can be organized by aggregating coping strategies into coping dimensions (Skinner, Edge, Altman, & Sherwood, 2003). A first dimension, task-oriented coping, aims at dealing directly with the stressor and the resulting thoughts and emotions. This dimension incorporates problem-focused (e.g., effort expenditure, logical analysis) and approach emotion-focused coping strategies (e.g., relaxation, thought control; Nes & Segerstrom, 2006). A second dimension, disengagement-oriented coping, also labeled avoidance-oriented coping, encompasses strategies like behavioral disengagement and venting of unpleasant emotions, through which athletes disengage themselves from the process of dealing with the actual demands of the situation. A third dimension, distraction-oriented coping, represents strategies like distancing and mental distraction, which are used to momentarily focus on stimuli unrelated to the stressful situation. Despite a tendency to merge disengagement and distraction into a single construct, these two dimensions were found to form distinct factors in confirmatory factor analyses (e.g., Ayers, Sandler, West, & Roosa, 1996). Moreover, these two coping dimensions differently relate to a myriad of positive and negative outcomes, thus providing support for their different adaptive function (Skinner et al., 2003).

Flett and colleagues (1991) argued that different dimensions of perfectionism promote the use of distinct situational coping actions. Dunkley and Blankstein (2000) showed that self-oriented perfectionism was significantly related to task-oriented coping, whereas socially prescribed perfectionism was related to avoidance-oriented coping. Furthermore, coping dimensions have recently been modeled as potential mediators of the relationship between perfectionism and emotional outcomes. Structural equation analyses demonstrated the mediating role of avoidance-oriented coping in the association of ECP with emotional distress (Dunkley & Blankstein, 2000; Dunkley et al., 2000). In addition, a study using daily measures of coping showed that avoidance-oriented coping and perceived stress both acted as partial mediators between ECP and negative affect (Dunkley et al., 2003).

The mediating role of coping in the association from perfectionism to affect has been identified. However, the mediating processes by which dimensions of perfectionism relate to coping remain unexplored. As displayed in Figure 1, and consistent with the proposition of Miquelon et al. (2005), it was hypothesized that motivational orientations should act as mediators between perfectionism and coping dimensions. It has been shown that self-determined motivation relates to a myriad of positive outcomes (Vallerand & Rousseau, 2001). Studies in the


academic (Knee & Zuckerman, 1998) and interpersonal domains (Knee, Patrick, Vietor, Nanayakkara, & Neighbors, 2002) have established that self-determined motivation positively correlates with task-oriented coping, on the one hand, and that non-self-determined motivation relates to higher usage of disengagement-oriented coping, on the other hand. Contextual motivation to participate in sport has also been related to situational coping during a sport competition (Amiot, Gaudreau, & Blanchard, 2004). Specifically, task-oriented coping fully mediated the relationship between self-determined motivation and positive affect whereas disengagement-oriented coping mediated the relationship between non-self-determined motivation and negative affect. Up to now, research on self-determination has yet to distinguish between disengagement and distraction-oriented coping. However, distraction has been found to relate to dispositional pessimism (Gaudreau & Blondin, 2004), which is a known antecedent of non-self-determined motivation (Shogren, Lopez, Wehmeyer, Little, & Pressgrove, 2006). Furthermore, orienting one’s attention on external or internal activities unrelated to the task at hand (i.e., distraction) is at the opposite end of good concentration and task absorption, which are both negatively associated with non-self-determined motivation (Brière, Vallerand, Blais, & Pelletier, 1995). Therefore, it can be expected that non-self-determined motivation should positively correlate to passive forms of self-regulation such as distraction-oriented coping.

Coping, Goal Attainment, and Life satisfaction

Studies in the field of coping have focused extensively on emotional outcomes while rarely considering the role of coping in achievement-related outcomes. As displayed in Figure 1, this study sought to extend the literature by examining the associations of coping with both goal attainment and life satisfaction. Personal goals serve as an important reference point for the self and the affective systems (e.g., Carver & Scheier, 1998). A meta-analytical review indicated that goal attainment relates concurrently and prospectively to higher subjective well-being (Koestner, Lekes, Powers, & Chicoine, 2002). Of particular interest, a single positive experience of goal attainment is sufficient to render sustainable increases in emotional adjust-ment (Wiese, 2007). Recent theorizing also posits that effortful goal striving is a key element in the maintenance of happiness (Lyubomirsky, Sheldon, & Schkade, 2005). In the same way, coping utilization during a sport competition has been proposed to influence both goal attainment and change in subjective well-being (Hoar et al., 2006). Consistent with the model displayed in Figure 1, studies in the sport domain have supported a sequence in which goal attainment mediates the relationship between coping and situation-specific subjective well-being (Amiot et al., 2004; Gaudreau & Blondin, 2004). Specifically, task-oriented coping pro-moted higher goal attainment, which increased the likelihood of positive affect. In contrast, disengagement-oriented coping related to lower goal attainment, which resulted in increased negative affect. Distraction-oriented coping was not expected to correlate to either goal attainment or emotional adjustment given that it relates to adaptive outcomes only when used in combination with more active self-regulation processes such as task-oriented coping (Gaudreau & Blondin, 2004).


The Present Study

In this study, perfectionism was conceived as a personality trait influencing behavior and emotional adjustment at the situational, contextual, and global levels of analysis (Flett & Hewitt, 2006). Although athletes’ perfectionism can be measured at the contextual level with sport-based questionnaires (e.g., Dunn et al., 2006), their dis-positional level of perfectionism may still predict motivational orientations, coping processes, and adjustment. Hence, this prospective structural equation modeling study aimed at extending previous research by proposing an integrative model of the differential associations of two broad dimensions of dispositional perfection-ism with contextual motivation, situational coping, goal attainment, and changes in general life satisfaction of athletes. As with studies on situational coping (e.g., Amiot et al., 2004), an important sport competition was targeted for this research to ensure that the motivational relevance of the situation would be sufficient to prompt the utilization of coping strategies. The hypothesized relationships between the variables are displayed in Figure 1.

This research entailed three sets of mediation hypotheses. Firstly, self-determined motivation was expected to mediate the relationship between PSP and task-oriented coping, whereas non-self-determined motivation was presumed to mediate the relationships of ECP with disengagement- and distraction-oriented coping. There are other factors that may mediate the relationship between person-ality dispositions and coping utilization, such as secondary cognitive appraisals (e.g., Lazarus, 1999). For this reason, partial rather than complete mediation was expected. Secondly, goal attainment was hypothesized to mediate the relation of task- and disengagement-oriented coping with changes in life satisfaction. Distraction-oriented coping was not expected, in and of itself, to relate significantly to goal attainment and life satisfaction, particularly after accounting for task- and disengagement-oriented coping (e.g., Gaudreau & Blondin, 2004). Finally, the relationship between ECP and changes in life satisfaction was expected to be mediated by a sequence leading from contextual motivation to situational coping to goal attainment. Two exploratory mediation hypotheses could also be investigated in the current model. On the basis of our review of 17 published studies, PSP was not expected to correlate significantly with life satisfaction. However, these studies were performed outside of the sport domain and the relationship between PSP and life satisfaction remains to be explored within the athletic population. Similarly, the model could assume the mediating role of coping in the relationships between motivational orientations and goal attainment. However, the literature on the link between contextual self-determination and goal attainment remains scarce in the sport domain (Vallerand & Rousseau, 2001). Therefore, these exploratory mediation hypotheses were tested only in the presence of a significant association between PSP and life satisfaction, on the one hand, and between motivational orientations and goal attainment, on the other hand.

Method

Participants

A sample of 209 French Canadian athletes was recruited to participate in this study. Of the initial sample, 23 athletes did not complete the Time 2 questionnaire


package. The final sample included 186 participants (57% male) ranging from 14 to 28 years of age (M = 18.30, SD = 3.25). On average, they had been competing in their sport for 7.76 years (SD = 4.12) and were participating at regional (39%), provincial (55%), and national levels of competition (6%). They were competing in ice hockey (16%), soccer (31%), volleyball (29%), broomball (7%), tennis (6%), badminton (6%), and alpine skiing (5%).

Procedure and Design

A research assistant contacted coaches to inform them of the main goals of the study and to obtain permission to recruit their players. Coaches were asked to choose an important competition as a target for the research. A majority of participants (73%) perceived the competition to be the most important of the season whereas the remainders perceived it as being one of the most important of their season. As a consequence, a majority of coaches required that the questionnaires be kept as short as possible in order not to interfere with the preparation for the competition. The study was presented to 18 groups of athletes during the week preceding the start of the study and the goal was to recruit all athletes who were attending the meeting. The size of the groups ranged from 6 to 17 athletes and the participation rate was always higher than 85%.

Participants completed the first questionnaire package during their last training session before their competition. The Time 1 questionnaire package measured life satisfaction, perfectionism, and motivation. On average, this questionnaire pack-age was completed 66 hr before the competition (SD = 24 hr; range 22–120 hr). The second questionnaire package, which was completed during the first training session after competition, entailed the recall of coping used during the competi-tion, as well as the concurrent report of goal attainment and life satisfaction. On average, this questionnaire package was completed 70 hr after competition (SD = 29 hr, range 24–144 hr). Variation in precompetition time delay had no significant association with the Time 1 variables, as did the variation in postcompetition time on the variables measured at Time 2 (ps > .10).

Measures2

Perfectionism (Time 1). Perfectionism was measured with the French version (Labrecque, Stephenson, Boivin, & Marchand, 1999) of the HF-MPS (Hewitt & Flett, 1991) and F-MPS (Frost et al., 1990). Brief versions from Cox et al. (2002) were used given the limited time available to complete the questionnaires. Results of a confirmatory factor analytical study revealed that the brief versions provide superior fit to the data compared with their original counterparts, with correlations between the brief and the original versions being above .85 (Cox et al., 2002).

The French version of the brief F-MPS was used to measure concerns over mistakes (five items), doubts about actions (four items),3 parental pressure (five items), personal standards (five items), and organization (four items).4 Items were rated using a scale ranging from 1 (strongly disagree) to 5 (strongly agree). Two scales from the French version of the brief HF-MPS were also used to measure self-oriented (five items) and socially prescribed perfectionism (five items). Items were rated using a scale from 1 (not at all agree) to 7 (totally agree). In all analyses reported in this study, the subscales from these two questionnaires were combined


into latent variables of PSP (i.e., personal standards and self-oriented perfectionism) and ECP (i.e., concerns over mistakes, doubts about actions, parental pressure, and socially prescribed perfectionism).

Sport-Related Motivation (Time 1). Selected subscales (i.e., intrinsic, identi-fied, extrinsic, and amotivation) from the French version of the Sport Motivation Scale (SMS; Brière et al., 1995) were used to assess contextual motivation. Each subscale contained three items selected through item-level and confirmatory factor analyses on the databases available in our research laboratory. In addition, items were selected to be nonredundant and to cover the entire conceptual domain of each type of motivation. In other words, an effort was made to maximize interitem correlation without sacrificing the content validity of the scales (Smith, McCarthy, & Anderson, 2000). The 12 items were rated on a scale from 1 (not at all agree) to 7 (totally agree). Participants indicated the extent to which each item represented a reason why they generally practice their sport. Self-determined and non-self-determined motivations were treated as separate variables. Such an approach was preferable given their differential relationships with coping (e.g., Amiot et al., 2004; Knee et al., 2002) and perfectionistic dispositions (Miquelon et al., 2005).5 Although latent variables can be identified with only two indicators (Bollen, 1989), it is preferable to include at least three indicators per latent construct to prevent empirical underidentification (Kenny, 1979). Only two subscales were available to model self-determined motivation (i.e., intrinsic and identified) and non-self-determined motivation (i.e., amotivation and extrinsic). Therefore, we relied on an item-level modeling approach in which pairs of individual items representing intrinsic and identified motivation, on the one hand (i.e., self-determined motiva-tion

1 = intrinsic

1 + identified

1), and extrinsic motivation and amotivation, on the

other hand (i.e., nonself-determined motivation1 = extrinsic

1 + amotivation

1), were

respectively summed unweighted. As such, the approach allowed for the utiliza-tion of three indicators for each of the self-determined and non-self-determined motivation latent variables.

Coping During the Sport Competition (Time 2). Coping was measured using the French version of the Coping Inventory for Competitive Sport (CICS; Gaudreau & Blondin, 2002). Using a scale from 1 (does not correspond at all) to 5 (corresponds very strongly), athletes rated the extent to which each item represented their actions or thoughts during the competition. The CICS contains nine four-item subscales and one three-item subscale that can be organized in second-order dimensions repre-senting task-oriented (mental imagery, thought control, effort expenditure, seeking support, logical analysis, and relaxation), distraction-oriented (mental distraction and distancing), and disengagement-oriented coping (venting of unpleasant emo-tions and disengagement/resignation). Previous studies provided support for the reliability and validity of the CICS (Gaudreau & Blondin, 2002, 2007). The latent variables of task-, distraction-, and disengagement-oriented coping were created based on the hierarchical framework of the CICS. Consistent with validation stud-ies, effort expenditure was set to cross-load negatively on disengagement-oriented coping whereas seeking support loaded on distraction-oriented coping rather than on task-oriented coping. These specifications were tenable given that effort expenditure prevents athletes from disengaging themselves from the task at hand and because seeking support during a competition enables momentary distraction.


Perceived Goal Attainment (Time 2). The Attainment of Sport Achievement Goal Scale (A-SAGS; Gaudreau, Amiot, Blondin, & Blanchard, 2002) was used to assess goal attainment. Using a 7-point Likert-type scale from 1 (does not cor-respond at all) to 7 (corresponds completely), athletes indicated the extent to which each item corresponded to their performance during the competition. The A-SAGS contains three four-item subscales measuring different, yet interrelated, criteria of goal attainment: mastery (e.g., “I mastered the difficulties of the situation”), self-referenced (e.g., “I did better than my usual performances”), and normative (e.g., “I outperformed other athletes”). Previous research lent credence to the factor validity of a hierarchical model in which the three first-order factors loaded significantly on a factor of goal attainment (Gaudreau et al., 2002). Consistent with tenets of the A-SAGS, the interscale correlations in this sample were moderately high (from .52 to .65). Hence, the three subscales were used as indicators of a latent variable representing goal attainment.

General Life satisfaction (Time 1 & 2). General life satisfaction was assessed using the French version (Blais, Vallerand, Pelletier, & Brière, 1989) of the Satis-faction with Life Scale (SWLS; Pavot & Diener, 1993). Using a scale from 1 (not at all agree) to 7 (totally agree), participants were explicitly asked to rate each item according to their life in general rather than specifically to their sport domain. The five items were used as indicators of the Time 1 and Time 2 latent variables of life satisfaction.

Overview of Statistical AnalysesStructural equation modeling (SEM) was performed using maximum likelihood estimation derived from the EQS 6.1 software. Even with a small sample size (n < 200), SEM was preferred to both path analysis with manifest variables and multiple regressions because this approach automatically corrects the unreliability of the mediators, which is crucial to increase the power of testing indirect effects (Hoyle & Kenny, 1999).

Firstly, a confirmatory factor analysis was performed to assess the suitability of the proposed measurement model and to estimate error-free correlations between latent variables. Secondly, as recommended by Holmbeck (1997), a preliminary test was performed to ascertain the existence of a significant association between the independent variable(s) and the dependent variable(s) without the presence of the mediator(s) in each of our three predetermined hypotheses of mediation (perfectionism → coping; coping → life satisfaction; ECP → life satisfaction). These incomplete and misspecified models were tested to set up the logical argu-ment of the mediation hypotheses rather than to provide evidence of goodness of fit. Thirdly, a series of latent path analyses were performed to assess the fit of the proposed structural model. In a first step, the full mediation model was tested (see Figure 1). In a second step, three predetermined partial mediation models were tested and compared against the full mediation model. The full mediation model is supported if the partial mediation model does not provide a better fit to the data (in other words, if the newly added direct paths are not significant). Finally, each newly added path was decomposed into direct and indirect effects (Mackinnon, Lockwood, Hoffman, West, & Sheets, 2002). The direct effect represents the association of an independent variable with a dependent variable whereas the


indirect effect corresponds to the effect of one or more mediating variables in that relationship. Evidence for a fully mediated relationship necessitates the indirect effect to be significant (Z > 1.96) and the direct effect to be nonsignificant (Z ≤ 1.96). In contrast, a partially mediated relationship requires both the indirect and the direct effects to be significant. It should be noted that the power of indirect effects remains limited (Hoyle & Kenny, 1999). Hence, a strict interpretation of statistical significance could render type-II errors as it does not consider the effect size of the indirect effect (i.e., indirect effect / total effect).

The fit of the model was tested using χ2 and other fit indices. Values between .90 and .94 for the CFI and NNFI indicate acceptable fit whereas values of .95 and higher indicate a relatively good fit (Hu & Bentler, 1999). Values smaller than .08 for the RMSEA indicate acceptable fit whereas values smaller than .06 indicate a relatively good fit. Furthermore, when the upper bound of the RMSEA 90% CI was below .06, the model was assumed to provide a close fit. The Satorra–Bentler rescaled χ2 (SBχ2), the robust CFI (RCFI), and the robust standard errors were reported given that multivariate normality could not be assumed (multivariate normalized kurtosis = 10.95). Nested models were compared using the formula of the difference in SBχ2 that takes into account the uncorrected χ2 (Satorra & Bentler, 2001).

Results

Preliminary Analyses

Attrition is a common issue in longitudinal research. In this study, the percentage of missing data at Time 2 (11%) was reasonable. However, results of ANOVAs revealed that participants (n = 186) who completed the Time 2 measures were significantly lower (p < .01) on PSP, F(1, 207) = 9.14, d = −0.60,) and ECP, F(1, 207) = 25.09, d = −0.90), and they were higher on life satisfaction at Time 1, F(1, 207) = 6.88, d = 0.48, compared with the participants who did not complete the Time 2 questionnaire (n = 23). As such, the differential attrition will be discussed more extensively as a limitation to the generalizability of the results of this study. For now, a listwise deletion of the missing participants was chosen because the data were not missing at random.

Box’s M tests were performed to assess homogeneity of covariance matrix across genders and types of sport (i.e., individual and team sports). Using the stringent alpha level (p < .001), recommended for this analysis (Meyers, Gamst, & Guarino, 2005), it was found that the covariance matrix was homogeneous across genders, Box’s M(406, 87,981) = 563.36, p > .01, and heterogeneous across types of sport, Box’s M(406, 9023) = 913.47, p < .001. As suggested by Stevens (1996), however, this latter result should be interpreted cautiously given the very unequal sample size of athletes from individual (17%) and team sports (83%). The data were pooled together for subsequent analyses.

Measurement Model

The descriptive statistics and reliability estimates of Time 1 and Time 2 measures are provided in Table 1. A measurement model with 10 interrelated latent variables was initially tested. The auto-correlated uniquenesses of the Time 1 and Time 2


Table 1 Descriptive Statistics and Reliability Estimates

Range M SD α

Life satisfaction Time 1 1–7 5.46 1.03 .83Life satisfaction Time 2 1–7 5.39 1.13 .88

Self-determined motivationIntrinsic motivation 1–7 5.91 1.18 .84Identified motivation 1–7 4.09 1.53 .79

Non-self-determined motivationExtrinsic motivation 1–7 2.77 1.41 .81Amotivation 1–7 1.46 0.97 .83

Evaluative concerns perfectionismSocially prescribed perfectionism (H-MPS) 1–7 3.04 1.15 .79Concern over mistakes (F-MPS) 1–5 2.24 0.81 .74Parental criticism (F-MPS) 1–5 2.10 0.92 .84Doubts about actions (F-MPS) 1–5 2.50 0.75 .63

Personal standards perfectionismSelf-oriented perfectionism (H-MPS) 1–7 5.23 1.03 .78Personal standards (F-MPS) 1–5 3.37 0.83 .80

Task-oriented copingMental imagery 1–5 2.91 0.93 .73Logical analysis 1–5 3.05 0.80 .61Relaxation 1–5 2.53 0.96 .82Thought control 1–5 3.20 0.86 .72Effort expenditure 1–5 3.99 0.77 .83

Distraction-oriented copingSeeking support 1–5 2.05 0.99 .84Mental distraction 1–5 1.82 0.83 .78Distancing 1–5 2.08 0.87 .76

Disengagement-oriented copingDisengagement/resignation 1–5 1.74 0.87 .83Venting of unpleasant emotions 1–5 2.57 1.17 .89

Goal attainmentMastery-oriented 1–7 4.95 1.07 .93Self-referenced 1–7 3.68 1.71 .86Normative-oriented 1–7 3.65 1.50 .86

indicators of life satisfaction were freely estimated. The first measurement model yielded an acceptable fit to the data (see Model 1 in Table 2). The second mea-surement model constrained the free factor loadings and uniquenesses of the life satisfaction latent variables to equality across time. These equality constraints did not yield a substantial increase in SBχ2 (see Model 2 in Table 2). This model was assumed to be the best fitting and most parsimonious model. Standardized factor loadings ranged from .66 to .83 for life satisfaction; from .76 to .77 for PSP; from .61 to .83 for ECP; from .78 to .83 for self-determined motivation; from .77 to .85 for non-self-determined motivation; from .43 to .76 for task-oriented coping; from .48 to .71 for distraction-oriented coping; from −.61 to .70 for disengagement-oriented coping;6 and from .60 to .95 for goal attainment. Error-free correlations between the latent variables are presented in Table 3. It should be reported that PSP did

368

Tab

le 2

Fi

t In

dic

es o

f th

e M

easu

rem

ent

and

Str

uct

ura

l Mo

del

s

χ2S

Bχ2

df

CFI

RC

FIN

NFI

RM

SE

A (9

0% C

I)∆S

Bχ2

∆df

∆CFI

Mea

sure

men

t mod

els

1. H

ypot

hetic

al75

4.88

701.

5250

9.9

17.9

23.9

03.0

51 (

.043

to .0

58)

2. I

nvar

ianc

ea76

4.21

705.

1751

8.9

17.9

25.9

05.0

51 (

.043

to .0

58)

M

odel

2 v

s. 1

6.15

9<

.000

Str

uctu

ral m

odel

s3.

Ful

l med

iatio

n82

6.68

763.

4754

8.9

06.9

14.8

98.0

53 (

.045

to .0

59)

4. P

artia

l med

iatio

n Ib

814.

0975

1.16

545

.910

.917

.901

.052

(.0

44 to

.059

)

Mod

el 4

vs.

313

.93*

*3

.004

5. P

artia

l med

iatio

n II

c81

3.70

751.

1554

3.9

09.9

16.9

00.0

52 (

.044

to .0

59)

M

odel

5 v

s. 4

0.29

2−.

001

6. P

artia

l med

iatio

n II

Id80

9.05

746.

7554

4.9

11.9

19.9

03.0

52 (

.044

to .0

58)

M

odel

6 v

s. 4

3.96

*1

.001

Not

e. A

ll χ2 a

nd S

Bχ2 w

ere

sign

ifica

nt a

t p <

.001

.a L

oadi

ngs

and

uniq

uene

ss o

f lif

e sa

tisfa

ctio

n w

ere

inva

rian

t acr

oss

mea

sure

men

t poi

nts.

b Thr

ee d

irec

t pat

hs f

rom

per

fect

ioni

sm to

cop

ing.

c Tw

o di

rect

pat

hs f

rom

cop

ing

to p

ostc

ompe

titio

n lif

e sa

tisfa

ctio

n.d D

irec

t pat

h fr

om E

CP

to p

ostc

ompe

titio

n lif

e sa

tisfa

ctio

n.

**p

< .0

1, *

p <

.05.


Tab

le 3

E

rro

r-Fr

ee C

orr

elat

ion

s B

etw

een

th

e L

aten

t Var

iab

les

in t

he

Mea

sure

men

t M

od

el

12

34

56

78

9

1. L

ife

satis

fact

ion

Tim

e 1

2. L

ife

satis

fact

ion

Tim

e 2

.86*

3. P

erso

nal s

tand

ards

per

fect

ioni

sm .0

8 –

.05

4. E

valu

ativ

e co

ncer

ns p

erfe

ctio

nism

–.3

7* –

.42*

.58*

5. S

elf-

dete

rmin

ed m

otiv

atio

n .1

6 .1

4 .4

3* .1

2

6. N

on-s

elf-

dete

rmin

ed m

otiv

atio

n –

.17*

–.1

6 .3

7* .4

9* .1

4

7. T

ask-

orie

nted

cop

ing

.22*

.24*

.30*

.17

.49*

.14

8. D

istr

actio

n-or

ient

ed c

opin

g –

.11

–.1

6 .1

6 .4

8* .1

4 .4

8* .5

9*

9. D

isen

gage

men

t-or

ient

ed c

opin

g –

.26*

–.3

0* .1

3 .4

3* –

.01

.53*

.01

.67*

10. G

oal a

ttain

men

t .2

0* .3

2* .1

0 –

.02

.13

–.1

0 .5

2* .0

2 –

.69*

Not

e. *

p <

.05.


not significantly correlate with Time 2 life satisfaction, on the one hand, and that motivational orientations were not significantly correlated with goal attainment, on the other hand.

Prerequisite Condition of Mediation

A first model examined the paths of perfectionism with coping dimensions. Task-oriented coping was predicted by PSP (β = .37, p < .01), whereas distraction- (β = .54, p < .01) and disengagement-oriented coping (β = .45, p < .01) were both predicted by ECP. As expected, a second model revealed the significant association between Time 2 life satisfaction and both task- (β = .22, p < .05) and disengagement-oriented coping (β = −.32, p < .01). Distraction-oriented coping was not significantly associated to Time 2 life satisfaction (β = −.01, p > .05). Finally, a third model indicated that ECP (β = −.16, p < .05) was significantly related to Time 2 life satisfaction. Altogether, these results met the prerequisite condition of Holmbeck (1997) for subsequent tests of the three mediation hypotheses. It should noted that the third model revealed that PSP was not significantly associated with Time 2 life satisfaction (β = −.13, p > .05), whereas an additional model indicated that both self-determined motivation (β = .14, p > .05) and non-self-determined motivation (β = −.11, p > .05) were not significantly related to goal attainment. Hence, these three exploratory paths were not included in the subsequent structural models.

Structural Models

Fully Mediated Model. The fully mediated model was specified as per Figure 1 but the residuals of the three coping dimensions were allowed to correlate with one another to take into consideration the expected interrelationships between the coping dimensions. The parameters were all significant (p < .05). The proposed model fit the data marginally, with the NNFI lower than .90 (see Model 3 in Table 2).

Direct Paths from Perfectionism to Coping. On the basis of our first set of medi-ating hypotheses, a partial mediation model added direct links between PSP and task-oriented coping and between ECP and both disengagement- and distraction-oriented coping. The fit of this model was acceptable and the newly added paths yielded a significant improvement in SBχ2 (see Model 4 in Table 2).

Results of the tests of direct and indirect effects are presented in Table 4. Of particular interest, the three indirect effects of perfectionism on coping were highly significant and they explained from 33% to 50% of the total effect. The direct effects of both PSP on task-oriented coping and ECP on distraction-oriented coping were significant, whereas the direct effect of ECP on disengagement-oriented coping was not significant although it approached significance (p < .07). In this case, a strict interpretation of statistical significance is not warranted given the relatively low power of direct and indirect effects in mediation analyses (Hoyle & Kenny, 1999). Including this direct effect of ECP on disengagement in the final model seemed more prudent than discarding this path. The presence of this non-negligible direct effect, albeit nonsignificant in pure statistical terms, provides substantial evidence against a fully mediated effect. Altogether, these results indicate that motivational styles partially mediated the relationships of perfectionism with coping. The three direct paths from perfectionism to coping were retained in subsequent models.

371

Tab

le 4

S

ign

ifica

nce

Tes

t o

f th

e U

nst

and

ard

ized

Dir

ect

and

Ind

irec

t E

ffec

ts

Tota

lD

irect

effe

ctIn

dire

ct e

ffect

BB

SE

ZB

SE

Z

Part

ial m

edia

tion

I: D

irec

t pat

h fr

om p

erfe

ctio

nism

to c

opin

g

Pers

onal

sta

ndar

ds p

erfe

ctio

nism

→ ta

sk-o

rien

ted

copi

ng0.

341

0.20

30.

077

2.65

**0.

138

0.04

43.

13**

Eva

luat

ive

conc

erns

per

fect

ioni

sm →

dis

trac

tion-

orie

nted

cop

ing

0.29

30.

195

0.07

02.

81**

0.09

70.

038

2.58

**

Eva

luat

ive

conc

erns

per

fect

ioni

sm →

dis

enga

gem

ent-

orie

nted

cop

ing

0.32

50.

163

0.08

81.

85t

0.16

10.

054

2.99

**

Part

ial m

edia

tion

II: D

irec

t pat

hs f

rom

cop

ing

to li

fe-s

atis

fact

ion

Task

-ori

ente

d →

Pos

tcom

petit

ion

life-

satis

fact

ion

0.06

8−

0.05

30.

085

−0.

620.

121

0.04

62.

61**

Dis

enga

gem

ent-

orie

nted

→ P

ostc

ompe

titio

n lif

e-sa

tisfa

ctio

n−

0.15

7−

0.05

20.

111

−0.

46−

0.10

50.

055

−1.

90t

Part

ial m

edia

tion

III:

Dir

ect p

ath

from

per

fect

ioni

sm to

life

-sat

isfa

ctio

n

Eva

luat

ive

conc

erns

per

fect

ioni

sm →

Pos

tcom

petit

ion

life-

satis

fact

ion

−0.

193

−0.

152

0.07

1−

2.15

*−

0.04

10.

017

−2.

38*

Not

e. *

*p <

.01,

*p

< .0

5, t p

< .0

7.


Direct Paths from Coping to Time 2 Life Satisfaction. On the basis of the second set of mediating hypotheses, a partial mediation model added the direct links of task- and disengagement-oriented coping with postcompetition life satisfaction. As expected, the newly added direct paths were nonsignificant and they did not yield a significant improvement in SBχ2 (see Model 5 in Table 2). These results suggest the presence of fully mediated effects. Consistent with this interpretation, the indirect effect of task-oriented coping was significant and it explained 70% of the total effect (see Table 4). The indirect effect of disengagement-oriented coping was not significant but was close to reaching significance (p < .07). In this case, a strict interpretation of statistical significance is not warranted as it would likely produce a type-II error given that goal attainment explained 67% of the relationship between disengagement-oriented coping and Time 2 life satisfaction. Altogether, these results indicate that goal attainment fully mediated the relationships of the two coping dimensions with changes in life satisfaction. The two nonsignificant direct paths were not retained in subsequent models.

Direct Path from ECP to Time 2 Life Satisfaction. Evaluative concerns perfec-tionism was the only dimension of perfectionism to significantly correlate with postcompetition life satisfaction (see Table 3), which is consistent with our review of 17 published studies. On the basis of the third set of mediating hypotheses, a partial mediation model was tested in which the direct path from ECP to Time 2 life satisfaction was added. The fit of this model was acceptable and the newly added path yielded a significant improvement in SBχ2 (see Model 6 in Table 2). The direct and indirect effects were both significant (see Table 4), with the indirect effect explaining 21% of the total effect. Thus, the sequence running from non-self-determination to disengagement-oriented coping to goal attainment partially mediated the relationship of ECP with changes in life satisfaction. This model was assumed to be the best fitting structural model (see Figure 2).

DiscussionResults of this study provided support for an integrative model in which the dimensions of perfectionism distinctly relate to motivational orientations and coping processes, which, in turn, relate to goal attainment and adjustment. This study replicated results from research outside of the sport domain by showing the distinct associations of athletes’ perfectionistic dimensions with both motivational orientations (Miquelon et al., 2005; Van Yperen, 2006) and coping processes (e.g., Dunkley et al., 2003). Moreover, this research lent credence for the three sets of mediation hypotheses included in the integrative model.

Perfectionism and Coping: The Mediating Role of Motivation

This study contributes to both the sport and the general psychology literature by being the first to establish that contextual motivation mediates the relationship of perfectionistic dispositions with situational coping. Of particular interest, self-deter-mined motivation toward sport was a partial mediator in the positive relationship between PSP and reported utilization of task-oriented coping during a competition. Likewise, non-self-determined motivation toward sport was a partial mediator in the

373

Fig

ure

2 —

Par

tially

med

iate

d m

odel

of t

he a

ssoc

iatio

ns b

etw

een

perf

ectio

nism

, mot

ivat

ion,

cop

ing,

goa

l atta

inm

ent,

and

life

satis

fact

ion.

The

fact

or

load

ings

and

uni

quen

esse

s of

life

sat

isfa

ctio

n w

ere

inva

rian

t acr

oss

times

. Alth

ough

not

illu

stra

ted,

the

dist

urba

nces

of

the

thre

e co

ping

dim

ensi

ons

corr

elat

ed f

reel

y. T

he d

istu

rban

ces

wer

e st

anda

rdiz

ed a

nd th

e va

rian

ce u

nacc

ount

ed f

or b

y th

e m

odel

is o

btai

ned

by s

quar

ing

this

val

ue. n.

s.p

> .1

0.

t p <

.07.

Oth

erw

ise,

all

para

met

ers

wer

e si

gnifi

cant

at p

< .0

5.


positive association between ECP and both distraction- and disengagement-oriented coping. These results indicate that motivational factors should be considered more systematically to understand how personality factors influence coping utilization during stressful situations such as sport competitions (Lazarus, 1999).

From a theoretical standpoint, the failure of motivational orientations to entirely account for the associations between perfectionism and coping (i.e., partial media-tion) was to be expected given that the relationships from personality to coping can theoretically be transmitted by both motivational and cognitive appraisal fac-tors. Yet, the conceptual linkage between motivation toward sport and situational coping utilization remains largely unexplored. Theoretically, it has been suggested that individuals with self-determined motivation may interpret the problems they encounter as challenges rather than as stressors (Blais, Sabourin, Boucher, & Vallerand, 1990), thus promoting the use of task-oriented forms of coping. Hence, future research should examine whether secondary cognitive appraisals mediate the relationship between motivation and coping.

This study provided evidence for the mediating role of motivation in the rela-tionship between perfectionism and coping. However, the processes at play in the association between perfectionism and motivational orientations remain largely unexplored both in sport and in other life domains. Self-determination theory posits that motivational orientations of athletes are fueled by the satisfaction of three basic psychological needs. Studies in the realm of sport (Amorose & Anderson-Butcher, 2007) and exercise behavior (Hagger, Chatzisarantis, & Harris, 2006) provide sup-port for this contention. From a theoretical standpoint, it would seem warranted to examine the relationship of dispositional perfectionism with the satisfaction of the need for autonomy, competence, and relatedness. Furthermore, future work should examine how both need satisfaction and motivation could combine to mediate the relationships between dispositional perfectionism and coping.

Coping and Adjustment: The Mediating Role of Goal Attainment

Athletes generally feel better when they successfully attain their goals (Amiot et al., 2004; Gaudreau & Blondin, 2002; Graham et al., 2002). In recent research, goal attainment has been proposed as a potential mediator of the relationship between coping and postcompetition affective states (e.g., Gaudreau & Blondin, 2004). Results of this study extended the support for this hypothesis with a measure of general life satisfaction. Although life satisfaction was quite stable across measure-ment points (β = .78), subjective evaluation of goal attainment after an important competition fully mediated the relationships of task- and disengagement-oriented coping with change in life satisfaction. It seems that coping not only plays an important role in goal attainment, but that goal attainment also explains why coping dimensions relate differently to athletes’ adjustment after a sport competition. This finding is consistent with a bottom-up conception of subjective well-being as athletes are becoming more satisfied while successfully progressing toward goals closely related to their identity (Wiese, 2007). Although goal attainment can be a satisfying experience in and of itself, its payoffs would be greater for the athletes who are pursuing sport-related goals closely aligned with their values and interests (Sheldon & Kasser, 1998). Consequently, future work should examine for whom


and under which features of the sport environment goal attainment could bolster subjective well-being of athletes.

Perfectionism and Adjustment: The Mediating Role of Coping

Along with relating to both goal attainment and adjustment, coping has been theoretically conceived as a mediating mechanism through which perfectionism influences psychological adjustment (e.g., Flett et al., 1991). Results of this study contribute to a growing body of literature demonstrating the intervening role of disengagement-oriented coping in the relationship between ECP and maladjust-ment (e.g., Dunkley & Blankstein, 2000). In this sample, ECP predicted a decrease in athletes’ life satisfaction after a sport competition because non-self-determined motivation toward sport promoted reliance on disengagement-oriented coping, which, in turn, prevented goal attainment. As with earlier findings (Dunkley et al., 2000; Dunkley et al., 2003), the relationship between ECP and changes in life satisfaction remained significant over and above the mediating sequence presented in the model (perfectionism → contextual motivation → situational coping → goal attainment). Nonetheless, the indirect effect of this sequential pathway accounted for 22% of the relationship between ECP and change in life satisfaction.

Besides yielding support for a partially mediated effect, these results illustrate both the direct pervasive effect of ECP and the sequential pathways by which it leads to lower levels of life satisfaction. Intervening on motivational orientations toward sport and coping utilization could reduce some but not all of the debilita-tive effects of ECP. Of particular concern, planning interventions designed to help people determine how, where, and when to work on their personal goals have been found to be less effective with socially prescribed perfectionists, presumably because they are delivered in an unsupportive or directive manner. However, these interventions become highly successful when delivered in a way that fosters feelings of autonomy and competence (Koestner et al., 2006). The modification of athletes’ motivational orientation, through interventions delivered by sport psychologists in a way that promotes a sense of competence and personal ownership, could perhaps circumvent the tendency of athletes with an ECP predisposition to experience lower levels of subjective well-being.

The Case of Personal Standards Perfectionism: Healthy, Unhealthy, or Neutral?

In recent years, PSP has been conceived as a healthy (Stoeber & Otto, 2006), an unhealthy (Flett & Hewitt, 2005), or a neutral dimension of perfectionism (Bieling et al., 2004). Results of this study illustrate the complexity of categorizing PSP as being either healthy or unhealthy. Although PSP plays a more neutral role in predicting life satisfaction, it nonetheless relates to self-determined motivation and task-oriented coping of athletes, which, in turn, promote goal attainment and life satisfaction. Altogether, these results indicate that focusing exclusively on subjec-tive well-being would render an incomplete and somewhat inaccurate portrait of PSP, thus highlighting the value of examining integrative models such as the one tested in this study.


Why is it that PSP does not significantly correlate concurrently and pro-spectively with life satisfaction? One possible explanation is that PSP relates to antagonist mediation processes that are respectively promoting and thwarting one’s subjective well-being. It has been argued that the positive effect of a mediator, when combined with the negative effect of a second mediator, can produce a nonsignificant direct effect between the predictor and the outcome variable (Frazier, Tix, & Barron, 2004). Pursuing this further, the measurement model revealed a positive bivariate association of PSP with both self-determined (r = .43) and non-self-determined motivations toward sport (r = .37),7 which are known to diverge in their associations to emotional adjustment (Vallerand & Rousseau, 2001). Such a result resonates well with recent claims that PSP entails a mixture of antagonist approach and avoidance tendencies, responsible for promoting the need for achievement and the need to prevent failures, respectively (Flett & Hewitt, 2006). In line with this reasoning, a recent study has shown the positive linkage of self-oriented perfectionism with both approach and avoidance performance goals (Van Yperen, 2006).

Limitations and Future DirectionsDespite the strengths of this prospective study, the correlational design precludes definitive inference regarding the causal associations between the variables. In future studies, some of the variables could be manipulated through interventions fostering self-determined motivation and the utilization of task-oriented coping. Future studies could also use larger samples with diversified subgroups of athletes to assess the invariance of the proposed measurement and structural models across gender and types of sport. This issue is particularly important given the reported heterogeneity in the variance–covariance matrix between athletes from individual and team sports. The design could also be improved by including a third measure-ment point to rule out alternative interpretations about the directionality of the reported effects (Frazier et al., 2004). Incidentally, 11% of the participants did not complete the questionnaires at Time 2 and they reported higher scores on perfec-tionism and lower scores on life satisfaction at Time 1 compared with athletes who have participated at both waves. Hence, an effort would have to be made to limit the loss of generalizability caused by nonrandom attrition in future research with multiple waves of data collection (Ribisl et al., 1996).

This study was also limited by the exclusive reliance on self-report measures. A recent study demonstrated the invariance of the relationship between perfec-tionism and depression across self-report and informant ratings (Flett, Besser, & Hewitt, 2005). Nonetheless, future research should try to obtain both self-report and informant ratings of dispositional perfectionism from coaches, teammates, and friends. A recent study incorporated a non-self-reported measure of implicit self-determination in the prediction of academic performance (Burton, Lydon, D’Alessandro, & Koestner, 2006). Similarly, sport-related motivational orientations could be measured with a computerized lexical decision task in which athletes associate as rapidly as possible the word sport with cued words representing dif-ferent forms of motivation: fun, importance, guilt, rewards, and boredom. In future research, the inclusion of such measures of implicit motivation would be useful to limit the social desirability and the shared method variance pitfalls of relying exclusively on self-reported data.


According to the tenets of the hierarchical model of motivation (Vallerand, 1997), motivation and outcomes should be more strongly correlated when measured at the same level of analysis. In that case, it can be hypothesized that situational coping and goal attainment would relate even more strongly to motivation for a spe-cific sport competition (i.e., situational motivation) than to the contextual measure of sport motivation used in this study. Similarly, this model would posit that goal attainment would influence sport satisfaction, which, in turn through a bottom-up effect, would influence general life satisfaction. Future studies should incorporate general, contextual, and situational measures of motivation and well-being to exam-ine the postulates of the hierarchical model in the sport domain. Using a similar line of reasoning, it would be useful to compare the predictive power of dispositional versus contextual measures of perfectionism toward sport (e.g., Dunn et al., 2006). In accordance with the hierarchical model, it could be presumed that contextual sport motivation would relate even more strongly to sport-specific perfectionism than to the dispositional measure of perfectionism used in this study.

As a final concern, it would be preferable to measure all types of motivation from the self-determination continuum, including integrated and introjected motiva-tion. The decision to exclude introjected motivation was motivated by a requirement from the coaches to keep the questionnaire as short as possible given the two waves of data collection during the week of an important sport competition. Previously unpublished data with college students (n > 1000) provided empirical foundation for this decision. The partial correlations between PSP (controlling for ECP) and both introjected and extrinsic motivation were identical (rs = .05). Similarly, the partial correlations between ECP (controlling for PSP) and both introjected (r = .16) and extrinsic motivation (r = .14) were almost identical. Therefore, based on the above data, a confident decision was made that the relationship between perfectionism and non-self-determined motivation would be highly similar with or without the inclusion of the introjected motivation subscale. Regrettably, this decision assumed that the association of perfectionism and introjected motivation would be comparable across the sport and academic domains, a matter which deserves future empirical examination.

ConclusionOverall, this research suggests that an integrative perspective can be helpful in testing and presenting the complex associations between perfectionism, motiva-tion, coping, and adjustment of athletes. This research supplements the literature on dispositional perfectionism by bringing together a network of variables that have been studied independently and almost exclusively with samples of college students. As a closing remark, PSP seems to be a relatively neutral factor when it comes to predicting athletes’ goal attainment and life satisfaction. Yet, where does it leave sport psychologists in terms of intervening with athletes scoring high on PSP? Although this pivotal question remains unanswered, this study does inform sport researchers that PSP and ECP distinctively relate to motivational orientations and situational coping strategies, which, in turn, are associated with goal attain-ment and general life satisfaction. Furthermore, the contradictory associations of PSP with antagonistic motivational processes (i.e., self-determined and non-self-determined motivation) indicate the need of examining whether characteristics of


both the athletes and the sport environment could moderate the relationship of PSP with athletes’ subjective well-being and performance.

Notes

1. The 17 references can be obtained by contacting the first author.

2. A series of confirmatory factor analyses provided evidence for the acceptable fit of the instru-ments used in this study. These results are available upon request.

3. A fourth item was added, given that doubt about actions had a low internal consistency in the validation study of Cox et al. (2002; α = .63). For this reason, 23 items were used rather than the 22 items proposed by Cox et al. (2002).

4. Results of confirmatory factor analyses suggest that organization should not be conceived as an indicator of PSP (for a review, see Stoeber & Otto, 2006). In this study, a measurement model indicated that the standardized factor loading leading from PSP to organization was lower than .30, and modification indices suggested that organization could cross-load on ECP. Hence, this subscale of perfectionism was not used in the reported analyses.

5. Researchers are routinely using a self-determination index, and therefore we have tried this approach. However, the bivariate relation between PSP and the self-determination index was nonsignificant in the measurement model. However, the same relationship became significant in the structural model (which is controlling for the effect of ECP). This result outlined the presence of a suppression effect, suggesting that the self-determination index did not account for the distinct relationship between dimensions of perfectionism and different types of motivation.

6. In this sample, seeking support loaded significantly on distraction-oriented coping (stan-dardized λ = .48) and effort expenditure cross-loaded on both disengagement-oriented coping (standardized λ = −.61) and task-oriented coping (standardized λ = .43).

7. An alternative structural model added a path from PSP to non-self-determined motivation. The fit of this model was not superior to the best fitting model. The added path was not significant, B = .10, Z = 0.98. Nonetheless, the bivariate correlation between PSP and non-self-determination warrants further empirical exploration.

Acknowledgments

This research was supported by regular research grants from the Social Sciences and Humanities Research Council of Canada and from the University of Ottawa Scholars’ Fund awarded to Patrick Gaudreau.

ReferencesAmiot, C.E., Gaudreau, P., & Blanchard, C.M. (2004). Self-determination, coping, and goal

attainment in sport. Journal of Sport and Exercise Psychology, 26, 386–411.Amorose, A.J., & Anderson-Butcher, D. (2007). Autonomy-supportive coaching and self-

determined motivation in high school and college athletes: A test of self-determination theory. Psychology of Sport and Exercise, 8, 654–670.

Ayers, T.S., Sandler, I.N., West, S.G., & Roosa, M.W. (1996). A dispositional and situ-ational assessment of children’s coping: Testing alternative models of coping. Journal of Personality, 64, 923–958.

Bieling, P.J., Israeli, A., & Antony, M.M. (2004). Is perfectionism good, bad, or both? Examining models of the perfectionism construct. Personality and Individual Differ-ences, 36, 1373–1385.


Bieling, P.J., Israeli, A., Smith, J., & Antony, M.M. (2003). Making the grade: The behav-ioral consequences of perfectionism in the classroom. Personality and Individual Differences, 35, 163–178.

Blais, M.R., Sabourin, S., Boucher, C., & Vallerand, R.J. (1990). Toward a motivational model of couple happiness. Journal of Personality and Social Psychology, 59, 1021–1031.

Blais, M.R., Vallerand, R.J., Pelletier, L.G., & Brière, N.M. (1989). L’échelle de satisfac-tion de vie: Validation canadienne-française de “Satisfaction With Life Scale” [French Canadian validation of the Satisfaction With Life Scale]. Canadian Journal of Behav-ioural Science, 21, 210–223.

Bollen, K.A. (1989). Structural equations with latent variables. New York: Wiley.Brière, N.M., Vallerand, R.J., Blais, M.R., & Pelletier, L.G. (1995). Développement et

validation d’une mesure de motivation intrinsèque, extrinsèque et d’amotivation en contexte sportif: L’Échelle de Motivation dans les Sports (ÉMS). [On the development and validation of the French form of the Sport Motivation Scale]. International Journal of Sport Psychology, 26, 465–489.

Burton, K.D., Lydon, J.E., D’Alessandro, D.U., & Koestner, R. (2006). The differential effects of intrinsic and identified motivation on well-being and performance: Prospec-tive, experimental, and implicit approaches to self-determination theory. Journal of Personality and Social Psychology, 91, 750–762.

Carver, C.S., & Scheier, M.F. (1998). On the self-regulation of behavior. New York: Cam-bridge University Press.

Cox, B.J., Enns, M.W., & Clara, I.P. (2002). The multidimensional structure of perfection-ism in clinically distressed and college student samples. Psychological Assessment, 14, 365–373.

Deci, E.L., & Ryan, R.M. (2002). An overview of self-determination theory: An organismic-dialectical perspective. In E.L. Deci & R.M. Ryan (Eds.), Handbook of self-determina-tion research (pp. 3–36). Rochester, NY: University of Rochester Press.

Dunkley, D.M., & Blankstein, K.R. (2000). Self-critical perfectionism, coping, hassles, and current distress: A structural equation modeling approach. Cognitive Therapy and Research, 24, 713–730.

Dunkley, D.M., Blankstein, K.R., Halsall, J., Williams, M., & Winkworth, G. (2000). The relation between perfectionism and distress: Hassles, coping, and perceived social sup-port as mediators and moderators. Journal of Counseling Psychology, 47, 437–453.

Dunkley, D. M., Blankstein, K. R., Masheb, R. M., & Grilo, C. M. (2006). Personal standards and evaluative concerns dimensions of “clinical” perfectionism: A reply to Shafran et al. (2002, 2003) and Hewitt et al. (2003). Behavior Research and Therapy, 44, 63-84.

Dunkley, D.M., Zuroff, D.C., & Blankstein, K.R. (2003). Self-critical perfectionism and daily affect: Dispositional and situational influences on stress and coping. Journal of Personality and Social Psychology, 84, 234–252.

Dunn, J.G., Dunn, J.C., Gotwals, J.K., Vallance, J.K., Craft, J.M., & Syrotuik, D.G. (2006). Establishing construct validity evidence for the Sport Multidimensional Perfectionism Scale. Psychology of Sport and Exercise, 7, 57–79.

Flett, G.L., Besser, A., & Hewitt, P.L. (2005). Perfectionism, ego defense styles, and depres-sion: A comparison of self-reports versus informant ratings. Journal of Personality, 73, 1355–1396.

Flett, G.L., & Hewitt, P.L. (2005). The perils of perfectionism in sports and exercise. Current Directions in Psychological Science, 14, 14–18.

Flett, G.L., & Hewitt, P.L. (2006). Positive versus negative perfectionism in psychopathol-ogy: A comment on Slade and Owens’s dual process model. Behavior Modification, 30, 472–495.

Flett, G.L., Hewitt, P.L., Blankstein, K.R., & O’Brien, S. (1991). Perfectionism and learned resourcefulness in depression and self-esteem. Personality and Individual Differences, 12, 61–68.


Frazier, P.A., Tix, A.P., & Barron, K.E. (2004). Testing moderator and mediator effects in counseling psychology. Journal of Counseling Psychology, 51, 115–134.

Frost, R.O., Heimberg, R.G., Holt, C.S., Mattia, J.I., & Neubauer, A.L. (1993). A com-parison of two measures of perfectionism. Personality and Individual Differences, 14, 119–126.

Frost, R.O., & Henderson, K.J. (1991). Perfectionism and reactions to athletic competition. Journal of Sport and Exercise Psychology, 13, 232–235.

Frost, R.O., Marten, P., Lahart, C., & Rosenblate, R. (1990). The dimensions of perfection-ism. Cognitive Therapy and Research, 14, 449–468.

Gaudreau, P., Amiot, C.E., Blondin, J-P., & Blanchard, C.M. (2002, November). A hierar-chical model of goal attainment in sport competition. Paper presented at the annual meeting of the Canadian Society for Psychomotor Learning and Sport Psychology, Vancouver, BC.

Gaudreau, P., & Blondin, J-P. (2002). Development of a questionnaire for the assessment of coping strategies employed by athletes in competitive sport settings. Psychology of Sport and Exercise, 3, 1–34.

Gaudreau, P., & Blondin, J-P. (2004). The differential associations of dispositional optimism and pessimism with coping, goal attainment, and emotional adjustment during a sport competition. International Journal of Stress Management, 11, 245–269.

Gaudreau, P., & Blondin, J.-P. (2007). Testing the between-group and longitudinal invari-ance of sport-related coping strategies using structural equation modeling. Manuscript submitted for publication.

Gotwals, J.K., Dunn, J.G.H., & Wayment, H.A. (2003). An examination of perfectionism and self-esteem in intercollegiate athletes. Journal of Sport Behavior, 26, 17–38.

Gould, D., Udry, E., Tuffey, S., & Loehr, J. (1996). Burnout in competitive junior tennis players: I. A quantitative psychological assessment. The Sport Psychologist, 10, 322–340.

Graham, T.R., Kowalski, K.C., & Crocker, P.R. (2002). The contributions of goal charac-teristics and causal attributions to emotional experience in youth sport participants. Psychology of Sport and Exercise, 3, 273–291.

Hagger, M.S., Chatzisarantis, N.L., & Harris, J. (2006). From psychological need satisfaction to intentional behavior: Testing a motivational sequence in two behavioral contexts. Personality and Social Psychology Bulletin, 32, 131–148.

Hall, H.K., Kerr, A.W., & Matthews, J. (1998). Precompetitive anxiety: The contribution of achievement goals and perfectionism. Journal of Sport and Exercise Psychology, 20, 194–217.

Hewitt, P.L., & Flett, G.L. (1991). Perfectionism in the self and social contexts: Conceptu-alization, assessment, and association with psychopathology. Journal of Personality and Social Psychology, 60, 456–470.

Hewitt, P.L., Flett, G.L., & Mikail, S.F. (1995). Perfectionism and relationship adjustment in pain patients and their spouses. Journal of Family Psychology, 9, 335–347.

Hoar, S.D., Kowalski, K.C., Gaudreau, P., & Crocker, P.R. (2006). A review of coping in sport. In S. Hanton & S.D. Mellalieu (Eds.), Literature reviews in sport psychology (pp. 47–90). New York: Nova Science.

Holmbeck, G. (1997). Toward terminological, conceptual, and statistical clarity in the study of mediators and moderators: Examples from the child-clinical and pediatric psychology literatures. Journal of Consulting and Clinical Psychology, 65, 599–610.

Hoyle, R.H., & Kenny, D.A. (1999). Sample size, reliability, and tests of statistical media-tion. In R.H. Hoyle (Ed.), Statistical strategies for small sample research. Thousand Oaks, CA: Sage Publications.

Hu, L-T., & Bentler, P.M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Model-ing, 6, 1–55.


Kenny, D.A. (1979). Correlations and causality. New York: Wiley.Knee, C.R., Patrick, H., Vietor, N.A., Nanayakkara, A., & Neighbors, C. (2002). Self-

determination as growth motivation in romantic relationships. Personality and Social Psychology Bulletin, 28, 609–619.

Knee, C.R., & Zuckerman, M. (1998). A nondefensive personality: Autonomy and control as moderators of defensive coping and self-handicapping. Journal of Research in Personality, 32, 115–130.

Koestner, R., Horberg, E.J., Gaudreau, P., Powers, T.A., Di Dio, L., Bryan, C., et al. (2006). Bolstering implementation plans for long haul: The benefits of simultaneously boost-ing self-concordance and self-efficacy. Personality and Social Psychology Bulletin, 32, 1547–1558.

Koestner, R., Lekes, N., Powers, T.A., & Chicoine, E. (2002). Attaining personal goals: Self-concordance plus implementation intentions equals success. Journal of Personality and Social Psychology, 83, 231–244.

Labrecque, J., Stephenson, R., Boivin, I., & Marchand, A. (1999). Validation de l’échelle multidimensionnelle du perfectionnisme auprès de la population francophone du Québec [Validation of the multidimensional perfectionism scale with the French population from Quebec]. Revue Francophone de Clinique Comportementale et Cognitive, 3, 1–14.

Lazarus, R.S. (1999). Stress and emotion: A new synthesis. New York: Springer.Lyubomirsky, S., Sheldon, K.M., & Schkade, D. (2005). Pursuing happiness: The architecture

of sustainable change. Review of General Psychology, 9, 111–131.Mackinnon, D.P., Lockwood, C.M., Hoffman, J.M., West, S.G., & Sheets, V. (2002). A

comparison of methods to test mediation and other intervening effects. Psychological Methods, 7, 83–104.

Martin, T.R., Flett, G.L., Hewitt, P.L., Krames, L., & Szanto, G. (1996). Personality cor-relates of depression and health symptoms: A test of a self-regulation model. Journal of Research in Personality, 30, 264–277.

Meyers, L.S., Gamst, & Guarino, A. J. (2005). Applied multivariate research: Design and interpretation. Thousand Oaks, CA: Sage Publications.

Miller, P.S., & Kerr, G.A. (2003). The role experimentation of intercollegiate student athletes. The Sport Psychologist, 17, 196–219.

Miquelon, P., Vallerand, R.J., Grouzet, F.M.E., & Cardinal, G. (2005). Perfectionism, aca-demic motivation, and psychological adjustment: An integrative model. Personality and Social Psychology Bulletin, 31, 913–924.

Nes, L.S., & Segerstrom, S.C. (2006). Dispositional optimism and coping: A meta-analytic review. Personality and Social Psychology Review, 10, 235–251.

Pavot, W., & Diener, E. (1993). Review of the Satisfaction With Life Scale. Psychological Assessment, 5, 164–172.

Powers, T.A., Koestner, R., & Topciu, R.A. (2005). Implementation intentions, perfectionism and goal progress: Perhaps the road to hell is paved with good intentions. Personality and Social Psychology Bulletin, 31, 902–912.

Ribisl, K.M., Walton, M.A., Mowbray, C.T., Luke, D.A., Davidson, W.S., & Bootsmiller, B.J. (1996). Minimizing participant attrition in panel studies through the use of effec-tive retention and tracking strategies: Review and recommendations. Evaluation and Program Planning, 19, 1–25.

Satorra, A., & Bentler, P.M. (2001). A scaled difference chi-square test statistic for moment structure analysis. Psychometrika, 66, 507–514.

Schwartz, B., Ward, A., Monterosso, J., Lyubomirsky, S., White, K., & Lehman, D.R. (2002). Maximizing versus satisficing: Happiness is a matter of choice. Journal of Personality and Social Psychology, 83, 1178–1197.

Sheldon, K.M., & Kasser, T. (1998). Pursuing personal goals: Skills enable progress, but not all progress is beneficial. Personality and Social Psychology Bulletin, 24, 1319–1331.


Shogren, K.A., Lopez, S.J., Wehmeyer, M.L., Little, T.D., & Pressgrove, C.L. (2006). The role of positive psychology constructs in predicting life satisfaction in adolescents with and without cognitive disabilities: An exploratory study. Journal of Positive Psychol-ogy, 1, 37–52.

Skinner, E.A., Edge, K., Altman, J., & Sherwood, H. (2003). Searching for the structure of coping: A review and critique of category systems for classifying ways of coping. Psychological Bulletin, 129, 216–269.

Slade, P.D., & Owens, G.R. (1998). A dual process model of perfectionism based on rein-forcement theory. Behavior Modification, 3, 372–390.

Smith, G.T., McCarthy, D.M., & Anderson, K.G. (2000). On the sins of short-form develop-ment. Psychological Assessment, 12, 102–111.

Stevens, J. (1996). Applied multivariate statistics for the social sciences (3rd ed.). Mahwah, NJ: Lawrence Erlbaum.

Stoeber, J., & Otto, K. (2006). Positive conceptions of perfectionism: Approaches, evidence, challenges. Personality and Social Psychology Review, 10, 295–319.

Vallance, J.K., Dunn, J.G.H., & Dunn, J.L. (2006). Perfectionism, anger, and situation criticality in competitive youth ice hockey. Journal of Sport and Exercise Psychology, 28, 383–406.

Vallerand, R.J. (1997). Toward a hierarchical model of intrinsic and extrinsic motivation. In M. Zana (Ed.), Advances in experimental social psychology (Vol. 29, pp. 271–360). New York: Academic Press.

Vallerand, R.J., & Rousseau, F.L. (2001). Intrinsic and extrinsic motivation in sport and exercise: A review using the hierarchical model. In R.N. Singer, H.A. Hausenblas, & C.M. Janelle (Eds.), Handbook of sport psychology (2nd ed., pp. 389–416). New York: Wiley.

Van Yperen, N.W. (2006). A novel approach to assessing achievement goals in the context of the 2 x 2 framework: Identifying distinct profiles of individuals with different dominant achievement goals. Personality and Social Psychology Bulletin, 32, 1432–1445.

Wiese, B.S. (2007). Successful pursuit of personal goals and subjective well-being. In B.R. Little, K. Salmela-Aro, & S.D. Phillips (Eds.), Personal project pursuit: Goals, action, and human flourishing (pp. 301–328). Mahwah, NJ: Lawrence Erlbaum.

Manuscript submitted: July 11, 2006Revision accepted: February 3, 2008

athletes’ broad dimensions of dispositional perfectionism: … · 2017-10-11 · coping in the...

Documents