UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Longitudinal investigation of the role of coping on the relationship between dimensions of perfectionism… Riendeau, Coralie 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2014_november_riendeau_coralie.pdf [ 9.03MB ]
Metadata
JSON: 24-1.0166994.json
JSON-LD: 24-1.0166994-ld.json
RDF/XML (Pretty): 24-1.0166994-rdf.xml
RDF/JSON: 24-1.0166994-rdf.json
Turtle: 24-1.0166994-turtle.txt
N-Triples: 24-1.0166994-rdf-ntriples.txt
Original Record: 24-1.0166994-source.json
Full Text
24-1.0166994-fulltext.txt
Citation
24-1.0166994.ris

Full Text

      LONGITUDINAL INVESTIGATION OF THE ROLE OF COPING ON THE RELATIONSHIP BETWEEN DIMENSIONS OF PERFECTIONISM AND ATHLETIC BURNOUT IN SPORT  by CORALIE RIENDEAU  B.Sc., McGill University, 2012   A THESIS SUBMITTED IN PARTIAL FULFILMENT OF  THE REQUIREMENTS FOR THE DEGREE OF   MASTER OF ARTS  in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Kinesiology)  The University of British Columbia (Vancouver)  September 2014  © Coralie Riendeau, 2014    ii Abstract   Perfectionism in athletes is a personality disposition that influences cognitive processes and behaviour (Hall, Hill, & Appleton, 2012). Perfectionism involves two dimensions: personal standards perfectionism (PSP) and evaluative concerns perfectionism (ECP). These dimensions are typically associated with positive and negative outcomes, respectively (Gotwals, Stoeber, Dunn, & Stoll, 2012). Perfectionism has shown important relationships with athlete burnout (Hill, 2013), a psychological symptom that involves emotional and physical exhaustion, perceived reduced accomplishment, and sport devaluation (Raedeke, 1997). One mechanism that is thought to influence the perfectionism-burnout relationship is coping. Cross-sectional research has found that task-oriented coping (TOC) is typically positively associated with PSP and negatively associated with burnout whereas disengagement oriented coping (DOC) is associated negatively with PSP and positively with ECP and burnout (Hill, Hall, & Appleton, 2010a). The purpose of this study was to investigate the mediation effect of coping strategies on the relationship between dimensions of perfectionism and burnout over the course of an athletic season. University level varsity athletes (nfemale = 90; nmale = 35) participated in a longitudinal study involving four time points, each approximately 4-5 weeks apart. The findings revealed that PSP was a negative predictor of burnout whereas ECP was a positive predictor of athletic burnout. Mediation analyses at the within- and between-individual level supported the mediation effect of DOC on the relationship between ECP and burnout and the mediation effect of TOC on the relationship between PSP and burnout. Secondary longitudinal analyses revealed linear decrease in burnout and TOC over a four month period. Quadratic growth models accounted for the change in PSP and DOC during the athletic season. Finally, test of the 2 x 2   iii model of dispositional perfectionism (Gaudreau & Thompson, 2010) indicated that pure PSP was associated with lower burnout than non-perfectionism, pure ECP was linked with the highest burnout level, and mixed perfectionism was associated with higher burnout than pure PSP. The results revealed consistent relationships between perfectionism, coping, and burnout over time. Overall the study highlighted the role of coping in the relationship between dimensions of perfectionism and athletic burnout.       iv Preface This thesis was based on data collected as part of the project Perfectionism and coping. Dr. Peter Crocker was the principal investigator on this project. I was involved in participant recruitment, amendment to the ethics application, data collection and analyses. Ethical approval was granted by the University of British Columbia Behavioural Research Ethics Board (H12-02521).    v Table of contents Abstract ...................................................................................................................................... ii Preface ...................................................................................................................................... iv Table of contents ....................................................................................................................... v List of tables ............................................................................................................................. ix List of figures ............................................................................................................................. x Acknowledgements .................................................................................................................. xi Chapter 1: Introduction and literature review ............................................................................ 1 1.1 Introduction ..................................................................................................................... 1 1.2 Literature review .............................................................................................................. 4 1.2.1 Perfectionism ............................................................................................................ 4  1.2.1.1 Models of perfectionism .................................................................................... 6 1.2.1.2 Perfectionism in sport ...................................................................................... 11  1.2.1.3 Outcomes associated with perfectionism in sport ............................................ 12  1.2.1.4 Stability of perfectionism over time ................................................................ 15 1.2.2 Burnout ................................................................................................................... 15  1.2.2.1 Definition ......................................................................................................... 15 1.2.2.2 Burnout in sport ............................................................................................... 16  1.2.2.3 Burnout and perfectionism ............................................................................... 18 1.2.3 Mediators of the relationship between perfectionism and burnout ........................ 20  1.2.3.1 Coping .............................................................................................................. 21  1.2.3.2 Coping and the relationship between perfectionism and burnout .................... 23   vi 1.2.4 Longitudinal design to examine mediation ............................................................ 24 1.3 Purpose and research questions ..................................................................................... 25 1.3.1 Relationships between perfectionism, burnout, and coping ................................... 25 1.3.2 Change in burnout, perfectionism, and coping over time ....................................... 27 1.3.3 Test of the 2 x 2 model of dispositional perfectionism .......................................... 28 Chapter 2: Methods ................................................................................................................. 29 2.1 Participants .................................................................................................................... 29 2.2 Procedures ..................................................................................................................... 29 2.3 Measures ........................................................................................................................ 31 2.4 Data analyses ................................................................................................................. 35 2.4.1 Data screening  ....................................................................................................... 35 2.4.2 Descriptive statistics  .............................................................................................. 36 2.4.3 Test of the 2 x 2 model of dispositional perfectionism  ......................................... 36 2.4.4 Longitudinal analyses  ............................................................................................ 38 2.4.5 Multilevel modeling and test of the mediation model  ........................................... 39 Chapter 3: Results .................................................................................................................... 45 3.1 Data screening  .............................................................................................................. 45 3.2 Descriptive statistics  ..................................................................................................... 46 3.3 Test of the 2 x 2 model of dispositional perfectionism  ................................................ 52 3.4 Longitudinal analyses  ................................................................................................... 56 3.4.1 Linear growth models  ............................................................................................ 57 3.4.2 Quadratic growth models  ....................................................................................... 60 3.5 Multilevel modeling and test of the mediation models  ................................................ 61   vii 3.5.1 Multilevel modeling at the within-individual level  ............................................... 61 3.5.2 Mediation at the within-individual level  ............................................................... 64 3.5.3 Multilevel modeling at the between-individual level  ............................................ 67 3.5.4 Mediation at the between-individual level  ............................................................ 68 Chapter 4: Discussion .............................................................................................................. 70 4.1 Discussion  ..................................................................................................................... 70 4.1.1 The role of coping  .................................................................................................. 71 4.1.2 Progression of study variables over time  ............................................................... 73  4.1.2.1 Burnout  ........................................................................................................... 73  4.1.2.1 Perfectionism  .................................................................................................. 73  4.1.2.1 Coping  ............................................................................................................. 75 4.1.3 Burnout, perfectionism, and the 2 x 2 model of perfectionism  ............................. 76 4.2 Strengths and limitations  .............................................................................................. 78 4.3 Future directions ............................................................................................................ 80 4.4 Practical implications .................................................................................................... 82 References ............................................................................................................................... 85 Appendices .............................................................................................................................. 99 Appendix A: Coach contact letter ........................................................................................ 99 Appendix B: Athlete information letter ............................................................................. 100 Appendix C: Athlete consent form .................................................................................... 101 Appendix D: Time 1 questionnaire ................................................................................... 103 Appendix E: Time 2, 3, and 4 questionnaire ..................................................................... 114 Appendix F: Pre-notice email ............................................................................................ 125   viii Appendix G: Questionnaire email ..................................................................................... 126 Appendix H: Reminder email ............................................................................................ 127 Appendix I: Thank you email ............................................................................................ 128 Appendix J: Results of the hierarchical regression for perfectionism predicting athletic burnout ............................................................................................................................... 129 Appendix K: Conditional growth models predicting athletic burnout .............................. 130     ix List of tables Table 1: Response patterns to the four phases of the study ..................................................... 45 Table 2: Summary of descriptive statistics for study variables ............................................... 47 Table 3: Summary of bivariate correlations for scores of study variables at each assessment  ................................................................................................................................................. 50 Table 4: Results of the moderated regression analyses predicting athletic burnout ................ 53 Table 5: Hierarchical regression models for athletic burnout with multiple imputations approach ................................................................................................................................... 55 Table 6: Intraclass correlations for study variables ................................................................. 57 Table 7: Unconditional growth models: fixed effects ............................................................. 59 Table 8: Unconditional growth models: random effects ......................................................... 59 Table 9: Quadratic growth models: fixed effects .................................................................... 60 Table 10: Quadratic growth models: random effects .............................................................. 61 Table 11: Estimates of fixed effects of the level-1 predictors ................................................. 63 Table 12: Random effects of the model including PSP, ECP, TOC, and DOC as level-1 predictor of burnout ................................................................................................................. 64 Table 13: Estimates of the fixed effects of the level-2 predictors ........................................... 68	       x List of figures Figure 1: The tripartite model of perfectionism ........................................................................ 7 Figure 2: The 2 x 2 model of perfectionism ............................................................................ 10 Figure 3: Visual representation of expected relationships ....................................................... 27 Figure 4: Timeline for data collection ..................................................................................... 31 Figure 5: Equations to obtain values of athletic burnout  ........................................................ 38 Figure 6: Linear and quadratic growth equations  ................................................................... 39 Figure 7: Three step approach to test the within-individual multilevel model ........................ 41 Figure 8: Three step approach to test the between-individual multilevel model ..................... 42 Figure 9: Predicted values of athletic burnout using the main effect of PSP and ECP ........... 56 Figure 10: Within- and between-individual effects  ................................................................ 66  	         xi Acknowledgements  First I want to acknowledge all the athletes who participated in this study, without them we could not answer many scientific questions. I also want to thank the coaches who believed in this project and facilitated contact with their athletes.   Thank you to my supervisor Dr. Peter Crocker who has believed in me and provided opportunities for academic growth. My lab family has supported me throughout this whole process and for that I am extremely grateful: Carolyn McEwen, Erica Bennett, Dani Wilson, Katie Gunnell, Amber Mosewich, and Katherine Tamminen. A special thank you to Louisa Scarlett and Sarah Kiengersky, you were with me every step of the way, I could not have done it without you.   I want to acknowledge my committee members: Dr. Mark Beauchamp, thank you for coming into the lab on a regular basis to remind us to “push the boundaries of science”. Your feedback during this process helped go beyond the obvious and make new connections between ideas. Dr. Patrick Gaudreau, thank you for guiding me thought some complex statistical analyses and providing a channel of communication in French.   Financial support for this project was provided by scholarships from the Social Sciences and Humanities Research Council of Canada and The University of British Columbia Faculty of Education.  Finally, ‘merci’ to my family and friends back home for your unconditional support. Many thanks to my friends here in Vancouver for providing encouragement and allowing me to take my mind off school when needed, especially Natasha Pestonji and Amy Obetkoff.      1 Chapter 1: Introduction and literature review  1.1 Introduction  It has been argued that perfectionism is a requirement for success in athletic domains (Gould, Dieffenbach, & Moffett, 2002); however, others have claimed that perfectionism can lead to negative consequences such as psychological distress (Flett & Hewitt, 2005). Perfectionism is a personality disposition that has received extensive coverage in the athletic domain, both in the scientific community and in the media. Some world-renowned athletes who have achieved success at the highest level of competition have been said to be perfectionist – notable examples include rugby player John Wilkinson and soccer player Roy Keane (Hall, 2006). Some people use this link between perfectionism and high athletic achievement to claim that perfectionism is a necessary requirement for success in sport (Gould et al., 2002). Although there is some acceptance among sport researchers that aspects of perfectionism may facilitate achievement (Gotwals, Stoeber, Dunn, & Stoll, 2012; Stoeber, Uphill, & Hotham, 2009), there is evidence that maladaptive aspects can also lead to anxiety, shame, dysfunctional cognitions, and burnout (Flett, Greene, & Hewitt, 2004; Hill, 2013; Hill & Appleton, 2011; Kaye, Conroy, & Fifer, 2008). Presently there appears to be no clear definition of perfectionism that is accepted by all researchers (see Hall, Hill, & Appleton, 2012 for a review). Typically, perfectionism is said to involve a personal striving component that represents aspirations towards very high achievement standards and a harsh self-criticism component (see Hall et al., 2012 for a review). Some researchers argue that personal striving in isolation does not capture the construct of perfectionism (Flett & Hewitt, 2002). There seems to be a consensus on two main dimensions of perfectionism: personal standards perfectionism (PSP) and evaluative concerns perfectionism (ECP; Gaudreau & Thompson, 2010; Stoeber & Otto, 2006).    2 Researchers in sport have examined how dimensions of perfectionism are related to psychological and emotional variables (see Hall et al., 2012 for a review). Recent work by Hill and colleagues (Appleton, Hall, & Hill, 2009; Hill, Hall, Appleton, & Kobuz, 2008; Hill, Hall, & Appleton, 2010a; Hill, 2013) has focused on the relationship between perfectionism and burnout. Burnout in sport is a psychological symptom associated with prolonged stress and comprises three components: emotional or physical exhaustion, devaluation of sport, and reduced achievement (Raedeke, 1997). Stress associated with training and competing could create an environment favourable for the development of burnout. Moreover, athletes’ perfectionistic tendencies can also be associated with training- and competition-related stress and in turn influence athletic burnout. Research has found significant relationships between dimensions of perfectionism and burnout in young soccer players; PSP was negatively associated with burnout and ECP was positively associated with burnout (Hill et al., 2008; Hill, 2013).  In order to better understand the complex relationship between perfectionism and burnout, mediation effects were explored (Hill et al., 2008; Hill et al., 2010a, 2010b). Mediators are variables that help clarify the mechanism by which one factor (perfectionism) influences a second factor (burnout; Hayes, 2013). Presence of a statistically significant mediator, or partial mediator, indicates that the impact of the first variable on the second variable occurs at least in part because of its link with the mediator (Hayes, 2013). One potential mediator of the relationship between perfectionism and burnout is coping (Hill et al., 2010a).  Coping is defined as an active process that involves behaviours and cognitive processes to manage stress (Lazarus & Folkman, 1984). Sport is an achievement domain associated with multiple potential stressors such as injury, coach pressure, performance    3 demands, personal and others’ expectations, and interpersonal conflicts (Hoar, Kowalski, Gaudreau, & Crocker, 2006; Nicholls, Polman,, Levy, Taylor, & Cobley, 2007). Athletes who experience difficulties dealing, or coping, with those stressors are more likely to suffer performance decrement in competitive contexts (Lazarus, 2000a). Tied to coping is cognitive appraisal, a subjective interpretation of the interplay between the athlete and the situation (Lazarus, 2000a), which can be divided into two components – primary and secondary appraisal (Lazarus, 1991, 2000a, 2000b). Primary appraisal is the evaluation of what is at stake for the individual in a stressful situation (Lazarus, 1991). Secondary appraisal involves the assessment of the different strategies that the person perceives being able to engage in to resolve the stressful situation (Lazarus, 1991). Appraisal and coping have the potential to impact the development or persistence of burnout symptoms for athletes by influencing their perception of threat or challenge and control over stressful situations (Lazarus, 2000a).  The present research project sought to extend findings about the mediating role of coping strategies on the relationship between perfectionism and burnout in athletes (Hill et al., 2010a). The study extended Hill and colleagues’ (2010a) findings regarding coping as a mediator to the relationship between perfectionism and burnout to a longitudinal research design and to a sample that differed in terms of age and cultural background. The study adopted a longitudinal design to follow change over time and to explore stability of the relationships between the variables studied. The 2 x 2 model of dispositional perfectionism (Gaudreau & Thompson, 2010) was used as a conceptual framework of perfectionism, which led to the testing of the hypotheses put forth by this recent model.       4 1.2 Literature review 1.2.1 Perfectionism   Perfectionism is a psychological construct that has been studied extensively (see reviews by Gotwals et al., 2012; Lo & Abbott, 2013; Stoeber & Otto, 2006; Stoeber, 2011). It has historically been associated with a number of maladaptive constructs such as anxiety (Kawamura, Hunt, Frost, & Marten DiBartolo, 2001), depression (Sherry et al., 2013), shame (Ashby, Rice, & Martin, 2006), negative affect (as cited in Lo & Abbott, 2013) and low self-esteem (Ashby et al., 2006). Some aspects of perfectionism have been associated with adaptive constructs such as high achievement (Stoeber & Rambow, 2007), self-serving attributions (Stoeber & Becker, 2008), higher perseverance (as cited in Stoeber & Otto, 2006), and positive affect and life satisfaction (as cited in Stoeber & Otto, 2006). Whereas researchers once conceptualized perfectionism as unidimensional (Burns, 1980), it is now more commonly regarded as a multidimensional construct (Gotwals et al., 2012; Flett & Hewitt, 2005).  Although researchers seem to agree on the multidimensional nature of perfectionism, there is no consensus on the exact definition of the construct. Past definitions of perfectionism have included components such as: harsh self-evaluation following failure, parental expectation and pressure, irrational beliefs and dysfunctional attitudes, selective attention to personal shortcomings, and narrow margin between success and failure (Frost & Henderson, 1991; Hall et al., 2012). Perfectionism in sport is typically defined as a psychological disposition that encompasses striving for extremely high standards, self-critical tendencies, and associating self-worth with personal achievement (Hall et al., 2012). The definition of perfectionism that was used in this research project was as follows: a psychological disposition that involves striving towards high standards that are self-imposed    5 and concerns over mistakes that are tied with a belief that worth is tied to high achievement (Stoeber, 2011). There is a debate in the literature about the nature of perfectionism – more specifically whether perfectionism is adaptive or maladaptive (Flett & Hewitt, 2005; Gotwals et al., 2012; Stoeber & Otto, 2006). The multidimensional conceptualization of perfectionism leads to diverging ideas about the benefit or detriment of different facets of the construct. Two types of perfectionism have been differentiated; however, the terms and specific definition vary slightly between studies resulting in some confusion in the perfectionism literature. In this document, personal standards perfectionism (PSP) and evaluative concerns perfectionism (ECP) will be used systematically throughout the text. The first dimension, ECP, involves facets such as concerns over mistakes, doubts about actions, socially prescribed perfectionism, and perceived discrepancy between personal achievements and desired achievements (Stoeber & Otto, 2006). Socially prescribed perfectionism is suggested to represent perceived external pressure to achieve high standards and is generally tied to strong self-criticism (Hewitt & Flett, 1991). The body of research on perfectionism reveals that ECP is associated with maladaptive outcomes such as neuroticism, maladaptive coping, and negative affect (see Stoeber, 2011 for a review). Evaluative concerns perfectionism has also been referred to as: maladaptive perfectionism, negative perfectionism, dysfunctional perfectionism, unhealthy perfectionism, self-evaluative perfectionism, and maladaptive evaluation perfectionism (Stoeber & Otto, 2006).  The second perfectionism dimension, PSP, typically involves high personal standards and self-oriented perfectionism (Stoeber & Otto, 2006). Self-oriented perfectionism is proposed to represent the pursuit of extremely high personal standards, stringent evaluation of one’s behaviour, and motivation to attain perfection (Hewitt & Flett, 1991). Research    6 evidence on whether PSP is associated with adaptive or maladaptive outcomes is equivocal. Some studies have found positive correlations between PSP and adaptive outcomes, such as conscientiousness, adaptive coping, and positive affect (see Stoeber, 2011 for a review), whereas others have found inconclusive relationships between PSP and adaptive outcomes (see Stoeber & Otto, 2006 for a review). A recent review by Gotwals and colleagues (2012) investigated the mixed evidence relating to PSP in the sport domain and concluded that PSP is generally related to positive outcomes and rarely associated with negative outcomes. The PSP dimension is also known as: positive perfectionism, functional perfectionism, positive striving perfectionism, healthy perfectionism, personal standards perfectionism, and conscientious perfectionism (Stoeber & Otto, 2006).   Studies investigating perfectionism are numerous and branch out to diverse areas outside of clinical psychology such as sport (Flett & Hewitt, 2005; Gotwals et al., 2012; Hall et al., 2012; Stoeber, 2011), school (Stoeber & Rambow, 2007) and the workplace (Childs & Stoeber, 2012). The last two decades have seen an increase in the research on perfectionism in the sport domain (see reviews by Flett & Hewitt, 2005; Gotwals et al., 2012; Hall et al., 2012; Stoeber, 2011). The next section will review two prominent models of perfectionism commonly used in the sport literature. 1.2.1.1 Models of perfectionism  The tripartite model of perfectionism  Stoeber and Otto (2006) proposed a two-dimension model of perfectionism, the tripartite model (see Figure 1). The first dimension, perfectionistic concerns1, includes                                                 1 In order to stay true to Stoeber and Otto’s (2006) tripartite model of perfectionism, the two dimensions of perfectionism described in this paragraph will be the ones used by these authors. The retention of the original dimensions (perfectionistic concerns and perfectionistic strivings) has been favored over the use of ECP and PSP in this context to allow comparison    7 constant worries about one’s achievement and how they relate to one’s high performance standards. This dimension is also characterized by self-worth being dependent on the achievement of the perceived high expectations of significant others. A review of perfectionism in sport has revealed perfectionistic concerns to be maladaptive for athletes (Gotwals et al., 2012). The second dimension in the tripartite model (Stoeber & Otto, 2006) is perfectionistic strivings, referring to the excessively high performance standards that one aspires to achieve based on personal desire to be without flaw. This latter dimension of   Figure 1 The tripartite model of perfectionism Note. Two dimensions combine to create two groups of perfectionists and one group of non-perfectionists. Adapted from Stoeber & Otto, 2006.                                                                                                                                                    of the two main models of perfectionism. The use of ECP and PSP will be prioritized throughout the text with the exception of the section on the tripartite model.    8 perfectionism has been associated with more adaptive outcomes in athletes even though it is linked to mixed findings when it comes to sport and exercise (Gotwals et al., 2012). According to the tripartite model, perfectionistic concerns are generally maladaptive, whereas perfectionistic strivings are considered adaptive, especially when the overlap with perfectionistic concerns is taken into account. Perfectionistic strivings have been shown to be adaptive when an athlete’s abilities are matched to his or her expectations (Gotwals et al., 2012). When an athlete’s abilities fall short of their expectations and the athlete experiences failure, perfectionistic strivings are maladaptive (Gotwals et al., 2012). The tripartite model classifies people into one of three groups (Figure 1). The first group consists of healthy perfectionists, associated with low perfectionistic concerns and high perfectionistic striving. The second group is unhealthy perfectionists and includes people high on both dimensions of perfectionism. The last group, non-perfectionists, includes individuals low on perfectionistic strivings regardless of how high or low they are on perfectionistic concerns. The tripartite model of perfectionism has received support in the sport domain (Gucciardi, Mahoney, Jalleh, Donovan, & Parkes, 2012). The 2 x 2 model of dispositional perfectionism Gaudreau and Thompson (2010) proposed a model of perfectionism that makes different assumptions and predictions about the dimensions of perfectionism than those suggested by Stoeber and Otto’s (2006) tripartite model. The 2 x 2 model of dispositional perfectionism (Gaudreau & Thompson, 2010) is based on two basic dimensions of perfectionism that combine to create four subtypes (see Figure 2). Evaluative concerns perfectionism includes doubts about one’s actions and a tendency to worry about mistakes. It also involves the belief that others set high standards for an individual and that the individual    9 has to reach those in order to be accepted by others. The second dimension, PSP, represents the striving for high self-imposed standards. ECP and PSP combine to create four subtypes: (I) non-perfectionism (low PSP and low ECP), (II) pure PSP (high PSP and low ECP), (III) pure ECP (low PSP and high ECP), and (IV) mixed perfectionism (high PSP and high ECP; Gaudreau, 2013). These four subtypes of perfectionism are associated with four hypotheses: (1) pure PSP can be associated with either a) better, b) worse, or c) equivalent outcomes than non-perfectionism, (2) pure ECP will be the most damaging type of perfectionism, (3) mixed perfectionism will be less harmful than pure ECP, and (4) mixed perfectionism will be associated with lower well-being than pure PSP (Gaudreau & Thompson, 2010). The 2 x 2 model also holds a number of assumptions (Gaudreau & Verner-Filion, 2012). First, it assumes that both PSP and ECP can coexist in all individuals. Second, the specific combination of both dimensions allows computation of the probability that each person is part of one of the four subtypes proposed by the model. Third, it is this pattern of internal combination of PSP and ECP that determines the specific correlates of perfectionism for each individual. Fourth, the four subtypes of perfectionism proposed by the 2 x 2 model are theoretically distinct. Fifth, the subtypes are distinct etiologically and functionally; they are differently associated with antecedents, processes, and consequences. Sixth, the labels of the four subtypes of perfectionism are generic, they do not assume a healthy/unhealthy nature of the subtypes. Finally, the 2 x 2 model of perfectionism assumes that the broad dimensions, ECP and PSP, include facets of perfectionism (socially prescribed perfectionism, and self-oriented perfectionism respectively).     10  Figure 2 The 2 x 2 dispositional model of perfectionism (Gaudreau & Thompson, 2010).  Note. The figure also includes the hypotheses suggested by the 2 x 2 model of perfectionism. Adapted from Gaudreau, 2013.  One of the key advantages of the 2 x 2 model of perfectionism is that it acknowledges that the dimensions can interact within an individual, which is why it is important to consider both PSP and ECP simultaneously (Gaudreau, 2013). Research in the academic domain has employed the 2 x 2 model of perfectionism to look at outcome variables associated with perfectionism (e.g., Franche, Gaudreau, & Miranda, 2012; Gaudreau & Thompson, 2010). They found support for hypotheses 1a, 2, 3, and 4 of the 2 x 2 model (Gaudreau & Thompson, 2010). Pure PSP was associated with better outcomes than non-perfectionism:    11 higher academic satisfaction, positive affect, academic goal progress, and self-determination. Pure ECP was associated with the lowest level of self-determination compared to all other subtypes. Mixed perfectionism was associated with better outcomes than pure ECP but worse outcomes than pure PSP. A recent study applied the 2 x 2 model of perfectionism in the sport domain to study the relationship between perfectionism and burnout (Hill, 2013), this study will be reviewed in the next section. The 2 x 2 model shows a promising framework within which to investigate perfectionism in the sport domain, especially to tease apart some equivocal findings about personal strivings (Stoeber & Otto, 2006). 1.2.1.2 Perfectionism in sport Research in the sport domain has investigated the role that perfectionism plays in the lives of athletes (see Flett & Hewitt, 2005; Gotwals et al., 2012; Hall, 2006). Some sport researchers have argued that perfectionism can be domain specific; that is, a person’s perfectionistic tendencies can be different depending on the life domain (Dunn, Gotwals, & Causgrove Dunn, 2005). For example, a student-athlete can set excessively high standards in sport but set more moderate standards in school. The tendency to have high perfectionism in a domain might depend on how valuable the domain is to one’s sense of self, or on how much external pressure (coach, parents, peers) one perceives in a particular domain. This has lead to the development of sport-specific measures such as the Sport-Multidimensional Perfectionism Scale (Dunn, Causgrove Dunn, & Syrotuik, 2002) and the Sport-Multidimensional Perfectionism Scale-2 (Sport-MPS-2; Gotwals & Dunn, 2009). However, it should be noted that both dispositional and sport-specific measures are used in sport research (Gotwals et al., 2012). Research on perfectionism in sport has generated a number of empirical findings which will be reviewed next.     12 1.2.1.3 Outcomes associated with perfectionism in sport  In the last decade, a number of reviews of perfectionism in the sport domain have been published (Flett & Hewitt, 2005; Hall, 2006; Hall, Hill, & Appleton, 2012; Gotwals et al., 2012; Stoeber, 2011). These reviews have shifted from a mainly debilitative view of perfectionism in sport (Flett & Hewitt, 2005; Hall, 2006) to a more balanced view of the potential benefits and costs of perfectionism (Hall et al., 2012; Gotwals et al., 2012; Stoeber, 2011). Past evidence largely suggested that perfectionism was essentially maladaptive in sport (Hall, 2006). For example, findings such as the fact that athletes high in perfectionism perceived achievement situations as more threatening than athletes low in perfectionism encouraged this bias (Hall, Kerr, & Matthews, 1998). A recent review by Gotwals and colleagues (2012) highlights studies concluding that ECP is maladaptive in sport and PSP is generally adaptive even though there are mixed findings about this last dimension of perfectionism. Negative outcomes associated with perfectionism in sport will be reviewed next, followed by positive outcomes.  For a number of years, researchers perceived that perfectionism was maladaptive for athletes’ emotional and psychological adjustment, although negative cognitions and behaviours associated with perfectionism can sometime be conducive to athletic success (Flett & Hewitt, 2005). Athletes high in ECP were found to experience anxiety, low confidence, negative reactions to mistakes, to adopt a failure orientation, and an ego orientation (Frost & Henderson, 1991). Appleton and colleagues (2009) found that for male athletes, ECP was positively associated with burnout. ECP has been associated with fear of failure (Stoeber & Becker, 2008) and with negative body image (Dunn, Craft, Causgrove Dunn, & Gotwals, 2011). Athletes who scored high in ECP reported lower life-satisfaction following an important competitive event in their respective sport (Gaudreau & Anlt, 2008).    13 High ECP was also associated with lower goal achievement and higher use of disengagement-oriented coping (Gaudreau & Antl, 2008). For athletes, ECP was associated with mastery-avoidance and performance-avoidance goal orientations, which are perceived as generally maladaptive goal orientations (Kaye et al., 2008). Evaluative concerns perfectionism was also positively associated with negative affect and inversely associated with positive affect (Kaye et al., 2008). Finally, Flett and Hewitt (2005) suggested that perfectionist athletes who lack the ability to achieve their high standards are especially at risk to experience maladaptive outcomes and emotional distress. The research evidence converges towards a consensus that ECP is associated with mainly negative outcomes in the sport domain.  Several sport studies support the association of PSP with adaptive outcomes. Female soccer players high in PSP with hopes to succeed made internal attributions for the reason or cause of success once success was achieved, which represents a self-serving attribution (Stoeber & Becker, 2008). Junior athletes high in PSP reported higher use of task-oriented coping (Hill et al., 2010a) and male junior athletes high in PSP reported lower levels of burnout (Appleton et al., 2009). When the negative reaction to imperfection was statistically controlled, PSP was associated with hopes of success and inversely related to fear of failure and self-depreciating attributions in female soccer players (Stoeber & Becker, 2008). In athletes, PSP was associated with less concentration disruption (Gotwals, Dunn, Causgrove Dunn, &Gamache, 2010) and with increased positive affect and adaptive goal orientation such as mastery-approach and performance-approach (Kaye et al., 2008). Although multiple studies in the sport domain suggests that PSP is associated with adaptive outcomes, some studies looking at the association of PSP with maladaptive outcomes have revealed both neutral and inverse relationships (see Gotwals et al., 2012 for a review).    14  As can be seen in recent reviews of the evidence on perfectionism in the sport domain (Gotwals et al., 2012; Stoeber, 2011), there seems to be a growing consensus about the adaptive nature of PSP. The 2 x 2 model of perfectionism offers a means to not only explore the relationship of PSP and ECP with outcomes of perfectionism but also to explore the unique combination of PSP and ECP that resides within each individual (Gaudreau, 2013; Gaudreau & Thompson, 2010). This model involves specific hypotheses about the combined contribution of PSP and ECP to athletes’ cognitions, affect, and performance. For example, Gaudreau and Verner-Filion (2012) found that pure PSP was not associated with statistically significantly higher positive affect, subjective vitality, and life-satisfaction than non-perfectionism in a sample of athletes (hypothesis 1c). Pure ECP was associated with statistically significant lower level of positive affect, subjective vitality, and life-satisfaction than non-perfectionism (hypothesis 2; Gaudreau & Verner-Filion, 2012). Mixed perfectionism was associated with higher levels of positive affect, subjective vitality, and life-satisfaction than pure ECP (hypothesis 3; Gaudreau & Verner-Filion, 2012). Results for the fourth hypotheses of the 2 x 2 model were partly congruent with the predicted relationship; mixed perfectionism was not lower on positive affect and subjective vitality than pure PSP, but consistent with hypothesis 4, mixed perfectionism was associated with lower life-satisfaction in athletes than pure PSP (Gaudreau & Verner-Filion, 2012). Gaudreau and Verner-Filion’s (2012) results represent a detailed picture of how both dimensions of perfectionism combine to create different predictions of associated outcomes. Research on perfectionism in sport could benefit from this approach as it could help explain inconsistent results found in other studies.      15 1.2.1.4 Stability of perfectionism over time Longitudinal studies involving perfectionism typically assess perfectionism solely at the first time point and assume that it remains stable throughout the study. This trend holds in research in the sport domain (e.g., Chen, Kee, & Tsai, 2008; Hill, Stoeber, Brown, & Appleton, 2014) and academic domain (e.g., Enns, Cox, Sareen, & Freeman, 2001). However, some researchers have attempted to examine perfectionism over time. For example, Boone and colleagues (2012) used daily diaries to measure PSP and ECP on seven consecutive days in a sample of adolescents. They found daily variations in PSP and ECP; 43.79% and 36.98% (respectively) of this daily variation was associated with within-individual change. They also found that PSP was a statistically significant positive predictor of ECP within each day. Research about stability of perfectionism in sport has not been conducted to date. 1.2.2 Burnout 1.2.2.1 Definition Burnout is a psychological syndrome that involves emotional exhaustion, reduced sense of accomplishment, and depersonalization (Maslach, 1982). Burnout has been increasingly studied in relation to sport, with Raedeke (1997) defining burnout in sport as a psychological syndrome that involves emotional or physical exhaustion, reduced accomplishment in sport, and sport devaluation. The first symptom of emotional or physical exhaustion results from the effect of the prolonged stressful demands of sport (Raedeke & Smith, 2001). The second symptom is reduced accomplishment, which refers to perceived reduced skills and abilities within the practice of one’s sport (Raedeke & Smith, 2001). The last symptom is sport devaluation, represented by lack of interest and care about training and performance; it can also present itself as reduced importance of sport (Raedeke & Smith,    16 2001). The first two dimensions of burnout are similar to the ones in Maslach’s definition of burnout in the workplace (1982). The third dimension, sport devaluation, represents an adaptation to the sport context of the depersonalization dimension proposed by Maslach (1982).  1.2.2.2 Burnout in sport Research has provided insight into athletic burnout, revealing symptoms as well as associated and restraining factors of burnout (i.e., Gould, Tuffey, Udry, & Loehr, 1996b; Gould, Tuffey, Udry, & Loehr, 1997; Gustafsson, Hassmen, Kenta, & Johanson, 2008). Symptoms associated with athletic burnout include: lack of athletic success, emotional and physical exhaustion, devaluation of one’s sport, lack of motivation, mental distancing from one’s sport, mood swings, depressed mood, and irritability. Gustafsson and colleagues (2008) identified a number of factors associated with burnout such as: excessive training schedules, low autonomy in one’s training, insufficient recovery period between training sessions, difficulties balancing school and sport, worries about the future, stressful social relationships, and lack of social support. Finally, their interviews revealed that athletic identity and entrapment were associated with continued involvement and burned out athletes. Gould, Udry, Tuffey, and Loehr (1996a) examined burnout in junior tennis players and found that compared to a non-burnout group, burned out tennis players were higher in external motivation and amotivation, reported more parental criticisms and expectations, a higher need for organization, and greater concerns over mistakes (Gould, et al., 1996a). A review of burnout in sport (Goodger, Gorely, & Lavallee, 2007) revealed positive associations between perceived stress and burnout and negative associations between coping and burnout. The study findings reviewed do not differentiate the different type of coping strategies and their relationship with burnout (Goodger et al., 2007), although studies    17 published after this review revealed problem-focussed coping to be negatively associated with burnout and disengagement-coping to be positively associated with burnout in athletic population (Hill et al., 2010a; Schellenberg, Gaudreau, & Crocker, 2013). Further relevant findings about burnout in sport include a study with swimmers, which revealed that the relationship between coping and burnout (when social support is controlled) is mediated by perceived stress associated with involvement in swimming (Raedeke & Smith, 2004). The measure of coping used did not differentiate between different coping function such as task-oriented coping and disengagement-oriented coping (Raedeke & Smith, 2004). More research needs to be done to understand the effect of coping behaviour on burnout.  While athletic burnout has been primarily investigated with cross-sectional research design (e.g., Appleton et al., 2009; Hill et al., 2010a; Hill, Hall, Appleton, & Kozub, 2008), a few researchers have adopted longitudinal approaches (Cresswell & Eklund, 2005, 2006; Schellenberg et al., 2013). For example, Cresswell and Eklund (2005) measured burnout in rugby players at three occasions over a 12-week period. They found that all three dimensions of burnout varied slightly over time but the only dimension of burnout that displayed a statistically significant change during the study period was reduced accomplishment (Cresswell & Eklund, 2005). A second investigation with rugby players tracked burnout over a 30-week period, as the researchers postulated that the 12-week study had not been long enough to capture change in burnout. Athletic burnout was measured before, during, and after a competitive year. The two main findings were that burnout likely does not vary following a linear pattern and that there are differences in the pattern of change of the three dimensions of burnout (Cresswell & Eklund, 2006). In an investigation of collegiate volleyball players Schellenberg and colleagues (2013) found that between the start and end of the competitive season, the change in athletic burnout was positively predicted by    18 disengagement-oriented coping and negatively predicted by task-oriented coping strategies. This last study only involved two time points (Schellenberg et al., 2013) which was not enough to capture the rate of change in athletic burnout. 1.2.2.3 Burnout and perfectionism Personality factors that influence appraisal processes, such as perfectionism, are thought to be antecedents of burnout (Hill et al., 2010a). At the conceptual level, reduced accomplishment might be related to ECP. The perception of failure to achieve objectives that is associated with ECP can also result in reduced athletic accomplishment. Moreover, ECP involves a social component to the effect that athletes high on ECP typically associate failure with social rejection. Trying to cope with social rejection in situations of perceived reduced accomplishment might be linked with sport devaluation as a means to distance one’s identity from the area of social rejection. Conceptual links between PSP and separate dimensions of burnout are not as clear.  In the late 1990s, the seminal studies on burnout in junior elite tennis players by Gould and colleagues (1996a, 1996b, 1997) included perfectionism as one of their variables of interest because of observed perfectionistic tendencies in elite athletes. They expected that the burned out athletes would display higher levels of perfectionism than the non-burnout athletes. They found that the burnout group scored higher on two dimensions of perfectionism – parental expectations and concerns over mistakes (Gould et al., 1996a). In the second phase of their study, they used a qualitative approach to investigate causes or antecedents of burnout (Gould et al., 1996b). A key finding was that there appear to be two distinct paths that lead to burnout. The first one involves athletes who are high in perfectionism and lack appropriate coping strategies to deal with their perfectionistic tendencies. This results in high perceived stress from training demands that would not    19 typically be considered overwhelming in non-perfectionists. The second stream, labelled “situational pressure,” involves athletes that are not necessarily high in perfectionism but are under high pressure from others (especially parents) to perform, which causes high stress (Gould et al., 1996b).  The relationship between burnout and perfectionism in sport has been investigated in several studies (e.g., Appleton et al., 2009; Gould et al., 1996a; Hill et al., 2010a, 2010b). It has been shown that PSP is negatively associated with burnout; that is, athletes who score high in PSP tend to report less burnout symptoms, whereas athletes who score low in PSP tend to have higher burnout scores (Appleton et al., 2009). The reverse relationship exists for ECP; that is, ECP is positively associated with burnout (Appleton et al., 2009; Hill et al., 2008). Two recent studies used the 2 x 2 model of perfectionism, or a similar framework, to explore the relationship between perfectionism and athletic burnout. Chen and colleagues (2008) used a perfectionism model similar to the 2 x 2 model to investigate its relationship with burnout in collegiate athletes in a cross-sectional study. They found a negative association between burnout and high PSP and a positive association between burnout and high ECP. Additionally, combinations of high PSP and low ECP were associated with lower burnout scores. Combinations of high PSP and high ECP as well as combinations of high ECP and low PSP were associated with higher burnout scores. These results are consistent with hypotheses 1a and 4 of the 2 x 2 model of perfectionism.  Hill (2013) used the 2 x 2 model of dispositional perfectionism to test the relationship between perfectionism and burnout in athletes in a cross-sectional study. Moderated hierarchical regressions revealed that PSP was negatively associated with total burnout and ECP was positively associated with total burnout. This provides support for hypothesis 1a, 2,    20 3, and 4 of the 2 X 2 model (Hill, 2013). These studies suggest that the 2 x 2 model of dispositional perfectionism is appropriate to study the relationship between perfectionism and burnout in athletes. Although Hill’s (2013) study provided a necessary test of the 2 x 2 model within the relationship between perfectionism and burnout, further research needs to be done. For example, Hill acknowledged some shortcomings of his research project, such as the limited generalizability of the results and the cross-sectional nature of the data collection. Moreover, following Gaudreau and Verner-Filion’s (2012) recommendation to test the 2 x 2 model using different perfectionism scales, Hill (2013) suggested that research using scales other than the Hewitt-MPS (such as the Sport-MPS-2) would help expand his findings.  1.2.3 Mediators of the relationship between perfectionism and burnout Recent work on the links between dimensions of perfectionism and burnout has explored possible mediating factors of this relationship. Potential mediators include self-acceptance, validation-seeking, and coping. For example, in addition to the direct positive effect between ECP and burnout and direct negative effect between PSP and burnout, unconditional self-acceptance was a partial mediator in those relationships in young male soccer players (Hill et al., 2008). That is, ECP was negatively associated with unconditional self-acceptance, which in turn was negatively associated with burnout. Personal standard perfectionism was negatively related to unconditional self-acceptance, which was negatively related to burnout. A study with canoe-polo and kayak slalom athletes explored the mediation effect of validation- and growth-seeking (Hill et al., 2010b). Findings revealed partial mediation of validation-seeking in the relationship between ECP and burnout. This suggests that the ECP dimension involves the desire to please others (validation-seeking) and that this pursuit is associated with increased burnout symptoms. Research has also investigated the    21 effect of coping in the perfectionism and burnout relationship (Hill et al., 2010a). A brief review of coping will be presented before exploring the theoretical reasons for the proposed meditation effect of coping.  1.2.3.1 Coping A discussion of coping requires an understanding of stress and appraisal. Stress is the personal experience of a transactional reaction between a person and his or her environment that causes physiological or psychological distress (Aldwin, 2007). Some physiological processes are typically associated with the stress response such as activation of the hypothalamic-pituitary-adrenal (HPA) axis and heightened sympathetic activity (Buckworth & Dishman, 2002). An important cognitive component of stress is appraisal. Lazarus (1991, 2000b) divides the appraisal process into two components, primary and secondary appraisal. Primary appraisal reflects the evaluation of a situation that seeks to determine what is at stake, that is, whether the situation can affect personal goals. Secondary appraisal involves an assessment of what can be done in this situation, which is influenced by perceived control, available resources, and expectations about the future (Lazarus, 1991). Coping resources and perceived coping effectiveness can therefore influence the stress process.  Coping is an active cognitive and behavioural process that seeks to manage stress (Lazarus & Folkman, 1984). Psychological stress can be experienced when the demands of a situation are perceived as exceeding one’s resources (Lazarus, 1991). Lazarus’ (1991) cognitive-motivational relational theory is based on a transactional relationship between the person and the environment. In this transactional relationship, appraisal and coping are part of the stress process. Individuals bring personality factors and cognitive knowledge to a situation and these influence how the situation is perceived and dealt with. Moreover, how    22 one reacts to a situation has the potential to influence the situation, which in return will influence later appraisal of the situation.   According to Lazarus and Folkman (1984), coping has two functions: emotional regulation (or emotion-focused coping) and management of a threatening or challenging situation (or problem-focused coping). In a review of coping in sport, Hoar and colleagues (2006) identified avoidance coping as a third coping function that has been commonly used by researchers. Coping strategies are classified as problem-focused coping when they seek to modify the stressful situation or one’s interpretation of the situation (Hoar et al., 2006). On the other hand, emotion-focused coping involves strategies that are aimed at dealing with the emotional distress resulting from the stressful situation (Hoar et al., 2006). Finally, avoidance coping can be displayed through either physical or psychological removal from the stressful situation (Hoar et al., 2006).   Gaudreau and Blondin (2002) developed a slightly different taxonomy to classify ways of coping. They regrouped micro-level coping strategies into three macro-level dimensions of coping: task-oriented, disengagement-oriented, and distraction-oriented coping (Gaudreau & Blondin, 2002, 2004). In this model, task-oriented coping (TOC) involves strategies that seek to modify the stressful situation or change one’s perception of the situation such as thought control, mental imagery, relaxation, effort expenditure, logical analysis, and seeking social support. Disengagement-oriented coping (DOC) regroups strategies such as venting unpleasant emotions and disengagement from the situation or resignation. Distraction-oriented coping focuses on dealing with the emotional distresses caused by stressful situations and involves strategies such as mental distraction and distancing.      23 1.2.3.2 Coping and the relationship between perfectionism and burnout Personal factors, such as perfectionistic disposition and its related cognitive processes, are likely related to appraisal and coping. This means that the relationship between the dimensions of perfectionism and the use of specific coping strategies could influence burnout. Hill and colleagues (2010a) investigated the potential mediating effect of coping on the relationship between perfectionism and burnout in junior elite athletes. Perfectionism was divided into two components, ECP and PSP. Coping was assessed based on stress associated with competitive and training demands. The modified COPE questionnaire was used to measure the use of coping strategies and the data was analysed based on Dunkley and colleagues’ (Dunkley & Blankstein, 2000; Dunkley, Zuroff, & Blankstein, 2003) two coping factors, problem-focused coping and avoidant coping. Finally, burnout was assessed with the total burnout score from the Athlete Burnout Questionnaire (ABQ; Raedeke & Smith, 2001). Findings revealed that the relationship between ECP and burnout was mediated by avoidant coping; that is, higher ECP was associated with higher avoidant coping that was associated with higher burnout. Moreover, the negative relationship between PSP and burnout was mediated by problem-focused coping. Higher PSP was linked to higher problem-focused coping that was linked to lower burnout. Avoidance coping was also a mediator of the relationship between PSP and burnout; that is higher PSP was associated with lower avoidant coping that was associated with higher burnout. This study did, however, have a few limitations. First, the study used a cross-sectional design. The mediation effect of coping as well as the direct effect of perfectionism on burnout would be better studied with a longitudinal design. Second, the interactive effect of both dimensions of perfectionism was not examined. Third, the sample was composed of junior elite athletes from England; therefore, the findings should be expanded to other sporting populations.     24 1.2.4 Longitudinal design to examine mediation Research examining mediators of the relationship between perfectionism and burnout has been limited by the use of cross-sectional designs. Cross-sectional designs provide static information about a phenomenon, measuring a number of variables at one occasion. Although such cross-sectional designs are useful to answer specific questions, they offer limited information on mediating variables.  Longitudinal panel is a type of research design that assesses the same people at two or more occasions (Rogosa, 1995). Longitudinal studies with two time points provide information about the amount of change in the measured variables over time. Although change scores based on two time points are informative, longitudinal studies with more than two time points allow for analyses of rate of change. Multiple measurements, (four or more) allow the researchers to observe whether the relationships are linear or not (Stoolmiller, 1995). Growth curve analysis looks at individual slopes on a given variable measured repeatedly over time (Stoolmiller, 1995). This technique provides information on individual developmental trajectory and on the stability or instability of a construct. Moreover, longitudinal design allows investigation of the effect of different predictors on the construct of interest. Predictors can influence the strength and direction of the individual slopes on the variable of interest (Stoolmiller, 1995). For example, perfectionism might be a predictor variable that would influence the starting point and rate of change in athletic burnout. In this research project, the use of longitudinal design was preferred over cross-sectional design because the former provides insights on the covariation of perfectionism, coping, and burnout as well as on the developmental process of these variables over time.      25 1.3 Purpose and research questions  The purpose of this research project was to examine the role of coping on the relationship between perfectionism and burnout in Canadian university varsity athletes. More specifically, the relationships between perfectionism, coping, and burnout were assessed in relation to the intense training demands that accompany high performance sport (Gustafsson et al., 2008). Furthermore, a longitudinal design was adopted to allow for an in-depth analysis of those relationships. The use of a longitudinal design allowed to investigate stability or instability of the constructs, and rate of change or individual and group-level developmental trajectory (Stoolmiller, 1995).  1.3.1 Relationships between perfectionism, burnout, and coping The main purpose of this study was to investigate the contribution of coping to the relationship between dimensions of perfectionism and the level of athletic burnout during a competitive season. Taking into consideration the influence of personal factors such as perfectionism and coping could facilitate understanding of the phenomenon of athletic burnout. It was hypothesized that burnout would be positively associated with ECP and negatively associated with PSP as those dimensions of perfectionism are typically related to maladaptive and adaptive outcomes respectively (Hall et al., 2012; Stoeber & Otto, 2006). Research on perfectionism and burnout in athletes supports these predictions (Appleton et al., 2009; Hill et al., 2010a,b; Hill, 2013).  Theoretically, engaging in task-oriented coping offers long-term benefits because the athlete actively tries to modify stressful situations or one’s appraisal of the stressful situations, therefore, preventing repeated encounters with the same stressor. Task-oriented coping has been associated with long-term positive adaptation in the workplace (Carmona, Buunk, Peiro, Rodriguez, & Bravo, 2006) and negatively associated with maladaptive    26 outcomes in athletes (e.g., Schellenberg et al., 2013). These findings suggest that task-oriented coping would be negatively associated with burnout, with frequent use of task-oriented coping strategies associated with lower burnout symptoms. Disengagement-oriented coping can be associated with positive outcomes in the short term when dealing with factors outside of one’s control (Anshel & Wells, 2000), but could also be associated with low coping effectiveness when dealing with recurring stressors (as suggested in a review by Richards, 2012). Studies looking at coping and athletic burnout revealed positive relationships between disengagement-oriented coping and burnout (Hill et al., 2010a; Schellenberg et al., 2013). It is advantageous to study disengagement-oriented coping longitudinally because of dual-prediction associated with its short- versus long-term effectiveness. In terms of the association between perfectionism and coping, there is limited empirical evidence in the sport domain, but it has been hypothesized that PSP should be associated with greater use of task-oriented coping and ECP should be associated with greater use of disengagement-oriented coping. Empirical evidence supports these predictions (Dunkley et al., 2003; Gaudreau & Anlt, 2008; Hill et al., 2010a) but also suggests that PSP is negatively associated with disengagement-oriented coping (Gaudreau & Anlt, 2008; Hill et al., 2010a).  Mediation analyses were used to explore the potential mediating effect of coping strategies on the relationship between perfectionism and burnout (see Figure 3). The relationship between PSP and burnout was expected to be mediated by TOC and DOC. On the other hand, the relationship between ECP and burnout was expected to be mediated by DOC but not TOC as previous studies have shown no associations between ECP and TOC (Hill et al., 2010a).      27  Figure 3 Visual representation of expected relationships. Note. Grey circle represent the hypothesised nature (+/-) of the relationships.  1.3.2 Change in burnout, perfectionism, and coping over time Another purpose of this study was to look at burnout in high performance athletes over the course of a competitive season. Individual growth curves were examined to gain a better understanding of the temporal pattern of burnout. Although there is no consensus on the origin of athlete burnout, recent theoretical models such as the failure-adaptation model (Tenebaum, Jones, Kitsantas, & Berwick, 2003) and the total-quality-recovery model (Kentta & Hassmen, 1998) share aspects with Silvia’s model (1990) that suggests that burnout results from maladaptation to training or insufficient recovery (Goodger et al., 2007). A longitudinal approach to the study of burnout seemed appropriate based on Silva’s (1990) theoretical framework that propose that burnout results from maladaptation to training or insufficient recovery. Studying burnout over an extended period provided information about temporal change in this construct. Furthermore, in the last decade researchers studying burnout in sport    28 have increasingly adopted longitudinal quantitative designs to study this phenomenon (see Cresswell & Eklund, 2005, 2006; DeFreese, 2013; Schellenberg et al., 2013). Longitudinal approach to the study of burnout reframes this psychological construct as a chronic process rather than a state and provides opportunities to explore causes of athletic burnout (Goodger et al., 2007). Athletic burnout was expected to vary over the course of the study based on prior longitudinal studies in sport (Cresswell & Eklund, 2005, 2006). More specifically, burnout was hypothesized to increase over the course of the season (Schellenberg et al., 2013). The two dimensions of perfectionism were expected to show small fluctuations across assessments based on recent evidence that both PSP and ECP fluctuate on a daily basis (Boone et al., 2012). Coping strategies were expected to vary over time although empirical evidence in the sport domain did not allow to make specific hypotheses about rate of change (Louvert, Gaudreau, Menaut, Genty, & Deneuve, 2007; Tamminen & Holt, 2010).  1.3.3 Test of the 2 x 2 model of dispositional perfectionism To increase understanding of the complex relationship between different types of perfectionism and burnout, the hypotheses associated with the 2 x 2 model of dispositional perfectionism were tested (see Figure 2; Gaudreau & Thompson, 2010). Pure PSP would be associated with lower burnout (hypothesis 1a), higher burnout (hypothesis 1b), or equal burnout level as individuals who exhibit non-perfectionism (hypothesis 1c). Additionally, pure ECP would be associated with the highest level of burnout of all four subtypes of perfectionism (hypothesis 2). Third, pure ECP should be associated with higher burnout than mixed perfectionism (hypothesis 3). And finally, pure PSP would be associated with lower burnout than mixed perfectionism (hypothesis 4).     29 Chapter 2: Methods 2.1 Participants Athletes from university level varsity sport teams were recruited for this study. A total of 155 participants took part in the first phase of the study. Teams were selected on the criterion that they train on a regular basis in October through to the beginning of March because the study focuses on stress experienced in relation to training demands. Recruitment emails were sent to coaches of university level sports such as swimming, track and field, cross-country, basketball, volleyball, rugby, golf, and ice hockey.  The final sample included 125 athletes. The sample was 72% female (nfemale = 90; nmale = 35). The athletes ranged in age from 18 to 36 years, with an average age of 20.65 years (SD = 2.53). All of the athletes participated at the university varsity sport level: volleyball (n = 37), basketball (n = 28), rugby (n = 28), ice hockey (n = 13), track & field or cross country (n = 9), swimming (n = 6), and golf (n = 4). Athletes’ year of eligibility ranged from 1 to 6 (n1year = 36; n2year = 21; n3year = 27; n4year = 21; n5year = 15; n6year = 2; three athletes did not report their year of eligibility). Thirty athletes (24%) reported having been, or being on a national team, and four athletes (3.2%) reported having been, or being, on an Olympic team. 2.2 Procedure Coaches and athletic directors from Canadian universities in large city centres were contacted to obtain permission to talk to their athletes (appendix A). A research assistant, or the principal investigator, met with the athletes at a time determined by the coaches. A brief description of the study was provided as well as a letter of invitation (appendix B). Athletes who desired to participate were handed a consent form (appendix   30 C) and a paper copy of the first questionnaire (appendix D). The athletes were free to choose between completing the questionnaire during the meeting or on their own time and return the questionnaire within the following week.  At the end of the first questionnaire, participants were asked to provide an email address at which they could be contacted to complete phase 2, 3, and 4. Phases 2-4 involved the completion of online questionnaires (see appendix E for time 2-4 questionnaire). Participants were contacted via email following Dillman’s (2007) Tailored Design Method: pre-notice emails were sent to alert them of the upcoming survey (appendix F), three days later questionnaire emails that included the link to the secure online survey were sent to the participant (appendix G), then up to three reminder emails were sent four, seven, and nine days following the questionnaire emails (appendix H). Finally, thank you emails were sent to the athletes who engaged in the online portion of the study to thank them for their time (appendix I).  Data were collected following the same schedule as the first and second year to avoid seasonal effects (see Figure 4). The four assessments were each approximately 4-5 weeks apart, taking into consideration exam week and the holiday break. The first time point was around mid-October, the second time point in late-November, the third time point in mid-January, and the fourth and final time point in late-February.   31   Figure 4 Timeline for data collection Note. P-N: pre-notice email, R: reminder email. Athletes completed a paper version of the questionnaire at Time 1, and electronic version of the questionnaire at Time 2, 3, and 4. The data collection cycle was adapted to the academic calendar to avoid overloading students with work during midterm and end of term examinations.   Participants were offered a stipend of ten dollars for submitting the first questionnaire. Additionally, their email addresses were entered into a draw for a chance to win one of ten bookstore gift cards valued at $50 each2. Participants’ email addresses were entered one more time for each online assessment that they took part in. Participants were only eligible to win one of the ten prizes. Upon completion of data collection, the ten prizes were offered to randomly selected athletes. The athletes had up to two weeks to claim their prizes. Two athletes failed to claim their prizes within the designated timeframe, two alternate winners were randomly selected.  2.3 Measures  Demographics: Relevant demographic information was collected; gender, age, sport, year of eligibility, membership on national or Olympic team, ethnicity, and first                                                 2 The additional incentive was included at the beginning of the second data collection cycle to improve retention rates. Participants who took part in the first data collection cycle were added in the draw retroactively.   32 language. An email address was requested in order to contact the participants for phase 2, 3, and 4 of the study.  Perfectionism: The Sport-Multidimensional Perfectionism Scale-2 (Sport-MPS-2; Gotwals & Dunn, 2009) was used to assess sport-specific perfectionism. This scale has 42 items and is composed of six subscales. The subscales are: personal standard (PS; I hate being less than the best at things in my sport), concern over mistakes (COM; If I fail in competition, I feel like a failure as a person), perceived parental pressure (PPP; My parents set very high standards for me in my sport), perceived coach pressure (PCP; I fell like I can never quite live up to my coach’s standards), doubts about actions (DAA; I rarely fell that my training fully prepares me for competition), and organization (Org; I have and follow a pre-competitive routine). Items are rated on a Likert-scale ranging from 1 (strongly disagree) to 5 (strongly agree). Participants are asked to indicate to what extent they agree with the items based on their training in the past week. Scores on the PSP dimension of perfectionism were represented by the PS subscale and scores on the COM subscale served as indicator of the ECP dimension. This conceptualization of the two dimensions has been suggested by Stoeber (2011) and has been used in a recent study (Crocker, Gaudreau, Mosewich, & Kljajic, in press). External validity for the Sport-MPS-2 was established in two separate studies (Gotwals & Dunn, 2009; Gotwals, et al., 2010). Reliability of the relevant subscales of the Sport-MPS-2 has been shown to be good (αPS = .74, αCOM = .79; Phase 3; Gotwals & Dunn, 2009). Cronbach’s alphas for the subscales used in this study were satisfactory at all time points (αPS = 0.81-0.84; αCOM = 0.86-0.92).   33  Coping: Coping was assessed with a modified version of the Coping Inventory for Competitive Sport (CICS; Gaudreau & Blondin, 2002). This scale includes 39 items divided into ten subscales. The subscales are part of three higher order coping dimensions. The first coping dimension is task-oriented coping (TOC). TOC is composed of: thought control (TC; I tried to get rid of my doubts by thinking positively), mental imagery (MI; I visualized that I was in total control of the situation), relaxation (RE; I tried to relax my body), effort expenditure (EE; I gave my best effort), logical analysis (LA; I thought about possible solutions in order to manage the situation), and seeking support (SS; I asked someone for advice concerning my mental preparation). The second coping dimension, disengagement-oriented coping (DOC) includes: venting of unpleasant emotion (VE; I expressed my frustrations) and disengagement/resignation (DR; I let myself feel hopeless and discouraged). The last coping dimension, distraction-oriented coping is represented by distancing (DI; I kept all people at a distance) and mental distraction (MD; I occupied my mind in order to think about other things than the training). Each item represents a coping behaviour and athletes are asked to rate to what extent they have used this coping behaviour over the last week on a five points Likert-scale ranging from 1 (does not correspond at all to what I did or thought) to 5 (corresponds very strongly to what I did or thought). Four items were modified to refer to behaviour used in relation to “training” instead of the original items that referred to behaviour in “competition”.  Gaudreau and Blondin (2002) found evidence of reliability and validity in a sample of French Canadian athletes. Confirmatory factor analysis supported the use of a 10 factor model with three higher-order dimensions. A recent review of measurement in   34 sport and exercise recommended the use of the CICS above other existing scales (Lidor, Crocker, & Mosewich, 2012).  The participants completed the full version of the modified CICS. In accordance with the method used in a recent study using the CICS to assess training demands (Schellenberg et al., 2013), two items from the TOC dimension were removed from the analysis because they do not apply in the context of training demands (i.e., I tried not to be intimidated by other athletes and I analysed the weaknesses of my opponents). Moreover, only the TOC and DOC dimensions were used in the analyses3. To compute the TOC and DOC score, first, subscales scores were computed by averaging the items within the subscales, then, the relevant subscales were averaged to obtain TOC and DOC scores. Cronbach’s alphas for the subscales within TOC and DOC were satisfactory across all four assessments (see Table 2 for exact values).   Burnout: The Athlete Burnout Questionnaire was used to assess burnout (Raedeke & Smith, 2001). This scale contains 15 items rated on a Likert-type scale ranging from 1 (I almost never feel this way) to 5 (I feel this way most of the time). Two items were reversed coded. Three subscales represent the dimensions of burnout in sport; emotional/physical exhaustion (I feel physically worn out from my sport), reduced sense of accomplishment (I am not performing up to my ability in my sport), and sport devaluation (I have negative feelings toward my sport). In a series of three validation studies conducted by Raedeke and Smith (2001), study 2 revealed that the scale showed positive relationships with stress, and negative relationships with coping, social support,                                                 3 Exploratory factor analysis in SPSS version 20.0 revealed poor factor structure for distraction-oriented coping. Moreover, distraction-oriented coping focus on strategies used during sport competition and did not make sense in relation to training demands.   35 and enjoyment. Study three examined the psychometric properties of the ABQ with a different sample from the one in the first two studies revealed good factor structure, acceptable reliability (α ≥ .71), and test-retest reliability (R2 ≥ .86; Raedeke & Smith, 2001).  In this study, the subscales had good reliability across all four assessments (α ≥  .83; see Table 2). The total burnout score was preferred to subscales scores in order to reflect the entire construct; this approach was used in a recent study involving athletic burnout (Schellenberg et al., 2013). A total athletic burnout score was computed in two steps; first, the three subscales scores were obtained by averaging the five items in each subscale, second, the total score was computed using the average of the three subscales.4 2.4 Data analysis.  Data analysis proceeded in five parts.  2.4.1 Data screening  The data from the paper version of the time 1 questionnaire were entered in an excel spreadsheet. Data for time 2, 3, and 4 were imported in excel from the secure online Edudata database. Participants’ data for each time point were matched using the email address they provided as well as demographic information. The data were then imported into SPSS version 20.0.  Participants’ data were examined for missing values, more specifically, percent of missing data was examined. The missing data at the item level were replaced using                                                 4 Participants also completed measures of contingent self-worth, goal progress (Dugas, Gaudreau, & Carraro, 2012), and appraisal (Stress Appraisal Measure; Peacock & Wong, 1990). These measures are not included in the analysis for this research project. The order in which the measures were completed was as follow: Sport-MPS-2, contingent self-worth, Stress Appraisal Measure, CICS, ABQ, and goal progress.   36 person median replacement based on the non-missing items of the subscale. This strategy is deemed appropriate in cases where few data points are missing and the items are moderately to highly correlated (Acuna & Rodriguez, 2004). However, data missing at the occasion level were not replaced. Hierarchical linear model software such as HLM7 student version handles missing data at the occasion level without major impact on the estimates produced when the level-1 model is well specified (Zaidman-Zait & Zumbo, 2013). Data were screened for outliers, data point with z-score above 3.29 were considered outliers (Tabachnick & Fidell, 2007). The normality assumption was tested based on skewness, kurtosis, visual inspection of the data (histogram, box plot, normal Q-Q plot, and normal probability plot), and the Komogorov-Smirnov test (Burdenski, 2000). 2.4.2 Descriptive statistics  Descriptive statistics were computed, such as mean and standard deviation, skewness, and kurtosis. Cronbach’s alphas were computed for each subscale to evaluate scale internal consistency. Pearson bivariate correlations were computed to examine the hypothesized associations between variables.  2.4.3 Test of the 2 x 2 model of dispositional perfectionism One purpose of this study was to test the 2 x 2 model of dispositional perfectionism to see whether its four hypotheses (Gaudreau & Thompson, 2010) were supported in this sample of university athletes. To test the predictive value of perfectionism on burnout, moderated hierarchical regression following Cohen, Cohen, West, and Aiken’s (2003) procedure was used as recommended by Gaudreau and Thompson (Gaudreau, 2012; Gaudreau & Thompson, 2010). The first step was to   37 determine whether there was a statistically significant interaction between PSP and ECP (Gaudreau, 2012). First, PSP and ECP scores were mean centered. Then the interaction term was created by multiplying the centered PSP and ECP scores. The dependent variable, athletic burnout remained uncentered. The first step of the regression model included PSP and ECP as predictors (centered values) and athletic burnout as the dependent variable. The second step of the hierarchical regression model tested the moderation effect by including the interaction term (PSP x ECP) in the model. The R2 change between step 1 and step 2 was examined to determine whether the addition of the interaction term improved the fit of the model. Moreover, the statistical significance of the interaction term was observed. In the presence of a statistically significant interaction simple slope analysis would be examined. In the absence of a significant interaction between PSP and ECP, the main effects of PSP and ECP would be examined. The presence of an interaction between PSP and ECP is neither a necessary nor sufficient condition of the 2 x 2 model of perfectionism (Gaudreau, 2012). Moderated hierarchical models were tested for each time point.  The procedure to evaluate the four hypotheses of the 2 x 2 model of dispositional perfectionism in the absence of an interaction term has been discussed by Gaudreau (2012). The main effects of PSP and ECP help determine whether the different subtypes of perfectionism are associated with different level of burnout. More specifically, the regression equation is used to determine the predicted value associated with the four subtypes of perfectionism. Value for high PSP is determined by adding one standard deviation to the group average on PSP, whereas low PSP is obtained by subtracting one standard deviation from the average. The equations provided by Gaudreau (2012) are   38 presented in Figure 5. The predicted values of burnout are then presented in a graph to facilitate interpretation.  Figure 5 Equations to obtain predicted values of athletic burnout Note. Ŷ = predicted value of athletic burnout; bPSP = b-weight for PSP; bECP = b-weight for ECP; low = one standard deviation below the mean; high = one standard deviation above the mean.  2.4.4 Longitudinal analyses  Intraclass correlations were calculated for perfectionism, coping, and athletic burnout to determine whether a multilevel model was needed (Garson, 2013). The intraclass correlation measures the percentage of variance attributable to between-individual variance as opposed to within-individual variance. It was computed by dividing the variation between individual (τ) by the sum of τ and the variance (σ2). The intraclass correlation is always between zero and one (Garson, 2013). In this study the different data collections points were grouped at the individual level.  Growth curve modelling in HLM 7 student version was used to investigate the progression of the variables over time. This analysis provided information about the rate of change of the different variables. Before testing models that included a combination of variables, unconditional latent growth curve models were tested for each variable separately. This technique provides information on individual progression over time. The (1) Ŷ of Non-perfectionism = Intercept + (bPSP x lowPSP) + (bECP x lowECP) (2) Ŷ of Pure PSP = Intercept + (bPSP x highPSP) + (bECP x lowECP) (3) Ŷ of Pure ECP = Intercept + (bPSP x lowPSP) + (bECP x highECP) (4) Ŷ of Mixed Perfectionism = Intercept + (bPSP x highPSP) + (bECP x highECP)   39 unit of analysis is each individual, a growth curve is provided for every participant and their trajectory on a given variable is compared to other participants’ trajectories. Both linear and quadratic models were fitted to the data to determine which function best fit the observed data (see formulas 1.1 and 1.2 in Figure 6). Time was centred at baseline for both the linear model (Time 1 = 0; Time 2 = 1; Time 3 = 2; Time 4 = 3) and quadratic model (Time 1 = 0; Time 2 = 1; Time 3 = 4; Time 4 = 9). Figure 6 Linear and quadratic growth equations 2.4.5 Multilevel modelling and test of the mediation model Finally, multilevel modelling was used to test the relationships between the dimensions of perfectionism, coping dimensions, and athletic burnout. PSP and ECP were included in the analysis following one of the assumptions of the 2 x 2 model of dispositional perfectionism that stipulates both dimensions coexist to some extent in every individual (Gaudreau & Verner-Filion, 2012). Including only one dimension would be inconsistent with the tenets of the theoretical framework. Thus, the results of the moderated regression analysis in section 2.1.4.3 informed inclusion or exclusion of the interaction term in further analysis. 1.1 Linear model    Level-1 𝑌™ =   𝜋?? +   𝜋??(𝑇𝐼𝑀𝐸™ ) +   𝑒™     Level-2 𝜋?? =   𝛽℡ +   𝑟?? 𝜋?? =   𝛽™ +   𝑟?? 1.2 Quadratic model    Level-1 𝑌™ =   𝜋?? +   𝜋??(𝑇𝐼𝑀𝐸™ ) + 𝜋??(𝑇𝐼𝑀𝐸𝑆𝑄™ )  +  𝑒™     Level-2 𝜋?? =   𝛽℡ +   𝑟?? 𝜋?? =   𝛽™ +   𝑟?? 𝜋?? =   𝛽™ +   𝑟??    40 Analyses were conducted at two different levels, within-individual and between-individual. Analyses at the within-individual level provided information about the pattern of change around an individual’s mean score. For example, when athlete Y scores higher than his average on ECP does he also score higher than his personal average on athletic burnout? Analyses at the between-individual level provided information about the group trend. For example, athletes who have higher ECP scores on average might also have higher athletic burnout scores. It is important to look at the relationships at the within- and between-individual level because they answer different research questions; analyses at the within-individual level provided information about personal changes associated with an increase or decrease in the other variable variables, whereas analyses at the between-individual level provided information about where an athlete falls within the distribution on one variable based on their score on the other variables. The strength and direction of the relationships at the within- and between-individual level can differ.  The model presented in Figure 35 was tested with HLM.7 student version in three separate steps at the within-individual level (see Figure 7), followed by similar analyses at the between-individual level (see Figure 8). The predictor variables were entered at level-1 to test the model at the within-individual level, whereas, the predictors were entered at level-1 and level-2 when testing the model at the between-individual level. First, the total effects of PSP and ECP on athletic burnout were estimated (step I). Second, the effects of PSP and ECP on TOC were estimated (step II.A), as well as the effects of PSP and ECP on DOC (step II.B). Third, the effects of PSP, ECP, TOC, and DOC on athletic burnout were estimated simultaneously (step III). The parameters                                                 5 The analysis used does not allow for testing this exact model, a path between ECP and TOC was estimated even though we did not expect it to be statistically significant.    41 estimated in those three steps were then used to evaluate the mediating effect of coping dimensions between dimensions of perfectionism and athletic burnout.  Figure 7 Three step approach to test the within-individual multilevel model Note. Level-1 predictors are group mean centered; BO, athletic burnout; PSP, personal standard perfectionism; ECP, evaluative concerns perfectionism; TOC, task-oriented coping; DOC, disengagement oriented coping.    Step I   Level-1              𝐵𝑂™ =   𝜋?? +   𝜋??(𝑃𝑆𝑃™ ) +   𝜋??(𝐸𝐶𝑃™ ) + 𝑒™    Level-2             𝜋?? =   𝛽℡ +   𝑟??             𝜋?? =   𝛽™ +   𝑟??              𝜋?? =   𝛽™ +   𝑟?? Step II.A   Level-1              𝑇𝑂𝐶™ =   𝜋?? +   𝜋??(𝑃𝑆𝑃™ ) +   𝜋??(𝐸𝐶𝑃™ ) + 𝑒™    Level-2             𝜋?? =   𝛽℡ +   𝑟??             𝜋?? =   𝛽™ +   𝑟??              𝜋?? =   𝛽™ +   𝑟?? Step II.B   Level-1              𝐷𝑂𝐶™ =   𝜋?? +   𝜋??(𝑃𝑆𝑃™ ) +   𝜋??(𝐸𝐶𝑃™ ) + 𝑒™    Level-2             𝜋?? =   𝛽℡ +   𝑟??             𝜋?? =   𝛽™ +   𝑟??              𝜋?? =   𝛽™ +   𝑟?? Step III   Level-1              𝐵𝑂™ =   𝜋?? +   𝜋??(𝑃𝑆𝑃™ ) +   𝜋??(𝐸𝐶𝑃™ ) + 𝜋??(𝑇𝑂𝐶™ )  +  𝜋??(𝐷𝑂𝐶™ ) + 𝑒™    Level-2             𝜋?? =   𝛽℡ +   𝑟??             𝜋?? =   𝛽™ +   𝑟??              𝜋?? =   𝛽™ +   𝑟??             𝜋?? =   𝛽™ +   𝑟??             𝜋?? =   𝛽™ +   𝑟??     42  Figure 8 Three step approach to test the between-individual multilevel model Note. Level-1 predictors are group mean centered; level-2 predictors are grand mean centered; BO, athletic burnout; PSP, personal standard perfectionism; ECP, evaluative concerns perfectionism; TOC, task-oriented coping; DOC, disengagement oriented coping; AV, individual average across all four time points.  Simple mediation models were tested using Sobel’s approach (1982). This approach is preferred to the causal step approach (Baron & Kenny, 1986) that requires four conditions for mediation to be present. Although Baron & Kenny’s approach to Step I   Level-1              𝐵𝑂™ =   𝜋?? +   𝜋??(𝑃𝑆𝑃™ ) +   𝜋??(𝐸𝐶𝑃™ )+ 𝑒™    Level-2             𝜋?? =   𝛽℡ +   𝛽™ (𝑃𝑆𝑃𝐴𝑉?) + 𝛽™ (𝐸𝐶𝑃𝐴𝑉?)  +  𝑟??             𝜋?? =   𝛽™ +   𝑟??             𝜋?? =   𝛽™ +   𝑟??  Step II.A   Level-1              𝑇𝑂𝐶™ =   𝜋?? +   𝜋??(𝑃𝑆𝑃™ ) +   𝜋??(𝐸𝐶𝑃™ )+ 𝑒™    Level-2             𝜋?? =   𝛽℡ +   𝛽™ (𝑃𝑆𝑃𝐴𝑉?) + 𝛽™ (𝐸𝐶𝑃𝐴𝑉?)  +  𝑟??             𝜋?? =   𝛽™ +   𝑟??             𝜋?? =   𝛽™ +   𝑟?? Step II.B   Level-1              𝐷𝑂𝐶™ =   𝜋?? +   𝜋??(𝑃𝑆𝑃™ )+   𝜋??(𝐸𝐶𝑃™ ) + 𝑒™    Level-2             𝜋?? =   𝛽℡ +   𝛽™ (𝑃𝑆𝑃𝐴𝑉?) + 𝛽™ (𝐸𝐶𝑃𝐴𝑉?)  +  𝑟??             𝜋?? =   𝛽™ +   𝑟??             𝜋?? =   𝛽™ +   𝑟?? Step III   Level-1              𝐵𝑂™ =   𝜋?? +   𝜋??(𝑃𝑆𝑃™ ) +   𝜋??(𝐸𝐶𝑃™ )+ 𝜋??(𝑇𝑂𝐶™ )  +  𝜋??(𝐷𝑂𝐶™ )+ 𝑒™    Level-2             𝜋?? =   𝛽℡ +   𝛽™ (𝑃𝑆𝑃𝐴𝑉?) + 𝛽™ (𝐸𝐶𝑃𝐴𝑉?) + 𝛽™ (𝑇𝑂𝐶𝐴𝑉?) + 𝛽™ (𝐷𝑂𝐶𝐴𝑉?)  +  𝑟??             𝜋?? =   𝛽™ +   𝑟??             𝜋?? =   𝛽™ +   𝑟??             𝜋?? =   𝛽™ +   𝑟??             𝜋?? =   𝛽™ +   𝑟??    43 mediation analysis remains the most used in the social science (Rucker, Preacher, Tormala, & Petty, 2011). Hayes (2013) outlines four reasons why it should be abandoned. First, the causal step approach fails to quantify the indirect effect and to infer its presence in the population of interest (Hayes, 2013). Second, three null hypotheses need to be rejected to even consider presence of a mediation effect. If any of those hypotheses are rejected it vetoes the mediation test (Hayes, 2013). Third, Baron and Kenny’s (1986) approach places too much emphasis on the relationship between the independent and dependent variable (Hayes, 2013; Rucker et al., 2011). Rucker and colleagues (2011) demonstrated in a simulation study that in about 50% of cases where there was no direct effect between the independent and dependent variable an indirect effect was detected. Hayes (2013) made a case that the predictor and outcome variable are not required to be associated in order to test for mediation. Fourth, the lack of quantification of the indirect effect encourages researchers to think of mediation effect in qualitative terms (Hayes, 2013). This fourth concern was echoed by Rucker and colleagues (2011) who reported problems with the use of terms such as full, complete, or perfect mediation versus partial mediation. They illustrated this problem with an example by Tormala, Falces, Briñol, and Petty (2007) where in the presence of a complete mediation, a second mediator was found to mediate the relationship between the independent and dependent variable. Consequently, Rucker and colleagues (2011) recommend that evaluations of total or partial mediation be based on theory and the magnitude of the indirect effect (Rucker et al., 2011). The indirect effect, or mediation effect, represented the impact of the predictor variable on the outcome variable through the mediator variable. This effect was evaluated   44 based on the product of path a and b (ab). Path a represents the effect of the independent variable on the mediating variable. Path b represents the effect of the mediator variable on the outcome variable. The model tested in Step II (A and B) provided coefficient estimates for path a. The combined model tested in Step III provided the values of path b. Furthermore, Sobel’s test (1982) allows for the computation of a standard error. Significance of the indirect effect was assessed with the ratio of ab and its standard error compared to a critical Z value. This technique also calls for a normal theory approach (Hayes, 2013), which provides a strategy to make mediation inference about the population.     45 Chapter 3: Results 3.1 Data screening  Before proceeding with the data analyses, the data set was screened for missing values. One hundred and fifty five athletes took part in the first phase of the study. One participant failed to provide an email address to be contacted for the following three phases of the study, this participant was removed from the analyses. An additional 29 athletes only completed the first phase of the study and therefore, they were also removed from the analyses because having only one data point does not allow for the examination of change over time. Data from one participant was removed from the fourth phase of the study because the participant answered ‘3’ for every item and completed the survey in less than three minutes and thus it was assumed the participant did not consider the questions. The final sample included 125 athletes who completed at least two phases of the study. Of the final 125 participants, 66 completed all four time points, 32 completed three time points, and 27 completed two time points (see Table 1). Table 1  Response patterns to the four phases of the study Time 1 Time 2 Time 3 Time 4 Assessments completed Number of participant 1 1 1 1 4 66 1 1 1 0 3 18 1 1 0 1 3 4 1 0 1 1 3 10 1 1 0 0 2 16 1 0 1 0 2 9 1 0 0 1 2 2 Note. 1 = participated; 0 = did not participate.  Missing data at the item level (< 1.0 %) was dealt with using median replacement at the person level. Only two cases involved more than one item missing for any given   46 subscale. In the first case, less than 50% of the items were missing for the subscale and median replacement was used. In the second case, more than 50% of the items were missing and the observed data of that subscale for the participant were removed from the analyses (Hawthorne & Elliott, 2005).  Data were screened for outliers. Of all the variables measured at the four time points, only one data point was found to be an outlier. One participant reported a score with a z-score above 3.29 on DOC at the second time point (i.e., 3.294), this data point was not removed from the analyses. The tests of the normality assumption revealed normal distribution of the data for all variables except DOC. Komogorov-Smirnov test was statistically non-significant for PSP, ECP, TOC, and burnout at all four time points, thus the data was normally distributed. Komogorov-Smirnov test for DOC at all four time points was significant indicating violation of the normality assumption. Other normality indicators confirmed this assessment: skewness of the data was low for PSP, ECP, TOC, and burnout (-0.229 to 0.423) but higher for DOC (0.611 to 1.002). Kurtosis was acceptable for all variables (-0.877 to 0.978). Visual examination of histogram, box plot, normal Q-Q plot, and normal probability plot resulted in congruent conclusion; PSP, ECP, TOC and burnout were normally distributed at all four time points, whereas DOC was not normally distributed. 3.2 Descriptive statistics  Descriptive statistics for perfectionism, coping, and burnout are presented in Table 2.      47 Table 2  Summary of descriptive statistics for study variables  Range M SD Skewness Kurtosis α N       Perfectionism        Personal standard perfectionism T1 1-5 3.68 0.61 0.14 -0.41 0.81 125 Personal standard perfectionism T2 1-5 3.58 0.61 -0.02 -0.67 0.82 108 Personal standard perfectionism T3 1-5 3.51 0.64 -0.23 0.08 0.84 104 Personal standard perfectionism T4 1-5 3.62 0.59 -0.22 0.10 0.83 83 Evaluative concerns perfectionism T1 1-5 2.94 0.79 0.04 -0.45 0.86 125 Evaluative concerns perfectionism T2 1-5 2.93 0.85 0.01 -0.88 0.90 108 Evaluative concerns perfectionism T3 1-5 2.91 0.89 -0.10 -0.64 0.92 104 Evaluative concerns perfectionism T4 1-5 3.03 0.90 -0.22 -0.73 0.92 83       Task-oriented coping        Thought control T1 1-5 3.18 0.81 -0.00 -0.41 0.61 125 Thought control T2 1-5 3.15 0.70 -0.04 -0.54 0.63 105 Thought control T3 1-5 3.14 0.79 -0.06 -0.31 0.73 101 Thought control T4 1-5 3.14 0.79 -0.05 -0.30 0.71 82 Mental imagery T1 1-5 3.09 0.94 0.06 -0.58 0.78 125 Mental imagery T2 1-5 3.15 0.81 -0.32 -0.26 0.70 105 Mental imagery T3 1-5 3.11 0.94 -0.21 -0.42 0.79 101 Mental imagery T4 1-5 3.16 0.91 -0.09 -0.32 0.82 82 Relaxation T1 1-5 2.95 1.04 -0.01 -0.84 0.85 125 Relaxation T2 1-5 2.90 0.91 -0.06 -0.48 0.83 105 Relaxation T3 1-5 2.92 0.94 -0.08 -0.65 0.89 101 Relaxation T4 1-5 2.86 0.98 0.14 -0.54 0.88 82 Effort expenditure T1 1-5 3.93 0.61 -0.14 -0.28 0.72 125 Effort expenditure T2 1-5 3.87 0.59 -0.21 0.09 0.68 105 Effort expenditure T3 1-5 3.84 0.66 -0.49 0.43 0.80 101 Effort expenditure T4 1-5 3.82 0.71 -0.40 0.30 0.78 82 Logical analysis T1 1-5 3.24 0.83 -0.37 -0.04 0.59 125 Logical analysis T2 1-5 3.06 0.87 0.01 -0.58 0.68 105 Logical analysis T3 1-5 2.93 0.84 -0.01 -0.75 0.66 101 Logical analysis T4 1-5 3.01 0.77 0.09 -0.48 0.57 82    48  Range M SD Skewness Kurtosis α N Seeking support T1 1-5 2.82 1.03 0.18 -0.73 0.76 125 Seeking support T2 1-5 2.53 0.99 0.33 -0.73 0.81 105 Seeking support T3 1-5 2.63 1.04 0.20 -1.04 0.85 101 Seeking support T4 1-5 2.67 1.01 0.23 -0.72 0.85 82 Task-oriented coping T1 1-5 3.20 0.62 -0.02 -0.16 0.80 125 Task-oriented coping T2 1-5 3.11 0.55 0.03 0.95 0.75 105 Task-oriented coping T3 1-5 3.10 0.63 -0.15 -0.46 0.81 101 Task-oriented coping T4 1-5 3.11 0.61 0.20 0.28 0.79 82       Disengagement-oriented coping        Disengagement/resignation T1 1-5 1.71 0.77 1.42 1.85 0.79 125 Disengagement/resignation T2 1-5 1.65 0.81 1.51 2.13 0.85 105 Disengagement/resignation T3 1-5 1.68 0.87 1.46 1.40 0.89 101 Disengagement/resignation T4 1-5 1.76 0.81 1.11 0.15 0.86 82 Venting of emotion T1 1-5 2.61 0.95 0.25 -0.62 0.78 125 Venting of emotion T2 1-5 2.46 0.94 0.41 -0.82 0.84 105 Venting of emotion T3 1-5 2.36 0.95 0.52 -0.73 0.83 101 Venting of emotion T4 1-5 2.53 0.94 0.39 -0.90 0.84 82 Disengagement oriented coping T1 1-5 2.16 0.71 0.61 0.50 0.53 125 Disengagement oriented coping T2 1-5 2.06 0.78 0.85 0.31 0.74 105 Disengagement oriented coping T3 1-5 2.02 0.81 1.00 0.40 0.74 101 Disengagement oriented coping T4 1-5 2.15 0.79 0.60 -0.86 0.76 82       Athletic burnout        Reduced accomplishment T1 1-5 2.50 0.83 0.26 -0.31 0.83 125 Reduced accomplishment T2 1-5 2.49 0.82 0.34 -0.54 0.85 104 Reduced accomplishment T3 1-5 2.52 0.84 0.40 -0.43 0.86 102 Reduced accomplishment T4 1-5 2.63 0.81 0.30 -0.44 0.87 82 Emotional/physical exhaustion T1 1-5 2.88 1.01 0.31 -0.59 0.92 125 Emotional/physical exhaustion T2 1-5 2.89 0.91 0.25 -0.10 0.93 105 Emotional/physical exhaustion T3 1-5 2.91 0.90 0.18 -0.46 0.92 102 Emotional/physical exhaustion T4 1-5 3.05 0.91 -0.15 -0.59 0.94 82       49   Range M SD Skewness Kurtosis α N Sport devaluation T1 1-5 2.02 0.90 0.92 0.21 0.83 125 Sport devaluation T2 1-5 2.10 0.92 0.71 -0.32 0.90 105 Sport devaluation T3 1-5 2.24 1.01 0.71 -0.42 0.92 102 Sport devaluation T4 1-5 2.25 1.02 0.64 -0.48 0.93 82 Burnout total T1 1-5 2.47 0.73 0.42 -0.21 0.72 125 Burnout total T2 1-5 2.49 0.72 0.29 -0.30 0.74 104 Burnout total T3 1-5 2.56 0.76 0.29 -0.49 0.76 102 Burnout total T4 1-5 2.64 0.74 0.12 -0.98 0.72 82 Note. T1 = time one; T2 = time two; T3 = time three; T4 = time four. M = mean; SD = standard deviation; α = Cronbach’s alpha; N = number of participant.  The means for PSP and ECP are consistent with values reported in other studies with athletes of similar age and level of sport involvement (Crocker, et al., in press; Dunn et al., 2006). Athletic burnout in this study was higher than that reported in junior male soccer players (Hill, 2013). Task-oriented coping and disengagement-oriented coping were not compared to values obtained in different studies because a modified version of the CICS was used in this study. Bivariate correlations (see Table 3) between the variables of interest were calculated using SPSS version 20.0.   50 Table 3  Summary of bivariate correlations for scores of study variables at each assessment   Time 1 Time 2 Time 3 Time 4   1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 Time 1 1. PSP --                    2. ECP .46** --                   3. TOC .15 -.08 --                  4. DOC .31** .45** .15 --                 5. Burnout .04 .39** -.16 .53** --                Time 2 1. PSP .78** .50** .03 .41** .17 --               2. ECP .47** .86** -.04 .48** .46** .54** --              3. TOC .25* .07 .57** .14 -.01 .24* .07 --             4. DOC .36** .45** .18 .61** .39** .44** .60** .32** --            5. Burnout .17 .45** .01 .52** .73** .12 .48** .00 .57** --           Time 3 1. PSP .69** .42** .06 .27** .03 .79** .50** .18 .35** .14 --          2. ECP .40** .86** -.10 .47** .42** .49** .90** .00 .54** .55** .48** --         3. TOC .16 .00 .66** .19 -.15 .13 -.03 .75** .05 -.13 .22* .03 --        4. DOC .22* .38** .11 .56** .41** .27* .48** .06 .64** .50** .15 .53** .18 --       5. Burnout .07 .42** -.07 .52** .72** .12 .49** -.04 .59** .83** -.04 .51** -.10 .62** --      Time 4 1. PSP .75** .32** .06 .28** .09 .78** .41** .26* .29* .07 .82** .44** .31** .27* .02 --     2. ECP .45** .85** -.14 .52** .45** .53** .89** .03 .55** .51** .49** .91** .05 .49** .47** .44** --    3. TOC .19 .04 .65** .18 -.11 .17 .05 .58** .16 -.03 .34** .05 .74** .18 -.07 .34** .04 --   4. DOC .35** .39** .20 .58** .30** .43** .43** .38** .78** .49** .34** .46** .33** .63** .49** .41** .52** .20 --  5. Burnout .19 .40** .00 .52** .69** .25* .45** .04 .54** .75** .09 .50** .00 .46** .83** .11 .47** -.13 .57** -- Note. PSP = personal standard perfectionism; ECP = evaluative concerns perfectionism; TOC = task-oriented coping; DOC = disengagement-oriented coping. * = p < .05; ** = p < .01.  51 Personal standard perfectionism had large positive correlations with ECP at each of the time points6. PSP had small significantly positive correlations with TOC at the second, third, and fourth time points but was not significantly correlated with TOC at the first time point. PSP had significant moderate positive correlations with DOC at the first, second, and fourth time points, and had a small non-significant positive correlation with DOC at the third time point. ECP was not significantly related to TOC and had significant moderate positive correlations with DOC at Time 1 and significant large positive correlations at Time 2, 3, and 4. Burnout was not significantly related to PSP and had moderate positive correlations with ECP at Time 1, 2, and 4, and large positive correlation at Time 3. Moreover, athletic burnout was consistently not significantly related to TOC and had large positive correlations with DOC.  Bivariate correlations reported in Table 3 provide preliminary evidence of relative stability of the relationships between some variables over time. Within each dimension of perfectionism, the reported scores were strongly correlated between time points, there was a trend indicating slightly higher correlations between proximal time points than between distal time points (i.e., T1 with T2 showed higher correlations than T1 with T3). Correlations between ECP at time 1 and DOC at the different time points were similar (i.e., 0.45; 0.45; 0.38; 0.39). This pattern also existed for the correlations between ECP at time 1 and burnout at all four time points, as well as between DOC reported at the first time point and burnout reported throughout the four phases of the study.                                                 6 Effect size of the correlation coefficients were evaluated based on Cohen’s (1988) guidelines: 0.1 = small; 0.3 = moderate; 0.5 = large.   52 3.3 Test of the 2 x 2 model of dispositional perfectionism   The results of the moderated hierarchical regression models are presented in Table 4. The interaction term (PSP x ECP) was not statistically significant in any of the four moderated hierarchical regression models tested. The interaction term was removed from the models in any further analyses on the basis of parsimony and new models were tested with uncentered scores of PSP and ECP as predictors of athletic burnout (see appendix J). Simple main effects were examined to test the four hypotheses of the 2 x 2 model of dispositional perfectionism. Evaluative concerns perfectionism was consistently a significant positive predictor of athletic burnout across all time points. For example, higher self-reported ECP at Time 1 predicted higher athletic burnout at Time 1 (appendix J). The predictive value of PSP was less consistent. At Time 1 and Time 3, PSP was a statistically significant negative predictor of athletic burnout, such that higher PSP scores were associated with lower burnout scores. There was a main effect of PSP at Time 1 and Time 3, although, at Time 2 and Time 4, PSP was not a statistically significant predictor of athletic burnout. Next, imputation procedure for missing data was used to help determine whether this finding could be explained by the loss of statistical power associated with reduced sample size at Time 2, 3, and 4.      53 Table 4  Results of the moderated regression analyses predicting athletic burnout Effects  t b p-value Time 1        Step 1 R2 = 0.181  F(2, 122) = 13.492, p < 0.001          PSP  -1.998 -0.223 0.048       ECP  5.176 0.446 0.000    Step 2 Δ R2 = 0.000 ΔF(1, 121) = 0.002, p = 0.967          PSP x ECP  0.041 0.005 0.967 Time 2        Step 1 R2 = 0.254 F(2, 101) = 17.221, p < 0.001          PSP  -1.978 -0.243 0.051       ECP  5.695 0.497 0.000    Step 2 Δ R2 = 0.005 ΔF(1, 100) = 0.669, p = 0.415          PSP x ECP  0.818 0.100 0.415 Time 3        Step 1 R2 = 0.373 F(2, 98) = 29.189, p < 0.001          PSP  -4.124 -0.437 0.000       ECP  7.620 0.592 0.000    Step 2 Δ R2 = 0.015 ΔF(1, 97) = 2.362, p = 0.128          PSP x ECP  -1.537 -0.144 0.128 Time 4        Step 1 R2 = 0.236 F(2, 79) = 12.199, p < 0.001          PSP  -1.108 -0.153 0.271       ECP  4.800 0.429 0.000    Step 2 Δ R2 = 0.029 ΔF(1, 78) = 3.128, p = 0.081          PSP x ECP  -1.769 -0.238 0.081  Note. PSP = personal standard perfectionism; ECP = evaluative concerns perfectionism. t = t-test value for the b-weight; b = unstandardized b-weight; p-value = p-value at the 95% confidence interval.     54 Multiple imputation is a technique used to handle missing data. It creates M >1 complete sets of data using imputation strategies to fill in the missing data points (Collins, Schafer, & Chi-Ming, 2001). Multiple imputation creates M equivalent sets of possible values based on the information available. Auxiliary variables such as gender or age can be included when generating the M data sets to help inform imputation. Typical multiple imputation applications involve the creation of 5 to 10 equivalent data sets (Collins et al., 2001). Analyses are run on M data sets and results are then averaged and reported as pooled results.   The multiple imputation technique in SPSS was used to create five equivalent data sets with complete data to preserve the power of the complete sample size. Age and gender were included in the multiple imputation as auxiliary variables. The other variables included in the multiple imputation were PSP, ECP, as well as the three dimensions of burnout and total burnout at each phase of the study. Hierarchical regressions were tested with the data after multiple imputations (see Table 5). Results revealed a statistically significant main effect of ECP across all time points and statistically significant main effects of PSP at Time 1, 2, and 3 but only marginally significant main effect of PSP at Time 4. The next step involved evaluating the four hypotheses of the 2 x 2 model of dispositional perfectionism. The equations provided by Gaudreau (2012) were used to test the hypotheses2 (see Figure 2) and the predicted values of athletic burnout were                                                 2 SPSS does not provide standard deviations values for the pooled data. Standard deviations were calculated with the standard error of the mean provided based on the following formula, where N represent the total sample size (Altman & Bland, 2005):  𝑠𝑒™?? = ™ ?    or   𝑆𝐷 = 𝑠𝑒™?? ∗    𝑁   55 presented in Figure 9. At Time 1, 2, and 3, pure PSP was associated with statistically significant lower athletic burnout than non perfectionism (H1a). At Time 4, pure PSP was Table 5  Hierarchical regression models for athletic burnout with multiple imputations approach Effects  t b p-value Time 1 R2 = 0.181, F(2, 122) = 13.492, p < 0.001       PSP  -2.00 -0.22 0.046    ECP  5.18 0.45 <0.001 Time 2 R2 = 0.228, F(2, 122) = 18.280, p < 0.001       PSP  -2.49 -0.34 0.017    ECP  4.78 0.51 <0.001 Time 3 R2 = 0.353, F(2, 122) = 33.518, p < 0.001       PSP  -4.46 -0.45 <0.001    ECP  7.57 0.59 <0.001 Time 4 R2 = 0.193, F(2, 122) = 14.571, p < 0.001       PSP  -1.84 -0.23 0.071    ECP  5.48 0.40 <0.001 Note. PSP = personal standard perfectionism; ECP = evaluative concerns perfectionism. t = t-test value for the b-weight; b = unstandardized b-weight; p-value = p-value at the 95% confidence interval. N = 125. SPSS version 20.0 does not provide R2 and F values for the hierarchical regression completed on the pooled analyses (time 2-4). The R2 and F values reported in the table for time 2-4 were obtained by calculating the average of the analyses on the five imputed datasets.   associated with marginally lower levels of athletic burnout than non perfectionism (H1a). At all four time points, pure ECP was associated with statistically significant higher levels of athletic burnout than non perfectionism (H2). At Time 1, 2, and 3, pure ECP was associated with statistically higher athletic burnout than mixed perfectionism (H3). Burnout level differed marginally between pure ECP and mixed perfectionism at the fourth time point (H3). Across all four time points, pure PSP was associated with lower burnout than mixed perfectionism (H4).      56   Figure 9 Predicted values of athletic burnout using the main effect of PSP and ECP Note. PSP = personal standard perfectionism; ECP = evaluative concerns perfectionism. Solid lines represent high PSP, dashed lines represent low PSP. Main effect of ECP are displayed horizontally whereas main effect of PSP are displayed vertically.  3.4 Longitudinal analyses  The longitudinal analyses examined within-individual as well as between-individual change over time. First, the intraclass correlation was calculated for all of the variables (see Table 6). The intraclass correlation for athletic burnout indicated that 78.2% of the variance was associated to inter-individual differences (between-  57 individual), and 21.8% was associated to intra-individual differences (within-individual). The intraclass correlation between zero and one indicated a need for the data to be grouped at the individual level, thus, taking into consideration the dependence of the data provided by the same individuals. The intraclass correlation for PSP was similar to athletic burnout, revealing a comparable ratio of intra- to inter-person variance. The intraclass correlation of ECP showed that there was less variation within the individuals, and that most of the difference in scores resided between the athletes. Both coping dimensions had intraclass correlation coefficients reflecting a more balanced divide between intra- and inter-individual differences.  Table 6  Intraclass correlations for study variables Variable ICC Burnout 0.782  PSP 0.767 ECP 0.867 TOC 0.653 DOC 0.607 Note. ICC = intraclass correlation; PSP = personal standard perfectionism; ECP = evaluative concerns perfectionism; TOC = task-oriented coping; DOC = disengagement-oriented coping.   3.4.1 Linear growth models Results from the unconditional growth model revealed that athletes’ initial burnout scores were different from zero (Table 7) and there were individual differences in their initial scores, as reflected by the random effect of the intercept (see Table 8). Burnout showed a statistically significant linear increase over time (i.e., slope). Three level-2 predictors were tested separately, age, PSP, and ECP (see Appendix J). Neither   58 age nor PSP had an influence on the initial burnout value or on the rate of change in burnout. ECP was the only predictor that had an influence on athletic burnout, it was associated with higher burnout values at the beginning of the season, but it did not influence the rate of change in burnout.  Athletes reported initial PSP and ECP scores that were different from zero (i.e., intercepts were significant). PSP showed a statistically significant decrease over time (i.e., slope) when a linear model was fitted to the data whereas ECP did not significantly increase or decrease following a linear trajectory over the four time points (i.e., slope was non-significant). Both PSP and ECP showed between person variation (i.e., random effects were significant). In terms of coping strategies, participants scores on TOC and DOC had initial values significantly different than zero (i.e., intercepts were significant) that revealed inter-individual differences on participants initial use of TOC and DOC (i.e., random effects were significant). There was a statistically significant linear decrease in the use of TOC over the season (i.e., negative slope was significant). The use of DOC did not significantly follow a linear trajectory (i.e., slope was non-significant). There was non-significant inter-individual variation in the reported use of DOC (i.e., random effect was non-significant). The linear decrease in TOC varied between athletes (i.e., random effect was significant when evaluated with a less stringent alpha level, p = 0.15, to palliate the low statistical power of the random effect as recommended by Nezlek, 2012).     59 Table 7 Unconditional growth models: fixed effects  Variable  Coefficient Standard error t-ratio DF p-value Burnout Intercept 𝛽℡  2.469 0.063 39.199 123 <0.001 TimeSlope 𝛽™  0.062 0.017 3.607 123 <0.001 PSP Intercept 𝛽℡  3.650 0.053 68.383 124 <0.001 TimeSlope 𝛽™  -0.047 0.016 -2.929 124 0.004 ECP Intercept 𝛽℡  2.930 0.069 42.603 124 <0.001 TimeSlope 𝛽™  0.008 0.016 0.498 124 0.619 TOC Intercept 𝛽℡  3.171 0.051 62.200 124 <0.001  TimeSlope 𝛽™  -0.038 0.017 -2.247 124 0.026 DOC Intercept 𝛽℡  2.121 0.061 34.882 124 <0.001  TimeSlope 𝛽™  -0.025 0.022 -1.138 124 0.257 Note. PSP = personal standard perfectionism; ECP = evaluative concerns perfectionism; TOC = task-oriented coping; DOC = disengagement-oriented coping. The fixed effects estimates reported are with robust standard errors.  Table 8 Unconditional growth models: random effects   SD Variance component DF 𝒳? p-value Burnout Intercept 𝑟? 0.647 0.419 122 758.021 <0.001 TimeSlope 𝑟? 0.093 0.009 122 163.448 0.007 PSP Intercept 𝑟? 0.549 0.301 123 727.380 <0.001  TimeSlope 𝑟? 0.100 0.010 123 210.104 <0.001 ECP Intercept 𝑟? 0.731 0.534 123 1173.452 <0.001  TimeSlope 𝑟? 0.092 0.008 123 183.665 <0.001 TOC Intercept 𝑟? 0.487 0.237 123 436.724 <0.001  TimeSlope 𝑟? 0.058 0.003 123 145.987 0.077 DOC Intercept 𝑟? 0.546 0.299 123 322.906 <0.001  TimeSlope 𝑟? 0.046 0.002 123 127.127 0.381 Note. PSP = personal standard perfectionism; ECP = evaluative concerns perfectionism; TOC = task-oriented coping; DOC = disengagement-oriented coping.      60 3.4.2 Quadratic growth models Analyses of the potential quadratic nature of the change in burnout, ECP, and TOC over time did not reveal quadratic relationships (Table 9). On the other hand, the PSP trajectory was consistent with a quadratic model; the parabola had a negative linear component and opened upward. Disengagement-oriented coping also had a significant quadratic component to its trajectory; the parabola had a negative linear component and opened upward. The quadratic curve may account for part of the intra-individual variation in DOC that could not be explained by a linear model. Both quadratic models (PSP and DOC) had individual variance in the initial values, but the growth curve was similar across individuals (Table 10). Table 9 Quadratic growth model: fixed effects Variable  Coefficient Standard error t-ratio DF p-value Burnout Intercept 𝛽℡  2.468 0.065 37.973 124 <0.001 Time Slope 𝛽™  0.083 0.057 1.454 124 0.149  TimeSQ Slope 𝛽™  -0.009 0.018 -0.520 124 0.604 PSP Intercept 𝛽℡  3.685 0.054 67.988 124 <0.001  Time Slope 𝛽™  -0.175 0.046 -3.776 124 <0.001  TimeSQ Slope 𝛽™  0.045 0.014 3.185 124 0.002 ECP Intercept 𝛽℡  2.936 0.070 42.088 124 <0.001  Time Slope 𝛽™  -0.014 0.046 -0.293 124 0.770  TimeSQ Slope 𝛽™  0.008 0.015 0.525 124 0.600 TOC Intercept 𝛽℡  3.196 0.054 58.772 124 <0.001  Time Slope 𝛽™  -0.120 0.058 -2.068 124 0.041  TimeSQ Slope 𝛽™  0.029 0.019 1.531 124 0.128 DOC Intercept 𝛽℡  2.157 0.063 34.029 124 <0.001  Time Slope 𝛽™  -0.154 0.072 -2.132 124 0.035  TimeSQ Slope 𝛽™  0.046 0.023 2.030 124 0.045 Note. PSP = personal standard perfectionism; ECP = evaluative concerns perfectionism; TOC = task-oriented coping; DOC = disengagement-oriented coping. The fixed effects estimates reported are with robust standard errors.    61 Table 10 Quadratic growth model: random effects   SD Variance component DF 𝒳? p-value Burnout Intercept 𝑟? 0.675 0.455 97 619.821 <0.001 TimeSlope 𝑟? 0.417 0.174 97 180.061 <0.001  TimeSqSlope  𝑟? 0.117 0.014 97 160.994 <0.001 PSP Intercept 𝑟? 0.557 0.310 100 509.349 <0.001  TimeSlope 𝑟? 0.295 0.087 100 116.593 0.123  TimeSqSlope  𝑟? 0.071 0.005 100 102.302 0.417 ECP Intercept 𝑟? 0.736 0.541 100 858.581 <0.001  TimeSlope 𝑟? 0.238 0.056 100 116.174 0.129  TimeSqSlope  𝑟? 0.054 0.003 100 108.184 0.271 TOC Intercept 𝑟? 0.538 0.290 97 394.904 <0.001  TimeSlope 𝑟? 0.4130 0.171 97 144.034 0.002  TimeSqSlope  𝑟? 0.126 0.016 97 141.440 0.002 DOC Intercept 𝑟? 0.557 0.310 97 267.637 <0.001  TimeSlope 𝑟? 0.316 0.100 97 104.412 0.285  TimeSqSlope  𝑟? 0.078 0.006 97 93.280 >0.500 Note. PSP = personal standard perfectionism; ECP = evaluative concerns perfectionism; TOC = task-oriented coping; DOC = disengagement-oriented coping.  3.5 Multilevel modelling and test of the mediation models  3.5.1 Multilevel modelling at the within-individual level  Step I: In this step a multilevel model was tested to investigate the effect of PSP and ECP as level-1 predictors on athletic burnout at the within-individual level. The model explained 18.5% of the variance in burnout (R2 = 0.185) experienced by athletes. Personal standards perfectionism was a statistically significant negative predictor of athletic burnout (𝛽™  = -0.194, se = 0.078, t (124) = -2.479, p = 0.015) and there were individual differences in the influence of PSP on burnout (𝑟? = 0.135, SD = 0.367, 𝒳? (93) = 123.861, p = 0.018). Thus, on days when athletes scored higher than their personal average on PSP, they scored lower than their average on burnout. Moreover, the extent to   62 which higher PSP scores are linked to lower burnout score varied between individuals. Evaluative concerns perfectionism was a statistically significant positive predictor of burnout (𝛽™  = 0.217, se = 0.072, t (124) = 3.015, p = 0.003) and the predictive effect of ECP on burnout was similar across individuals (𝑟? = 0.066, SD = 0.257, 𝒳? (93) = 97.169, p = 0.363). On average, athletes scored 0.217 units higher on burnout when they scored one unit higher than their personal average on ECP.  Step II.A: The model tested investigated the effect of both dimensions of perfectionism on TOC. The model explained 10.1% of the intra-individual variance in TOC (R2 = 0.101). Personal standards perfectionism was a positive predictor of TOC (𝛽™  = 0.260, se = 0.072, t (124) = 3.593, p < 0.001); the effect of PSP on TOC was similar across individuals (𝑟? = 0.054, SD = 0.233, 𝒳? (94) = 82.297, p > 0.500). Athletes tend to report an increase of 0.260 units in TOC on days where they score one PSP unit higher than their personal average. Evaluative concerns perfectionism was not a significant predictor of TOC (𝛽™  = -0.072, se = 0.068, t (124) = -1.054, p = 0.294) and this was consistent for most athletes (𝑟? = 0.019, SD = 0.136, 𝒳?2 (94) = 82.297, p > 0.500). This means that ECP did not help explain significant within-individual variation in TOC.  Step II.B: This model tested the effect of PSP and ECP on DOC. The model explained 12.7% of the intra-individual variance in DOC (R2 = 0.127). Disengagement-oriented coping was not predicted significantly by PSP (𝛽™  = 0.013, se = 0.085, t (124) = 0.149, p = 0.882) and there were no individual differences in this effect (𝑟? = 0.063, SD = 0.252, 𝒳? (94) = 85.326, p > 0.500). On the other hand, ECP was a statistically significant positive predictor of DOC (𝛽™  = 0.385, se = 0.094, t (124) = 4.087, p < 0.001) and this effect was constant across individuals (𝑟? = 0.085, SD = 0.292, 𝒳? (94) =   63 108.008, p = 0.153). Thus, on days when athletes reported higher ECP than their personal average they typically scored higher on DOC than usual, the magnitude of this effect was similar between participants.  Step III: The combined model3 that included four level-1 predictors, PSP, ECP, TOC, and DOC, accounted for 37.9% of the variance in athletic burnout (R2 = 0.379). In this model PSP, TOC, and DOC were significant predictors of athletic burnout (see Table 11). When athletes scored higher than their personal average on either PSP or TOC, they reported lower burnout levels than usual. On the other hand, reporting higher DOC than their average was associated with higher burnout. The random effects revealed small individual differences (see Table 12). The fixed effects of step I-III are also reported in Figure 10.  Table 11 Estimates of fixed effects of the level-1 predictors Fixed effect Coefficient Standard error t-ratio DF p-value Intercept 𝛽℡  2.550 0.062 42.274 124 <0.001 PSP Slope 𝛽™  -0.144 0.066 -2.186 124 0.031 ECP Slope 𝛽™  0.069 0.066 1.056 410 0.292* TOC Slope 𝛽™  -0.169 0.062 -2.728 124 0.007 DOC Slope 𝛽™  0.274 0.050 5.435 124 <0.001 Note. PSP = personal standard perfectionism; ECP = evaluative concerns perfectionism; TOC = task-oriented coping; DOC = disengagement oriented coping. * indicates a rough approximation. The fixed effects estimates reported are with robust standard errors.                                                    3 The original model (see Figure 10) was unable to converge because every level-1 predictor matrix was near singular. Four models were tested, in each model the random effect of one of the level-1 predictors was removed. The estimates of four models were compared and found to be similar. It was decided that the model reported would be the one where the random effect of ECP was removed to allow the model to converge on a solution. This decision was based on the consistently non-significant random effect of ECP.    64 Table 12 Random effects of the model including PSP, ECP, TOC, and DOC as level-1 predictor of burnout  SD Variance component DF 𝒳? p-value Intercept 𝑟? 0.674 0.454 62 1168.815 <0.001 PSP Slope 𝑟? 0.207 0.043 62 59.007 >0.500 TOC Slope 𝑟? 0.260 0.067 62 67.216 0.303 DOC Slope 𝑟? 0.241 0.058 62 70.803 0.207 Note. PSP = personal standard perfectionism; TOC = task-oriented coping; DOC = disengagement oriented coping.   3.5.2 Mediation at the within-individual level  Based on the results of step I-III three mediation effects were tested. The first one was the mediation effect of TOC on the relationship between PSP and burnout. The second was the mediation by DOC on the relationship between ECP and burnout. The third one was the mediation effect of DOC on the relationship between PSP and burnout.  The first mediator tested was TOC. Following Rucker and colleagues’ (2011) suggestion, the mediation effect was evaluated with ab and Sobel’s test (1982). The coefficients involved in the test of mediation were obtained from Step II.A and Step III. Path a was the predictive effect of PSP on TOC. Path b was the predictive effect of TOC on burnout. The relationship between PSP and athletic burnout was mediated by TOC (ab = -0.040, Zab = -2.076, p = 0.038). When athletes report higher levels of PSP than their average, they tend to report greater use of TOC, which partially accounts for the decrease in burnout compared to the athletes’ personal average.  The second mediator to be evaluated was DOC. Path a was obtained in Step II.B, it represented the relationship between ECP and DOC. The global model tested in Step III revealed that DOC was a predictor of athletic burnout (path b). The relationship   65 between ECP and athletic burnout was mediated by DOC (ab = 0.110, Zab = 3.050, p = 0.002). This means that on days where athletes report higher ECP than their personal average, they tend to also report higher use of DOC, which result in increased burnout level compared to their typical average.  The third mediation effect at the within-individual level was that of DOC on the relationship between PSP and athletic burnout. Step II.B evaluated the effect of PSP on DOC (path a). Path b, or the effect of DOC on burnout, was obtained in the test of the general model in Step III. There was no significant mediation effect of DOC on the relationship between PSP and burnout at the within-individual level (ab = 0.003, Zab = 0.149, p = 0.882).  66  Figure 10 Within- and between-individual effects Note. The dotted lines represent statistically non-significant (p > .05) associations. The values reported are unstandardized fixed effects with robust standard errors. The double arrow represents a correlation.   67 3.5.3 Multilevel modelling at the between-individual level Step I: At the inter-individual level the model including both dimensions of perfectionism explained 28.9% of the variance in athletic burnout (R2 = 0.289). At level-2, PSP was a statistically significant negative predictor of athletic burnout (𝛽™  = -0.372, se = 0.106, t (122) = -3.501, p < 0.001). This means that athletes who scored lower in burnout typically reported higher levels of PSP. The other dimension of perfectionism, ECP, was a statistically significant positive predictor of burnout (𝛽™  = 0.548, se = 0.074, t (122) = 7.430, p < 0.001). Thus, the athletes who scored high on ECP also reported high burnout symptoms.  Step II.A: The next model examined the influence of PSP and ECP on TOC at the between individual level. The model explained 3.8% of the inter-individual variance in TOC (R2 = 0.038). Personal standards perfectionism was a positive predictor of TOC (𝛽™  = 0.300, se = 0.101, t (122) = 2.970, p = 0.004). The athletes who reported higher levels of PSP were likely to also report higher levels of TOC. On the other hand, ECP was not a significant predictor of TOC (𝛽™  = -0.090, se = 0.069, t (122) = -1.302, p = 0.195). This means that only PSP, not ECP, had a significant influence on the reported use of TOC, but this effect was weak.  Step II.B: A large portion of the between-individual variance in DOC was explained by the model that investigated the effect of PSP and ECP on DOC (R2 = 0.381). Disengagement-oriented coping was not predicted by PSP (𝛽™  = 0.107, se = 0.091, t (122) = 1.170, p = 0.224). On the other hand, ECP was a statistically significant positive predictor of DOC (𝛽™  = 0.422, se = 0.068, t (122) = 6.236, p < 0.001). Generally athletes who report high levels of ECP also report frequent use of DOC.     68  Step III: The combined model4 that included PSP, ECP, TOC, and DOC as level-1 and level-2 predictors revealed that level-2 PSP, ECP, TOC, and DOC were significant predictors of athletic burnout and accounted for 49.8% of the variance in burnout (see Table 13). Two level-2 predictors, PSP and TOC, were negatively associated with burnout. Athletes who score low on either or both TOC and PSP tend to report higher levels of athletic burnout than athletes who score high on either or both TOC and PSP. On the other hand, ECP and DOC were positive predictors of burnout. Those who scored high in athletic burnout also scored high in DOC and ECP. The fixed effects at the between-individual level, step I-III, are also reported in Figure 10.   Table 13 Estimates of fixed effects of the level-2 predictors Fixed effect Coefficient Standard error t-ratio DF p-value Intercept 𝛽℡  2.55 0.043 59.003 120 <0.001 PSP 𝛽™  -0.367 0.086 -4.283 120 <0.001 ECP 𝛽™  0.281 0.075 3.754 120 <0.001 TOC 𝛽™  -0.225 0.086 -2.628 120 <0.001 DOC 𝛽™  0.631 0.090 6.998 120 <0.001 Note. PSP = personal standard perfectionism; ECP = evaluative concerns perfectionism; TOC = task-oriented coping; DOC = disengagement-oriented coping. The fixed effects estimates reported are with robust standard errors.  3.5.4 Mediation at the between-individual level The same three mediation models that were tested at the intra-individual level were tested at the inter-individual level. The first mediator model to be tested was the effect of TOC on the relationship between PSP and athletic burnout. Path a was obtained                                                 4 The original model (see Figure 10) was unable to converge because every level-1 predictor matrix was near singular. Level-1 ECP was fixed to allow the model to converge; this was the same modification that was made in the analyses at the intra-individual level.     69 through Step II.A, the effect of PSP on TOC. Step III provided the coefficient for path b that represented the effect of TOC on burnout. The mediation effect of TOC on the relationship between PSP and burnout at the between-individual level was marginally statistically significant (ab = -0.064, Zab = 1.930, p = 0.054). The second mediator to be tested was DOC. Step II.B provided the coefficient for the effect between ECP and DOC (path a). The model tested in Step III involved path b, or the relationship between DOC and athletic burnout. At the between-individual level, DOC was a statistically significant mediator of the relationship between ECP and athletic burnout (ab = 0.270, Zab = 4.721, p < 0.001). This means that athletes who typically report high levels of ECP also report frequent use of DOC and this is associated with reported high athletic burnout. The third mediation model tested was the effect of DOC on the relationship between PSP and burnout. Path a was obtained from Step II.B where the effect of PSP on DOC was estimated. The general model estimating the effect of PSP, ECP, TOC, and DOC on athletic burnout provided the coefficient for path b. At the between-individual level DOC was not a statistically significant mediator of the relationship between PSP and burnout (ab = 0.067, Zab = 1.156, p = 0.248).   70 Chapter 4: Discussion 4.1 Discussion  The main purpose of this study was to evaluate the potential mediating role of coping in the relationship between dimensions of perfectionism and burnout in university level varsity athletes. A longitudinal approach was used to look at the interplay of these variables over time. The results indicated that the negative relationship between personal standards perfectionism (PSP) and burnout was mediated by task-oriented coping (TOC). Moreover, the positive association between evaluative concerns perfectionism (ECP) and burnout was mediated by disengagement-oriented coping (DOC). These findings were similar within- and between-individuals and highlighted the significant role that coping plays in the relationship between personality-like constructs and burnout in the sport domain.   Secondary purposes of this thesis were to track perfectionism, coping, and burnout over time and to test the predictions of the 2 x 2 model of perfectionism as it relates to athlete burnout. Longitudinal analyses revealed a significant linear increase in athletic burnout and a significant linear decrease in the use of TOC over a 20 week period. Quadratic curves with negative linear component and upward bend predicted the trajectory in both PSP and DOC during the period studied. On the other hand, ECP was stable during the study. Implications of these findings are discussed later in this section. Test of the hypotheses of the 2 x 2 model of perfectionism revealed that at the first three time points, ECP and PSP were significant predictors of athletic burnout. More specifically, PSP was a negative predictor and ECP was a positive predictor of athletic burnout. Hypotheses 1a, 2, 3, and 4 of the 2 x 2 model of perfectionism were supported   71 by the data at the first three time points. At the fourth and final time point, only ECP was a statistically significant predictor of burnout, in which case only hypotheses 1c, 2, and 4 were supported. Issues related to the 2 x 2 model are examined in more detail later in this section. 4.1.1 The role of coping  Coping strategies were found to play a major role in the relationship between a personality construct and a psychological outcome. More specifically, dimensions of perfectionism were found to be predictors of athletic burnout and coping acted as a mediator to these relationships. Personal standards perfectionism was associated with lowered levels of athletic burnout and this relationship was strengthen by the mediation effect of TOC. Thus, high level of PSP were associated with higher levels of TOC which, in turn, were associated with lower burnout symptoms. This extends previous cross-sectional findings to longitudinal design and to a sample of varsity university level athletes (Hill et al., 2010a). This mediation effect was supported at two levels: within- and between-individuals. At the individual level, increases in PSP compared to one’s personal average predicted lower levels of burnout within the examined timeframe and this was partially explained by the use of TOC which was also associated with lower burnout. Analyses at the between-individual level indicated that athletes who score higher than the grand mean on PSP typically score higher than average on TOC and lower than the norm on burnout.   Based on previous findings (Hill et al., 2010a) it was hypothesized that the relationship between PSP and burnout would be mediated by DOC. This mediation effect was not supported by the present results. Contrary to past evidence (Hill et al., 2010a),   72 results indicated that PSP and burnout were not significantly correlated at each time point. Moreover, there were positive moderate correlation between PSP and DOC at Time 1, 2, and 4, and no significant correlation at Time 3, whereas, previous evidence indicated negative correlation between PSP and DOC (Hill et al., 2010a). Furthermore, PSP was not a predictor of DOC in the multilevel models.  Multilevel modelling revealed a positive relationship between ECP and athletic burnout. This relationship was mediated by DOC. Thus, high ECP was associated with high DOC which, in turn, predicted high burnout symptoms. The mediation effect of DOC was similar at the within- and between-individual level. At the intra-individual level, the use of DOC explained part of the increase in burnout as compared to athletes’ personal average on occasions where they reported higher levels of ECP than usual. At the between individual level, athletes who scored higher on ECP than the average had a tendency to report greater use of DOC than the mean and higher burnout symptoms. The mediation effect of DOC on the relationship between ECP and burnout at the between-individual level was consistent with past cross-sectional research evidence (Hill et al., 2010a).   This study represented the first test of the mediation effect of coping on the relationship between perfectionism and athletic burnout at the within-individual level. Group level association between variables do not always translate into similar within-individual relationships. This is why the mediation effect of coping on the relationship between perfectionism and athletic burnout was examined within- and between- individual in this study. Within-individual analyses provide information about systematic covariation between the variables, thus whether change in one variable is associated with   73 change in another variable. In the present study, within-individual analyses indicated that deviation from one’s mean perfectionism score was associated with change from one’s mean burnout score. More specifically, increase in PSP was associated with decrease in burnout and this relationship was mediated by TOC. Increase in ECP was associated with increase in burnout compared to an individual’s average score and DOC mediated this relationship.  4.1.2 Progression of study variables over time  4.1.2.1 Burnout  The current study was one of the first to investigate athletic burnout repeatedly during approximately 20 weeks of training in university level varsity athletes. Analyses of burnout over time revealed a statistically significant increase in burnout over the course of the study. This was consistent with past empirical evidence (Schellenberg et al., 2013); however, this previous research only used two time points. The growth curve of athletic burnout varied between participants, which means that it is possible that some athletes reported increasingly higher levels of burnout whereas some athletes reported similar burnout throughout the study, and some reported lower burnout at the end of the study compared to the beginning of the study. Contrary to suggestions by Cresswell and Eklund (2006) that burnout might not follow a linear trajectory, the present study found that the change in burnout was better explained by a linear model than an alternative quadratic model.  4.1.2.2 Perfectionism   Change in perfectionism over time was different for PSP and ECP. Whereas PSP significantly decreased over time, ECP was relatively stable. Most variance in both PSP   74 and ECP was at the between-individual level. Since test of the stability of perfectionism over time in the sport domain had not previously been tested, two models of change were investigated. Both linear and quadratic model accounted for change in PSP over time, with the quadratic model represented the data better. The parabola of the quadratic curve in PSP started with a negative linear component and had an upward opening. Change in ECP over the four time points did not follow either a linear or quadratic trajectory.  The stability of ECP differed from previous findings that showed increase in ECP during a similar timeframe (Damian, Stoeber, Negru, & Babān, 2013). The divergent findings might be related to the population studied; in the current study the sample was composed of university student-athletes (Mage = 20.65; SD = 2.53) whereas Damian and colleagues (2013) studied a population of non-athlete high school students (Mage = 16.7; SD = 0.9). Evaluative concerns perfectionism might be based on internalized cognitive schemas that are less likely to change over a 20 weeks period in more developmentally advanced population such as university student-athletes as opposed to high school students. Personal standards perfectionism displayed a quadratic curve with an initial decrease and an upward opening. The PSP trajectory might be tied to change in the training demands throughout the season or it could reflect developmental change in perfectionism. Whereas recent empirical evidence revealed daily variations in PSP and ECP (Boone et al., 2012), the current findings differed because they suggest an organized pattern of change in PSP, quadratic curve, as opposed to random variation on a daily basis. Future research could investigate the relationship between change in PSP and athletic goals. Maybe athletes begin their season with high standards for themselves, but   75 as the season progress they lower their standards or the importance that they attach to their achievement in order to account for their performance. When the season begins, everything is possible, thus it might be adaptive to set high personal standards, but as the end of the season gets closer it is less likely that one will achieve their high standards, therefore athletes may disengage with those previously high standards and set more realistic standards.  4.1.2.3 Coping  Coping was less stable than burnout or perfectionism. Even though TOC and DOC displayed more within-individual variation than burnout and perfectionism, coping dimensions did not present large increases or decreases over time. Task-oriented coping followed a significant linear decreased over time, whereas DOC followed a quadratic curve with an initial negative linear component and subsequent upward acceleration. Contrary to past findings (Louvet et al., 2007), the pattern of change in coping strategies was similar between athletes in this study. In accordance to results from qualitative investigation of female basketball players, the use of TOC was consistently higher than the use of DOC (Tamminen & Holt, 2010). The quadratic growth curve describing change in DOC during a 20 week period shared some similarities with Taminnen and Holt’s (2010) findings. In both cases DOC was high at the beginning of the study followed by a decrease. Differences arose in the level of DOC reported at the end of both studies. In the present study athletes reported an increase in the use of DOC at the fourth time point, whereas in the other study the use of DOC remained low at the end of the competitive season (Tamminen & Holt, 2010). The discrepancy in the results might be due to the temporal location of the last time point in   76 both studies; towards the end of the competitive season for the present study versus two weeks post-season for Tamminen and Holt (2010). The difference could also be due to the methodologies used, Tamminen and Holt’s (2010) findings were obtained through a series of interviews.  4.1.3 Burnout, perfectionism, and the 2 x 2 model of perfectionism  Consistent with the literature on ECP and burnout (Appleton et al., 2009; Chen et al., 2008; Gould et al., 1996a; Hill et al., 2010a; Hill, 2013), the present results revealed that ECP was consistently positively correlated with athletic burnout. However, contrary to previous findings (Appleton et al., 2009; Hill et al., 2010a; Hill, 2013; Chen et al., 2008), there were no significant correlation between PSP and total athletic burnout in this study. Only one other study reported non significant correlation between PSP and separate dimensions of burnout (Hill et al., 2010b). Even though PSP and burnout did not have significant bivariate correlation, PSP can still be a significant predictor of athletic burnout in multivariate regression, because multiple regressions use series of partial correlations to create a linear model that predicts the independent variable. The four hypotheses associated with the 2 x 2 model of dispositional perfectionism (Gaudreau & Thompson, 2010) were tested to explore the relationship between dimensions of perfectionism and athletes burnout. Evaluative concerns perfectionism was a positive predictor of burnout at all four time points whereas PSP was a statistically significant negative predictor of burnout at the first three time points, but only a marginally significant predictor of burnout at Time 1. Similar to Hill’s (2013) results, support was found for hypotheses 1a, 2, 3, and 4 at the first three time points. This suggests that pure PSP was the combination of perfectionism associated with the   77 lowest level of burnout, mixed perfectionism was associated with higher burnout than pure PSP but lower burnout than pure ECP, and finally, pure ECP was the combination linked with the highest burnout levels. At the fourth time point, hypotheses 2, and 4 were supported, and hypotheses 1a and 3 were marginally supported. The marginal effect of PSP on burnout at the fourth time point was likely related to participant attrition. The present study provided further support for the use of the 2 x 2 model of perfectionism in the sport domain. Reviews of perfectionism in sport have struggled to determine whether PSP can be associated with adaptive outcomes. It is possible that, as Gaudreau and Verner-Filion (2012) suggested, both dimensions of perfectionism are best studied together because they exist to some extent within each individual. This model might be even more relevant in cases where the associations between PSP or ECP with the outcome variable are unclear. The data seemed to illustrate that both dimensions of perfectionism need to be studied in combination to best predict associated athletic burnout.    The present study extended Hill’s (2013) findings whereby the 2 x 2 model of perfectionism was used to predict athletic burnout. It extends past cross-sectional findings (Hill, 2013) to longitudinal research design. Longitudinal design provided information about the consistency of relationships among dimensions of perfectionism and burnout over a 20 week period (Crocker, Mosewich, Kowalski, & Besenski, 2010). Moreover, the study extended Hill’s (2013) findings to a sample of university level varsity athletes. In addition, a different scale was employed in this study to assess perfectionism, the sport-MPS-2 scale, to follow recommendation to test the 2 x 2 model of perfectionism with different measure of perfectionism (Hill, 2013).    78 4.2 Strengths and limitations  The longitudinal research designed used in this study was a major strength; it allowed to examine burnout, perfectionism, and coping over time. These variables were measured at each time point to track their progression during four months. Burnout and coping in relation with training demands had previously been examined in a longitudinal study with only two time points (Schellenberg et al., 2013). The present study was stronger because it measured coping and burnout at four times, allowing a test of quadratic curve models. It was also the first study to conduct a longitudinal investigation of the mediation effect of coping on the relationship between perfectionism and athletic burnout. Moreover, the multiple data points per participants permitted analyses of the mediation effect of coping at both the within- and between-individual level. Other strengths of this study included strong psychometric properties of the measures and sample including athletes involved in varied sports. The measures of athletic burnout (ABQ; Raedeke & Smith, 2001) and sport-specific perfectionism (Sport-MPS-2; Gotwals & Dunn, 2009) had strong psychometric properties. The measure of coping strategies in the context of training demands was adapted from a psychometrically sound scale (CICS; Gaudreau & Blondin, 2002) that assesses coping in competitive context. The population included a variety of university level varsity sport (e.g., basketball, ice hockey, rugby, track & field, swimming), the results were not restricted to one specific sport.  The use of longitudinal research design was a major strength of this study but this design is typically associated with missing data. Missing data in longitudinal research design is common occurrence and multilevel modelling handles missing data with low   79 impact on parameter estimates as long as the level-1 model is well specified (Zaidman-Zait & Zumbo, 2013). This is because there are no assumptions in longitudinal multilevel modelling about equal number of data point for every participant (Hox, 2000). Therefore, no missing data replacement strategies are necessary when using longitudinal multilevel modelling techniques as long as the level-1 model is correctly specified.  A limitation of this study was the sample size. The number of athletes who took part in the study was sufficient to run key analyses although some analyses were not explored due to the lack of statistical power. For example, when a latent variable is created, different weighs can be attributed to the indicators of the construct (Little, 2013). In this study, coping dimensions, TOC and DOC, could have benefited from a latent variable structure as different coping subscales loaded to varying degree on either TOC or DOC. The sample of athletes who took part in the study did not allow to model the multilevel mediation analyses with latent variables. Moreover, joint analysis of growth curve for perfectionism, coping, and burnout could reveal temporal pattern in the interplay of those variables.   The limited number of teams sampled restricted the richness of the data. Athletes on any given team share a common training environment with their teammates, it could be informative to model this shared experience with a nested model at the team level. Furthermore, although all the teams were training on a regular basis during the sampled time period, some teams were in the midst of their competitive season (e.g., basketball) whereas other teams were not competing at the time of the study (e.g., golf). Thus some athletes may have faced more intense training demands than others.     80 4.3 Future directions  The research design of the present study provided new and insightful knowledge about perfectionism, coping, and athletic burnout. The focus of this study was on the role of coping on the relationship between a personality-like construct and psychological maladjustment, athletic burnout. New advances in psychology have seen the emergence of positive psychology. At the subjective level, positive psychology is interested in understanding how individuals flourish (Park-Perin, 2013). Positive psychology proponents orient their research towards two objectives: improving strength and performance of people within the general population and optimizing human potential (Seligman & Csikszentmihalyi, 2000). Positive psychology is concerned with the role of positive emotions (e.g., satisfaction, happiness, contentment) and optimal experiences (e.g., flow and engagement; Park-Perin, 2013). Future studies could investigate positive outcomes (e.g., positive affect) or performance outcomes in relation with perfectionism, coping, and burnout in the sport domain. Adopting this approach could shed light on the positive aspect of athletes’ psychological profile. Tracking athletic performance in relation with perfectionism, coping, and burnout might provide further incentives to explore the benefits provided by the use of TOC to alleviate burnout symptoms in sport. Another avenue of research would include investigation of entrapment in sport in relation with burnout. From a commitment perspective, people can engage in an activity because they want to or because they have to (Schmidt & Stein, 1991). Athlete commitment is determined by three factors (Schmidt & Stein, 1991). First, athletes’ satisfaction with the balance of cost and rewards associated with sport participation. Second, presence or absence of attractive alternative to sport engagement. Third,   81 resources that athletes have invested in sport such as money, time, physical and emotional resources. Entrapment happens when athletes perceive low rewards and high costs associated with sport involvement, furthermore they have invested too much to quit and perceive a lack of alternatives (Schmidt & Stein, 1991). Athletes who fit an entrapment profile are more likely to experience burnout (Raedeke, 1997). Longitudinal investigation of entrapment in the context of athletic burnout would be a promising research avenue.  Goal engagement and disengagement might also be related to the change in PSP during the season. Close examination of personal goals as well as goal engagement and disengagement during the competitive season might reveal an avenue of intervention. As PSP predicted lower athletic burnout, teaching adaptive goal setting technique might help adjust athletes’ expectations to the training demands and help alleviate burnout symptoms.  Future studies of the progression of athletic burnout could benefit from investigating the effect of type of sport and position within the team. Previous findings suggest that position on the rugby field was associated with different temporal pattern in burnout symptoms (Cresswell & Eklund, 2006). It is possible that the between-individual differences in burnout trajectory over time were associated with the sport one was engaged in or with one’s position on the team. This could be tested by including sport or position within the team as a moderator variable of the growth curve.   In terms of research on perfectionism, longitudinal studies that extend over the 20 weeks timeframe of the current study would provide insight into the stable or developmental trajectory of perfectionism in the sport context. Investigation of   82 perfectionism over several years would be informative in terms of the magnitude and rate of change in perfectionism. Moreover, longer studies might shed some light on the cause of change in perfectionism, such as change in (or adaptation to) the training environment or normal developmental process.   Future research should also consider studying perfectionism, coping, and burnout in relation to more dynamic events such as major competitive events or sport transitions. The present study focused on coping in relation to training demands that represent a more or less stable environment for athletes. Investigation of the effect of coping in more dynamic context such as competitive events or transitions (e.g., selection process, injury, retirement from sport) could help determine whether coping takes over in acute situations or perfectionism remains influential on burnout symptoms. Another avenue for future research would be the stability or instability of perfectionism in relation with transitions in sport. Transitions in sport are coping processes that result in positive or negative outcomes (Taylor & Oglivie, 1994). Schlossberg (1981) proposed that transitions involve change in the assumptions that one holds about oneself and the world, this cognitive shift results in a change in behaviour. Transitions could have significant impact on perfectionism (i.e., cognition about the self and the world) and burnout (i.e., negative outcome), moreover coping would be a key factor of this process.   4.4 Practical implications  This study highlighted two key constructs, perfectionism and coping, to consider in discussion of athletic burnout. The study revealed relationships between those three variables. People in charge of improving athletes’ performance or sport-related experience could benefit from learning more about the dynamics of burnout as well as   83 factors that can influence burnout symptoms. These people include but are not limited to coach, parent, mental skill consultant, and trainer.  Research on athletic burnout has benefits such as: the potential to first sensitize people to the symptoms associated with burnout. Second, it allows to identify covariates of athletic burnout that can then be targeted in interventions designed to lower or control this psychological syndrome. For example, this study revealed that dimensions of perfectionism and coping strategies (TOC and DOC) are associated with athletic burnout. Third, longitudinal studies help understand temporal change in burnout. More specifically, the present study revealed that athletes already reported moderate level of athletic burnout towards the beginning of their season (middle of October in this case). This raises concerns about the training environment that university athletes are experiencing. One potential intervention avenue would be to sensitize coaches and athletic trainers about athletic burnout. It might be the case that building longer or more frequent rest period into the training cycle of university athletes would help lower burnout symptoms.   The mediation analyses conducted in this study emphasized the strong relationship between coping and athletic burnout. Coping strategies that seek to modify the stressful situation (i.e., TOC) were associated with lower levels of athletic burnout, whereas strategies designed to disengage oneself from a stressful situation (i.e., DOC) were indicative of higher reported burnout. Interventions or programs designed to teach athletes a variety of coping strategies that they can draw on have been promoted for a number of years (e.g., stress management training, Smith, 1980; stress inoculation   84 training, Meichenbaum, 1993). Results from this study highlighted the potential beneficiary effect of teaching athletes good coping skills.  The role of coping as a mediator of the relationship between dimensions of perfectionism and athletic burnout suggests that interventions should focus on this variable in order to decrease burnout level. More specifically, coping as an active process (Lazarus & Folkman, 1984) is malleable and would offer a good intervention avenue. Coping has the potential to not only have a direct effect on burnout level but also to impact the influence of dimensions of perfectionism on burnout. Successful coping interventions rely on the development of psychological skills (Smith, Schultz, Smoll, & Ptacek, 1995). Athletes who use adaptive coping skills such as TOC would be better equipped to manage the stress associated with training demands and might experience lower level of burnout.     85 References Acuna, E., & Rodriguez, C. (2004). The treatment of missing values and its effect on classifier accuracy. In Classification, Clustering, and Data Mining Applicationg (pp. 639-647). Sprigner Berlin Heidelberg, Aldwin, C. M. (2007). Stress, coping, and development: an integrated perspective (2nd ed., p. 432). New York: Guilford Press. Altman, D. G., & Bland, M. J. (2005). Standard deviations and standard errors. BJM, 331, 903. Anshel, M. H., & Wells, B. (2000). Personal and situational variables that describe coping with acute stress in competitive sport. The Journal of Social Psychology, 140, 434–450. Appleton, P. R., Hall, H. K., & Hill, A. P. (2009). Relations between multidimensional perfectionism and burnout in junior-elite male athletes. Psychology of Sport and Exercise, 10, 457–465.  Ashby, J. S., Rice, K. G., & Martin, J. L. (2006). Perfectionism, shame, and depressive symptoms. Journal of Counselling & Development, 84, 148–156. Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182. Buckworth, J., & Dishman, R. K. (2002). Exercise psychology (p. 344). Champaign, IL: Human Kinetics. Boone, L., Soenens, B., Mouratidis, A., Vansteenkiste, M., Vestuyf, J., & Braet, C. (2012). Daily fluctuations in perfectionism and their relation to eating disorder symptoms. Journal of Research in Personality, 46, 678–687.    86 Burdenski Jr, T. K. (2000). Evaluating univariate, bivariate, and multivariate normality using graphical procedures. Burns, D. D. (1980). The perfectionist’s script for self-defeat. Psychology Today, 34–52. Carmona, C., Buunk, A. P., Peiro, J. M., Rodriguez, I., & Bravo, M. J. (2006). Do social comparison and coping styles play a role in the development of burnout? Cross-sectional and longitudinal findings. Journal of Occupational and Organizational Psychology, 79, 85–99.  Chen, H. L., Kee, H. Y., Chen, M.-Y., & Tsai, Y.-M. (2008). Relation of perfectionism with athletes’ burnout: further examination. Perceptual and Motor Skills, 106, 811–820. Childs, J. H., & Stoeber, J. (2012). Do you want me to be perfect? Two longitudinal studies on socially prescribed perfectionism, stress, and burnout in the workplace. Work & Stress, 26, 347–364. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale: NJ: Erlbaum. Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple regression/correlation analysis for the behavioral sciences (3rd ed., p. 703). New York, NY: Routledge. Collins, L. M., Schafer, J. L., & Kam, C.-M. (2001). A comparison of inclusive and restrictive strategies in modern missing data procedures. Psychological Methods, 6, 330–351. Cresswell, S. L., & Eklund, R. C. (2005). Changes in athlete burnout and motivation over a 12-week league tournament. Medicine and Science in Sports and Exercise, 9, 1957–1966.    87 Cresswell, S. L., & Eklund, R. C. (2006). Changes in athlete burnout over a thirty-week “rugby year.” Journal of Science and Medicine in Sport, 9, 125–134. Crocker, P. R. E., Mosewich, A. D., Kowalski, K. C., & Besenski, L. J. (2010). Coping: Research design and analysis issues. In A. R. Nicholls (Ed.), Coping in Sport: Theory, Method, and Related Constructs (pp. 53–76). New York, NY: Nova Science Publishers. Crocker, P. R. E., Gaudreau, P., Mosewich, A. D., & Kljajic, K. (in press). Perfectionism and the stress process in intercollegiate althetes: Examining the 2 x 2 model of perfectionism in sport competition. International Journal of Sport and Exercise Psychology. Damian, L. E., Stoeber, J., Negru, O., & Babān, A. (2013). On the devolpment of perfectionism in adolescence: Perceived parental expectations predict longitudinal increases in socially prescribed perfectionism. Personality and Individual Differences, 55, 688–693. DeFreese, J. D. (2013). Advances in sport-based burnout research--Social support, negative social interactions, and the perceived stress-athlete burnout relationship. In North American Society for the Psychology of Sport and Physical Activity (pp. S13–S13). New Orlean. Dillman, D. A. (2007). Mail and internet surveys: The tailored design method. Hoboken, NJ: John Wiley & Sons, Inc. Dugas, M., Gaudreau, P., & Carraro, N. (2012). Implementation planning and progress on physical activity goals: The mediating role of life-management strategies. Research Quarterly for Exercise and Sport, 83, 77-85.    88 Dunkley, D. M., & Blankstein, K. R. (2000). Self-critical perfectionism, coping, hassles, and current distress: A structural equation modelling approach. Cognitive Therapy and Research, 24, 713–730. 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. H., Causgrove Dunn, J., Gotwals, J. K., Vallance, J. K. H., 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. Dunn, J. G. H., Causgrove Dunn, J., & Syrotuik, D. G. (2002). Relationship between multidimensional perfectionism and goal orientations in sport. Journal of Sport & Exercise Psychology, 24, 376–395. Dunn, J. G. H., Craft, J. M., Causgrove Dunn, J., & Gotwals, J. K. (2011). Comparing a domain-specific and global measure of perfectionism in competitive female figure skaters. Journal of Sport Behavior, 34, 25–46. Dunn, J. G. H., Gotwals, J. K., & Causgrove Dunn, J. (2005). An examination of the domain specificity of perfectionism among intercollegiate student-athletes. Personality and Individual Differences, 38, 1439–1448. Enns, M. W., Cox, B. J., Sareen, J., & Freeman, P. (2001). Adaptive and maladaptive perfectionism in medical students: a longitudinal investigation. Medical Education, 35, 1034–1042. Flett, G. L., & Hewitt, P. L. (2002). Perfectionism and maladjustment: an overview of theoretical, definitional, and treatment issues. In G. L. Flett & P. L. Hewitt (Eds.),    89 Perfectionism: Theory, Research, and Treatment (pp. 5–31). Washington, DC: American Psychological Association. 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., Greene, A., & Hewitt, P. L. (2004). Dimensions of perfectionism and anxiety sensitivity. Journal of Rational-Emotive & Cognitive-Behavior Therapy, 22, 39–57. Franche, V., Gaudreau, P., & Miranda, D. (2012). The 2 x 2 model of perfectionism: a comparison across Asian Canadians and European Canadians. Journal of Counselling & Development, 59, 567–574. Frost, R. O., & Henderson, K. J. (1991). Perfectionism and reactions to athletic competition. Journal of Sport & Exercise Psychology, 13, 323–335. Garson, D. G. (2013). Introductory guide to HLM with HLM 7 software. In D. G. Garson (Ed.), Hierarchical Linear Modeling (pp. 55–96). United States of America: SAGE Publications. Gaudreau, P. (2013). The 2 x 2 model of perfectionism: Commenting the critical comments and suggestions of Stoeber (2012). Personality and Individual Differences, 53, 351–355.  Gaudreau, P., & Antl, S. (2008). Athletes’ broad dimensions of dispositional perfectionism: examining changes in life satisfaction and the mediating role of sport-related motivation and coping. Journal of Sport & Exercise Psychology, 30, 356–382. 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.     90 Gaudreau, P., & Blondin, J.-P. (2004). Differential associations of dispositional optimism and pessimism with coping, goal attainment, and emotional adjustment during sport competition. International Journal of Stress Management, 11, 245–269. doi:10.1037/1072-5245.11.3.245 Gaudreau, P., & Thompson, A. (2010). Testing a 2 x 2 model of dispositional perfectionism. Personality and Individual Differences, 48, 532–537.  Gaudreau, P., & Verner-Filion, J. (2012). Dispositional perfectionism and well-being: A test of the 2 x 2 model of perfectionism in the sport domain. Sport, Exercise, and Performance Psychology, 1, 29–43.  Goodger, K., Gorely, T., Lavallee, D., & Harwood, C. (2007). Burnout in sport: a systematic review. The Sport Psychologist, 21, 127–151. Gotwals, J. K., & Dunn, J. G. H. (2009). A multi-method multi-analytic approach to establishing internal construct validity evidence: the Sport Multidimensional Perfectionism Scale-2. Measurement in Physical Education and Exercise Science, 13, 71–92. Gotwals, J. K., Dunn, J. G. H., Causgrove Dunn, J., & Gamache, V. (2010). Establishing validity evidence for the Sport Multidimensional Perfectionism Scale-2 in intercollegiate sport. Psychology of Sport and Exercise, 11, 423–432. Gotwals, J. K., Stoeber, J., Dunn, J. G. H., & Stoll, O. (2012). Are perfectionistic strivings in sport adaptive? A systematic review of confirmatory, contradictory, and mixed evidence. Canadian Psychology, 53, 263–279.  Gould, D., Dieffenbach, K., & Moffett, A. (2002). Psychological characteristics and their development in Olympic champions. Journal of Applied Sport Psychology, 172–204.    91 Gould, D., Udry, E., Tuffey, S., & Loehr, J. (1996a). Burnout in competitive junior tennis players: I. A qualitative psychological assessment. The Sport Psychologist, 10, 322–340. Gould, D., Tuffey, S., Udry, E., & Loehr, J. (1996b). Burnout in competitive junior tennis players: II. Qualitative analysis. The Sport Psychologist, 10, 341–366. Gould, D., Tuffey, S., Udry, E., & Loehr, J. (1997). Burnout in competitive junior tennis players: III. Individual differences in the burnout experience. The Sport Psychologist, 11, 257–276. Gucciardi, D. F., Mahoney, J., Jalleh, G., Donovan, R. J., & Parkes, J. (2012). Perfectionistic profiles among elite athletes and differences in their motivational orientations. Journal of Sport & Exercise Psychology, 34, 159–183. Gustafsson, H., Hassmen, P., Kentta, G., & Johansson, M. (2008). A qualitative analysis of burnout in elite Swedish athletes. Psychology of Sport and Exercise, 9, 800–816.  Hall, H. K. (2006). Perfectionism: a hallmark quality of world class performers, or a psychological impediment to athletic development? In D. Hackfort & G. Tenebaum (Eds.), Essential Processes for Attaining Peak Performance (pp. 178–211). Oxford: Meyer & Meyer Sport.  Hall, H. K., Hill, A. P., & Appleton, P. R. (2012). Perfectionism: a foundation for sporting excellence or an uneasy pathway toward purgatory? In G. Roberts & D. Treasure (Eds.), Advances in motivation in sport and exercise (3rd ed., pp. 129–168). Champaign, IL: Human Kinetics. Hall, H. K., Kerr, A. W., & Matthews, J. (1998). Precompetitive anxiety in sport: the contribution of achievement goals and perfectionism. Journal of Sport & Exercise Psychology, 20, 194–217.    92 Hawthorne, G., & Elliott, P. (2005). Imputing cross-sectional missing data: Comparison of common techniques. Australian and New Zealand Journal of Psychiatry, 39, 583–590. Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis: a regression based approach (p. 507). New York: The Guilford Press. Hewitt, P. L., & Flett, G. L. (1991). Perfectionism in the self and social contexts: conceptualization, assessment, and association with psychopathology. Journal of Personality and Social Psychology, 60, 456–470. Hill, A. P. (2013). Perfectionism and burnout in junior soccer players: A test of the 2 x 2 model of dispositional perfectionism. Journal of Sport & Exercise Psychology, 35, 18–29. Hill, A. P., & Appleton, P. R. (2011). The predictive ability of the frequency of perfectionistic cognitions, self-oriented perfectionism, and socially prescribed perfectionism in relation to symptoms of burnout in youth rugby players. Journal of Sports Sciences, 29, 695–703.  Hill, A. P., Hall, H. K., & Appleton, P. R. (2010a). Perfectionism and athlete burnout in junior elite athletes: the mediating role of coping tendencies. Anxiety, Stress, & Coping, 23, 415–430.  Hill, A. P., Hall, H. K., Appleton, P. R., & Murray, J. J. (2010b) Perfectionism and burnout in canoe polo and kayak slalom athletes: The mediating influence of validation and growth-seeking. The Sport Psychologist, 24, 16–34. Hill, A. P., Hall, H. K., Appleton, P. R., & Kozub, S. A. (2008). Perfectionism and burnout in junior elite soccer players: The mediating influence of unconditional self-acceptance. Psychology of Sport and Exercise, 9, 630–644.    93 Hill, A. P., Stoeber, J., Brown, A., & Appleton, P. R. (2014). Team perfectionism and team performance: a prospective study. Journal of Sport & Exercise Psychology, 36, 303–315. Hoar, S. D., Kowalski, K. C., Gaudreau, P., & Crocker, P. R. E. (2006). A review of coping in sport. In S. Hanton & S. D. Mellalieu (Eds.), Literature reviews in sport psychology (pp. 47–90). Hauppauge, NY: Nova Science Publishers. Hox, J. J. (2000). Multilevel analyses of grouped and longitudinal data. In T. D. Little, K. U. Schnabel & J. Baumert (Eds.), Modeling Longitudinal and Multilevel Data: Practical Issues, Applied Approaches, and Specific Examples. (pp. 15-32). Mahwah, NJ: Lawrence Erlbaum Associates.  Kawamura, K. Y., Hunt, S. L., Frost, R. O., & Marten DiBartolo, P. (2001). Perfectionism, anxiety, and depression: Are the relationships independent? Cognitive Therapy and Research, 25, 291–301. Kaye, M. P., Conroy, D. E., & Fifer, A. (2008). Individual differences in incompetence avoidance. Journal of Sport & Exercise Psychology, 30, 110–132. Kentta, G., & Hassmen, P. (1998). Overtraining and recovery. Sport Medicine, 26(1), 1–16. Lazarus, R. S. (1991). Emotion and adaptation. New York: Oxford University Press. Lazarus, R. S. (2000a). Cognitive-motivational-relational theory of emotion. In Y. L. Hanin (Ed.), Emotions in sport (pp. 39–63). Champaign, IL: Human Kinetics. Lazarus, R. S. (2000b). How emotions influence performance in competitive sports. The Sport Psychologist, 14, 229–252. Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New York, NY: Springer Publishing.    94 Lidor, R., Crocker, P. R. E., & Mosewich, A. D. (2012). Coping in Sport and Exercise. In G. Tenenbaum, R. C. Eklund, & A. Kamata (Eds.), Measurement in Sport and Exercise Psychology (pp. 393–407). Human Kinetics. Little, T. D. (2013). Longitudinal structural equation modeling (p. 386). New York: The Guilford Press. Lo, A., & Abbott, M. J. (2013). Review of the theoretical, empirical, and clinical status of adaptive and maladaptive perfectionism. Behaviour Change, 30, 96–116. Louvet, B., Gaudreau, P., Menaut, A., Genty, J., & Deneuve, P. (2007). Longitudinal patterns of stability and change in coping across three competitions: A latent class growth analysis. Journal of Sport & Exercise Psychology, 29, 100–117. Maslach, C. (1982). Burnout: the cost of caring. New York: Prentice-Hall. Meichenbaum, D. (1993). Stress innoculation training: a 20-year update. In P. M. Lehrer & R. L. Woolfolk (Eds.), Principles and practices of stress management (2nd ed., pp. 373–406). New York: Guilford. Nezlek, J.B. (2012). Multilevel modeling analyses of diary-style data. In M. R. Mehl & T. S. Conner (Eds.), Handbook of research methods for studying daily life (pp. 357-383). New York, NY: Guilford. Nicholls, A. R., Polman, R. C. J., Levy, A. R., Taylor, J., & Cobley, S. (2007). Stressors, coping, and coping effectiveness: gender, type of sport, and skill differences. Journal of Sport Sciences, 25, 1521–1530. Park-Perin, G. (2013). Positive psychology. In S. Hanrahan & M. B. Andersen (Eds.), Routledge Handbook of Applied Sport Psychology: a comprehensive guide for students and practitioners (pp. 141–149). New York, NY: Routledge.    95 Peacock, E. J., & Wong, P. T. (1990). The stress appraisal measure (SAM): A multidimensional approach to cognitive appraisal. Stress Medicine, 6, 227–236. Raedeke, T. D. (1997). Is athlete burnout more than just stress? A sport commitment perspective. Journal of Sport & Exercise Psychology, 19, 396–417. Raedeke, T. D., & Smith, A. L. (2001). Development and preliminary validation of an athlete burnout measure. Journal of Sport & Exercise Psychology, 23, 281–306. Raedeke, T. D., & Smith, A. L. (2004). Coping resources and athletes burnout: an examination of stress mediated and moderation hypotheses. Journal of Sport & Exercise Psychology, 26, 525–541. Richards, H. (2012). Coping processes in sport. In J. Thatcher, M. Jones, & D. Lavallee (Eds.), Coping and emotion in sport (Second ed., pp. 1–32). New York and London: Routledge. Rogosa, D. (1995). Myths and methods: “myths about longitudinal research” plus supplemental questions. In J. M. Gottman (Ed.), The analysis of change (pp. 3–66). Mahwah, New Jersey: Lawrence Erlbaum Associates. Rucker, D. D., Preacher, K. J., Tormala, Z. L., & Petty, R. E. (2011). Mediation analysis in social psychology: Current practices and new recommendations. Social and Personality Compass, 5, 359–371. Schellenberg, B. J. I., Gaudreau, P., & Crocker, P. R. E. (2013). Passion and coping: relationships with changes in burnout and goal attainment in collegiate volleyball players. Journal of Sport & Exercise Psychology, 35, 270–280. Schmidt, G. W., & Stein, G. L. (1991). Sport commitment: A model integrating enjoyment, dropout, and burnout. Journal of Sport & Exercise Psychology, 13, 254–265.    96 Schlossberg, N. K. (1981) A model for analyzing human adaptation to transition. The Counseling Psychologist, 9, 2-18. Seligman, M. E. P., & Csikszentmihalyi, M. (2000). Positive psychology: An introduction. American Psychologist, 55, 5–14. Sherry, S. B., MacKinnon, A. L., Fossum, K.-L., Antony, M. M., Stewart, S. H., Sherry, D. L., Nealis, L. J., & Mushquash, A. R. (2013). Perfectionism, discrepancies, and depression: Testing the perfectionism social disconnection model in a short-term, four-wave longitudinal study. Personality and Individual Differences, 54, 692–697. Silva, J. M. (1990). An analysis of the training stress syndrome in competitive athletics. Journal of Applied Sport Psychology, 8, 5–20. Smith, R. E. (1980). A cognitive-affective approach to stress management training for athletes. In C. Nadeau, W. Halliwell, K. Newell, & G. Roberts (Eds.), Psychology of motor behavior and sport (pp. 54–73). Champaign, IL: Human Kinetics. Smith, R. E., Schutz, R. W., Smoll. F. L., & Ptacek, J. T. (1995). Development and validation of a multidimensional measure of sport-specific psychological skills: The Athletic Coping Skills Inventory-28. Journal of Sport and Exercise Psychology, 17, 379-398. Sobel, M. E. (1982). Asymptotic confidence intervals for indirect effects in structural equation models. Sociological Methodology, 13, 290-312. Stoeber, J. (2011). The dual nature of perfectionism in sports: relationships with emotion, motivation, and performance. International Journal of Sport and Exercise Psychology, 4, 128–145. Stoeber, J., & Becker, C. (2008). Perfectionism, achievement motives, and attribution of success and failure in female soccer players. International Journal of Psychology, 43, 980–987.     97 Stoeber, J., & Otto, K. (2006). Positive conceptions of perfectionism: approaches, evidence, challenges. Personality and Social Psychology Review, 10, 295–319. Stoeber, J., & Rambow, A. (2007). Perfectionism in adolescent school students: relations with motivation, achievement, and well-being. Personality and Individual Differences, 42, 1379–1389. Stoeber, J., Uphill, M., & Hotham, S. (2009). Predicting race performance in triathlon: the role of perfectionism, achievement goals, and personal goal setting. Journal of Sport & Exercise Psychology, 31, 211–245. Stoolmiller, M. (1995). Using latent growth curve models to study developmental processes. In J. M. Gottman (Ed.), The analysis of change (pp. 103–138). Mahwah, New Jersey: Lawrence Erlbaum Associates.  Tabachnick, B. G., & Fidell, L. S. (2007). Multivariate analysis of variance and covariance. Using multivariate statistics, 3, 402-407. Tamminen, K. A., & Holt, N. L. (2010). Female adolescent athletes’ coping: A season-long investigation. Journal of Sports Sciences, 28, 101–114. Taylor, J., & Oglivie, B. C. (1994) A conceptual model of adaptation to retirement among athletes. Journal of Applied Sport Psychology, 6, 1-20. Tenebaum, G., Jones, C. M., Kitsantas, A., Sacks, D. N., & Berwick, J. P. (2003). Failure adaptation: An investigation of the stress response process in sport. International Journal of Sport Psychology, 34, 27–62. Tormala, Z. L., Falces, C., Briñol, P., & Petty, R. E. (2007). Ease of retrieval effects in social judgment: The role of unrequested cognitions. Journal of Personality and Social Psychology, 93, 143–157.    98 Zaisman-Zait, A., & Zumbo, B. D. (2013). Can multilevel (HLM) models of change over time adequately handle missing data? Journal of Educational Research and Policy Studies, 13, 18–31.   99 Appendices Appendix A: Coach contact letter    School of Kinesiology Exercise and Sport Psychology Lab Rm. 220b, War Memorial Gym 6081 University Blvd. Vancouver, BC V6T 1Z1  Coach Contact  Perfectionism and coping with stress in competitive sport   September 9, 2013  Dear [coach’s name], I am a Professor in the School of Kinesiology at UBC. I am currently recruiting competitive university and college athletes to participate in a study focused on how aspects of perfectionism (e.g., excessive concerns with evaluation, striving for high achievement standards) influence how athletes evaluate and cope with competitive stress and the subsequent impact on athlete satisfaction and burnout. We are also interested in gender differences in the athletic stress process.  This research will help clarify whether all aspects of perfectionism are harmful or whether some facets can be beneficial in competitive sport. This study is funded by a Social Sciences and Humanities Research Council of Canada research grant and has been approved by UBC Behavioural Research Ethics Board.  In order to conduct this study, I am hoping that you can assist me in providing a venue for a recruiting talk with athletes on your team. If so, could you please let me know the dates and times that work best for you and we can schedule a suitable date to conduct the presentation. I understand if you cannot make time for this presentation to your athletes.  The study involves four separate assessments, three which are online. In each assessment, the athletes will be asked to complete a questionnaire containing scientifically validated measures of perfectionism, self-esteem, cognitive appraisal (perceptions of threat, challenge, and control), coping, athlete burnout, and athlete satisfaction, along with some general demographic information. We are asking athletes to focus on competitive training demands. This takes about 15-20 minutes. After the first assessment, athletes will be contacted every 4-5 weeks to complete online assessments of a similar questionnaire. Athletes will receive a small stipend for participating in the study.   It is important to note that all information is confidential and only the research team members will know the identity of the athletes and their responses to the questionnaire.  Please respond to this letter if you can or cannot help me with this study. You can contact me by email (peter.crocker@.ubc. or phone (604-822-5580). Thank you for your time and consideration.  Sincerely  Dr. Peter Crocker    100 Appendix B: Athlete information letter    School of Kinesiology Exercise and Sport Psychology Lab Rm. 220b, War Memorial Gym 6081 University Blvd. Vancouver, BC V6T 1Z1  Player Contact  Perfectionism and coping with stress in competitive sport   September 4, 2013   I am a Professor in the School of Kinesiology at UBC in the area of sport and exercise psychology. I am currently recruiting competitive university and college athletes to participate in a study focused on how aspects of perfectionism (e.g., excessive concerns with evaluation, striving for high achievement standards), influence how athletes evaluate and cope with competitive stress and their subsequent impact on athlete satisfaction and burnout. We are also interested in gender differences in the athletic stress process.  This research will help clarify whether all aspects of perfectionism are harmful or whether some facets can be beneficial in competitive sport. This study is funded by a Social Sciences and Humanities Research Council of Canada research grant and has been approved by UBC Behavioural Research Ethics Board.  The study involves four separate assessments, three which are online. In each assessment, athletes will be asked to complete a questionnaire containing scientifically validated measures of perfectionism, self-esteem, cognitive appraisal (perceptions of threat, challenge, and control), coping, athlete burnout, and athlete satisfaction, along with some general demographic information. We are asking athletes to focus on competitive training demands. This takes about 15-20 minutes. After the first assessment, athletes will be contacted every 4-5 weeks (taking into account first term exams and winter break) to complete online assessments of a similar questionnaire. Athletes will receive a small stipend ($10) for participating in the first assessment of the study. Additionally, your email address will be entered into a random draw after submitting the first questionnaire for a chance to win 1 of 10 bookstore gift cards (from your respective school) valued at $50 each. Your email address will be entered in the draw one more time for each online assessment that you submit. For example, if you submit the first questionnaire and all three online questionnaires your name will be entered 4 times into the draw.  It is important to note that all information is confidential and only the research team members will know the identity of the athletes and their responses to the questionnaire.  You can contact me by email (peter.crocker@.ubc.ca) or phone (604-822-5580). Thank you for your time and consideration.  Sincerely  Dr. Peter Crocker    101 Appendix C: Athlete consent form    School of Kinesiology Exercise and Sport Psychology Lab Rm. 220b, War Memorial Gym 6081 University Blvd. Vancouver, BC V6T 1Z1  Sept 4, 2013 Participant Consent Form  Perfectionism and coping with stress in competitive sport.  Principal Investigator:        Peter Crocker, Ph.D.  School of Kinesiology, University of British Columbia    Contact number:  604-822-5580    peter.crocker@ubc.ca     Purpose of study: With funding support from a Social Sciences and Humanities Research Council of Canada Grant, we are conducting a study to examine how specific personality factors influence how athletes evaluate and cope with stressful competitive training experiences over the course of a season. We are asking for your participation in this research so that we can better understand how aspects of perfectionism impact the stress process in female and male competitive athletes. This understanding will help in the development of psychological skills training programs for athletes as well as assist in future research that examines stress, personality, and gender. Results from this study will be presented at scientific conferences and will be published in academic journals.  Study Procedures: • The study involves four separate assessments. In the first assessment, you will be asked to complete a questionnaire. The next three assessments involve completing the questionnaire using online methods through the internet.  • The questionnaire will take approximately 15-20 minutes to complete, and will require you to complete measures of various aspects of perfectionism, how you evaluate the demands of training, how you cope, and how you feel about your sport, along with some general demographic information. We will also require some personal identifiers and an email contact.  • Approximately every 4 weeks after the first assessment, you will be contacted by email and asked to complete an online questionnaire that is similar to the first questionnaire. This will take approximately 15-20 minutes to complete. Each subsequent assessment is approximately 4-5 weeks apart (taking in consideration first term exams and break between academic terms).   102 Risks to Participants: • There are no foreseeable risks associated with your involvement in this study.You may experience some discomfort when recalling the stressful competitive demands for each assessment. You can discontinue your involvement in the study at any time, again resulting in no penalty. In the event that you would like to further discuss your feelings regarding the topics discussed in the interviews, UBC Counselling Services (604-822-3811), UBC Student Health (604-822-7889), or SFU Health & Counselling Services (778-782-4615) can be of assistance.  Benefits to Participants • You will receive a $10.00 stipend after returning the first completed questionnaire to the researchers. Additionally, your email address will be entered into a random draw after submitting the first questionnaire for a chance to win 1 of 10 bookstore gift cards (from your respective school) valued at $50 each. Your email address will be entered in the draw one more time for each online assessment that you submit. For example, if you submit the first questionnaire and all three online questionnaires your name will be entered 4 times into the draw. A participant can only win 1 cash prize. Participants will still be eligible to win the prize if they withdraw from the study at any point.  • You will be provided with a summary report of the findings from this study upon request.  There is the potential that participation might help increase understanding about how competitive athletes deal with competition stress related to training. This is a step towards promoting the most positive sport experiences for athletes possible.    Confidentiality: Information gathered on the questionnaire will be used for research purposes only, and the identity of individual participants will only be known to members of the research team. Once both questionnaires are completed, they will be identified by code number only and will be securely stored for a minimum of five years as required by the University of British Columbia guidelines.  • Results of this study will be analyzed in group form and will be used in the preparation of a presentation and an academic research publication, all of which are public documents. A summary of the results will be available upon request. • You do not waive any legal rights by reading or agreeing to consent to participate in this study. • You are free to withdraw from this study at any time with absolutely no penalty. The decision to withdraw will NOT result in any loss of services or any other negative consequences.  Contact information about the rights of research subjects:  If you have any concerns about the treatment or rights of research participants, you may contact the Research Subject Information Line at 604-822-8598.   Contact information about the study:  If you have any questions concerning the procedures of this study or desire further information please contact Dr. Peter Crocker at (604) 822-5580. Consent: • By returning the first completed questionnaire you are consenting to participate in this study. Please keep a copy of this consent form for your records.    103 Appendix D: Time 1 questionnaire    School of Kinesiology Exercise and Sport Psychology Lab Rm. 220b, War Memorial Gym 6081 University Blvd. Vancouver, BC V6T 1Z1  Perfectionism and coping with stress in competitive sport  Time 1: Questionnaire  Thank you for participating in this study. This questionnaire will ask you questions about yourself concerning your thoughts, feelings, and behaviours about training demands in competitive sport. Answer each question as honestly and accurately as possible, and note that there is no right or wrong answer. There are also questions related to basic demographics.  Your responses will be kept confidential; only members of the research team will have access to specific questions.  However, we will require your university / college email address so that we can contact you for the other three assessments over the internet.  This questionnaire should take you approximately 15-20 minutes to complete.  Please return the questionnaire to the researchers.   104 Section 1. Personal Information  1. What is your gender (check one)?  ? Male  ? Female  2. How old are you? ______________years  3. Sport  a) What university sport are you competing in? _____________________  b) Year of eligibility?  1 2 3 4 5  c) Have you been a member of a senior National Team? Y N  d) Have you been a member of an Olympic team?   Y N  4. Based on these categories from the Canadian Census, how do you describe yourself? PLEASE CHECK ALL THAT APPLY:   ? White/Caucasian ? Chinese ? Japanese ? Korean ? Aboriginal/First Nation (e.g., North American Indian, Metis, Inuit) ? Filipino   ? South Asian (e.g., East Indian, Pakistani, Punjabi, Sri Lankan) ? South East Asian (e.g., Cambodian, Indonesian, Vietnamese) ? Black (e.g., African, Haitian, Jamaican, Somali) ? West Asian/Middle East (e.g., Afgani, Arab, Iranian) ? Other ethnic/cultural group, please specify: ___________  5. What is your first language? ? English ? Punjabi ? Spanish ? French ? German ? Korean ? Cantonese ? Mandarin ? Other _________________ (identify)  6. We require an email contact for the second assessment.   Primary Email address: __________________________________   Secondary Email address: _______________________________    105 Competitive Orientations Scale (Sport-MPS-2)  Instructions The purpose of this questionnaire is to identify how players view certain aspects of their competitive experience in sport. Please help us to more fully understand how players view a variety of their competitive experiences by indicating the extent to which you agree or disagree with the following statements. (Circle one response option to the right of each statement). Some of the question relate to your sport experiences in general, while others relate specifically to experiences on the team that you have most recently played with. There are no right or wrong answers so please don’t spend too much time on any one statement; simply choose the answer that best describes how you view each statement.   To what extent do you agree or disagree with the following statements? Strongly Disagree Disagree Neither Agree Nor Disagree Agree Strongly Agree 1 If I do not set the highest standards for myself in my sport, I am likely to end up a second-rate player. 1 2 3 4 5 2 Even if I fail slightly in competition, for me, it is as bad as being a complete failure. 1 2 3 4 5 3 I usually feel uncertain as to whether or not my training effectively prepares me for competition. 1 2 3 4 5 4 My parents set very high standards for me in my sport. 1 2 3 4 5 5 On the day of competition I have a routine that I try to follow. 1 2 3 4 5 6 I feel like my coach criticizes me for doing things less than perfectly in competition. 1 2 3 4 5 7 In competition, I never feel like I can quite meet my parents’ expectations. 1 2 3 4 5 8 I hate being less than the best at things in my sport. 1 2 3 4 5 9 I have and follow a pre-competitive routine.  1 2 3 4 5 10 If I fail in competition, I feel like a failure as a person. 1 2 3 4 5 11 Only outstanding performance during competition is good enough in my family. 1 2 3 4 5 12 I usually feel unsure about the adequacy of my pre-competition practices. 1 2 3 4 5 13 Only outstanding performance in competition is good enough for my coach. 1 2 3 4 5 14 I rarely feel that my training fully prepares me for competition. 1 2 3 4 5 15 My parents have always had higher expectations for my future in sport than I have. 1 2 3 4 5 16 The fewer mistakes I make in competition, the more people will like me. 1 2 3 4 5   106   To what extent do you agree or disagree with the following statements? Strongly Disagree Disagree Neither Agree Nor Disagree Agree Strongly Agree 17 It is important to me that I be thoroughly competent in everything I do in my sport. 1 2 3 4 5 18 I follow pre-planned steps to prepare myself for competition. 1 2 3 4 5 19 I feel like I am criticized by my parents for doing less than perfectly in competition. 1 2 3 4 5 20 Prior to competition, I rarely feel satisfied with my training. 1 2 3 4 5 21 I think I expect higher performance and greater results in my daily sport-training than most players. 1 2 3 4 5 22 I feel like I can never quite live up to my coach’s standards. 1 2 3 4 5 23 I feel that other players generally accept lower standards for themselves in sport than I do. 1 2 3 4 5 24 I should be upset if I make a mistake in competition. 1 2 3 4 5 25 In competition, I never feel like I can quite live up to my parents’ standards. 1 2 3 4 5 26 My coach sets very high standards for me in competition. 1 2 3 4 5 27 I follow a routine to get myself into a good mindset going into competition. 1 2 3 4 5 28 If a team-mate or opponent (who plays a similar position to me) plays better than me during competition, then I feel like I failed to some degree. 1 2 3 4 5 29 My parents expect excellence from me in my sport. 1 2 3 4 5 30 My coach expects excellence from me at all times: both in training and competition. 1 2 3 4 5 31 I rarely feel that I have trained enough in preparation for a competition. 1 2 3 4 5 32 If I do not do well all the time in competition, I feel that people will not respect me as an athlete. 1 2 3 4 5 33 I have extremely high goals for myself in my sport. 1 2 3 4 5 34 I develop plans that dictate how I want to perform during competition. 1 2 3 4 5 35 I feel like my coach never tries to fully understand the mistakes I sometimes make. 1 2 3 4 5    107   To what extent do you agree or disagree with the following statements? Strongly Disagree Disagree Neither Agree Nor Disagree Agree Strongly Agree 36 I set higher achievement goals than most athletes who play my sport. 1 2 3 4 5 37 I usually have trouble deciding when I have practiced enough heading into a competition. 1 2 3 4 5 38 I feel like my parents never try to fully understand the mistakes I make in competition. 1 2 3 4 5 39 People will probably think less of me if I make mistakes in competition. 1 2 3 4 5 40 My parents want me to be better than all other players who play my sport. 1 2 3 4 5 41 I set plans that highlight the strategies I want to use when I compete. 1 2 3 4 5 42 If I play well but only make one obvious mistake in the entire game, I still feel disappointed with my performance. 1 2 3 4 5    108 How I feel about myself when competing and training in sport  Instructions:  Please respond to each of the following statements by circling your answer using the scale from “1 = Strongly disagree” to “7 = Strongly agree.”     Strongly Disagree Disagree Disagree Somewhat Neutral Agree Somewhat Agree Strongly Agree 1 I feel worthwhile when I perform better than others in sport. 1 2 3 4 5 6 7 2 Knowing that I am better than others in sport raises my self-esteem. 1 2 3 4 5 6 7 3 Doing better than others in sport gives me a sense of self-respect. 1 2 3 4 5 6 7 4 My self-worth is affected by how well I do when I am competing with others in sport. 1 2 3 4 5 6 7 5 My self-worth is influenced by how well I do on competitive tasks in sport. 1 2 3 4 5 6 7    109 Athlete Evaluation Measure  Respond to each of these questions with respect to how you typically thought and felt during the training in your sport in the last week.  Please use the following scale for each question 1: not at all true  2: slightly true   3: moderately true   4: very true   5: extremely true    Not at All true  Moderately true  Extremely true 1 I felt I had the ability to overcome this stressful training. 1 2 3 4 5 2 I believed that managing this training could have a benefit for me. 1 2 3 4 5 3 I perceived this training as threatening. 1 2 3 4 5 4 I focused on the positive aspects when tackling this training. 1 2 3 4 5 5 I felt this training could be a negative experience for me. 1 2 3 4 5 6 I believed I had the necessary skills to manage this stressful training. 1 2 3 4 5 7 I was excited about the potential outcome of this training. 1 2 3 4 5 8 I believed I had what it takes to overcome this stressful training. 1 2 3 4 5 9 I felt this training would have negative consequences for me. 1 2 3 4 5 10 I believed I could become a stronger person after experiencing this stressful training. 1 2 3 4 5 11 I believed this stressful training would negatively impact me greatly. 1 2 3 4 5 12 I believed this stressful training was beyond my control. 1 2 3 4 5    110 SPECIFIC  STRATEGIES USED IN THE TRAINING IN THE LAST WEEK      Instructions  High performance sport often involves high training demands. We are interested in your thoughts and behaviours related to these demands. For each of the items, indicate the extent to which it corresponds to the way you TYPICALLY managed these training demands in the last week. There are no right or wrong answers.  We are interested in what YOU actually did or thought during the training.   Use the following scoring for each item.  1 Does not correspond at all to what I did or thought 2 Corresponds a little to what I did or thought  3 Corresponds moderately to what I did or thought   4 Corresponds strongly to what I did or thought 5 Corresponds very strongly to what I did or what I thought        Not at all A little Moderately Strongly Very strongly 1) I visualized that I was in total control of the situation……………………….. 1 2 3 4 5 2) I used swear-words loudly or in my head in order to expel my anger……….. 1 2 3 4 5 3) I  distanced myself from other athletes…………………………………….. 1 2 3 4 5 4) I committed myself by giving a consistent effort……………………………. 1 2 3 4 5 5) I occupied my mind in order to think about other things than the training….. 1 2 3 4 5 6) I tried not to be intimidated by other athletes………………………………... 1 2 3 4 5 7) I asked someone for advice concerning my mental preparation……………... 1 2 3 4 5 8) I tried to relax my body………………………………………………………. 1 2 3 4 5 9) I analyzed my past performances…………………………………………….. 1 2 3 4 5 10) I lost all hope of attaining my goal…………………………………………. 1 2 3 4 5 11) I mentally rehearsed the execution of my movements……………………… 1 2 3 4 5 12) I got angry…………………………………………………………………... 1 2 3 4 5 13) I retreated to a place where it was easy to think……………………………. 1 2 3 4 5 14) I gave a relentless effort…………………………………………………….. 1 2 3 4 5   111   Not at all A little Moderately Strongly Very strongly 15) I thought about my favourite activity in order not to think about the training…..………………………………………………………………………  1  2  3  4  5 16) I tried to get rid of my doubts by thinking positively………………………. 1 2 3 4 5 17) I asked other athletes for advice…………………………………………….. 1 2 3 4 5 18) I tried to reduce the tension in my muscles…………………………………. 1 2 3 4 5 19) I analyzed the weaknesses of my opponents………………………………... 1 2 3 4 5 20) I let myself feel hopeless and discouraged…………………………………. 1 2 3 4 5 21) I visualized myself doing a good performance……………………………... 1 2 3 4 5 22) I expressed my discontent…………………………………………………... 1 2 3 4 5 23) I kept all people at a distance……………………………………………….. 1 2 3 4 5 24) I gave my best effort………………………………………………………... 1 2 3 4 5 25) I entertained myself in order not to think about the training….…………… 1 2 3 4 5 26) I replaced my negative thoughts by positive ones………………………….. 1 2 3 4 5 27) I talked to a trustworthy person…………………………………………….. 1 2 3 4 5 28) I did some relaxation exercises…………………………………………….. 1 2 3 4 5 29) I thought about possible solutions in order to manage the situation……….. 1 2 3 4 5 30) I wished that the training would end immediately……..…………………... 1 2 3 4 5 31) I visualized my all-time best performance………………………………….. 1 2 3 4 5 32) I expressed my frustrations…………………………………………………. 1 2 3 4 5 33) I searched for calmness and quietness……………………………………… 1 2 3 4 5 34) I tried not to think about my mistakes……………………………………… 1 2 3 4 5 35) I talked to someone who is able to motivate me……………………………. 1 2 3 4 5 36) I relaxed my muscles……………………………………………………….. 1 2 3 4 5 37) I analyzed the demands of the training…..…………………………………. 1 2 3 4 5 38) I stopped believing in my ability to attain my goal………………………… 1 2 3 4 5 39) I thought about my family, my friends, or others to distract myself……….. 1 2 3 4 5      112 Athlete emotional experience in the last week  INSTRUCTIONS: Please read each statement carefully and decide if you ever feel this way about your sport participation in the last week. This includes all the training you have completed. Please indicate how often you have had this feeling or thought in the last week by circling a number 1 to 5, where 1 means “I almost never feel this way”, and 5 means “I feel this way most of the time.”  Almost Never Rarely Sometimes Frequently Almost Always 1 2 3 4 5    Almost Never  Sometimes  Almost Always 1 I’m accomplishing many worthwhile things in sport. 1 2 3 4 5 2 I feel so tired from training that I have trouble finding energy to do other things. 1 2 3 4 5 3 The effort I sport in my sport would be better spent doing other things. 1 2 3 4 5 4 I feel overly tired from my participation in my sport. 1 2 3 4 5 5 I am not achieving much in my sport. 1 2 3 4 5 6 I don’t care as much about my sport performance as I used to. 1 2 3 4 5 7 I am not performing up to my ability in my sport. 1 2 3 4 5 8 I feel “wiped out” from my sport. 1 2 3 4 5 9 I’m not into my sport like I used to be. 1 2 3 4 5 10 I feel physically worn out from my sport. 1 2 3 4 5 11 I feel less concerned about being successful in my sport than I used to. 1 2 3 4 5 12 I am exhausted by the mental and physical demands of my sport. 1 2 3 4 5 13 It seems that no matter what I do, I don’t perform as well as I should. 1 2 3 4 5 14 I feel successful at my sport. 1 2 3 4 5 15 I have negative feelings toward my sport. 1 2 3 4 5    113 Athlete Satisfaction in the Last Week  INSTRUCTIONS:  We are interested in how you felt about your sport in the last week. Please read each statement carefully and decide how you feel about participating in your sport in the last week.  Using the scale below, indicate to what extent you agree with the following items by circling the appropriate number.   Not at all agree Very slightly agree Slightly agree Moderately agree Strongly agree Very strongly agree Totally agree 1 2 3 4 5 6 7   In the last week,  1 My life as an athlete corresponds closely to my goals. 1 2 3 4 5 6 7 2 The conditions of my life as an athlete are excellent. 1 2 3 4 5 6 7 3 I am satisfied with my life as an athlete. 1 2 3 4 5 6 7 4 I get the important things that I want in my sport. 1 2 3 4 5 6 7 5 I would change almost nothing to my life as an athlete. 1 2 3 4 5 6 7      114 Appendix E: Time 2, 3, and 4 questionnaire    School of Kinesiology Exercise and Sport Psychology Lab Rm. 220b, War Memorial Gym 6081 University Blvd. Vancouver, BC V6T 1Z1  Perfectionism and coping with stress in competitive sport  Time 2, 3, and 4: Questionnaire  Thank you for participating in the second phase of this study. This questionnaire will ask you questions about yourself concerning your thoughts, feelings, and behaviours about competitive training in the last week.  Answer each question as honestly and accurately as possible, and note that there is no right or wrong answer. There are some basic demographic questions so we make sure we match your responses to your previous assessment.  Your responses will be kept confidential; only members of the research team will have access to specific questions.  This questionnaire should take you approximately 15-20 minutes to complete.    115 Section 1. Personal Information  1. What is your gender (check one)?  ? Male  ? Female  2. How old are you? ______________years  3. Sport  a) What university sport are you competing in? _____________________  b) Year of eligibility?  1 2 3 4 5  c) Have you been a member of a senior National Team? Y N  d) Have you been a member of an Olympic team?   Y N    4. Please provide your email address below (to enable us to match responses to your previous assessment.   Primary Email address: __________________________________   Secondary Email address: _______________________________    116 Competitive Orientations Scale (Sport-MPS-2)  Instructions The purpose of this questionnaire is to identify how players view certain aspects of their competitive experience in sport. Please help us to more fully understand how players view a variety of their competitive experiences by indicating the extent to which you agree or disagree with the following statements. (Circle one response option to the right of each statement). Some of the question relate to your sport experiences in general, while others relate specifically to experiences on the team that you have most recently played with. There are no right or wrong answers so please don’t spend too much time on any one statement; simply choose the answer that best describes how you view each statement.   To what extent do you agree or disagree with the following statements? Strongly Disagree Disagree Neither Agree Nor Disagree Agree Strongly Agree 1 If I do not set the highest standards for myself in my sport, I am likely to end up a second-rate player. 1 2 3 4 5 2 Even if I fail slightly in competition, for me, it is as bad as being a complete failure. 1 2 3 4 5 3 I usually feel uncertain as to whether or not my training effectively prepares me for competition. 1 2 3 4 5 4 My parents set very high standards for me in my sport. 1 2 3 4 5 5 On the day of competition I have a routine that I try to follow. 1 2 3 4 5 6 I feel like my coach criticizes me for doing things less than perfectly in competition. 1 2 3 4 5 7 In competition, I never feel like I can quite meet my parents’ expectations. 1 2 3 4 5 8 I hate being less than the best at things in my sport. 1 2 3 4 5 9 I have and follow a pre-competitive routine.  1 2 3 4 5 10 If I fail in competition, I feel like a failure as a person. 1 2 3 4 5 11 Only outstanding performance during competition is good enough in my family. 1 2 3 4 5 12 I usually feel unsure about the adequacy of my pre-competition practices. 1 2 3 4 5 13 Only outstanding performance in competition is good enough for my coach. 1 2 3 4 5 14 I rarely feel that my training fully prepares me for competition. 1 2 3 4 5 15 My parents have always had higher expectations for my future in sport than I have. 1 2 3 4 5 16 The fewer mistakes I make in competition, the more people will like me. 1 2 3 4 5   117   To what extent do you agree or disagree with the following statements? Strongly Disagree Disagree Neither Agree Nor Disagree Agree Strongly Agree 17 It is important to me that I be thoroughly competent in everything I do in my sport. 1 2 3 4 5 18 I follow pre-planned steps to prepare myself for competition. 1 2 3 4 5 19 I feel like I am criticized by my parents for doing less than perfectly in competition. 1 2 3 4 5 20 Prior to competition, I rarely feel satisfied with my training. 1 2 3 4 5 21 I think I expect higher performance and greater results in my daily sport-training than most players. 1 2 3 4 5 22 I feel like I can never quite live up to my coach’s standards. 1 2 3 4 5 23 I feel that other players generally accept lower standards for themselves in sport than I do. 1 2 3 4 5 24 I should be upset if I make a mistake in competition. 1 2 3 4 5 25 In competition, I never feel like I can quite live up to my parents’ standards. 1 2 3 4 5 26 My coach sets very high standards for me in competition. 1 2 3 4 5 27 I follow a routine to get myself into a good mindset going into competition. 1 2 3 4 5 28 If a team-mate or opponent (who plays a similar position to me) plays better than me during competition, then I feel like I failed to some degree. 1 2 3 4 5 29 My parents expect excellence from me in my sport. 1 2 3 4 5 30 My coach expects excellence from me at all times: both in training and competition. 1 2 3 4 5 31 I rarely feel that I have trained enough in preparation for a competition. 1 2 3 4 5 32 If I do not do well all the time in competition, I feel that people will not respect me as an athlete. 1 2 3 4 5 33 I have extremely high goals for myself in my sport. 1 2 3 4 5 34 I develop plans that dictate how I want to perform during competition. 1 2 3 4 5 35 I feel like my coach never tries to fully understand the mistakes I sometimes make. 1 2 3 4 5    118   To what extent do you agree or disagree with the following statements? Strongly Disagree Disagree Neither Agree Nor Disagree Agree Strongly Agree 36 I set higher achievement goals than most athletes who play my sport. 1 2 3 4 5 37 I usually have trouble deciding when I have practiced enough heading into a competition. 1 2 3 4 5 38 I feel like my parents never try to fully understand the mistakes I make in competition. 1 2 3 4 5 39 People will probably think less of me if I make mistakes in competition. 1 2 3 4 5 40 My parents want me to be better than all other players who play my sport. 1 2 3 4 5 41 I set plans that highlight the strategies I want to use when I compete. 1 2 3 4 5 42 If I play well but only make one obvious mistake in the entire game, I still feel disappointed with my performance. 1 2 3 4 5    119 How I feel about myself when competing and training in sport  Instructions:  Please respond to each of the following statements by circling your answer using the scale from “1 = Strongly disagree” to “7 = Strongly agree.”     Strongly Disagree Disagree Disagree Somewhat Neutral Agree Somewhat Agree Strongly Agree 1 I feel worthwhile when I perform better than others in sport. 1 2 3 4 5 6 7 2 Knowing that I am better than others in sport raises my self-esteem. 1 2 3 4 5 6 7 3 Doing better than others in sport gives me a sense of self-respect. 1 2 3 4 5 6 7 4 My self-worth is affected by how well I do when I am competing with others in sport. 1 2 3 4 5 6 7 5 My self-worth is influenced by how well I do on competitive tasks in sport. 1 2 3 4 5 6 7    120 Athlete Evaluation Measure  Respond to each of these questions with respect to how you typically thought and felt during the training in your sport in the last week.  Please use the following scale for each question 1: not at all true  2: slightly true   3: moderately true   4: very true   5: extremely true    Not at All true  Moderately true  Extremely true 1 I felt I had the ability to overcome this stressful training. 1 2 3 4 5 2 I believed that managing this training could have a benefit for me. 1 2 3 4 5 3 I perceived this training as threatening. 1 2 3 4 5 4 I focused on the positive aspects when tackling this training. 1 2 3 4 5 5 I felt this training could be a negative experience for me. 1 2 3 4 5 6 I believed I had the necessary skills to manage this stressful training. 1 2 3 4 5 7 I was excited about the potential outcome of this training. 1 2 3 4 5 8 I believed I had what it takes to overcome this stressful training. 1 2 3 4 5 9 I felt this training would have negative consequences for me. 1 2 3 4 5 10 I believed I could become a stronger person after experiencing this stressful training. 1 2 3 4 5 11 I believed this stressful training would negatively impact me greatly. 1 2 3 4 5 12 I believed this stressful training was beyond my control. 1 2 3 4 5    121 SPECIFIC  STRATEGIES USED IN THE TRAINING IN THE LAST WEEK      Instructions  High performance sport often involves high training demands. We are interested in your thoughts and behaviours related to these demands. For each of the items, indicate the extent to which it corresponds to the way you TYPICALLY managed these training demands in the last week. There are no right or wrong answers.  We are interested in what YOU actually did or thought during the training.   Use the following scoring for each item.  1 Does not correspond at all to what I did or thought 2 Corresponds a little to what I did or thought  3 Corresponds moderately to what I did or thought   4 Corresponds strongly to what I did or thought 5 Corresponds very strongly to what I did or what I thought        Not at all A little Moderately Strongly Very strongly 1) I visualized that I was in total control of the situation……………………….. 1 2 3 4 5 2) I used swear-words loudly or in my head in order to expel my anger……….. 1 2 3 4 5 3) I  distanced myself from other athletes…………………………………….. 1 2 3 4 5 4) I committed myself by giving a consistent effort……………………………. 1 2 3 4 5 5) I occupied my mind in order to think about other things than the training….. 1 2 3 4 5 6) I tried not to be intimidated by other athletes………………………………... 1 2 3 4 5 7) I asked someone for advice concerning my mental preparation……………... 1 2 3 4 5 8) I tried to relax my body………………………………………………………. 1 2 3 4 5 9) I analyzed my past performances…………………………………………….. 1 2 3 4 5 10) I lost all hope of attaining my goal…………………………………………. 1 2 3 4 5 11) I mentally rehearsed the execution of my movements……………………… 1 2 3 4 5 12) I got angry…………………………………………………………………... 1 2 3 4 5 13) I retreated to a place where it was easy to think……………………………. 1 2 3 4 5 14) I gave a relentless effort…………………………………………………….. 1 2 3 4 5   122   Not at all A little Moderately Strongly Very strongly 15) I thought about my favourite activity in order not to think about the training…..………………………………………………………………………  1  2  3  4  5 16) I tried to get rid of my doubts by thinking positively………………………. 1 2 3 4 5 17) I asked other athletes for advice…………………………………………….. 1 2 3 4 5 18) I tried to reduce the tension in my muscles…………………………………. 1 2 3 4 5 19) I analyzed the weaknesses of my opponents………………………………... 1 2 3 4 5 20) I let myself feel hopeless and discouraged…………………………………. 1 2 3 4 5 21) I visualized myself doing a good performance……………………………... 1 2 3 4 5 22) I expressed my discontent…………………………………………………... 1 2 3 4 5 23) I kept all people at a distance……………………………………………….. 1 2 3 4 5 24) I gave my best effort………………………………………………………... 1 2 3 4 5 25) I entertained myself in order not to think about the training….…………… 1 2 3 4 5 26) I replaced my negative thoughts by positive ones………………………….. 1 2 3 4 5 27) I talked to a trustworthy person…………………………………………….. 1 2 3 4 5 28) I did some relaxation exercises…………………………………………….. 1 2 3 4 5 29) I thought about possible solutions in order to manage the situation……….. 1 2 3 4 5 30) I wished that the training would end immediately……..…………………... 1 2 3 4 5 31) I visualized my all-time best performance………………………………….. 1 2 3 4 5 32) I expressed my frustrations…………………………………………………. 1 2 3 4 5 33) I searched for calmness and quietness……………………………………… 1 2 3 4 5 34) I tried not to think about my mistakes……………………………………… 1 2 3 4 5 35) I talked to someone who is able to motivate me……………………………. 1 2 3 4 5 36) I relaxed my muscles……………………………………………………….. 1 2 3 4 5 37) I analyzed the demands of the training…..…………………………………. 1 2 3 4 5 38) I stopped believing in my ability to attain my goal………………………… 1 2 3 4 5 39) I thought about my family, my friends, or others to distract myself……….. 1 2 3 4 5      123 Athlete emotional experience in the last week  INSTRUCTIONS: Please read each statement carefully and decide if you ever feel this way about your sport participation in the last week. This includes all the training you have completed. Please indicate how often you have had this feeling or thought in the last week by circling a number 1 to 5, where 1 means “I almost never feel this way”, and 5 means “I feel this way most of the time.”  Almost Never Rarely Sometimes Frequently Almost Always 1 2 3 4 5    Almost Never  Sometimes  Almost Always 1 I’m accomplishing many worthwhile things in sport. 1 2 3 4 5 2 I feel so tired from training that I have trouble finding energy to do other things. 1 2 3 4 5 3 The effort I sport in my sport would be better spent doing other things. 1 2 3 4 5 4 I feel overly tired from my participation in my sport. 1 2 3 4 5 5 I am not achieving much in my sport. 1 2 3 4 5 6 I don’t care as much about my sport performance as I used to. 1 2 3 4 5 7 I am not performing up to my ability in my sport. 1 2 3 4 5 8 I feel “wiped out” from my sport. 1 2 3 4 5 9 I’m not into my sport like I used to be. 1 2 3 4 5 10 I feel physically worn out from my sport. 1 2 3 4 5 11 I feel less concerned about being successful in my sport than I used to. 1 2 3 4 5 12 I am exhausted by the mental and physical demands of my sport. 1 2 3 4 5 13 It seems that no matter what I do, I don’t perform as well as I should. 1 2 3 4 5 14 I feel successful at my sport. 1 2 3 4 5 15 I have negative feelings toward my sport. 1 2 3 4 5    124 Athlete Satisfaction in the Last Week  INSTRUCTIONS:  We are interested in how you felt about your sport in the last week. Please read each statement carefully and decide how you feel about participating in your sport in the last week.  Using the scale below, indicate to what extent you agree with the following items by circling the appropriate number.   Not at all agree Very slightly agree Slightly agree Moderately agree Strongly agree Very strongly agree Totally agree 1 2 3 4 5 6 7   In the last week,  1 My life as an athlete corresponds closely to my goals. 1 2 3 4 5 6 7 2 The conditions of my life as an athlete are excellent. 1 2 3 4 5 6 7 3 I am satisfied with my life as an athlete. 1 2 3 4 5 6 7 4 I get the important things that I want in my sport. 1 2 3 4 5 6 7 5 I would change almost nothing to my life as an athlete. 1 2 3 4 5 6 7      125 Appendix F: Pre-notice email  To: [Participant’s email address]  Subject: Pre-notice: Coping in Sport Study  Dear [Insert name of athlete],  My name is Coralie and I’m a graduate student at UBC in Kinesiology. You have recently been involved in the study about Perfectionism and Coping with Stress in Competitive Sport. Thank you again for your participation. The research team and I really appreciate it!  A few days from now you will receive an email containing information about the second phase of this study, and a link to a BRIEF ONLINE QUESTIONNAIRE. The questionnaire is similar in length to what you have completed in the past, so it should take you about 10-20 minutes to complete. The questions will focus on how you evaluate situations in sport, how you coped, and your emotional state during the training.  I am contacting you in advance because many people prefer to know ahead of time when they will receive an online questionnaire. It is important that all phases of the study are completed, as it makes it possible to understand how important constructs in sport change over the course of a season. Without all of the time points, it is difficult to have a complete understanding. Your continued participation would be appreciated and would be very helpful.   Thank you for your involvement in this study. If you have any questions, please don’t hesitate to ask!  Coralie [Study email address]    126 Appendix G: Questionnaire email  To: [Participant’s email address]  Subject: Coping in Sport Study  Dear [Insert name of athlete],  A few days ago, an email was sent to you inviting you to participate in the second phase of the Perfectionism and Coping with Stress in Competitive Sport Study. Your participation in the first phase of this study was GREATLY appreciated, but it is important for all phases of this study to be completed to get a full account of the experience of high performances athletes.   Your participation in this phase of the study will involve completing a BRIEF ONLINE QUESTIONNAIRE. It is approximately the same length as what you have completed in the past, and should take you about 10-20 minutes to complete. By taking part in this phase of the study your name (email address) will be added one more time into a draw to win 1 of 10 $50 bookstore gift cards.  As this study follows a timeline, as we are trying to get a sense of experiences throughout the course of a season, completing this questionnaire in the near future (ideally within the next day or two) is essential.  The questionnaire can be accessed by clicking on the link below. When you click on the link, you will be redirected to the first page of the questionnaire (consent form), which contains information about the study such as the potential risks (none), benefits, contact information, and the purpose of the study. Clicking on the link does not automatically involve you in this phase of the study, you must read this page first and then click “continue” to participate.  TO ACCESS THE CONFIDENTIAL QUESTIONNAIRE, CLICK ON THE FOLLOWING LINK:  http:// gggggggggggggggggggggggggggggggggg  Again, thank you for your involvement in this study. If you have any questions, please don’t hesitate to ask (you can reply to this email, or use the contact information provided below).  Coralie [Study email address]    127 Appendix H: Reminder email  To: [Participant’s email address]  Subject: Reminder: Coping in Sport Study  Dear [Insert name of athlete],  You have been receiving emails inviting you to participate in the second phase of the Perfectionism and Coping with Stress in Competitive Sport Study, and this is a follow-up. Your participation in the previous phase of this study was GREATLY appreciated, but it is important for all phases of this study to be completed to get a full account of the experience of high performances athletes. I have received some responses from other high performance athletes involved in this study, and your participation in this phase of this study is extremely important.   In order for the study to be a success, more responses from athletes at this time point are needed.  Your participation in this phase of the study will involve completing a BRIEF ONLINE QUESTIONNAIRE. It is approximately the same length as what you have completed in the past, and should take you about 10-20 minutes to complete. By taking part in this phase of the study your name (email address) will be added one more time into a draw to win 1 of 10 $50 bookstore gift cards.  As this study follows a timeline, as we are trying to get a sense of experiences throughout the course of a season, completing this questionnaire in the near future (ideally within the next few days) is essential.  The questionnaire can be accessed by clicking on the link below. When you click on the link, you will be redirected to the first page of the questionnaire (consent form), which contains information about the study such as the potential risks (none), benefits, contact information, and the purpose of the study. Clicking on the link does not automatically involve you in this phase of the study, you must read this page first and then click “continue” to participate.  TO ACCESS THE CONFIDENTIAL QUESTIONNAIRE, CLICK ON THE FOLLOWING LINK:  http:// gggggggggggggggggggggggggggggggggg  Again, thank you for your involvement in this study. If you have any questions, please don’t hesitate to ask (you can reply to this email, or use the contact information provided below).  Coralie [Study email address]   128 Appendix I: Thank you email  To: [Participant’s email address]  Subject: Thank you!  Dear [Insert name of athlete],  I am writing to thank you for your participation in the second phase of the Perfectionism and Coping with Stress in Competitive Sport Study. Your participation was extremely helpful and very much appreciated!  Your next online questionnaire will be sent to you in January, and will be in the same format as what you have done previously.  If you have any questions or comments regarding the study, please don’t hesitate to contact me.  I hope your training is going well!  Coralie [Study email address] 	  	   129 Appendix J: Results of the hierarchical regression for perfectionism predicting athletic burnout    Results of the hierarchical regression for perfectionism predicting athletic burnout Effects  t b p-value Time 1 R2 = 0.181, F(2, 122) = 13.492, p < 0.001       PSP  -2.00 -0.22 0.048    ECP  5.18 0.45 <0.001 Time 2 R2 = 0.254, F(2, 101) = 17.221, p < 0.001       PSP  -1.98 -0.24 0.051    ECP  5.70 0.50 <0.001 Time 3 R2 = 0.373, F(2, 98) = 29.189, p < 0.001       PSP  -4.12 -0.44 <0.001    ECP  7.62 0.59 <0.001 Time 4 R2 = 0.236, F(2, 79) = 12.199, p < 0.001       PSP  -1.11 -0.15 0.271    ECP  4.80 0.43 <0.001 Note. PSP = personal standard perfectionism; ECP = evaluative concerns perfectionism.  Ntime1 = 125; Ntime2 = 104; Ntime3 = 101; Ntime4 = 82. t = t-test value for the b-weight; b = unstandardized b-weight; p-value = p-value at the 95% confidence interval.    	  	   130 Appendix K: Conditional growth models predicting athletic burnout   Conditional growth model with Age as level-2 predictor: Fixed effects Variable  Coefficient Standard error t-ratio DF p-value Intercept Intercept β ℡  2.811 0.551 5.100 122 <0.001 Age β ™  -0.017 0.027 -0.621 122 0.536 Slope Intercept β™  0.193 0.144 1.343 122 0.182 Age β℡  -0.006 0.007 -0.929 122 0.354 Note. The fixed effects estimates reported are with robust standard errors.  Conditional growth model with PSP as level-2 predictor: Fixed effects Variable  Coefficient Standard error t-ratio DF p-value Intercept Intercept β ℡  2.398 0.413 5.799 123 <0.001 PSP β ™  0.021 0.113 0.189 123 0.851 Slope Intercept β™  0.034 0.132 0.259 123 0.796 PSP β℡  0.006 0.036 0.166 123 0.868 Note. PSP = personal standard perfectionism. PSP was averaged across all four time points. The fixed effects estimates reported are with robust standard errors.     1.1	  Linear	  model	  	  	  	  Level-­‐1	   𝑌™ =   𝜋?? +   𝜋??(𝑇𝐼𝑀𝐸™ ) +   𝑒™ 	  	  	  	  Level-­‐2	   𝜋?? =   𝛽℡ +   𝛽™ (𝑋?)+   𝑟??	  𝜋?? =   𝛽™ +   𝛽℡ (𝑋?) +   𝑟?? 	  	  	  	  X	  =	  predictor	  	  	   131 Conditional growth model with ECP as level-2 predictor: Fixed effects Variable  Coefficient Standard error t-ratio DF p-value Intercept Intercept β ℡  1.316 0.210 6.254 123 <0.001 ECP β ™  0.394 0.070 5.614 123 <0.001 Slope Intercept β™  0.019 0.069 0.271 123 0.787 ECP β℡  0.012 0.022 0.548 123 0.585 Note. ECP = evaluative concerns perfectionism. ECP was averaged across all four time points. The fixed effects estimates reported are with robust standard errors.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0166994/manifest

Comment

Related Items