Gemini IA a noté :
Dafotec Récupération De Données

Le 2025-04-26 04:32:09

Being Mindful of Perfectionism and Performance Among Athletes in a Judged Sport Being Mindful of Perfectionism and Performance Among Athletes in a Judged Sport Erika D. Van Dyke edv0001@mix.wvu.edu Follow this and additional works at: https://researchrepository.wvu.edu/etd Part of the Other Psychology Commons Recommended Citation Van Dyke, Erika D., "Being Mindful of Perfectionism and Performance Among Athletes in a Judged Sport" (2019). Graduate Theses, Dissertations, and Problem Reports. 7428. https://researchrepository.wvu.edu/etd/7428 This Dissertation is protected by copyright and/or related rights. It has been brought to you by the The Research Repository @ WVU with permission from the rights-holder(s). You are free to use this Dissertation in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you must obtain permission from the rights-holder(s) directly, unless additional rights are indicated by a Creative Commons license in the record and/ or on the work itself. This Dissertation has been accepted for inclusion in WVU Graduate Theses, Dissertations, and Problem Reports collection by an authorized administrator of The Research Repository @ WVU. For more information, please contact researchrepository@mail.wvu.edu. Being Mindful of Perfectionism and Performance Among Athletes in a Judged Sport Erika D. Van Dyke Dissertation submitted to the College of Physical Activity and Sport Sciences at West Virginia University in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Sport, Exercise, and Performance Psychology Sam J. Zizzi, Ed.D., Chair Scott Barnicle, Ph.D. Edward F. Etzel, Ed.D. Aaron Metzger, Ph.D. Department of Sport Sciences Morgantown, West Virginia 2019 Keywords: mindfulness, perfectionism, performance, personality, elite athletes, sport psychology Copyright 2019 Erika D. Van Dyke ABSTRACT Being Mindful of Perfectionism and Performance Among Athletes in a Judged Sport Erika D. Van Dyke Literature pertaining to mindfulness and perfectionism in sport has expanded greatly in recent years. However, little research has integrated mindfulness and perfectionism, particularly within sports where athletes are judged on performance to a standard of perfection. The current study had two primary aims: (1) to explore profiles of mindfulness and perfectionism among intercollegiate gymnasts through a person-centered approach, and (2) to analyze differences in objective performance measures across the resulting profiles. The analytic sample consisted of 244 NCAA gymnasts representing NCAA Division I, II, and III institutions. Gymnasts completed self-report measures of mindfulness and perfectionism. Competitive performance records (i.e., national qualifying scores) were then gathered for participating gymnasts. Cluster analyses revealed a three-cluster solution: a moderate mindfulness, high perfectionism profile; a low mindfulness, low/moderate perfectionism profile; and a high mindfulness, very low perfectionism profile. Although competitive performance differences were not observed among the three profiles, exploratory post hoc pairwise comparisons indicated potential performance differences on vault and bars. Interestingly, gymnasts in different profiles performed more favorably on each event. Small to moderate effect size estimates provide some evidence that perfectionism may be adaptive to gymnastics performance. Elite level gymnasts were represented across three distinct profiles, suggesting that more than one profile of characteristics may be adaptive for reaching high levels of performance. Further, the sport context might be considered when interpreting the practical significance of the findings. The results can be used to help coaches, researchers, and practitioners better understand how mindfulness and perfectionism are expressed among athletes in a judged sport, and how these tendencies may be impactful in different ways. Future research exploring determinants of performance and mental health concurrently could provide further understanding of whether the characteristics that facilitate performance are congruent with those that facilitate wellbeing. BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE iii Acknowledgements Many people have crossed the path of my life during this project, and I will be forever grateful for the indelible footprints they have left there. Some of those people have simply left deeper imprints than others. These words acknowledge my gratitude for your presence on my journey. I would like to share an immense amount of appreciation for my advisor, Dr. Sam Zizzi. Thank you for extending me an invitation to join this esteemed doctoral program. Your supportive guidance and mentorship throughout this climb up research mountain has been invaluable, and has allowed me to grow as a scholar and more importantly as a person. Thank you for always inviting my curiosity, for your patience with my “thorough” process, for your white board check boxes of accountability, and for your gentle nudges to step back and get some perspective every so often. You have a remarkable way of seeing each of your students for who they are as learners. Thank you for seeing me. You have helped me to embrace the value in letting go, in imperfection, and in embracing my sunshine. Thank you for walking every step of the path along with me. The view from the top of this climb is spectacular. I would also like to thank Dr. Aaron Metzger for his incredible knowledge of the world of multivariate statistics. You provided the foundational learning that inspired many of the statistical directions taken throughout the course of this research project. Thank you for challenging me to explore interesting and complex questions, for making statistics accessible, and for giving me the skills and confidence necessary to carry out the analyses contained within these pages. I so appreciate you joining me for this glimpse into the experiences of athletes. My dissertation committee members, Dr. Etzel and Dr. Barnicle, have been such wonderful sources of support throughout this project and more broadly throughout my time here in the halls of CPASS. Thank you, Dr. Etzel , for greeting me each day with, “Ah little miss sunshine.” Your endearing reminders to “be well, and do good work,” to “walk between the raindrops,” and to know that “we’ll keep the light on for you” always made me smile and feel a valued member of our learning community. Your lovely words of wisdom and encouragement have and continue to mean so much to me. I am grateful to have been among the many students whose lives you have touched. Thank you, Dr. Barnicle, for always keeping your door open to my many many questions – often multiple in a given day. You have been such a supportive mentor, both in this project and in my growth as a teacher and consultant. I appreciate you being there for me throughout this learning journey. I am forever grateful that Dr. Zizzi crossed our paths, Candice (Clay) Brown. Thank you so incredibly much for the generous time and energy that you devoted to helping with my research project. From thoughtfully entering and checking data, to playing with statistical analyses, to reading drafts of the document, to presenting a poster together during CPASS research day. Your enthusiasm, dedication, and desire to learn and grow is inspiring, and I appreciate you lending some of your sparkle to our collaboration. Your presence throughout this project has been a blessing, and has brought me the joy of beginning my journey as a mentor. Thank you for your support. I am so excited to see what your very bright future will hold. BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE iv Thank you to all of the faculty in our Sport, Exercise, and Performance Psychology program for your belief in me, and for giving me the opportunity to pursue this path. One of the things that drew me here to WVU was the learning community cultivated by faculty and embraced by students in our program. I appreciate the legacy that you have built, and am so glad to have had the opportunity to be a part of this place. To my SEPP family. Thank you for your endless support, encouragement, smiles, and hugs throughout this journey, especially on days when the process felt daunting. A special note of appreciation for my cohort members, Adam, Carra, and Seth. What an amazing group of people with whom to share this experience. I am so grateful to have created so many memories with each of you – thank you for touching my life. I will continue to do my best to supply the sunshine and rainbows. A very special thank you must be extended to the roommates who have helped make this place feel like home during these past four years. Tammy, you led the way and I can only hope to follow in your footsteps to become a real professor someday. Carra, I am beyond grateful to be doing this journey along with you. Thank you for sharing all the laughs, tears, and moments of growth with me. I will forever remember our road trip to New York just after our first semester as doctoral students, and making a pact to do this thing together all the way through. We are so close now #HancockPact. Matt and Jordan (and Hunter), thank you for welcoming me into your little family these past three years, and for making Morgantown feel a little more like Northern California. I love you all. Thank you to my partner, Ben, for his love, laughter, and support. What a ride this has been! Whether living together or at a distance, your belief in me has shaped the course of this journey in a beautiful way. I appreciate you for your love of learning, endless curiosity, and for always being there by my side to go chasing waterfalls. Here’s to the adventure ahead. On a final note, to the people who gave me life, who instilled in me a joy of learning, and who have been my foundation through it all – my parents, Ivan and Anne Marie. I appreciate your unconditional love and support of my aspirations more than these simple words can express. Thank you for always believing that I can accomplish anything I set my mind and heart toward. You two are the best cheerleaders and consultants a daughter could ever wish for. I am so lucky you picked me. “And now that you don’t have to be perfect, you can be good.” – John Steinbeck BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE v Table of Contents Introduction…………………………………………………………………………….………….1 Pilot Study…………………………………………………………………………………………7 Method……………………………………………………………………………………….……8 Research Design and Sampling……………………………………………………………8 Instruments………………………………………………………………………………...8 Mindfulness……………………………………………………………………….8 Perfectionism………………………………………………………………….…...9 Demographic Questionnaire……………………………………………………...10 Competitive Gymnastics Performance…………………………………………...10 Procedures……………………………………………………………………………...…11 Statistical Analyses……………………………………………………………………….11 Results……………………………………………………………………………………………13 Determining the Analytic Sample………………………………………………………13 Descriptive Statistics on the Analytic Sample………………………………………….14 Cluster Interpretation…………………………………………………………………..15 Comparing Performance Across Clusters………………………………………………16 Discussion………………………………………………………………………………………..17 Tables and Figures………………………………………………………………………………..27 Literature Review……………………………………………………………………….………..31 Mindfulness: A Brief Backdrop………………………………………………………….31 Historical Roots…………………………………………………………………..32 Mindfulness-Based Approaches……………………………………………….....33 BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE vi Mindfulness in Sport and Performance Psychology…………………………………….34 Mindfulness-Based Approaches in Sport………………………………………..35 Linking Mindfulness-Based and Traditional PST Approaches…………………37 Mindfulness, Performance, and Psychological Aspects of Sport……………….39 Perfectionism in Sport and Performance Psychology…………………………………..56 Measurement……………………………………………………………………………66 Mindfulness………………………………………………………………….….66 Perfectionism…………………………………………………………………….70 Performance……………………………………………………………………...75 Directions for Future Research……………………………………………………..........77 References………………………………………………………………………………………..82 Table 1: Mindfulness Measurements…………………………………………………………….95 Table 2: Perfectionism Measurements…………………………………………………………...96 Table 3: Adapted Mindfulness-Flow-Performance Model in Sport………………………………97 Appendix A: Athlete Mindfulness Questionnaire………………………………………………..98 Appendix B: Sport Multidimensional Perfectionism Scale-2…………………………………...100 Appendix C: Demographic Questionnaire……………………………………………………...102 Appendix D: Email to Gymnastics Coaches…………………………………………………….104 Appendix E: Cover Letter and Informed Consent……………………………………………….105 Appendix F: IRB Approval……………………………………………………………………..106 BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE 1 Introduction Researchers studying mindfulness and acceptance-based evidence among athletes support a cautious yet optimistic view regarding the efficacy of such approaches in the context of sport (e.g., McAlarnen & Longshore, 2017; Noetel, Ciarrochi, Van Zanden, & Lonsdale, 2017; Sappington & Longshore, 2015). In a recent systematic review, Noetel et al. (2017) analyzed over 60 studies of mindfulness and acceptance-based approaches intended to promote positive sport outcomes, including athletic performance. Despite finding large effect sizes for the performance benefits of such interventions, the findings were deemed low in quality, lacking precision in effect sizes and consistency. Among the individual studies reviewed, researchers found preliminary support for mindfulness and acceptance-based interventions across a variety of sport outcomes (e.g., performance, flow, present-moment awareness, confidence, injury prevention, competitive anxiety, and burnout). Continued research efforts with increased rigor are thus needed to support mindfulness as an approach for enhancing sport performance and related outcomes. At their core, mindfulness and acceptance-based approaches focus on modifying the relationship one has with internal experiences (e.g., physical sensations, emotions, cognitions), rather than deliberately aiming to change, consciously control, suppress, or reduce internal experiences (e.g., Gardner & Moore, 2004, 2007; Hayes, Strosahl, & Wilson, 1999; Kabat-Zinn, 1982). Such mindfulness approaches cultivate present-focused awareness and attention, a nonjudgmental and accepting approach to situations, openness and curiosity toward experience, and compassion for self and others – elements that comprise the flavor of mindfulness (Zizzi, 2017). Recognizing the applicability of mindfulness-based approaches to the context of sport, BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE 2 researchers examined the link between mindfulness and objective measures of sport performance using a variety of study designs. In support of a mindful approach to facilitating sport performance, researchers found that a greater number of athletes who engaged in a mindfulness and acceptance-based program improved their national performance ranking compared to those who took part in a traditional change-based program (Bernier, Thienot, Codron, & Fournier, 2009). Authors of another study found that elite shooters in a mindfulness meditation group experienced significant increases in shooting performance (i.e., mean performance score increase from 528 to 544, SD = 13) and significant decreases in pre-competition anxiety (i.e., mean salivary cortisol level decrease from 1.33 to 0.66, SD = 0.07) from pre- to post-test compared to a control group (John, Verma, & Khanna, 2011). In a non-intervention study, Gooding and Gardner (2009) found that collegiate athletes’ levels of mindfulness significantly predicted basketball free throw shooting percentage in games across the competitive season (i.e., one standard deviation increase in mindfulness scores resulted in a 5.75% increase in free throw shooting percentage). When competitive experience was controlled for, however, mindfulness no longer arose as a significant predictor of competitive performance. Gooding and Gardner noted that competitive experience and mindfulness may predict sport performance through shared variance. Taken together, researchers suggest that mindfulness may influence objectively measured performance in sport among high level athletes. Further research may provide additional clarity regarding the utility of mindfulness for real-world competitive performance. As mindfulness research continues to evolve, studies that take a more nuanced approach to studying the mindfulness-performance relationship are needed. These kinds of studies can clarify how, when, and by whom mindfulness could be most useful. In a sport like gymnastics BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE 3 for instance, in which athletes are judged to a standard of perfection in their skills and technique, perfectionism may be considered contextually relevant to the study of mindfulness-performance relationships. Thus, the study of mindfulness and perfectionism among judged sport athletes may provide more nuanced insights regarding how these constructs are experienced together among individuals, and how those unique experiences may be related to individual differences in sport performance. Perfectionism has been defined as “a personality disposition characterized by striving for flawlessness and setting exceedingly high standards for performance, accompanied by tendencies for overly critical evaluations” (Stoeber, 2012, p. 294). Although different frameworks of perfectionism have been proposed in the literature (e.g., Frost, Marten, Lahart, & Rosenblate, 1990; Hewitt & Flett, 1991), researchers acknowledge that perfectionism is a multidimensional construct. Stoeber and Otto (2006) provided a way to conceptually integrate different proposed frameworks based on two higher-order dimensions of perfectionism – perfectionistic strivings and perfectionistic concerns. Although a point of some debate among perfectionism researchers in recent years (e.g., Flett & Hewitt, 2005), perfectionistic strivings have often been considered adaptive and facilitative of performance, whereas perfectionistic concerns have often been considered maladaptive and debilitative of performance in sport (e.g., Gotwals, Stoeber, Dunn & Stoll, 2012; Stoeber, 2012). In a recent meta-analytic review of multidimensional perfectionism in sport, the researchers further highlighted the complexity of these relationships, noting that perfectionistic concerns seem to be clearly maladaptive, while perfectionistic strivings may be adaptive or maladaptive for athletes (Hill, Mallinson-Howard, & Jowett, 2018). Considering potential contrasts between elements of perfectionism (e.g., self-critical evaluations) and elements of mindfulness (e.g., acceptance and self-compassion), further research integrating the BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE 4 two constructs is needed to understand how mindfulness and perfectionism may interact in athletes’ experiences. Many researchers have examined relationships between perfectionism and competitive sport performance among high level athletes (e.g., Gotwals et al., 2012; Hill et al., 2018). Relatively few researchers, however, have explored the intersection of mindfulness and perfectionism in sport. In their follow-up study of the long-term impact of mindfulness-based programming for sport performance, Thompson, Kaufman, De Petrillo, Glass, and Arnkoff (2011) found significant performance improvements in long-distance runners’ mile times from pre- and posttest to follow-up. In addition, the authors found negative associations between performance improvements and aspects of perfectionism, including overall trait perfectionism (r = .74), concern over mistakes (r = .69), and doubts about actions (r = .75). It is important to note that negative relationships are reflected in the positive correlations reported because performance improvement is measured through decreased mile time. Although the links between mindfulness, perfectionism, and performance should be interpreted with care due to the correlational nature of the study, Thompson et al. (2011) support the notion that mindfulness may be related to performance benefits, and that certain dimensions of perfectionism may negatively influence athletic performance. A recent critical evaluation of mindfulness research has raised questions regarding the varied definitions and measurements of mindfulness found in the literature, as well as the proposed, seemingly unquestioned, benefits of mindfulness without regard for potential adverse effects (Van Dam et al., 2018). Rather than leaping with naïve enthusiasm into the application of mindfulness interventions within a new population, researchers might first aim to understand the typical mindfulness experiences of those athletes. Taking a person-centered analytic approach to BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE 5 the study of mindfulness may help bring that explicit attention to salient features of mindfulness among athletes in a specific sport setting. Further, new insights gained about the athletes’ mindfulness experiences will be highly influenced by the specific core concepts measured in the selected instruments. Considering the work of Van Dam et al. (2018), researchers are encouraged to clearly define the flavor(s) of mindfulness assessed to enhance interpretability of future research findings. Person-centered approaches to data analysis allow the researcher to better understand unique profiles of the key constructs measured among individuals. Although person-centered approaches have been used in the study of mindfulness (e.g., Kee & Wang, 2008) and perfectionism (e.g., Gucciardi, Mahoney, Jalleh, Donovan, & Parkes, 2012), these constructs have yet to be studied concurrently through cluster analytic approaches. In their person-centered approach to the study of mindfulness in sport, Kee and Wang (2008) identified a four-cluster solution based on university athletes’ mindfulness scores (Mindfulness/Mindlessness Scale, MMS; Bodner & Langer, 2001). Individuals in the profile highest in mindfulness showed the most frequent use of psychological skills in sport. Specifically, athletes in the cluster highest in mindfulness had significantly higher goal setting, positive self-talk, and imagery compared to those in clusters lower in mindfulness characteristics. In addition, the researchers found significant differences in flow dispositions across the four mindfulness clusters. Thus, Kee and Wang identified a stable mindfulness cluster solution, and found significant differences among those mindfulness profiles on the outcome variables of psychological skill use and flow dispositions. Along this line of person-centered research, Gucciardi et al. (2012) explored profiles of perfectionism within a heterogeneous sample of elite athletes through a cluster analytic BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE 6 approach. The researchers found three distinct profiles of perfectionism among athletes based on the Sport Multidimensional Perfectionism Scale (Sport-MPS; Dunn, Causgrove Dunn, & Syrotuik, 2002): (1) adaptive perfectionists – high personal standards, low concern over mistakes, moderate perceived parent/coach pressures, (2) maladaptive perfectionists – high concern over mistakes and perceived parent/coach pressures, moderate/high personal standards, and (3) non-perfectionists – low personal standards and concern over mistakes, moderate perceived parent/coach pressures. Gucciardi et al. further revealed significant differences in motivational orientations among the perfectionism profiles. Specifically, adaptive perfectionists reported significantly lower levels of fear of failure, performance approach and avoidance goals, and mastery avoidance goals, and significantly higher levels of mastery approach goals than did maladaptive perfectionists. Non-perfectionists reported significantly lower levels of the motivational orientations assessed than did maladaptive perfectionists, and lower levels of mastery approach goals, performance approach goals, and intrinsic motivation than did adaptive perfectionists. The researchers thus supported a stable three cluster conceptualization of perfectionism, and found differences in motivational outcomes across those perfectionism profiles. Gucciardi et al. highlighted that among these elite athletes both adaptive and maladaptive perfectionists had high levels of personal standards, and that it was primarily the presence or absence of overly critical self-evaluations that differentiated maladaptive from adaptive perfectionists, respectively. Many existing studies that have examined mindfulness and links to performance outcomes have done so following programming and interventions (e.g., John et al., 2011; Thompson et al., 2011). Little research has explored athletes’ typical tendencies toward mindfulness and perfectionism qualities, and how together these characteristics may relate to BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE 7 performance in unique ways. The current study will thus explore mindfulness, perfectionism, and performance among gymnasts – athletes in a judged sport. Aims of the research will be twofold: (1) to examine whether unique profiles of mindfulness and perfectionism constructs exist among the athlete participants, and (2) to assess whether objective measures of competitive performance differ across the unique mindfulness and perfectionism profiles. Pilot Study The purpose of the pilot study was to assess the psychometric properties of mindfulness and perfectionism measures, and to explore relationships between mindfulness and perfectionism among intercollegiate gymnasts. Participants were female gymnasts (N = 301), ranging in age from 18 to 22 years (M = 19.46, SD = 1.20), who attended NCAA Division I, II, or III colleges and universities in the United States. Gymnasts completed the Athlete Mindfulness Questionnaire (AMQ; Zhang, Chung, & Si, 2017), and the personal standards and concern over mistakes subscales of the Sport Multidimensional Perfectionism Scale-2 (i.e., Sport-MPS-2; Gotwals & Dunn, 2009). Results of the correlational analyses supported theoretically expected associations among constructs (e.g., present-moment attention and awareness, r = .62; acceptance and concern over mistakes, r = -.41), and internal reliability coefficients across the mindfulness and perfectionism subscales ranged from .74 to .88. Confirmatory factor analyses supported the original three-factor structure of the AMQ [RMSEA = .06, SRMR = .05, CFI = .92]. The two-factor perfectionism model did not show good fit to the data. Modification indices were reviewed to evaluate potential model fit improvements. When residual error terms of items in the concern over mistakes factor were allowed to covary, the two-factor model fit improved markedly [RMSEA = .08, SRMR = .08, CFI = .91]. Through the pilot study, the research team supported the use of the AMQ among intercollegiate level gymnasts, and provided further BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE 8 understanding about relationships between mindfulness and perfectionism among athletes in a judged sport context. Research Design and Sampling Method The current study built upon the pilot study to explore mindfulness, perfectionism, and performance using a quantitative, descriptive correlational, research design. Female gymnasts 18 years of age or older attending National Collegiate Athletic Association (NCAA) Division I, II, or III colleges and universities in the United States who took part in the pilot study also participated in the present study. Prior to the pilot study, convenience sampling was used to contact all NCAA collegiate women’s gymnastics coaches with available contact information to seek permission to collect data with their teams. Gymnasts who previously completed the survey and reported their name were included in the first part of the study exploring mindfulness and perfectionism profiles. Participants who performed on at least one event during the 2019 competition season from January to April, and for whom a National Qualifying Score (NQS) could be calculated, were included in the second part of the study assessing performance differences among the resulting profiles. Instruments Mindfulness. The Athlete Mindfulness Questionnaire (AMQ; Zhang et al., 2017) was used to measure mindfulness in the current study. The AMQ is a 16-item, 3-factor measure of mindfulness for athletes. Mindfulness is assessed based on the subscales present moment attention (e.g., When I find myself distracted, I gently bring my attention back to my training), awareness (e.g., I am aware that my emotions during training and competition can influence my thinking and behavior), and acceptance (e.g., During training and competition, it doesn’t matter BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE 9 if the situation is good or bad, I can accept myself for who I am). Items are rated on 5-point Likert scales ranging from 1 (never true) to 5 (always true). Confirmatory factor analysis (CFA) revealed satisfactory fit indices for the 16-item, 3-factor structure of the instrument, X2(101) = 221.28, p < .001, CFI = 0.95, TLI = 0.94, WRMR = 1.04, RMSEA = 0.06. Internal consistency reliabilities for the three AMQ subscales ranged from 0.64 to 0.76. In the present study, internal consistency reliabilities for the subscales were slightly higher, ranging from 0.75 to 0.77. Convergent validity for the AMQ was supported through significant positive associations between the present moment attention, awareness, and acceptance subscales of the AMQ and mindfulness as measured by the Mindful Attention Awareness Scale (MAAS; Brown & Ryan 2003). Concurrent validity for the three subscales of the AMQ was also supported through significant negative relationships with burnout and experiential avoidance, and significant positive relationships with well-being, positive affect, and dispositional flow. Perfectionism. The full version of the Sport Multidimensional Perfectionism Scale-2 (Sport-MPS-2; Gotwals & Dunn, 2009; Gotwals, Dunn, Causgrove Dunn, & Gamache, 2010) is a 42-item, 6-factor measure assessing the multidimensional nature of perfectionism in sport. The subscale dimensions include personal standards, concern over mistakes, perceived parental pressure, perceived coach pressure, doubts about actions, and organization. Items are rated on 5 point Likert scales ranging from 1 (strongly disagree) to 5 (strongly agree). Athletes are asked to report how they “view certain aspects of their competitive experiences in sport.” For the purpose of the present study, two subscales, namely personal standards (e.g., It is important to me that I be thoroughly competent in everything I do in my sport) and concern over mistakes (e.g., I should be upset if I make a mistake in competition), were used to measure the two higher-order constructs of perfectionistic strivings and perfectionistic concerns, respectively. The personal BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE 10 standards and concern over mistakes subscales were selected because they have been recommended as indicators of perfectionistic strivings and perfectionistic concerns in sport (Stoeber & Madigan, 2016). The reliability of the Sport-MPS-2 has been demonstrated among athletes, with internal consistencies of .74 and .79 for the personal standards and concern over mistakes subscales, respectively (Gotwals & Dunn, 2009). In the current study, internal consistency reliabilities for the personal standards and concern over mistakes subscales were .77 and .88, respectively. Demographic questionnaire. The demographic questionnaire assessed participants’ age, highest level in gymnastics attained before college, race, and ethnicity. Gymnast participants were asked to report their name, and the college or university they attend so the research team could access their publicly available competition scores from an online platform. Experience and satisfaction working with a sport psychology professional, as well as experience with mindfulness, were also assessed. All identifiable information was held confidential, and all data gathered for the study was reported in aggregate to protect the anonymity of participants. Competitive gymnastics performance. Measurement of competitive gymnastics performance was based on the National Qualifying Score (NQS). The NQS is used in collegiate gymnastics to determine placement in post-season competition, and is calculated based on the following criteria: (1) three highest away scores on a given event, plus (2) next three highest scores on a given event, whether home or away, (3) drop the highest of the six scores, (4) average the remaining five scores. An NQS was calculated for each gymnast who participated in the study for each event on which she competed during the regular meet season to allow for comparison of mindfulness – perfectionism – performance relationships across events (i.e., vault, uneven bars, balance beam, floor exercise). BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE 11 Procedures After gaining approval from the Institutional Review Board (IRB) to conduct the research methods necessary for both the pilot and current studies, the researcher contacted college gymnastics coaches through email correspondence to request permission to collect data with their athletes. Coaches had the option to receive either paper copies of the counterbalanced questionnaires via mail, or a Qualtrics link to an online version of the questionnaires via email. Coaches then made the surveys available to their athletes to complete on a voluntary basis. All athletes receiving the survey were initially presented with a cover letter description of the study, and were asked to provide their consent to participate. Survey data were collected prior to, or in the beginning of, the competition season for each participant. All data were reported in aggregate, and any identifying information collected in the surveys was used for the sole purpose of accessing the gymnasts’ publicly available competitive performance data. Event scores ranging from 0.00 to 10.00 were retrieved online post-season from https://roadtonationals.com/results/standings/ for gymnasts who completed the questionnaires and who performed in a sufficient number of competitions throughout the season to have an NQS on at least one event. Statistical Analyses For the pilot study, data cleaning and preliminary analyses were conducted and reviewed to determine whether necessary assumptions for the substantive analyses were met. Missing values across the mindfulness and perfectionism items were assessed using Little’s MCAR test (p = .769), and no single item exceeded 1.3% missing data. Due to the low number and random nature of missing values, Expectation-Maximization (EM) procedures were then used to impute missing values. As expected, the gymnastics performance data on each of the four events were BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE 12 negatively skewed. Reflection and log base 10 transformation approaches improved the distribution of scores for use in later parametric statistical analyses. For the current analytic sample, descriptive statistics were calculated on demographic and key study variables. Internal consistency reliabilities were also assessed for the measures, as were correlational analyses for the performance data and each of the subscales in the mindfulness and perfectionism instruments. For the current study, cluster analytic approaches were conducted to establish whether unique profiles of mindfulness and perfectionism were present among the gymnast participants. Gymnasts’ scores on the three mindfulness and two perfectionism subscales were submitted to cluster analysis in a two-step procedure: (1) Ward’s hierarchical cluster analysis with squared Euclidean distance was conducted to help determine an initial number of clusters present among the participants, and (2) k-means iterative cluster analysis was used to further refine the cluster solution suggested by the initial hierarchical cluster analysis. The use of both hierarchical and iterative approaches to cluster analysis is supported in the literature (e.g., Gucciardi et al., 2012; Hair, Anderson, Tatham, & Black, 1998; Kee & Wang, 2008). A series of chi square analyses were then carried out to assess the number of athletes in each cluster who performed on each gymnastics event. This step allowed the research team to determine whether a sufficient sample size for each cluster/profile was met prior to conducting subsequent analyses. One-way analysis of variance (ANOVA) was then used to assess differences among the resulting mindfulness/perfectionism clusters in an objective measure of gymnastics performance (i.e., NQS) on each competitive event. Thus, four one-way ANOVAs were conducted to examine competitive performance differences across the clusters on each gymnastics event (i.e., vault, uneven bars, balance beam, floor exercise). Effect size estimates are reported as partial eta BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE 13 squared for the ANOVA models, and as Cohen’s d for pairwise mean comparisons. All statistical analyses were computed in recent versions of SPSS. Results Of the total number of participants who completed the mindfulness and perfectionism survey measures (N = 301), 244 gymnasts provided their name. It was only possible to access competition results for athletes who provided their names. Thus, the analytic sample for the current study was 244 gymnasts. Determining the Analytic Sample Prior to conducting the substantive analyses for the study, the research team conducted a series of preliminary analyses to assess potential differences between participants who would be retained in the analytic sample (n = 244) and those who would be omitted from the sample (n = 57) on demographic and key study variables. T-tests indicated no statistically significant (p > .05) differences between the gymnasts on the three mindfulness subscales. Statistically significant (p < .05) differences were found, however, between the participants on the two perfectionism subscales. Gymnasts who did not report their name indicated slightly higher perfectionism scores than those who did report their name for both concern over mistakes (no name: M = 3.18, SD = 0.78; name: M = 2.87, SD = 0.82) and personal standards (no name: M = 3.88, SD = 0.55; name: M = 3.61, SD = 0.58) subscales. Gymnasts who did not report their name were also slightly higher in skill level than the larger group of participants who did provide their name (mean difference = 0.22). Given these differences, cluster analyses were conducted for both the full (N = 301) and analytic (N = 244) samples. Chi square analyses were then assessed to determine the proportion of athletes in each cluster who did and did not report their name. Although gymnasts who did not BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE 14 provide their name were disproportionately found in clusters high in perfectionism dimensions, parallel to the findings of the t-tests, the resulting cluster solutions were similar between the two samples, indicating that the gymnasts who omitted their name were likely not driving the cluster solution. The research team thus decided to move forward in reporting the cluster solution established with the analytic sample of gymnasts who provided their name (N = 244), and who would be included in the subsequent analyses incorporating performance data. Descriptive Statistics on the Analytic Sample Demographic information for the analytic sample of collegiate gymnasts is presented in Table 1. Gymnasts in the present study ranged in age from 18 to 22 years (M = 19.46, SD = 1.22) and were predominantly white (n = 196). A majority of the gymnasts reached level 10 prior to college (n = 204), had previous experience with a sport psychology professional (n = 162) and were satisfied with their experience (n = 130), and came into the study with no prior experience with mindfulness (n = 150). Correlations, descriptive statistics, and internal consistency coefficients for the mindfulness and perfectionism subscale scores are outlined in Table 2. As expected, statistically significant (p < .01) positive correlations arose among the three mindfulness subscales, and between the two perfectionism subscales. Significant, weak positive correlations were also found between personal standards perfectionism and both present moment attention (r = .28) and awareness (r = .16); yet there was no relationship between personal standards and acceptance (r < .01). Significant negative correlations arose between concern over mistakes perfectionism and each of the mindfulness subscales, though only one was moderate in size: present moment attention (r = -.14), awareness (r = -.15), and acceptance (r = -.42). The negative relationships between mindfulness dimensions and concern over mistakes perfectionism, as well as the BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE 15 positive relationships between aspects of mindfulness and personal standards perfectionism, align well with previous research and theoretical perspectives of the constructs. Cluster Interpretation The five subscales were entered first into Ward’s hierarchical cluster analysis. Based on visual inspection of the resulting dendogram and graphed coefficients from the agglomeration schedule, a three- or five-cluster solution appeared to provide the best description of the data. Together, the dendogram and agglomeration schedule help indicate points at which dissimilar clusters were being forced to merge, and thus provide information to determine relatively distinct groupings of individuals. Centroid values from both the three- and five-cluster solutions were then taken forward to be used as initial seed points in the k-means iterative cluster analyses. The hierarchical and iterative cluster analytic approaches were then compared for both the three- and five-cluster solutions to assess the stability of the two solutions. Through crosstabulation, the percentage of cases similarly assigned to each cluster across the two analytic approaches could be assessed. Specifically, case classification for both the Ward’s and k-means analyses indicated 69% similarity for the three-cluster solution, compared to a slightly improved 73% similarity for the five-cluster solution. Despite a small increase in stability for the five cluster solution, power for subsequent statistical analyses would decrease notably given the smaller number of athletes in each group when moving from three- to five-clusters. The three cluster solution was therefore retained as it provided a nice explanation of gymnast mindfulness and perfectionism characteristics while permitting heightened power for subsequent analyses. Cluster means, standard deviations, and standardized scores for the final cluster solution are shown in Table 3. A visual representation of the final cluster solution is depicted in Figure 1. BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE 16 Interpretation of the subscale means for gymnasts in each cluster revealed the presence of three distinct profiles of mindfulness and perfectionism tendencies. The first cluster consisted of 87 athletes (35.7%) with a moderate mindfulness and high perfectionism profile. The second cluster consisted of 71 athletes (29.1%) with a low mindfulness and low/moderate perfectionism profile. The third cluster consisted of 86 athletes (35.3%) with a high mindfulness and very low perfectionism profile. A series of four crosstabulation analyses were conducted to preliminarily assess the number of athletes in each profile who obtained an NQS on each competitive event during the season. Across the analyses, clusters ranged in size from 17 to 34 gymnasts. The group sizes were thus considered adequate for conducting subsequent one-way ANOVAs to assess potential performance differences across the three mindfulness and perfectionism profiles. Comparing Performance Across Clusters Prior to conducting parametric statistics using the performance data, transformations were made to improve the normality of the data. A series of four one-way ANOVAs were then computed using the transformed performance data to assess differences among the profiles. Event means, standard deviations, and sample sizes for each cluster are shown in Table 4. Results of the ANOVAs may be found in Table 5. No statistically significant (p < .05) differences in performance were found among the mindfulness and perfectionism profiles on the four competitive events. Effect sizes (h2 p) across the four analyses ranged from 0.03 to 0.06, and power estimates for this set of analyses was low. Exploratory post hoc t-tests were then conducted between groups with the greatest mean score differences on each event. The event NQS means compared, as well as the Cohen’s d effect size for each pairwise comparison, are indicated in Table 4. Significant mean score differences were found between clusters on vault and bars. On vault, the high mindfulness, very low BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE 17 perfectionism cluster performed significantly better than the low mindfulness, low/moderate perfectionism cluster. On bars, the moderate mindfulness, high perfectionism cluster performed significantly better than the high mindfulness, very low perfectionism cluster. Small to moderate effect sizes were found for each of the NQS comparisons across the four events. Statistically significant correlations also arose between vault performance and acceptance (r = .24, p < .05), between bars performance and both personal standards (r = .31, p < .01) and concern over mistakes (r = .23, p = .05), between beam performance and personal standards (r = .23, p = .05), and between floor performance and awareness (r = .24, p < .05). Thus, despite the lack of significant findings when performance differences among the three profiles were considered together, potential relationships may exist between the mindfulness and perfectionism tendencies and competitive gymnastics performance. Discussion Three distinct profiles of mindfulness and perfectionism were observed among the gymnasts. Previous researchers have contributed to our understanding of how athletes may be grouped on each of these constructs independently (e.g., Gucciardi et al., 2012; Kee & Wang, 2008); however, this is the first study to our knowledge that used a cluster analytic approach to understand how athletes in a judged sport experience mindfulness and perfectionism together. Previously, researchers have supported stable three- and four- cluster solutions for perfectionism and mindfulness, respectively. Thus, arriving at a three-cluster solution that helps to explain gymnasts’ propensities for perfectionism and mindfulness concurrently aligns closely with existing notions of the constructs. Furthermore, the current finding that gymnasts were classified quite evenly across three profiles indicates that mindfulness and perfectionism may be experienced in varied ways even BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE 18 within a relatively homogeneous group of elite athletes performing at a very high level. This gymnast-specific finding appears consistent with previous multi-sport research, as profiles of mindfulness and perfectionism tendencies have independently been found to vary markedly among high level athletes representing a broad range of sports (e.g., Gucciardi et al., 2012; Kee & Wang, 2008). Although exploring a diverse set of personality characteristics was beyond the scope of the present study, these findings could be used to support the idea that there may not be just one adaptive personality profile for attaining an elite level of gymnastics. Furthermore, a national qualifying score (i.e., successful and consistent performance across a season) was achieved by athletes with and without notable levels of perfectionism and mindfulness. When discussing differences observed across the mindfulness and perfectionism profiles, for instance gymnasts “high in mindfulness” or “low in perfectionism,” scores are considered relative to the other gymnasts who participated in the study rather than compared to some normative criteria. Still, a basic understanding of how the current participants compared on the measurements to athletes included in previous research can provide clarity and highlight points of similarity and difference across studies. Zhang et al. (2017) found during development of the AMQ that subscale means among team and individual sport athletes ranged from 3.54 to 3.73, with standard deviations in the 0.82 to 0.99 range. Comparatively, the gymnasts in our sample had slightly higher mindfulness scores overall, with means ranging from 3.63 to 4.11 across the three subscales, and lower standard deviations in the range of 0.48 to 0.59. Taken together, the single-sport sample of gymnasts who participated in the current study reported slightly higher mindfulness tendencies with lower variation across responses than the multi-sport sample of athletes used when developing the instrument. BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE 19 In their research examining the validity of the Sport-MPS-2, Gotwals and Dunn (2009) reported mean scores for intercollegiate team sport athletes on the personal standards and concern over mistakes subscales were 3.68 (SD = 0.52) and 2.87 (SD = 0.68), respectively. In the present sample, the perfectionism scores among the intercollegiate gymnasts were equivalent or just slightly lower, with means of 3.61 (SD = 0.58) and 2.87 (SD = 0.82) for the personal standards and concern over mistakes subscales, respectively. Previously, Dunn et al. (2006) examined perfectionism among a sample of female figure skaters, a group slightly younger than the athletes in our gymnast sample but in a similar individual, judged sport context. The researchers found that the figure skaters’ scores for personal standards and concern over mistakes were also lower than in the team sport sample, with subscale means of 3.33 (SD = 0.86) and 2.37 (SD = 0.97) for personal standards and concern over mistakes, respectively. Thus, the notion that certain sports may be considered more “perfectionistic” than others may not necessarily equate to higher individual reports of perfectionism tendencies. Perhaps athletes in such sport contexts have normalized the pursuit of perfection in a different manner than athletes in team sports might, and thus may self-report their own perfectionism relative to a higher standard resulting in lower scores. In any case, the gymnasts in the present study scored on the perfectionism dimensions in ways that align closely with intercollegiate athletes in various team sports studied previously. Another primary aim of the current study was to assess potential performance differences across the mindfulness and perfectionism clusters. When gymnastics performance was considered on each of the four events across the three profiles, no significant differences were observed. It is important to note that the performance metric selected for comparison was relatively restricted in range given the calculation of the NQS as a snapshot of the better BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE 20 performances for each gymnast during the competition season (i.e., best and worst performances are dropped). Still, the NQS was selected intentionally as it is a practical performance metric used in collegiate gymnastics for qualification/placement in post-season competition. Given the utility of this performance score in the sport of gymnastics, the current findings may have practical significance. Specifically, high level performance appears to be attainable for athletes across competitive events, regardless of their propensities for mindfulness and perfectionism. Gymnasts may learn coping strategies that allow them to perform successfully with varying degrees of mindfulness and perfectionism tendencies. This finding may have practical utility for coaches and athletes – that individuals with different personality characteristics are all capable of high quality performance. Despite the lack of statistically significant differences among the three profiles, noteworthy patterns in mean differences in event performance were observed. Researchers have argued that reliance on p values when determining the meaning of results may be problematic, as the statistic is highly contingent on sample size and says relatively little about the real-world meaning of findings (e.g., Gigerenzer, 2004; Kruschke, 2013; Wilkinson, 2014). In recent years, researchers have recommended interpreting test statistics through a contextual lens to better understand the practical significance of the results (Andersen, McCullagh, & Wilson, 2007). A similar process was recently adopted in a study of mindfulness-based programming for injury prevention (Ivarsson, Johnson, Andersen, Fallby, & Altemyr, 2015), and we have followed these recommendations. In the context of gymnastics, very small differences in performance scores may mean the difference between placing and not placing, between qualifying to post-season competition and not qualifying. BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE 21 For example, the difference in NQS for the top 10 athletes on each competitive event during the 2019 regular season was approximately five hundredths of a point, 0.05. The mean score differences in the one to two tenth of a point range, 0.1 to 0.2, observed between profiles established in the current study were therefore considered to have real-world meaning in the context of gymnastics. For the comparison across profiles, h2 p effect size values ranged from 0.03 to 0.06, indicating that approximately three to six percent of variance in event performance may be accounted for by profile membership. In our sample of gymnasts, NQS values had a range of approximately one point on each event. Given the explained variance in event performance we observed in the study, a three to six percent change in NQS may mean the difference between first and tenth in the nation. Additionally, when effect sizes were computed for event scores with maximum variation between profiles, Cohen’s d values ranged from 0.37 to 0.65 indicating the presence of some moderate effect sizes. Thus, it is possible that some qualities of mindfulness and perfectionism serve athletes in more adaptive ways on certain competitive events. In support of this finding, statistically significant mean differences arose between the profiles compared on vault and on bars in post hoc pairwise comparisons. On vault, gymnasts highest in mindfulness performed best; whereas, on bars, gymnasts highest in perfectionism performed best. When observing the mean score patterns on each event across profiles, gymnasts highest in perfectionism performed relatively better than gymnasts in profiles lower in perfectionism on bars, beam, and floor. These trends suggest that the degree of mindfulness and perfectionism qualities most favorable for performance may vary by event, and that in contrast to existing research (e.g., Hill et al., 2018) perfectionism may be adaptive for competitive gymnastics performance in the context of the study. BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE 22 It seems possible that the mindfulness and perfectionism qualities favorable for performance may also differ from those favorable for wellbeing. For instance, although the findings of the current study lend initial, tentative support for the physical performance benefits of perfectionism in gymnastics, high levels of concern over mistakes perfectionism has previously been considered maladaptive for the emotion and wellbeing of athletes (e.g., Hill et al., 2018). Researchers have provided some evidence of the potential benefits of mindfulness for the wellbeing of athletes, including reduced competitive anxiety, stress, and burnout; injury prevention; and increased confidence and self-efficacy (e.g., Noetel et al., 2017). Although the question of athlete wellbeing was not within the scope of the current research, future researchers might consider including such assessments when studying these constructs to better understand both the quality of athletes’ experiences alongside their objective performance. The current findings may also have practical implications for researchers and practitioners interested in delivering mindfulness-based interventions to athletes such as those who participated in our study. In light of some equivocal research findings in the mindfulness literature, researchers have begun to question the efficacy of mindfulness-based interventions for all participants, and have raised the potential for adverse effects of meditation-related experiences (Van Dam et al., 2018). Understanding individual characteristics that may predispose individuals to have more favorable versus adverse effects to meditation-related practices may help to direct our research and applied efforts toward participants who may benefit most from them. For instance, concern over mistakes perfectionism has been positively associated with anxiety and depressive symptoms (Hill et al., 2018) – experiences that align with the relatively rare albeit real meditation-related adverse effects that have been observed (Van Dam et al., 2018). Future researchers might aim to understand if differences exist across profiles BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE 23 of individual characteristics in athletes’ readiness or desire to engage in mindfulness-based interventions, or in the effectiveness of such interventions among distinct profiles. By assessing these constructs prior to interventions, researchers or practitioners might then have a clearer understanding of who may benefit most from mindfulness-based interventions rather than assuming that all athletes participating have the capacity to benefit equally from such practices. We selected the NQS as our performance measure in the current research based on its practical use in determining placement in post-season competition. We recognize, however, that there may be potential limitations to this metric in our study. For instance, given the way the NQS is calculated, only a portion of a gymnasts’ performances throughout the season may be taken into consideration. Further, that portion of the season that is calculated into the NQS would not account for poorer competitive performances if gymnasts competed in most meets during the season. This performance measure may therefore be overly reductive and restricted in range, potentially limiting our ability to notice performance differences across the profiles on the different apparatus using inferential statistics. Future researchers might consider using a different performance metric (e.g., true average of all performances or range of performance scores) or a mix of measures to study the relationship between these concepts and performance. Self-report measures were used to assess individual experiences of both mindfulness and perfectionism among the gymnasts. When self-report measures are involved, there is a potential for bias in the form of accurate recall or social desirability. The potential for social desirability bias in the reported experiences may be especially salient given that athletes were asked to provide identifying information. During the initial screening analyses, small but significant differences in perfectionism scores were noted between gymnasts who did and did not provide their name, with those who did not report their name scoring higher in dimensions of BEING MINDFUL OF PERFECTIONISM AND PERFORMANCE 24 perfectionism than those who did. Although provision of names was a necessary component for carrying out the current study with objective performance measures, it should be noted that authentic self-report of perfectionism may be influenced to some extent when identifying information is provided. We accessed a large sample of elite level gymnasts to participate in the current study. Still, only small effects were found across profiles relative to objective performance – likely due to the restricted range in performance scores observed among this relatively homogeneous, high level sample of athletes, and to the inherently small increments of change frequently observed in the scoring system within gymnastics. With a larger sample of athletes, future researchers may be able to heighten the power to detect small, meaningful differences in performance among the different profiles. Still, that differences in performance outcomes were found between profiles may hold meaning for coaches, athletes, and practitioners operating in a sport where exceedingly small differences in scores often can have large practical consequences. Several directions for future research stem from the present findings. In the current study, we aimed to take a nuanced approach to studying the relationship between mindfulness and performance among collegiate gymnasts by also considering perfectionism as a contextually salient factor in the sport. We found individual profiles of mindfulness and perfectionism, and assessed how those quantitative reports were associated with objective measures of gymnastics performance. Still, further nuance in the contextually situated experiences of gymnasts seems attainable. Researchers have previously acknowledged the importance of understanding impact mechanisms of various facets of mindfulness (Birrer, Röthlin, & Morgan, 2012). By taking a mixed method approach, future researchers might better understand not only athlete profiles of BEING MINDFUL OF