When change-oriented feedback enhances motivation, well-being and performance: A look at autonomy-supportive feedback in sport
چکیده
Objectives: Change-oriented feedback (aka negative feedback) serves two important functions: it motivates athletes and guides them towards performance improvement. However, it can also lead to negative consequences such as anxiety or a decrease in athletes’ self-esteem and in the quality of the coacheathlete relationship. We propose that change-oriented feedback quality is key in predicting athletes’ reaction to this type of feedback. Based on SDT, we further suggest that a high quality changeoriented feedback must be autonomy-supportive. To test this hypothesis, we first define and measure an autonomy-supportive change-oriented feedback. We then investigate the relative impact of changeoriented feedback’s quantity and quality on athletes’ phenomenological experiences and performance. Method: In total, 340 athletes and 58 coaches participated in this study. Coaches and athletes filled out a questionnaire after a training session. HLM analyses were used to take into consideration the hierarchical structure of the data. Results: HLM analyses first show that an autonomy-supportive change-oriented feedback is empathic, accompanied by choices of solutions, based on clear and attainable objectives known to athletes, avoids person-related statements, is paired with tips, and given in a considerate tone of voice. Results also show that feedback quality predicts athletes’ outcomes above and beyond feedback quantity and coaches’ other autonomy-supportive behaviours. Conclusion: Results are discussed in light of their contribution to self-determination theory, the feedback literature and the improvement of coaches’ training. 2013 Elsevier Ltd. All rights reserved. Coaches play a major role in athletes’ lives. Not only do they represent important authority figures, they also act as models, confidants and motivators. Through their behaviours adopted within each of these roles, they can have profound cognitive, behavioural and emotional impacts on their athletes (Smoll & Smith, 2002). Feedback, defined as information conveyed to athletes about the extent to which their behaviours and performance correspond to expectations (Cusella, 1987; Hein & Koka, 2007), is one of the most crucial coaching behaviours as it directly conveys information about athletes’ competence (Horn, Glenn et Wentzell, 1993). While promotion-oriented feedback aims at confirming and reinforcing desirable behaviours, change-oriented feedback indicates that performance is inadequate and that behaviours need to 5, þ1 514 343 2460; fax: þ1 (J. Carpentier), g.mageau@ All rights reserved. be modified in order to eventually achieve athletes’ goals (Bloom & Hautaluoma, 1987; Cusella, 1987). Although past literature has often referred to these two types of feedback as positive and negative feedback respectively (e.g. Deci, Koestner, & Ryan, 1999; Kluger & DeNisi, 1996; Latting, 1992; Weinberg & Gould, 2011), the terms “promotion-oriented feedback” and “change-oriented feedback” are preferred because they better differentiate between the goals and the consequences of the feedback. Whereas the terms “positive” and “negative” in the original terminology can be interpreted as qualifying either the goals or the possible consequences of receiving the two types of feedback, the new terms specifically designate the different goals underlying the different types of feedback (i.e., promoting or changing a targeted behaviour). Given that both promotion-oriented and change-oriented feedback can have positive or negative outcomes depending on the way that it is given (e.g., Deci et al., 1999; Mouratidis, Lens, & Vansteenkiste, 2010), using more precise terms to designate the goal of the feedback avoids unnecessary confusion. Promotion-oriented feedback is without a doubt more pleasant to give than change-oriented feedback. Research shows that people J. Carpentier, G.A. Mageau / Psychology of Sport and Exercise 14 (2013) 423e435 424 in a position of authority often tend to distort, delay or withhold change-oriented feedback (Fisher, 1979; Larson, 1989). However, avoiding this type of feedback prevents athletes from benefiting from it. Specifically, change-oriented feedback interventions serve two important functions (Weinberg & Gould, 2011). First, they motivate by informing athletes about the discrepancy between actual and desired performances, which can increase their desire to perform better in the future. This desire in turn can be translated into greater effort and energy expenditure. Second, they guide by focussing athletes on the specific changes they need to implement if they wish to improve future performances. However, changeoriented feedback can also have many negative consequences such as impairing athletes’ performances, motivation and selfesteem, as well as the quality of the coacheathlete relationship (Baron, 1988; Fisher, 1979; Jussim, Soffin, Brown, Ley, & Kohlhepp, 1992; Sansone, 1989; Tata, 2002). As change-oriented feedback in sport is both inevitable and hard to give, tools designed to help coaches provide such feedback in a way that maximizes its potential positive consequences, while minimizing the negative ones, are greatly needed. The goal of the present study is to investigate the relative impact of change-oriented feedback’s quantity and quality on athletes’ phenomenological experiences and performance. Based on past research on optimal coaches’ behaviours (Lafrenière, Jowett, Vallerand, & Carbonneau, 2011; Mageau & Vallerand, 2003) and on a recent study on autonomy-supportive change-oriented feedback in sport (Mouratidis et al., 2010), we first postulate that to be of high quality, change-oriented feedback must support athletes’ autonomy as defined by self-determination theory (SDT; Deci & Ryan, 1985, 2000). SDT posits that humans’ psychological health and optimal functioning are facilitated by interpersonal contexts that support the basic psychological need for autonomy, i.e., the universal desire to feel that one is at the origin of one’s actions and that one’s actions are concordant with one’s values. Accordingly, coaches’ interpersonal style has been described as either autonomy-supportive or controlling, which in turn has been found to be an important predictor of athletes’ outcomes (Frederick & Ryan, 1995). Autonomy-supportive coaches consider their athletes as separate individuals with unique needs and feelings (Deci & Ryan, 1985, 2000; Ryan & Grolnick, 1986). In contrast, controlling coaches have a tendency to pressure their athletes to think, feel or be in specific ways, thereby making their athletes feel like pawns controlled by external forces (deCharms, 1968; Deci & Ryan, 1985). Autonomysupportive behaviours (i.e., providing choice, giving a rational and acknowledging feelings) have been linked to many positive consequences such as more self-determined motivation, higher selfesteem and greater well-being as reported by athletes (Amorose & Anderson-Butcher, 2007; Gagné, Ryan, & Bargmann, 2003; Quested & Duda, 2010; Reinboth, Duda, & Ntoumanis, 2004). While numerous specific behaviours adopted by autonomysupportive coaches have been studied (see Mageau & Vallerand, 2003; for a review), relatively few research within SDT has looked at the way they provide feedback to their athletes. Research pertaining to change-oriented feedback within SDT has mainly either highlighted the negative impact of giving change-oriented feedback compared to providing promotion-oriented feedback or no feedback (Koka & Hein, 2003; Vallerand & Reid, 1984; Whitehead & Corbin, 1991), or documented the impact of the quantity of change-oriented feedback on various outcomes without considering its quality (Black & Weiss, 1992). In the present research, providing change-oriented feedback (i.e., its quantity) is conceptualized as a specific aspect of structure, which refers to coaching behaviours aimed at organising athletes’ environment in away that increases competence and predictability (e.g., limit and goal setting, rule reinforcement, guidance; Grolnick & Pomerantz, 2009; Jang, Reeve, & Deci, 2010). We also postulate that the way change-oriented feedback is provided (i.e., its quality) greatly influences its outcomes. As it is the case for other elements of structure such as communicating expectations, setting limits, or giving rewards (Deci et al., 1999; Grolnick & Ryan, 1987; Jang et al., 2010; Koestner, Ryan, Bernieri, & Holt, 1984; Mouratidis et al., 2010; Ryan, Mims, & Koestner,1983), it is expected that positive outcomes will ensue when change-oriented feedback is presented in an autonomy-supportive way. Autonomy-supportive change-oriented feedback A recent study (Mouratidis et al., 2010) has begun the investigation of an autonomy-supportive change-oriented feedback in sport and showed that change-oriented feedback is indeed beneficial for athletes’ optimal motivation and well-being when it is communicated in an autonomy-supportive fashion. The authors relied on the definition of the autonomy-supportive coaching style to identify characteristics that could define an autonomysupportive change-oriented feedback. Specifically, this type of feedback was defined as 1) providing rationales to explain why behaviours should be changed, 2) considering athletes’ perspective, 3) providing choices of solutions, and 4) avoiding the use of a controlling communication style, which induces shame, conveys conditional regard or includes threats of punishment. In the present research and in line with Mouratidis et al. (2010), an autonomy-supportive change-oriented feedback is first characterized by being empathic (1) and paired with choices of solutions (2). “Being empathic” is defined as taking into consideration athletes’ feelings and difficulties, while “being paired with choices of solutions” refers to giving athletes multiple solutions to correct the situation, whenever possible. The characteristic “providing rationales to explain why behaviours should be changed” proposed by Mouratidis et al. (2010) was also included but this dimension was operationalized more concretely as “being based on clear and attainable objectives known to athletes (3)”. By assessing whether or not athletes knowand understand the objectives of the feedback, and agree that these objectives are attainable, this dimension captures whether or not athletes know the rationale behind the coach’s feedback. In addition, by being more concrete than the original characteristic, this dimension may be more readily used to help coaches provide autonomy-supportive feedback. Five additional characteristics were included to measure an autonomy-supportive change-oriented feedback: avoiding personrelated statements (4), pairing the feedback with tips on how to improve future performances (5), being delivered promptly (6), privately (7) and in a considerate tone of voice (8). These characteristics come from the feedback literature and needed to be integrated to the present research for two important reasons. First, they have been shown to lead to positive outcomes, which suggests that they do characterize a high quality feedback. Second, they are autonomy-supportive according to SDT’s definition of autonomy support. Research pertaining to these specific characteristics is briefly reviewed below. Avoiding person-related statements Tenants of SDT (Koestner, Zuckerman, & Koestner, 1987; Plant & Ryan, 1985; Ryan, 1982) have proposed that, in an autonomysupportive context, attention of the athletes should be maintained on the task to avoid ego-involvement, an internally controlling state that occurs when athletes come to view their performance as an indicator of their worth as a person (Nicholls, 1989). Results of ameta-analysis by Kluger and DeNisi (1996) also show that, in order J. Carpentier, G.A. Mageau / Psychology of Sport and Exercise 14 (2013) 423e435 425 to enhance performances, feedback intervention should maintain the attention of the receiver on the task instead of shifting it towards the self. An efficient strategy to maintain the attention of the receiver on the task is to avoid person-related statements. In an experimental study with young children, Kamins and Dweck (1999) showed that person-related feedback (defined as feedback concerning the person’s abilities, goodness, or worthiness) can create vulnerability and a sense of contingent self-worth. To avoid linking people’s selfesteem to their performance, feedback should be kept task-related through the use of comments on the process (e.g. effort and strategy use) and on specific technical changes (Hong, Chiu, Dweck, Lin, & Wan, 1999; Kamins & Dweck, 1999). Avoiding person-related statements thus captures part of what Mouratidis et al. (2010) have termed “non-controlling communication style” and supports athletes’ autonomy by preventing egoinvolvement, which is considered an internally controlling state (Grolnick & Ryan, 1987; Mageau & Vallerand, 2003; Ryan, 1982). It also contributes to support athletes’ autonomy by preventing the experience of controlling emotions, such as shame and fears of rejection (Assor, Roth, & Deci, 2004; Ryan & Deci, 2002).