Introduction

Despite the fact that athletes are, by designation, sport performers, they are basically human beings whose physical, mental, and social health is mirrored in their well-being and ill-health (^{Giles et al., 2020}). Well-being and mental health are a key component of the development and performance of athletes (^{Schinke et al., 2018}). In fact, athletes’ mental health and well-being has received increased attention in the sports community (^{Chang et al., 2020}; ^{Giles et al., 2020}; ^{Oliveira et al., 2022}; ^{Rice et al., 2016}), and literature has highlighted that it is important to explore factors that may influence their mental health, well-being and performance (^{Poucher et al., 2019}; ^{Walton et al., 2019}, especially in young athletes (^{Gama et al., 2021}; ^{Swan et al., 2018}). When it comes to the psychological benefits of sports involvement, athletes often gain a variety of psychological abilities that contribute to increased self-esteem, motivation, and resilience, which assist not only their performance but also other aspects of their lives (^{Weinberg & Gould, 2019}). Organized sports activities provide a social medium for athletes to interact, create relationships and teamwork, and foster a sense of belonging (e.g., ^{Beauchamp & Eys, 2014}). Sport participation, in contrast to these advantages, can undermine well-being of the athletes. For example, unsupportive environment, underperformance, and the emergence of maladaptive psychological symptoms and illnesses can all have a negative impact on mental health and well-being (^{Rice et al., 2016}).

Therefore, a factor that has been identified as crucial for athletes’ well-being and sports performance is the relationship with their teammates (e.g., ^{Gavrilova et al., 2017}). A study conducted by ^{Gavrilova et al. (2017}) demonstrated that support from teammates influences improvements in sports performance. Simultaneously, there is evidence that teammate’s relationship is a crucial factor in an athletes’ psychological adjustment (^{Donohue et al., 2007}). For example, teammates’ approval promotes the appropriate expression of emotions, global self-worth, and motivation to perform pro-social behaviours (^{Harter, 1999}). According to the evolutionary perspective, these positive social interactions seem to stimulate the affiliation system, promoting feelings of connectedness, sense of soothing, and social safeness (e.g., ^{Depue & Morrone-Strupinsky, 2005}; ^{Gilbert, 2010}).

Social safeness is considered an emotional experience characterzed by feelings of warmth, connectedness, and acceptance within interpersonal relations (^{Gilbert et al., 2009}). In this sense, when someone perceives their social world as safe, warm, and soothing, they tend to manage problems and challenging events more effectively and act more adaptively (^{Gilbert, 2005}; ^{Gilbert et al., 2009}). In contrast, individuals who do not feel safe in their social contexts and perceive others as unsafe or threatening need to stay vigilant and be ready to engage in defensive responses (^{Gilbert, 2005}). Indeed, social safeness has been negatively associated with several psychopathological indicators, such as self-criticism (^{Kelly et al., 2012}).

Self-criticism is an emotional regulation process regarding how people relate to themselves when they fail or make mistakes (^{Gilbert et al., 2004}). Self-criticism is defined as a type of negative self-judgment and self-scrutiny, where one displays a punitive response in face of one’s errors, faults, or attributes that may origin rejection or social disapproval (^{Gilbert et al., 2004}). Indeed, self-criticism may be understood as a strategy to cope with the shortcomings of an inferior or inadequate perceived self (^{Gilbert et al., 2004}). In the sport context, it is common that athletes use self-scrutiny and self-criticism when they are faced with difficulties (^{Kowalski & Duckham, 2014}). However, this constant and cruel self-to-self harassment has been linked to psychopathology (e.g., ^{Gilbert et al., 2008}). For example, a recent study conducted by ^{Walton et al. (2020}) revealed that self-criticism was a strong predictor of athletes’ psychological distress. Self-criticism has empirical evidence of being associated with psychopathological indicators, such as anxiety and depression (e.g., ^{Oliveira, Cunha et al., 2021}), and lower levels of well-being in athletes (e.g., ^{Oliveira, Rosado et al., 2021}; ^{Oliveira, Trindade et al., 2021}).

Given the importance of quality of relationships between peers during adolescence (e.g., ^{Crosnoe & Johnson, 2011}), it may be essential to explore the effect of the relationship that athletes maintain with their teammates has on their mental health. Also, despite the fact that athletes tend to believe that self-criticim is necessary for success in sport (^{Ferguson et al., 2014}), researchers have suggested that this response to failure can be counter-productive (^{Ferguson et al., 2015}). This study adds to the existing literature by highlighting the associations among athletes’ feelings of acceptance and connectedness in the relationship with their teammates (social safeness), self-criticism, psychological well-being, and perceived performance.

The current study aimed to test a comprehensive model that explores the effect of athletes’ feelings of acceptance and connectedness in their relationship with their teammates (social safeness) on their psychological well-being and perceived performance. Additionally, it sought to analyze whether self-criticism significantly acts in these associations while controlling the effect of the age of adolescent athletes. It was hypothesized that athletes’ feelings of acceptance and connectedness in the relationship with their teammates (social safeness) might be associated with higher levels of psychological well-being and perceived performance through lower levels of self-criticism.

Material and methods

Data analyses

The software IBM SPSS Statistics 22.0 was used to conduct descriptive and correlation analyses. The values of Skewness (Sk) and kurtosis (Ku) were analyzed to test the normality of the distribution of the study variables (^{Kline, 2005}). The multivariate normality of the items was assessed by the Mahalanobis distance (D ²) and statistically by Mardia’s normalized estimate of multivariate kurtosis in the form of critical ratio of kurtosis in Amos. Critical ratio of kurtosis <5.0 indicates multivariate normality (^{Byrne, 2010}).

Student’s t-tests were used to identify differences across gender in demographic variables. Pearson correlation coefficient analyses were performed to examine the associations between age, social safeness, self-criticism, psychological well-being, and perceived performance. These coefficients were interpreted in accordance with the guidelines of ^{Cohen et al. (2003}).

Therefore, AMOS software was used to examine the proposed theoretical model (Figure 1). In order to examine this model, we performed an SEM analysis. Since all instruments were properly validated, the total scores of all the instruments were used to represent each of the variables in the study. Therefore, we used ^{Bentler and Chou’s (1987}) recommendations regarding the minimum sample size, and only considered latent variables. In accordance with ^{Bentler and Chou’s (1987}) recommendations, the minimum sample size to conduct an SEM analysis will be calculated by multiplying the number of parameters of the theoretical model by five, if the data are perfectly well-behaved (i.e., normally distributed, no missing data or outlying cases). Taking into account that in our theoretical and saturated model we will obtain 20 parameters, in accordance with these authors we only need 100 subjects (20x5).

Note. ***p<.001; **p<.005; *p<.05.

Figure 1 Final path model 

This model tested the hypothesis that athlete-related social safeness (exogenous, independent variable) would present a significant effect on the psychological well-being and perceived performance (endogenous, dependent variables), through the mediational effects of self-criticism (endogenous, mediator variable). The effect of age was controlled. The maximum likelihood method was used for the estimation of the regression coefficients and fit statistics in the path model. The adequacy of the model was examined considering the following goodness of fit indices: Chi-square (χ ² ), that when nonsignificant indicates a very good model fit; the normed Chi-square (CMIN/df), that indicates an acceptable fit when <5; the Comparative Fit Index (CFI) and Tucker Lewis Index (TLI), which indicate a very good fit with values above .95; and the Root Mean Square Error of Approximation index (RMSEA), which indicates an adequate fit when values are <.08 (^{Kline, 2005}). The bootstrap procedure (with 5000 samples) was used to create 95% bias-corrected confidence intervals around the standardized estimates of the significance of total, direct, and indirect effects. The effect is statistically significant (p<.05) if zero is not included between the lower and the upper bound of the 95 % bias-corrected confidence interval (^{Kline, 2005}).

Results

Descriptive analyses

The descriptive and correlations between the study variables are reported in Table 1.

Table 1 Means (M), Standard Deviations (SD), and correlations between the study variables (N=164) 

Note. Athlete-related Social Safeness=Social Safeness and Pleasure Scale; Self-criticism=dimension of self-criticism of Forms of Self-Criticizing & Self-Reassuring Scale; Psychological well-being=psychological dimension of WHOQOL-bref; Perceived Performance=Sports Performance Perception Questionnaire. *p<.05; ***p<.001.

Discussion

Mental health, defined as a state of well-being in which people realize their capacities, can cope with the usual stresses of life, can work productively, and is able to make a contribution to theit community (^{World Health Organization [WHO], 2004}), is one of the most important aspects of an individual’s well-being. As a result, it’s important to remember that mental health isn’t only the absence of mental illness; it’s a continuum that extends from languishing to flourishing in life (^{Keyes, 2002}, ²⁰⁰⁵). This important part of mental health and well-being has been recognized in the sporting world, which is increasingly encouraging mental wellness rather than just treating mental disease in athletes. Therefore, recent studies have highlighted the importance of exploring factors that may contribute to athletes’ mental health, well-being, and performance (^{Giles et al., 2020}; ^{Poucher et al., 2019}; ^{Walton et al., 2019}). Also, more emphasis in literature is necessary for youth sport where the potential of supporting mental health at an earlier age could have important effects later in life (^{Swan et al., 2018}). This study is in accordance with these needs, exploring the relationship between social safeness (feelings of belonging and connecting to the team), self-criticism, performance, and psychological well-being in young athletes. Specifically, the current study’s main aim was to test an integrative model that explored the effect of athletes’ feelings of acceptance and connectedness in the relationship with their teammates (social safeness) on psychological well-being and perceived performance, as well as the mediating role of self-criticism in this association, while controlling the effect of age of adolescent athletes.

Correlational results demonstrated that feelings of belonging and connecting to the team (athlete-related social safeness) were negatively associated with self-criticism and positively linked with psychological well-being and perceived performance. These results are in accordance with previous studies in other contexts. Feelings of affiliation (social safeness) have been positively linked with indicators of well-being (e.g., ^{Gilbert et al., 2009}; ^{Kelly & Dupasquier, 2016}) and negatively associated with psychopathological indicators (e.g., ^{Gilbert et al., 2008}; ^{Kelly & Dupasquier, 2016}). Moreover, in the current study, self-criticism presented a negative correlation with psychological well-being. This result follows a recent study conducted in adult athletes that demonstrated a negative association between self-criticism and well-being (^{Oliveira, Cunha et al., 2021}). Thus, these data confirm the same relationship in a sample of adolescent athletes. In the present study, self-criticism also revealed a negative association with perceived performance, which is contrary to the findings found by ^{Killham et al. (2018}) that reported that self-criticism was not significantly related to athletes’ perceived sport performance. There is the possibility that self-criticism was linked to perceived sport performance due to specificity differences in the measures and sample used in this study.

Self-criticism is a maladaptive emotional process where one displays a punitive response in face of one’s errors, faults, or attributes that can be a source of rejection or social disapproval (^{Gilbert et al., 2004}). In the context of sport, it is common that athletes use self-criticism when they are faced with troubles and difficulties (^{Kowalski & Duckham, 2014}). However, this process revealed to be a strong predictor of psychological distress in athletes in other studies (e.g., ^{Walton et al., 2020}). Taking into account that literature has strongly associated psychological distress and perceived performance (^{Moen et al., 2019}), it was expected for self-criticism to also present a negative association with perceived performance. Therefore, this study corroborated previous research and added new data to clinical sport psychology literature.

These relationships were further examined in the path analysis that hypothesized that athlete-related social safeness would have an effect on psychological health and perceived performance through lower levels of self-criticism. The proposed theoretical model explained 52% and 28% of the variance of athletes’ psychological quality of life and perceived performance, respectively. This model suggests that young athletes who present higher levels of feelings of belonging and connecting to the team tend to present higher levels of psychological well-being and perception of better performance through lower levels of self-criticism. These results seem to contribute to explaining athletes’ psychological well-being, since athletes’ well-being has received special attention in literature (e.g., ^{Chang et al., 2020}), especially young athletes. Athletes deal with difficult events and experiences, such as injuries, getting benched, balancing commitments, losing a game (e.g., ^{Galli & Vealey, 2008}), and feeling accepted and connected to their teammates may contribute to dealing with these negative experiences. In fact, a recent study conducted by ^{Frentz et al. (2020}) demonstrated that support from teammates helped athletes to deal with difficult epxeriences in a more adaptive way, specifically, setbacks or injuries. In particular, this recent study explored how self-critical athletes have shifted to practically apply and integrate a more kind and understanding attitude toward themselves within competitive environments. It showed that support from teammates helped them in this shift (^{Frentz et al., 2020}). In this sense, the current study corroborates previous data showing that feelings of acceptance and belonging in relationships with teammates have an effect on levels of self-criticism, but added new data to the literature exploring the relationship between these variables and psychological well-being and perceived performance, especially among young athletes. In fact, a direct effect of feelings of acceptance and belonging in the relationship with teammates on psychological well-being and perceived performance was found.

Nonetheless, these findings need to be interpreted taking into consideration some limitations. Although these data were supported by robust statistical analysis, its cross-sectional design does not allow for establishing causal conclusions. In this sense, longitudinal studies should be conducted to better understand the causal associations between the athlete-related social safeness, psychological well-being, and perceived performance, and how self-criticism may mediate these associations. Moreover, the questionnaire protocol was exclusively composed of self-report instruments, which may present some bias (e.g., social desirability) that may compromise the data’s validity. Future studies should combine self-reported data with data obtained in a more objective manner. In addition, the sample size of the current study is considered a limitation of this study. Therefore, future researches are needed to evaluate if these findings are replicable in higher samples in order to be able to carry out a full SEM. Finally, though this study’s main objective was to specifically address the athlete-related social safeness effect on their psychological well-being and perceived performance, the parsimonious model examined in the present study is considered incomplete and could be explored in a different way. More precisely, the role of other potentially significant emotional processes (e.g., acceptance, attention to the present moment, and self-compassion) should be explored. On the other hand, it would be relevant to test the bidirectional relationship of athlete-related social safeness and self-criticism. This would mean testing an alternative model that hypothesizes self-criticism as an independent variable and athletes-related social safeness as a mediating variable, aiming to test the effect of self-criticism on feelings of acceptance and belonging in the relationship with teammates and their effects on psychological well-being and perceived performance. Furthermore, the sample size of this study does not allow testing an invariance model between boys (n=94) and girls (n=70), however in future studies it would be interesting to test this theoretical model exploring gender differences.

On the other hand, it is necessary to take into account that the sample of this study comprises only athletes who practice team sports, where the group factor is predominant in the study variables. Thus, it would be pertinent to test this model in a sample with athletes from individual and direct opposition sports. Thus, the generalization of the results in the sports context is limited to the characteristics of this sample. In fact, in team sports, interactions between players on the same side or team are crucial to the game’s outcome (e.g., soccer, basketball, volleyball). In contrast, individual sports (e.g., swimming, running, and cycling) do not involve such interaction (^{Jones, 2013}) Individual versus team sport participants have been studied to see if there are any changes in different aspects related to well-being and mental health. For example, in accordance with ^{Pluhar et al. (2019}) team sport athletes had positive experiences with coaching, skill development, and peer support which contribute to feelings of social acceptance and fewer negative outcomes among adolescents. This may have to do with the fact that in team sports, athletes are sourrounded by other people to rely on and help pick you up after a mistake, while people in an individual sport only have themselves and may cause greater internal attribution (^{Pluhar et al., 2019}). One of the ways to deal with the errors, mistakes, defeats, is through self-criticism, which is a maladaptive psychological process, which may explain these data. Therefore, it would be pertinent to test these differences between sports. However, the current study did not allow us to explore this, so future studies will be able to make this comparison.

The current study indicated that an athlete’s psychological well-being and perceived performance can be explained by athlete-related social safeness, and that self-criticism is an important mediator in this association. Given the salience of peer relationships during adolescence (e.g., ^{Long et al., 2020}), this study offers new empirical data that may be relevant for clinical and sport psychology practitioners.

In fact, the present study offers important insights by suggesting that feelings of acceptance and connectedness in team relationships may be at the root of young athletes’ well-being and perceived performance. Indeed, these novel findings seem to suggest the importance of the adoption of supportive, warm, and safeness relationships between teammates on athletes’ well-being and perceived performance. Therefore, this study seems to suggest the pertinence of developing interventions directed towards young athletes based on the promotion of affiliative skills. Furthermore, self-criticism seems to be an important emotional regulation process that has an impact on athletes’ well-being and performance. Many athletes believe that this way of relating to themselves is crucial to their success in competitive sport. However, this study seems to demonstrate that self-criticism is associated with lower levels of well-being and performance. Therefore, it seems to be important to address self-criticism in young athletes. Since self-compassion has demonstrated to be an emotional regulation process which is more adaptive in dealing with failures and mistakes in sport (e.g., ^{Frentz et al., 2020}; ^{Killham et al., 2018}), psychologists could implement interventions focused on self-compassion.