Incivility and psychological safety in youth sport: the reciprocal effects and its impact on well-being and social outcomes

ABSTRACT

Youth can often be the subject of rude and discourteous behaviors in their sport setting, given their susceptibility to the social environments. Incivility refers to insensitive behavior that exhibits a lack of respect for others, namely, disrespectful and rude behaviors. Incivility is a significant issue in youth sport since it negatively influences teams and individuals. The present study aims to investigate 1) how incivility and psychological safety are associated with each other over time and 2) the influence of the initial value of psychological safety and the change in psychological safety on youth athletes’ well-being and social outcomes. Three-wave time-lagged data collection was employed, and the present study included 283 youth athletes who completed the survey three times. The hypotheses were tested in SEM with cross-lagged panel and growth latent curve modeling. The results showed that coach and teammate incivility were significantly associated with the change in psychological safety. In contrast, the initial value of psychological safety was a significant antecedent of the subsequent coach and teammate incivility, well-being, and social outcomes. Lastly, the change in psychological safety was significantly associated with youth athletes’ well-being and social outcomes. The findings suggest that incivility and psychological safety were reciprocally associated, and psychological safety rather than incivility was a significant predictor of youths’ well-being and social outcomes. The present study found a mechanism underlying the relationship between incivility, psychological safety, and essential outcomes (i.e., well-being and social outcomes) in youth sport.

HIGHLIGHTS

• Coach and teammate incivility influenced the change of psychological safety over time.

• Psychological safety predicted subsequent coach and teammate incivility.

• Psychological safety was significantly associated with youth athletes’ well-being and social outcomes.

KEYWORDS:

• Positive youth development
• sport for development
• youth sport
• psychological well-being
• burnout
• social responsibility
• prosocial behavior

Introduction

In youth sport, uncivil behaviors are frequent as young people are often the victims of rude and discourteous behaviors in their sport setting. Johnson and Indvik (2001) claim that incivility is at the low end of the continuum of abuse in the organization, suggesting that such maladaptive behaviors in an organization or a team would be the root of a tragedy. In particular, corporal punishments in Japan are still witnessed, and such an abuse of power has caused suicide mortality of a high school student-athlete in Japan (Story, 2018). Additionally, uncivil behaviors are contagious; thus, leaders’ uncivil behaviors would predict followers’ antisocial behaviors in the team, which eventually become a cause of bullying in the team (Porath et al., 2015). Consequently, incivility should provoke significant issues in youth sport.

Psychological safety is a significant concern in management research, and its impact on organizational and individual performance and well-being has been evidenced (cf., Newman et al., 2017). Although the significance is widely recognized, few studies have been conducted in sport organizations or teams and adopted a longitudinal data collection (cf., Newman et al., 2017). The results suggest that sport management scholars necessarily investigate how the dynamic nature of psychological safety would be predicted by significant factors and the impacts on athletes. Additionally, the present study mainly focuses on the sport for development paradigm (e.g., Coalter, 2013); thus, the outcomes of interests are related to the fundamental paradigm in youth sport. Specifically, well-being and social outcomes (e.g., social responsibility and prosocial behaviors) are of significant interest in the research terrain, examining how to facilitate youth development through the power of sport (i.e., development through sport: Coalter, 2013).

The present study contributes to developing theoretical and practical insight into sport management literature by clarifying a mechanism of how incivility is associated with a team-level factor and individual outcomes (i.e., psychological safety, well-being, and social outcomes). An incivility spiral would occur as the uncivil behavior of one party leads to the uncivil behavior of the second party (Andersson & Pearson, 1999). The maladaptive interactions would result in a gradual increase in counterproductive behaviors in the team, which eventually leads to violence and bullying (Johnson & Indvik, 2001). Thus, the present study paid attention to not only the coach-athlete dyad but also the teammate’s behavior and the impact of a team-level factor (i.e., psychological safety) on youth athletes. The investigation practically helps stakeholders of youth sport organizations or clubs understand the need to be aware of the impacts of coaches and teammates on the team and individuals. Therefore, the purpose of the present study was to investigate 1) how incivility and psychological safety are associated with each other over time (the reciprocal relationship between incivility and psychological safety) and 2) the influence of the initial value and the rate of change in psychological safety across three-time points on well-being (i.e., psychological well-being and burnout) and social outcomes (i.e., social responsibility and prosocial and antisocial behaviors) of youth athletes.

Theoretical framework

Previous studies (e.g., Gerstner & Day, 1997) theoretically explain the significant relationship between important social figures in the team (e.g., leader, supervisor, colleague) and team-level factors (e.g., team psychological commitment, psychological safety). The present study included coach and teammate incivility as social figures’ behaviors and psychological safety as a team-level factor in the hypothesized model (Figure 1). The model was generated by the research questions regarding how significant figures’ behavior and team-level factors are reciprocally associated, which is based on the framework of incivility spirals proposed by Andersson and Pearson (1999). The framework theoretically proposes the spiraling effect of incivility in which one’s disrespectful behaviors lead to another’s retaliatory disrespectful behaviors within the team through intensified negative feelings by applying the theory of coercive actions (Tedeschi & Felson, 1994). According to Andersson and Pearson (1999), although coercive actions (e.g., aggression, violence) are beyond incivility, coercive actions and incivility occur as an exchange process in social interaction (Tedeschi & Felson, 1994).

Figure 1. Hypothesized theoretical model.

The coach is an influential social factor for youth athletes as a leader is a significant role model for followers in the team (Gerstner & Day, 1997). Also, teammates are significant social agents impacting youths’ behavior and psychological development (cf., Sheridan et al., 2014). As such, social interactions occur between coach and athlete or among teammates in their sport environments, which connotates that there is a possibility that a negative spiral may happen between those socially interacting agents. In addition, the incivility spiral framework argues that a negative organizational climate is a significant factor facilitating the formation and escalation of the incivility spiral. It suggests that a team-level factor (e.g., psychological safety) significantly influences the generation and facilitation of the incivility spiral. Consequently, the present study adopted the framework of the incivility spiral to develop the hypothesized model (Figure 1).

Furthermore, conservation of resources (COR) theory (Hobfoll, 1989) can explain how the initial point of and the change in psychological safety affect team members’ mental well-being and positive psychosocial development. COR theory is a motivational theory that posits that people generally exert efforts on obtaining, retaining, and protecting what the individuals centrally value (Hobfoll, 1989). According to the theory, when individuals possess enough resources or when resources are sufficiently fulfilled in their environment, they invest the available resources to develop their future resources. Therefore, the surpluses of the resources in the team foster the members’ well-being and positive development (Obrenovic et al., 2020; Whitman et al., 2014). The assumption suggests that psychologically safe environments, which generate the accessibility to resources in the team (Edmondson, 1999), can enhance youths’ well-being and social outcomes. Moreover, COR theory also illustrates that the loss of resources is disproportionately stressful for humans, negatively impacting personal well-being and development. Given that incivility can be characterized as the disturbance diminishing psychological safety (e.g., Smittick et al., 2019), the threat of a loss in their resources (e.g., negative relationship between incivility and psychological safety) can negatively influence individual well-being and development (Hobfoll, 1989). As such, the theory provides logical explanations for why the incivility spiral negatively affects youth athletes obtaining greater well-being and social outcomes.

Incivility in sport

Incivility refers to insensitive behavior that exhibits a lack of respect for others; in other words, disrespectful and rude behaviors (Andersson & Pearson, 1999; Porath & Erez, 2009; Porath & Pearson, 2012). Incivility is a significant issue in organizations since it negatively influences the organization or team, such as reducing performance and creativity (Porath & Erez, 2009) and employee retention (Porath & Pearson, 2012). Similarly, review research showed that incivility is consistently associated with negative outcomes (e.g., reduced employee performance, burnout, work withdrawal, and perception of job insecurity) in the workplace (cf., Schilpzand et al., 2016). The mechanisms of the negative influence of incivility on such important outcomes are explained by the increased stress level leading individuals to cognitive dysfunction (e.g., Porath & Pearson, 2010). Individuals own limited cognitive resources and distribute their attention to various activities (Kahneman, 1973). As a result, the personal cognitive resources are worn out by distributing attention to stressful activities (Hobfoll, 1989). Thus, incivility depletes personal cognitive resources required for positive functioning in the team (Riskin et al., 2017).

Verbal aggression and abusive behaviors are commonly studied in the sport context (e.g., Stirling, 2009). However, incivility is different from those behaviors. Verbal aggression is an attack on a person’s self-concept to induce emotional pain through verbal communication (Infante & Wigley, 1986). However, incivility is disrespectful behaviors, including both verbal and non-verbal components (Cortina et al., 2001) rather than attacking behaviors. Also, abuse and violence are at the low end of incivility (Johnson & Indvik, 2001), meaning that incivility is less intensive and does not include abusive and violent behaviors (Andersson & Pearson, 1999). Additionally, incivility is often considered a similar concept to bullying; however, bullying is more intense and destructive than incivility (Stirling, 2009). Therefore, incivility is distinguished from the relevant concepts, and investigating incivility in youth sport may shed light on the roots of such problematic behaviors among coaches and teammates.

Incivility is not only the case in the workplace but also in sporting environments because incivility occurs in social interaction (Carter, 1998). As coach-athlete interaction is an essential practice in sport, coach incivility is negatively associated with team performance through psychological safety in the NCAA women’s basketball teams (Smittick et al., 2019). As incivility would occur from higher to lower status (e.g., Caza & Cortina, 2007), it should be common in youth sports. Especially, the people in which the communities have a culture of higher power distance (i.e., people believe that unequal distribution of power is normal) should have more frequently experienced incivility. Since Asian countries (e.g., Japan) have a relatively higher power distance culture (Hofstede, 2011), the present participants are more likely to experience incivility in their sporting environments.

Incivility is also perpetrated by colleagues (Reio & Sanders-Reio, 2011; Riskin et al., 2017), indicating that such behaviors are prevalent among group members. As a result, the uncivil behaviors in a particular group would be an antecedent of aggression and bullying (Farrell et al., 2016). In the education context, incivility occurs in the classroom setting, also called classroom incivility (Bjorklund & Rehling, 2010). Thus, such behaviors (e.g., incivility) may be frequent amongst youth athletes. Furthermore, incivility from teammates would be affected by their coach’s uncivil behaviors since incivility is contagious (Porath et al., 2015). From a cognitive psychological perspective, incivility manipulates a semantic network of related concepts (e.g., hostility and hate) in people’s minds (Foulk et al., 2016). As a result, the framework of the incivility spiral (Andersson & Pearson, 1999) explains that the activation of the incivility-related semantic network fosters the individuals’ hostile behaviors toward others, producing a negative spiral of uncivil behaviors in their team. Therefore, the present study also investigated how coach incivility would predict future teammate incivility and vice versa. Therefore, the hypotheses regarding incivility between coaches and teammates are as follows.

H1a: Coach incivility at T1 would predict coach incivility at T2

H1b: Teammate incivility at T1 would predict teammate incivility at T2

H2a: Coach incivility at T1 would predict teammate incivility at T2

H2b: Teammate incivility at T1 would predict coach incivility at T2

Psychological safety in sport

Previous research has overlooked the significance of team-level factors (e.g., team/club social climate) on individuals, although investigations on the coach-athlete dyad have been conducted (cf., Sheridan et al., 2014). Thus, considering a team-level construct such as psychological safety in youth sport is a significant concern in sport management literature. Psychological safety is defined as “a shared belief that the team is safe for interpersonal risk taking” (Edmondson, 1999, p. 354). In a psychologically safe environment, the members feel that they will not be rejected by their teammates if saying their honest opinions (Edmondson, 1999). Further, the members in a psychologically safe environment mutually respect each other’s competency and are interested in others as a person. Therefore, the tacit belief among team members leads them to make challenging decisions and facilitates learning behaviors (Edmondson, 1999; Edmondson & Lei, 2014). Consequently, psychological safety is a positive motivational state that often predicts team and individual performance and development (cf., Newman et al., 2017).

Psychological safety is a state variable, not stable in the team over time (Bradley et al., 2012; Burke et al., 2006), suggesting that the observation in the process of change in psychological safety can provide significant insight into the theoretical and practical development of the literature. However, as Newman et al. (2017) criticized that the investigation of the dynamic nature of psychological safety is still limited, the longitudinal investigation of psychological safety contributes to developing the comprehension of the concept. The review study specifically indicated that applying Latent Growth Curve Modeling (LGCM; e.g., Duncan & Duncan, 2009) to psychological safety research would be a significant practice to understand the dynamic nature of psychological safety. Thus, investigating the dynamic nature of psychological safety across time and the associations with the variables of interest can contribute to the body of literature.

It’s crucial to clarify how psychological safety differs from other relevant concepts, such as empowerment, engagement, and trust (Edmondson, 1999; Frazier et al., 2017). Empowerment, engagement, and psychological safety have some similarities as those represent positive motivational states (Frazier et al., 2017). However, the concepts are distinct in that empowerment and engagement are one’s perceptions regarding their specific tasks, while psychological safety pays particular attention to the social environment to understand how others (e.g., teammates) react to interpersonal risk-taking (Edmondson, 1999). Also, Edmondson (2004) distinguished psychological safety from trust by pointing out the focus of each construct. Trust traditionally refers to one’s willingness to be vulnerable to the actions of another party (e.g., Mayer et al., 1995). Therefore, trust is one’s willingness to deliver the benefits from uncertainties to the significant person (e.g., teammates). In contrast, psychological safety is the perception of the social environment; thus, the focus is on how others can benefit from the uncertainty when taking some risks (Edmondson, 2004). Although scholars have increasingly investigated psychological safety in the sport context (cf., Vella et al., 2022), the investigation in youth sport is limited. Hence, the examination of psychological safety in youth would contribute to understanding how the concept is associated with youth sport environments and essential outcomes.

Incivility and psychological safety

Leadership is one of the essential antecedents of psychological safety in the organization or team (cf., Edmondson & Lei, 2014; Newman et al., 2017). In line with the evidence, leaders’ uncivil behaviors were negatively associated with college athletes’ psychological safety (Smittick et al., 2019). As such, we expected that coach incivility would be associated with the change (i.e., slope) in psychological safety. Additionally, we assume that teammates’ behaviors would also affect the team members’ psychological safety since teammate behaviors are a significant social environment factor for youth development in sport (cf., Holt et al., 2017; Sheridan et al., 2014). Moreover, a relationship network (e.g., coach-athlete relationship, the relationship among teammates) would be a significant predictor of members’ psychological safety (Newman et al., 2017), representing that the relationship with peers is an essential factor for psychological safety. Hence, we hypothesized that coach and teammate incivility would be associated with the change in psychological safety.

H3a: Coach incivility at T1 would be associated with the slope of psychological safety

H3b: Coach incivility at T2 would be associated with the slope of psychological safety

H4a: Teammate incivility at T1 would be associated with the slope of psychological safety

H4b: Teammate incivility at T2 would be associated with the slope of psychological safety

In contrast, a scholar argued that psychological safety facilitates the climate in the organization, which is “characterized by interpersonal trust and mutual respect” (Edmondson, 1999, p. 354). Thus, in line with the incivility spiral framework (Andersson & Pearson, 1999), psychological safety can also be an antecedent of low incivility in the team. The present study tested beyond the previous findings in the college sport context (Smittick et al., 2019) as we examined the reciprocal effect of incivility and psychological safety (i.e., incivility spiral). Therefore, we expected that incivility and psychological safety would be reciprocally related. In the LGCM, the intercept shows the starting point (average of the first-time assessment with adjusted error) of psychological safety across times. The hypotheses are as follows:

H5a: The intercept of psychological safety would predict coach incivility at T2

H5b: The intercept of psychological safety would predict teammate incivility at T2

Well-being and social outcomes for youths in sport

In the sport for development paradigm (e.g., Blom et al., 20212021; Coalter, 2013; Inoue et al., 2015; Schulenkorf et al., 2016), well-being and social outcomes are significant assets that can be nurtured through sport participation. Psychological well-being, defined as overall individual functioning comprising self-acceptance, autonomy, environmental mastery, personal growth, meaning in life, and positive relations with others (Ryff, 1989), has increasingly been studied in the sport context (e.g., Kim et al., 2020), but few investigated in the youth sport research. Psychological well-being is different from subjective well-being as it is based on the ancient Greek philosophy called Nicomachean ethics (Rowe & Broadie, 2002). The perspective, also called eudaimonia (cf., Huta & Waterman, 2014), overlaps with the WHO’s definition of mental health (World Health Organization, 2014). Therefore, psychological well-being is a core concept to understand the mental health of student-athletes (Keyes, 2002). Additionally, burnout in sport organizations is a traditional research stream in sport management (e.g., Doherty, 1998). Nevertheless, few studies have examined the relationship between environmental factors and youths’ burnout, while the previous study suggested that burnout is a significant indicator of dropout in sport participation. Therefore, investigating the environmental factors affecting youth athletes’ psychological well-being and burnout is vital and provides essential implications.

In addition to youth athletes’ well-being, researchers claim that sport has the power to facilitate positive youth development. For example, Inoue et al. (2015) found that youths with higher intrinsic motivation for sport participation were significantly related to positive attitudes toward a healthy lifestyle and lower engagement in threatening behaviors. Lee et al. (2021) similarly suggested that sport-based positive youth development programs positively impacted youths’ self-regulation ability in sport and general life and self-worth. Therefore, these results indicate that sport participation can be a significant tool to improve youths’ assets in their life. As sport participation can facilitate youths’ psychosocial development (cf., Bailey, 2006), it may become a cause of reducing antisocial behaviors. Blom et al. (2021) found that sport participation programs nurtured youths’ social responsibility and reduced the attitude toward violence in a developing country. Accordingly, recent researchers empirically demonstrated that sport participation could nurture youths’ assets; however, the environmental factors critically affecting such social outcomes have been barely investigated. Although incivility negatively influences individuals in various domains (e.g., Smittick et al., 2019), we expect that psychological safety rather than incivility would be a significant precursor of youths’ well-being and social outcomes. The rationales are argued below.

Psychological safety is associated with well-being (e.g., psychological well-being, burnout) and social outcomes (e.g., prosocial behavior). For example, Erkutlu and Chafra (2016) found that employees with greater psychological safety were significantly associated with higher psychological well-being. Fransen et al. (2020) also recently found that psychological safety was negatively associated with athletes’ burnout. Accordingly, psychological safety would be a significant predictor of athletes’ well-being. Moreover, psychological safety is an important factor for youths’ psychosocial development (Côté et al., 2008), suggesting that psychological safety would promote prosocial behaviors in the team. In fact, psychological safety was significantly associated with prosocial behaviors (Leung et al., 2015). Therefore, psychological safety can facilitate human functioning and ethical behaviors in the team. Also, psychological safety plays a significant role in becoming an outcome of leadership (Hypotheses 3 & 4) but also the antecedent of team/individual outcomes (e.g., team performance, burnout, and health; cf., Edmondson & Lei, 2014; Newman et al., 2017). Those points are also supported by COR theory since psychologically safe environments create available resources (Hobfoll, 1989) to facilitate greater well-being and positive development. Since the previous empirical findings are consistent with the assumption of the COR theory, we expected that the intercept (a starting point: T1) and slope (the rate of change across three times) of psychological safety would be associated with important outcomes for youths’ development through sport.

H6a~e: The intercept of psychological safety would be associated with youth athlete well-being and social outcomes

H7a~e: The slope of psychological safety would be associated with youth athlete well-being and social outcomes

All hypotheses are presented in the hypothesized model (Figure 1)

Method

Data collection procedures and participants

Data were collected at the end of April (T1), the beginning of July (T2), and the beginning of September (T3). The time span was decided based on the Japanese school-based sports calendar. In Japan, an academic year starts in April; therefore, the first year of middle and high school students start playing their sports generally in April. Moreover, the biggest competitions for middle- and high-school student-athletes are typically held during July and August, depending on the region and sports. Therefore, we decided to collect the data at the three time points. To recruit the participants, the authors created a recruitment poster with a QR code to allow coaches of potential participants to contact us directly for the present research participation and posted it on the authors’ SNS accounts. Based on the contact from the coaches who are interested in the present research, the first author had an online meeting with the coaches to explain the study protocol. Only if the coach agreed with the research purposes and protocol, the online survey was delivered to the participants through their coaches. The distribution of the online survey link to each coach was executed at each time point to collect the longitudinal data. The present study only retained the participants who completed the online survey three times to proceed with the data analyses.

The initial number of participants was 358 youth athletes from high school or middle school-level baseball, basketball, and badminton teams. However, 75 participants were excluded (attrition rate = 20.9%) from the further data analyses as they did not complete the follow-up surveys (T2 or T3). The attrition only internally occurred in each team; it can be considered that the attrition is missing at random, indicating that the potential bias is low in the longitudinally collected data (Fitzgerald et al., 1998). Therefore, the 283 youth athletes aged 12-18 years (M_age = 16.04, SD = 1.40; male = 92.2%) were included in the final data analyses. The participants listed themselves as either local (23.3%), prefectural (43.1%), regional (14.8%), national (17.7%), or international (1.1%) level youth athletes, who weekly practice 29.36 (SD = 13.45) hours on average. The participants were provided consent forms approved by the authors’ university IRB (2021-090).

The attrition rate was expected before collecting the data as it is challenging to maintain study participants in longitudinal studies. We adopted a rule-of-thumb approach to estimate the appropriate sample size. Researchers expected to collect a sample size of more than 200 since previous research (e.g., Cheong, 2011) suggests that the sample size for the LGCM should be a minimum of 200 samples to perform the statistical analysis. Therefore, the present samples are large enough to perform the statistical analyses.

Measurements

Incivility in Coach and Teammate

Perceptions of coach and teammate incivility were assessed by the adapted version of the Workplace Incivility Scale (WIS; Cortina et al., 2001; Smittick et al., 2019). Participants were asked to evaluate six behaviors with a 7-point Likert scale ranging from 1 (Never) to 7 (Always) after the stem: “Has your coaches/teammates engaged in any of the following behaviors?”

Psychological Safety

We adopted six items from the Team Psychological Safety Questionnaire (Edmondson, 1999). The original scale was modified and validated the questionnaire structure in the sport context (Fransen et al., 2020). The present participants were asked the six items anchoring from 1 (Strongly disagree) to 7 (Strongly agree).

Psychological Well-being

Psychological well-being in their general life was measured by six items from the short form of the Mental Health Continuum (Keyes, 2002). The measurement asked participants to answer how frequently they felt about each item during the past month. The questions were asked with a 7-point Likert scale from 1 (Never) to 7 (Every day).

Burnout in Sport

We combined some questionnaires to measure burnout. Theoretically, burnout consists of three subdimensions: cynicism, exhaustion, and inadequacy (Sorkkila et al., 2020). The present study assessed each component with three items adopted from Sorkkila et al. (2020) for cynicism (e.g., “I feel that I am losing interest in my sport”), Kristensen et al. (2005) for exhaustion (e.g., “I feel worn out to play my sport”), and Raedeke and Smith (2001) for inadequacy (e.g., “I am not achieving much in sport”) with slight modifications to the present context. The questions were also asked to rate from 1 (Strongly disagree) to 7 (Strongly agree).

Social Responsibility

To measure the social responsibility of youth athletes, we used the 6-item Social Responsibility Scale (Flewelling et al., 1993). The questions were rated using a 7-point Likert scale from 1 (Strongly disagree) to 7 (Strongly agree).

Prosocial and Antisocial Behaviors in Sport

We measured prosocial and antisocial behaviors among teammates using three items each from the Prosocial and Antisocial Behavior in Sport Scale (PABSS; Kavussanu & Boardley, 2009). Although the short version of PABSS includes five times each to measure prosocial and antisocial behaviors in the sport context, a previous study removed one item from the PABSS to fit the Asian context. Also, in the translation process into Japanese, we decided to remove one more item since the meaning of the items is too similar. The questions were rated by a 7-point Likert scale from 1 (Strongly disagree) to 7 (Strongly agree).

Data analysis

Data analyses were executed using IBM SPSS 28 and AMOS 28 statistics software. We adopted Kline’s (2005) standard fit indices; thus, a good fit is indicated for values greater than CFI and TLI = .90 and less than .08 for RMSEA and SRMR, respectively. Further, we used the bias-corrected confidence interval based on 5000 bootstrapped samples (Preacher & Hayes, 2004). The average variance extracted (AVE) value and composite reliability (CR) greater than .50 and .70, respectively, represent sufficient construct reliability (Fornell & Larcker, 1981).

We performed structural equation modeling with a cross-lagged panel model and LGCM. A cross-lagged panel model can be applied to investigate how the variables of interest are associated with each other. The statistical procedure is more rigorous than general time-lagged analysis (e.g., mediation) as it can control the influence of the variable at the previous time point (e.g., Datu et al., 2017). This approach was used for repeated measurements of coach and teammate incivility T1 and T2. Additionally, we applied the LGCM to examine the dynamic nature of psychological safety and its relationship with other variables of interest. The LGCM requires at least three waves of data for a continuous variable to analyze the change in the variable. The LGCM calculates 1) the mean of the intercept and 2) the mean of the slope (Duncan & Duncan, 2009). First, the mean of the intercept represents the initial value of psychological safety (i.e., T1) when adjusted for the measurement error. The parameter of the intercept mean can answer research questions such as, “how the baseline psychological safety is associated with the variables of interest at T2 & T3?” Second, the means of slope represents the development (i.e., the rate of change) of psychological safety across time with adjustments for measurement error. The parameter of slope mean can answer research questions, such as “how are the variables of interest and the rate of change in psychological safety associated with?” Therefore, one of the strengths of the LGCM is its flexibility; thus, the intercept and slope can be both independent and dependent variables in the structural model depending on the research questions (e.g., Guglielmi & Brekke, 2018).

Results

A missing data analysis revealed that the missing data was only 0.6%, indicating the influence of these missing data is negligible (Tabachnick & Fidell, 2013). The missing data were treated by an expectation-maximization method (Tabachnick & Fidell, 2013). Descriptive statistics and correlations are presented in Tables 1 and 2.

Table 1. Description statistics and Pearson correlations.

		1	2	3	4	5	6	7	8	9	10	11	12	13
1	Age
2	Coach Incivility T1	.16**
3	Coach Incivility T2	.07	.57**
4	Teammate Incivility T1	−.02	.28**	.24**
5	Teammate Incivility T2	−.02	.19**	.41**	.66**
6	Psychological safety T1	.004	−.33**	−.33**	−.54**	−.44**
7	Psychological safety T2	−.02	−.24**	−.36**	−.44**	−.62**	.63**
8	Psychological safety T3	−.05	−.15*	−.24**	−.41**	−.42**	.53**	.56**
9	Psychological Well-being T3	.20**	−.04	−.21**	−.24**	−.34**	.32**	.32**	.36**
10	Burnout T3	.01	.10	.22**	.28**	.30**	−.30**	−.31**	−.59**	−.51**
11	Social Responsibility T3	−.18**	−.22**	−.19**	−.06	−.06	.09	.11	.30**	−.02	−.31**
12	Prosocial Behavior T3	.15*	.04	−.09	−.17**	−.21**	.24**	.26**	.28**	.48**	−.25**	.02
13	Antisocial Behavior T3	.08	.13*	.19**	.26**	.31**	−.27**	−.35**	−.50**	−.14*	.39**	−.33**	−.16**
	Mean	16.04	2.03	1.94	1.76	1.75	5.41	5.42	5.29	5.63	2.69	5.09	5.98	2.57
	SD	1.40	1.33	1.13	1.05	1.00	1.19	1.10	1.15	1.02	1.32	1.55	0.78	1.21
	Minimum	12	1.00	1.00	1.00	1.00	1.25	1.50	1.75	2.00	1.00	1.00	3.00	1.00
	Maximum	18	6.83	5.83	5.50	5.17	7.00	7.00	7.00	7.00	7.00	7.00	7.00	7.00
	Skewness	−0.46	1.60	1.24	1.48	1.47	−0.69	−0.61	−0.52	−0.59	0.87	−0.90	−0.60	0.86
	Krurtios	−0.39	2.09	0.75	1.39	1.54	0.02	0.13	−0.19	0.10	0.39	0.18	−0.14	0.71
	Cronbach’s alpha	-	0.91	0.90	0.93	0.94	0.72	0.69	0.72	0.89	0.92	0.86	0.77	0.75

Note. *p < .05, **p < .01.

Table 2. Confirmatory factor analysis.

		Mean	SD	λ	C.R.	AVE
Coach Incivility (T1 & T2)
	“Has your coaches engaged in any of the following behaviors?”	(T1/T2)
1	Put you down or was condescending to you?	2.31/2.18	1.79/1.51	.91/.83	.92/.91	.67/.64
2	Paid little attention to your statement or showed little interest in your opinion?	2.15/2.00	1.84/1.57	.71/.69
3	Made demeaning or derogatory remarks about you?	1.90/1.81	1.50/1.18	.89/.84
4	Addressed you in unprofessional terms, either publicly or privately?	2.29/2.20	1.67/1.6	.85/.88
5	Ignored or excluded you from professional camaraderie?	1.58/1.53	1.27/0.99	.78/.77
6	Doubted your judgment on a matter over which you have responsibility?	1.97/1.94	1.45/1.27	.75/.78
Teammate Incivility (T1 & T2)
	“Has your teammates engaged in any of the following behaviors?”	(T1/T2)
1	Put you down or was condescending to you?	1.87/1.91	1.30/1.30	.87/.88	.94/.94	.71/.73
2	Paid little attention to your statement or showed little interest in your opinion?	1.85/1.78	1.34/1.19	.88/.88
3	Made demeaning or derogatory remarks about you?	1.84/1.79	1.30/1.15	.87/.93
4	Addressed you in unprofessional terms, either publicly or privately?	1.81/1.82	1.21/1.19	.84/.86
5	Ignored or excluded you from professional camaraderie?	1.42/1.51	0.90/0.95	.76/.79
6	Doubted your judgment on a matter over which you have responsibility?	1.75/1.70	1.17/1.10	.84/.78
Psychological Safety (T1, T2, & T3)		(T1/T2/T3)
1	If I make a mistake on this team, it is often held against me (R)	5.22/5.15/4.93	1.65/1.61/1.67	.61/.65/.67	.73/.69/.72	.41/.36/.40
2	People on this team sometimes reject others for being different (R)	5.00/5.13/4.76	1.69/1.55/1.74	.69/.62/.57
3	It is safe to take a risk on this team	5.76/5.75/5.75	1.35/1.31/1.23	.71/.55/.49
4	It is difficult to ask other team members for help (R)	5.65/5.67/5.73	1.75/1.65/1.53	.53/.57/.77
Psychological Well-being (T3)
1	You liked most parts of your personality	5.06	1.54	0.73	0.89	0.60
2	Good at managing the responsibilities of your daily life	5.35	1.35	0.78
3	You had warm and trusting relationships with others	6.04	1.11	0.79
4	You had experiences that challenged you to grow and become a better person	5.99	1.12	0.82
5	confident to think or express your own ideas and opinions	5.66	1.20	0.74
6	your life has a sense of direction or meaning to it	5.69	1.27	0.77
Burnout (T3)					0.93	0.62
1	I feel a lack of motivation in my sport and often think of giving up	2.16	1.45	0.77
2	I feel that I am losing interest in my sport	2.20	1.53	0.82
3	I’m continually wondering whether my sport participation has any meaning	2.14	1.54	0.78
4	I am physically exhausted	3.64	1.79	0.65
5	I am emotionally exhausted	2.77	1.82	0.87
6	I feel worn out	2.38	1.62	0.86
7	I am not achieving much in sport	3.50	1.84	0.76
8	It seems that no matter what I do, I do not perform as well as I should	2.97	1.78	0.80
9	I (do not) feel successful at sport	2.42	1.63	0.77
Social Responsibility (T3)					0.87	0.62
1	It is hard to get ahead without breaking the law now and then (R)	4.58	1.93	0.73
2	If I want to risk getting into trouble, that is my business and nobody else’s (R)	5.56	1.77	0.86
3	I don’t owe the world anything (R)	5.39	1.84	0.86
4	What I do with my life won’t make much difference one way or another (R)	4.82	1.84	0.68
Prosocial Behavior in sport (T3)					0.79	0.67
1	Encouraged a teammate	5.86	0.99	0.80
2	Congratulated a teammate for good play	6.41	0.71	0.71
3	Gave constructive feedback to a teammate	5.67	1.10	0.72
Antisocial Behavior in Sport (T3)
1	Argued with a teammate	2.97	1.61	0.56	0.78	0.55
2	Criticized a teammate	2.27	1.33	0.87
3	Showed frustration at a teammate’s poor play	2.48	1.47	0.75

Note . All factor loadings are significant (p < .001). Measurement model fit indices: χ2/df = 2619.60/1631 = 1.61; p < .001; CFI = .92; TLI = .91; RMSEA =

05; and SRMR = .06

Measurement model

We first conducted a confirmatory factor analysis (CFA) to determine the psychometric property of each construct measured in the study. The first measurement model showed insufficient model fit indices (χ²/df = 3738.80/ 2137 = 1.75; p < .001; CFI = .88; TLI = .86; RMSEA = .05; and SRMR = .07; Kline, 2005). The results demonstrated that some items in psychological safety (item 2: “Members of this team are able to bring up problems and tough issues” & item 7: “When working together with team members, my unique skills and talents are valued and utilized”) and social responsibility (item 5: “I really care about how my actions might affect others” & item 6: “I have a responsibility to make the world a better place”) presented low factor loadings (< .50; Hair et al., 2005). Therefore, those items were removed from further analysis. The modified measurement model showed better and adequate model fit indices (χ²/df = 2619.60/ 1631 = 1.61; p < .001; CFI = .92; TLI = .91; RMSEA = .05; and SRMR = .06). The AVE and CR for the constructs except psychological safety were greater than .50 and .70, respectively.

Regarding psychological safety, the scale (Edmondson, 1999) was conceptually and statistically validated in various contexts, including youth sport (e.g., Fransen et al., 2020; Smittick et al., 2019); however, as reported in Fransen et al.’s study, the scale had some problems. The present study also showed relatively low AVE values, as the threshold for the AVE is .50 (Fornell & Larcker, 1981). However, Fornell and Larcker (1981) also argue that the construct validity is still adequate if the CR is higher than .60. Thus, the CR and also Cronbach’s alphas for the psychological safety measurements in the present study showed adequate values (> .60), presented in Table 2. Since the concept is well-adopted in literature and most studies use Edmondson’s (1999) definition and questionnaire, the present study decided to keep the construct in the further analysis.

Hypothesis testing

We performed the structural equation modeling to test the causal relationships among the variables in the hypothesized model (Figure 1) across three time points. The model showed acceptable model fit indices (χ²/df = 2023.86/1257 = 1.61; p < .001; CFI = .93; TLI = .92; RMSEA = .05; and SRMR = .07). The summary of the results, including standardized coefficients and R-square values, was presented in Figure 2. Since only age was significantly correlated with some of the dependent variables, gender, competitive level, and weekly practice hours were excluded from the structural model.

Figure 2. Results of SEM with cross-lagged panel and latent growth curve modeling.

First, we tested the reciprocal relationship between coach/teammate incivility across two time points. The model showed that the autoregressive path from coach incivility T1 to teammate incivility T2 (β = .61, 95% CI = [.47, .72]) and from teammates incivility T1 to teammates incivility T2 (β = .65, 95% CI = [.52, .76]) were both significant (H1a & H1b). However, the cross-lagged paths from coach incivility T1 to teammates incivility T2 (β = −.05, 95% CI = [−.15, .06]) and from teammates incivility T1 to coach incivility T2 (β = .04, 95% CI = [−.10, .17]) were not significant. Thus, Hypotheses 2a and 2b were not supported.

Next, we examined how incivility and the intercept and slope (the rate of change) of psychological safety were associated with each other. The present result found that coach incivility T1 (β = .11, 95% CI = [−.09, .32]) was not but teammates incivility T1 (β = −.26, 95% CI = [−.50, −.03]) was significantly and positively associated with the slope of psychological safety, partially supporting the hypotheses (H3b). As we hypothesized, the intercept of psychological safety significantly and negatively predicted coach incivility T2 (β = −.16, 95% CI = [−.29, −.03]) and teammates incivility T2 (β = −.15, 95% CI = [−.26, −.03]), supporting the hypotheses (H4a & H4b). Additionally, coach incivility T2 (β = −.26, 95% CI = [−.49, −.06]) was significant and negative while teammates incivility T2 (β = −.16, 95% CI = [−.41, .08], p = .07) was only marginally associated with the slope of psychological safety, partially consistent with our hypotheses (H5a). Lastly, we investigated how the intercept and slope of psychological safety were associated with well-being and social outcomes. The results found that the intercept and slope of psychological safety significantly predicted all outcome variables (please see, Table 3) as we hypothesized (H6 & H7).

Table 3. Results for hypothesis testing.

	Path	Estimate	boot S.E.	95% BC-CI	p -value
H1a	Coach Incivility T1 → Coach Incivility T2	0.61	0.06	[.47, .72]	***
H1b	Teammate Incivility T1 → Teammate Incivility T2	0.65	0.06	[.52, .76]	***
H2a	Coach Incivility T1 → Teammate Incivility T2	−0.05	0.05	[−.15, .06]	0.31
H2b	Teammate Incivility T1 → Coach Incivility T2	0.04	0.07	[−.10, 17]	0.45
H3a	Coach Incivility T1 → Psychological Safety Slope	0.11	0.11	[−.09, .32]	0.19
H3b	Coach Incivility T2 → Psychological Safety Slope	−0.26	0.11	[−.49 −.06]	**
H4a	Teammate Incivility T1 → Psychological Safety Slope	−0.26	0.11	[−.50, −.03]	**
H4b	Teammate Incivility T2 → Psychological Safety Slope	−0.16	0.12	[−.41, .08]	0.07
H5a	Psychological Safety Intercept → Coach Incivility T2	−0.16	0.07	[−.29, −.03]	**
H5b	Psychological Safety Intercept → Teammate Incivility T2	−0.15	0.06	[−.26, −.03]	**
H6a	Psychological Safety Intercept → Psychological Well-being T3	0.50	0.08	[.33, .64]	***
H6b	Psychological Safety Intercept → Burnout T3	−0.50	0.08	[−.65, −.33]	***
H6c	Psychological Safety Intercept → Social Responsibility T3	0.30	0.12	[.08, .53]	***
H6d	Psychological Safety Intercept → Pro-social Behavior T3	0.42	0.08	[.26, .56]	***
H6e	Psychological Safety Intercept → Anti-social Behavior T3	−0.45	0.09	[−.62, −.28]	***
H7a	Psychological Safety Slope → Psychological Well-being T3	0.43	0.11	[.22, .64]	**
H7b	Psychological Safety Slope → Burnout T3	−0.75	0.08	[−.87, −.55]	***
H7c	Psychological Safety Slope → Social Responsibility T3	0.74	0.11	[.47 .91]	***
H7d	Psychological Safety Slope → Pro-social Behavior T3	0.33	0.11	[.13, .54]	***
H7e	Psychological Safety Slope → Anti-social Behavior T3	−0.73	0.09	[−.89, −.54]	***

Note. Fit indices: χ2/df = 2023.86/1257 = 1.61; p < .001; CFI = .93; TLI = .92; RMSEA = .05; and SRMR = .07

*p < .05, **p < .01, ***p < .001.Running head: INCIVILITY IN YOUTH SPORT

Optional analyses

The present study supplementally examined the mediating role of the slope of psychological safety in the relationship between incivility and outcome variables and the relationship between the intercept and the slope through coach/teammate incivility at T2. The results of the mediation analyses are presented in Table 4. Additionally, we tested the direct associations of coach/teammate incivility with outcome variables, showing that only two paths were found to be significant: coach incivility T1 and social responsibility T3 (β = −.17, 95% CI = [−.33, −.01]), and teammate incivility T2 and social responsibility T3 (β = .25, 95% CI = [.10, .44]).

Table 4. Results for the indirect and direct effects (Optional analyses).

	Path	Estimate	boot S.E.	95% BC-CI	p -value
Indirect effects
	Coach Incivility T1 → Psychological Safety Slope → Psychological Well-being T3	0.05	0.05	[−.03, .16]	0.24
	Coach Incivility T1 → Psychological Safety Slope → Burnout T3	−0.09	0.09	[−.28, .07]	0.26
	Coach Incivility T1 → Psychological Safety Slope → Social Responsibility T3	0.09	0.09	[−.08, .28]	0.28
	Coach Incivility T1 → Psychological Safety Slope → Pro-social Behavior T3	0.03	0.03	[−.02, .11]	0.24
	Coach Incivility T1 → Psychological Safety Slope → Anti-social Behavior T3	−0.07	0.06	[−.20, .05	0.25
	Teammate Incivility T1 → Psychological Safety Slope → Psychological Well-being T3	−0.11	0.06	[−.28, −.03]	*
	Teammate Incivility T1 → Psychological Safety Slope → Burnout T3	0.23	0.11	[.03, .49]	*
	Teammate Incivility T1 → Psychological Safety Slope → Social ResponsibilityT3	−0.22	0.12	[−.48, −.03]	*
	Teammate Incivility T1 → Psychological Safety Slope → Pro-social Behavior T3	−0.07	0.04	[−.17, −.01]	*
	Teammate Incivility T1 → Psychological Safety Slope → Anti-social Behavior T3	0.16	0.09	[.02, .37]	*
	Coach Incivility T2 → Psychological Safety Slope → Psychological Well-being T3	−0.13	0.07	[−.30, −.03]	*
	Coach Incivility T2 → Psychological Safety Slope → Burnout T3	0.25	0.11	[.06, .52]	*
	Coach Incivility T2 → Psychological Safety Slope → Social Responsibility T3	−0.25	0.12	[−.52, −.06]	*
	Coach Incivility T2 → Psychological Safety Slope → Pro-social Behavior T3	−0.07	0.04	[−.19, −.01]	*
	Coach Incivility T2 → Psychological Safety Slope → Anti-social Behavior T3	0.18	0.08	[.04, .38]	*
	Teammate Incivility T2 → Psychological Safety Slope → Psychological Well-being T3	−0.09	0.08	[−.30, .03]	0.15
	Teammate Incivility T2 → Psychological Safety Slope → Burnout T3	0.18	0.14	[−.08, .47]	0.16
	Teammate Incivility T2 → Psychological Safety Slope → Social Responsibility T3	−0.18	0.16	[−.55, .07]	0.16
	Teammate Incivility T2 → Psychological Safety Slope → Pro-social Behavior T3	−0.05	0.05	[−.18, .02]	0.13
	Teammate Incivility T2 → Psychological Safety Slope → Anti-social Behavior T3	0.16	0.09	[.02, .37]	*
	Psychological Safety Intercept → Coach Incivility T2 → Psychological Safety Slope	0.04	0.02	[.01, .10]	*
	Psychological Safety Intercept → Teammate Incivility T2 → Psychological Safety Slope	0.02	0.02	[−.004, .07]	0.097
Direct effects
	Coaches Incivility T1 → Social Responsibility T3	−0.17	0.08	[−.33, −.01]	**
	Teammates Incivility T2 → Social Responsibility T3	0.25	0.08	[.10 .44]	***

Note. *p < .05, **p < .01, ***p <.001.

Discussion

The purposes of the present study were to investigate 1) how incivility and psychological safety are associated with each other over time and 2) the influence of the initial point (i.e., intercept) and the development (i.e., slope) of psychological safety on well-being and social outcomes (e.g., prosocial and antisocial behaviors) of the high-school and middle-school level youth athletes in Japan. The results of SEM found that coach incivility T2 and teammate incivility T1 were significantly associated with the slope (the rate of change) of psychological safety. In contrast, the intercept (initial value) of psychological safety predicted coach and teammate incivility T2. Lastly, both the intercept and the slope of psychological safety were significantly associated with youth athletes’ well-being and social outcomes T3. Therefore, the results contribute to understanding how the incivility spiral (Andersson & Pearson, 1999) occurs in youth sport in Japan and its impact on youth athletes’ well-being and social outcomes. The present results also support the COR theory’s (Hobfoll, 1989) assumption by showing that psychological safety significantly affects youths’ well-being and development in sport.

The present study found that only autoregressive paths (from T1 to T2 in the same construct) were significant (H1a & H1b). In other words, coach incivility T1 was only associated with coach incivility T2, and teammate incivility T1 was only associated with teammate incivility T2. Since previous research found that incivility is contagious (Foulk et al., 2016; Porath & Erez, 2009), we expected that coach incivility would predict future teammate incivility and vice versa. However, the hypotheses (H2a & H2b) were not supported in the study, although they were significantly correlated with each other. The reason may be that the coach and teammate incivility showed low scores with small variances in the present study. Accordingly, coach and teammate incivility was not prevalent among the current participants. The point is a significant limitation of the study since the present participants were recruited by contact from the coaches who are interested in the present study, meaning that those coaches are sincere in learning effective coaching and leadership, which is also discussed in the limitation section. The present findings contribute to the development of the concept of incivility in youth sport by considering both the coach and teammate aspect and its reciprocal relationship (H1 & H2).

The present results found that coach incivility T2 was significantly and negatively associated with the slope (change across three times) of psychological safety, while coach incivility T1 was not related (Hypotheses 3). Therefore, the assumption that leadership is a significant predictor of psychological safety (e.g., Newman et al., 2017) was supported in the youth sport context. More specifically, the present results demonstrated that coach incivility at T2 (right before the most important competition) rather than coach incivility at T1 (the beginning of the season) was negatively associated with the change in psychological safety. In contrast, teammate incivility T1 was significantly and negatively related to the slope of psychological safety, but teammate incivility T2 was not associated with it (Hypotheses 4). Therefore, as suggested by previous studies (e.g., Holt et al., 2017; Newman et al., 2017; Sheridan et al., 2014), teammate incivility was also a significant factor affecting psychological safety. In particular, the present results found that teammate incivility T1 rather than teammate incivility T2 was significantly related to the change in psychological safety over time. Regarding the incivility at the initial time point (T1), teammate incivility, compared to coach incivility, was significantly impactful on the development of psychological safety. In contrast, at the second time point (T2), coach incivility was significantly associated with the development of psychological safety, although teammate incivility was not related.

The results are understandable as many coaches gradually reduce the intensity of coaching behaviors toward the important competitions; for example, coaches exhibited high frequencies in training and instruction behavior and low frequencies in autocratic behaviors and social support in the preseason (Amorose & Horn, 2001). Thus, the team with relatively higher coach incivility, even right before the important competition (T2), was negatively associated with psychological safety. In fact, the mean score showed that coach incivility T2 was relatively lower than T1, also supporting the interpretation. Nevertheless, teammate incivility was more stable from T1 to T2, and incivility among teammates, relative to coach incivility, at the beginning of the season (T1) more strongly affected the development of psychological safety in the team. The findings suggest that teammate incivility, in general, has a significant negative impact on the team, specifically at the beginning of the season. Therefore, the results provided a unique insight into how coaches and teammates differently influence the development of psychological safety in the body of literature (H3 & H4).

The current findings also demonstrated that the intercept of psychological safety (T1) also predicted coach and teammate incivility T2 (Hypothesis 5). As Edmondson (1999) claims that psychological safety fosters the organizational climate characterized by mutual trust and respect and is consistent with the assumption of incivility spiral (Andersson & Pearson, 1999), psychological safety was negatively associated with both coach and teammate incivility. Hence, the present findings support the relationships between incivility and psychological safety are mutually influential. The results imply that other factors affecting psychological safety would indirectly influence incivility through psychological safety. Consequently, the present findings added significant evidence to the theoretical foundation of the incivility spiral (Andersson & Pearson, 1999) by demonstrating that psychological safety can be a significant antecedent of lower incivility (H5). The findings contribute to the development of the theoretical foundation of the incivility spiral.

Moreover, the results of Hypotheses 3-5 support the reciprocal relationship between incivility and psychological safety, namely the incivility spiral (Andersson & Pearson, 1999). The results are congruent with the assumption that incivility and the team’s climate are interactive (Andersson & Pearson, 1999). In addition to the previous findings demonstrating that psychological safety was predicted by leadership or relationship with others in the organization (Newman et al., 2017), the present results showed that psychological safety also predicted future coach and teammate incivility. Thus, findings of the reciprocal relationship contribute to developing the literature by showing the detailed mechanism of how incivility and psychological safety are associated with each other in youth sport using a longitudinal study design.

Lastly, the present study found that both the intercept and the slope of psychological safety were significantly associated with youth athletes’ well-being and social outcomes (Hypotheses 6 & 7). The results suggest that the initial point of psychological safety (i.e., T1) predicted the outcomes variables T3, and the development of psychological safety were also significantly associated with the outcomes T3. The results can be explained by COR theory (Hobfoll, 1989), arguing that individuals invest resources to enhance their future resources when sufficient resources are available. This assumption indicates that the affordability of the resources promotes well-being and positive development (Obrenovic et al., 2020; Whitman et al., 2014). Hence, psychologically safe environments are essential for obtaining, retaining, and protecting their valued resources. Since psychological safety provides youths with accessibility to resources, it is a significant factor for youths’ well-being and development (H6 & H7). Moreover, the present findings also added unique evidence that the accessibility of significant resources in the team can facilitate youth athletes’ ethical development.

Optional analyses revealed that psychological safety mediates the relationship between incivility and outcomes. As incivility can cause stronger stressors in the team by inducing the threat of a loss in their resources, it will cause decreasing cognitive resources of the team members (Duncan & Duncan, 2009; Hobfoll, 1989; Riskin et al., 2017). The threat of a loss in their resources (e.g., incivility) can harm individuals’ cognitive resources, in turn leading them to dysfunction through the decreased accessibility to resources in the team. Thus, it is reasonable that incivility can negatively affect the feeling of psychological safety, in turn decreasing their well-being and development. Consequently, the theory and our findings consistently indicate that it is imperative to prevent individuals from depleting their resources and the accessibility to resources to maximize their performance and development (Hobfoll, 1989).

Since the optional analyses revealed that only the direct relationships between incivility and social responsibility were significant, the results imply the more significant impact of psychological safety on outcomes over the impacts of incivility, providing evidence that psychological safety is an essential factor in youth sport. Still, it deserves theoretical interpretations for the significant direct relations. It is reasonable to observe the negative association between coach incivility and social responsibility as coach incivility can lead to a number of negative outcomes (Cunningham et al., 2013; Smittick et al., 2019). Interestingly, teammate incivility was positively associated with social responsibility. We argue that such responses could be regarded as youths’ defense mechanisms in relation to multiple in-group identities (Lock & Funk, 2016). As our data indicated that incivility was not contagious from coach to teammates and vice versa, youths might have formed at least two different groups with coach and teammates. If youth participants experience conflicts in a primary group they belong to (i.e., in-group with teammates), they may conveniently demonstrate a positive attitude in other domains (i.e., social responsibility) to maintain desirable self-identity. Nevertheless, our interpretations are speculation at best. Future studies should assess whether or not and how youths from multiple in-groups in team sport settings.

Practical implications

The present results suggest practical implications for the positive development of young people using sports as a tool. The present study found that the improvement of psychological safety is key in sport teams to nurture youths’ well-being and social outcomes. Additionally, the results found that psychological safety was related to future uncivil behaviors among coaches and teammates. Therefore, it is paramount for stakeholders in sport teams, clubs, or organizations to understand the importance of creating psychologically safe environments for youth athletes. Also, the present study suggested that the team-level factor is affected by coaches’ and teammates’ behaviors. The findings indicate that the facilitation of civil behaviors (e.g., respectful behaviors) in the team would significantly increase psychological safety in the team. For example, a previous study showed that the expression of gratitude to team members could improve team functioning (Riskin et al., 2019). Therefore, the coach and teammate can engage in a gratitude intervention program in the team (Gabana et al., 2019) so that a more psychologically safe environment may be developed in youth sport programs.

The present study also found how sport can be used to improve youth sport environments and foster positive youth development. In addition to the previous sport for development literature indicating the importance of sport-based development programs (Inoue et al., 2015; Lee et al., 2021) and the roles of coaches, parents, and teammates (e.g., Holt et al., 2017; Sheridan et al., 2014), the present study specified the significance of psychologically safe environments for acquiring more positive well-being and social outcomes. Therefore, sport-based development programs might need to consider fostering psychologically safe environments in sport teams/clubs or organizations. The present results suggest a better idea for the development of sport-based development programs.

Limitations and future directions

The present study showed relatively low measurement validity in terms of psychological safety. Although the scale was well-validated in various fields (cf., Edmondson & Lei, 2014) and the Japanese translation was performed by the experts on the concept, the measurements were developed and mostly tested in Western cultural contexts. It might be important to consider the culture of the present study context (Eastern culture). Therefore, future studies necessarily replicate the results in the different cultural contexts and may need to modify the items to adopt the culture. Next, future research may need to understand why uncivil behaviors occur more in a particular organization. For instance, coaches’ behaviors would be influenced by higher-order factors such as the school’s policy and governance (cf., Schilpzand et al., 2016). Moreover, the coaches of the present participants were interested in the aims of the present study. This may be a cause of bias; thus, future research needs to include more diverse populations, including competitive levels and types of sports, and a large sample size. Lastly, the present study only investigated the influence of incivility. Previous literature has indicated the significance of civility in organizations (e.g., Porath et al., 2015) based on positive organizational behavior literature. Hence, researchers also need to consider examining the significance of the positive side of coach and teammate behaviors in sport environments.

Conclusion

The present study found that coach and teammate incivility was negatively associated with the change in psychological safety, but the initial level of psychological safety predicted subsequent coach and teammate incivility. Also, psychological safety was associated with important outcomes for youth athletes (e.g., well-being and social outcomes). The results found a mechanism underlying the relationship between incivility, psychological safety, and important outcomes (i.e., well-being & social outcomes) in youth sport.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The present research was financially supported by Grant-in-Aid for Research Activity Start-up of the Japan Society for the Promotion of Science [JSPS KAKENHI Grant Number 21K21232]