Introduction
Considering the importance of fairness in a workplace or organization, numerous researchers have examined the relationships between individuals’ perceptions of fairness and organizational outcome variables, such as organizational commitment, job satisfaction, organizational citizenship behavior (OCB), turnover intention, trust, and work outcomes (e.g., Bakhshi, Kumar, & Rani, Reference Chelladurai2009; Khan, Abbas, Gul, & Raja, Reference Khan, Abbas, Gul and Raja2013; Rai, Reference Rai2013; Schilpzand, Martins, Kirkman, Lowe, & Chen, Reference Schilpzand, Martins, Kirkman, Lowe and Chen2013; Suliman & Kathairi, Reference Suliman and Kathairi2013). Indeed, several scholars in their meta-analyses found that organizational justice was found to be a significant predictor of the aforementioned outcome variables (e.g., Cohen-Charash & Spector, Reference Cohen-Charash and Spector2001; Colquitt, Conlon, Wesson, Porter, & Ng, Reference Colquitt, Conlon, Wesson, Porter and Ng2001; Li & Cropanzano, Reference Li and Cropanzano2009; Shao, Rupp, Sharlicki, & Jones, Reference Shao, Rupp, Skarlicki and Jones2013). Consequently, the significance of such relationships has led to the development of substantial knowledge in organizational justice within the disciplines of management and psychology. While there has been relatively little research focusing on the relationships between organizational justice, and organizational attitudes, and behaviors in a sport context, Jordan, Gillentine, and Hunt (Reference Jordan, Gillentine and Hunt2004) proposed that improving athlete’s perception of fairness in a team sport setting would increase their positive attitudes and behaviors toward their teams or coaches as well as team unity and commitment. This proposition needs to be further examined in a sport context as sport teams share many common characteristics with other types of organizations (see Chelladurai, Reference Dittmore, Mahony, Andrew and Hums2001) due to the fact that coaches should possess skills similar to those required in other management positions.
Perception of fairness as a key construct of organizational justice is considered to be one of the most important antecedents affecting organizational employees’ or members’ attitudes and behaviors (Greenberg & Colquitt, Reference Greenberg and Colquitt2005). For instance, it is plausible that an authority figure’s unfair treatment of members in an organization often give rises to a lack of effort and cooperation among them. In a team sport setting, likewise, it is critical to understand athletes’ perceptions of fairness since their attitudes and behaviors derived from the perceptions of fairness are likely to either negatively or positively influence goals/objectives of a sport team. For example, athletes having unfair negative perceptions of fairness toward their teams would show detrimental behaviors, such as group fragmentation, athlete drop out, and withholding effort (Jordan, Gillentine, & Hunt, Reference Jordan, Gillentine and Hunt2004), and may negatively affect the team’s resulting athletic performance (e.g., a win-loss record). In contrast, athletes having positive perceptions of fairness would reveal favorable attitudes and behaviors for their teams, such as increased satisfaction, commitment, performance, and group cohesion, and eventually contribute to the success of the team.
As noted above, organizational commitment is one of the most frequently employed outcome variables in relation to organizational justice. Organizational commitment can generally be defined as a psychological attachment between an employee or member and an organization (Meyer & Allen, Reference Meyer and Allen1991). Originally, organizational commitment was viewed as a single dimensional construct, but there was little agreement with this perspective (see Meyer, Allen, & Topolnytsky, Reference Meyer, Allen and Topolnytsky1998). Today, it is well documented that organizational commitment consists of three components: (a) affective, (b) continuance, and (c) normative commitment (Meyer & Allen, Reference Meyer and Allen1991). Affective commitment is concerned with an individual’s strong emotional attachment and level of involvement to an organization, while continuance commitment is related to the cost of leaving the organization. Normative commitment is associated with a feeling of obligation to stay with the organization. Among these components, most individuals’ work experiences in an organization would be primarily expected to have a relation with affective domain (Allen & Meyer, Reference Andrew, Kacmar, Blakely and Bucklew1996). In addition, affective commitment involves feelings of intrinsic motivation rather than feelings of pressure (continuance commitment) and obligation (normative commitment) and is likely to be related to positive attitudes and behaviors (Poon, Reference Poon2013). Furthermore, affective commitment out of the three components is the most widely used component in the research on organizational justice (Colquitt et al., Reference Colquitt, Conlon, Wesson, Porter and Ng2001). Thus, the current study focuses on examining the affective commitment.
According to a review of literature on organizational justice, when employees perceive that they are treated fairly in an organization, they are more likely to be committed and tend to identify with or emotionally attach to the organization (Meyer & Allen, Reference Meyer and Allen1997; Bakhshi, Kumar, & Rani, Reference Chelladurai2009; Rai, Reference Rai2013). Similarly, the results were also supported by meta-analytic reviews (Cohen-Charash & Spector, Reference Cohen-Charash and Spector2001; Li & Cropanzano, Reference Li and Cropanzano2009). Based upon these empirical findings from other disciplines, it would be expected that when athletes possess positive perceptions of fairness in a team sport setting, they are more likely to be committed to their team.
Organizational commitment (i.e., affective commitment) is not just affected by organizational justice (Colquitt et al., Reference Colquitt, Conlon, Wesson, Porter and Ng2001), but also affected by other variables, such as: perceived organizational support (Rhoades, Eisenberger, & Armeli, Reference Rhoades, Eisenberger and Armeli2001), OCB (Schilpzand et al., Reference Schilpzand, Martins, Kirkman, Lowe and Chen2013), and leadership style (Avolio, Zhu, Koh, & Bhatia, Reference Carron, Brawley and Widmeyer2004). Several scholars outside of the sport management field have called for the investigation of the impact of moderating variables in order to substantially improve our knowledge and understanding of the justice–commitment link (Ambrose & Schminke, Reference Bakhshi, Kumar and Rani2003; Nowakowski & Conlon, Reference Nowakowski and Conlon2005; Andrew, Kacmar, Blakely, & Bucklew, Reference Bies and Moag2008). Given that recent directions in justice research is that justice perceptions are mainly developed based on one’s work group (Liao & Rupp Reference Liao and Rupp2005; Colquitt & Jackson, Reference Colquitt and Jackson2006), one possible moderating variable between organizational justice and commitment is group cohesion.
In a sport context, Greenberg, Mark, and Lehman (Reference Greenberg, Mark and Lehman1985) also suggested that perceptions of unfairness (i.e., injustice) in a team sport setting can influence team/group cohesion. Group cohesion is defined as a dynamic process which reflects the tendency for a group to stick together and remain united in the pursuit of its instrumental objectives and/or for the satisfaction of a member’s affective need (Carron, Brawley, & Widmeyer, Reference Cropanzano, Bowen and Gilliland1998). Within an organization, the degree of group cohesion refers to the level of trust, cooperation, and friendship (Andrew et al., Reference Bies and Moag2008). Conceptually, individuals with high perceptions of group cohesion are likely to remain a part of one’s group, be loyal to the group, and identify with the group (Friedkin, Reference Friedkin2004). With its perceived importance in an organizational context, previous empirical studies found that group cohesion was positively associated with affective commitment (Wech, Mossholder, Steel, & Bennett, Reference Wech, Mossholder, Steel and Bennett1998) as well as team performance (Pillai & Williams, Reference Pillai and Williams2004). Further, Cuskelly (Reference Cuskelly1995) found that there was a moderate relationship between group cohesion and sport executive members’ commitment to the organization. Applying these findings to a team sport setting, it is plausible that athletes are more likely to become attached to their team and further show team loyalty when they experience positive perceptions of fairness about their coaches and have a high level of group cohesion in the team.
As stated above, organizational commitment among athletes can be strengthened through both the perception of fairness (i.e., organizational justice) and group cohesion. Considering that previous research outside of the sports contexts has already demonstrated the relationships between each of the two antecedent variables (organizational justice and group cohesion) and organizational commitment independently, it is essential to investigate how organizational commitment is affected when these two antecedent variables are taken into account together in a team sport setting. Indeed, Andrew et al. (Reference Bies and Moag2008) have recently explored the moderating effect of group cohesion between organizational justice and affective commitment among employees of a pharmaceutical company. According to the findings of the study, individuals with high levels of group cohesion would have a stronger relationship between organizational justice and affective commitment than those with low levels of group cohesion. Based upon the finding, the current study will examine the moderating role of group cohesion in the relationship between organizational justice and affective commitment in a team sport setting. We will begin by providing an overview of organizational justice and related research in a sport context.
OVERVIEW OF ORGANIZATIONAL JUSTICE
Studies about the perceptions of fairness have been conducted within the framework of organizational justice. Over the past 40 years, scholars in the field of organizational justice have worked with three dimensions of organizational justice: (a) distributive justice, (b) procedural justice, and (c) interactional justice. Even though numerous studies in the beginning primarily focused on distributive justice, it is not sufficient to fully account for the concept of justice, which led to the emergence of other justice principles (procedural and interactional justices). We will explain each of the three principles in relation to a team sport settings.
Distributive Justice
Distributive justice refers to the perceived fairness of outcomes one receives. This type of justice was originally grounded in Adams’ (Reference Adams1963, Reference Adams1965) theory of inequity. According to the theory, individuals in a group or organization compare the ratio of their inputs (contributions) to their outputs (rewards) with the ratio of inputs and outputs of other workers. If the ratio is equal, individuals would be satisfied with it and perceived fairness. In an athletic context, particularly in a team setting, the equity principle can be applied in two ways: (a) ability (e.g., skill, athletic performance) and (b) effort (Tornblom & Jonsson, Reference Tornblom and Jonsson1985). For example, an athlete may feel that because he or she dedicated a great amount of effort for the practice and made significant contribution toward the team’s success (i.e., high inputs) compared with other athletes, he/she deserves to be selected for a starting position (rewards). However, if the coach selected another athlete who is believed to have made less contribution or effort to the team as a starting member, the athlete’s level of group cohesion or commitment to the team would decrease in that his/her expected outcome did not match with the actual outcome.
Procedural Justice
While distributive justice focuses on the perception of fairness of outcomes or end results, procedural justice centers on the notion of perceived fairness of the process or procedures in which outcomes are determined, regardless of the outcomes the individuals or groups receive (Thibaut & Walker, Reference Thibaut and Walker1975). Greenberg (Reference Greenberg1990) connected the development of procedural justice with the study of Thibaut and Walker (Reference Thibaut and Walker1975). Thibaut and Walker compared participants’ reactions to simulated dispute-solution procedures with respect to two types of control in legal decisions: decision control (the amount of control they had over directly determining the outcomes) and process control (the amount of control they offered the disputants over the procedures used to settle their grievance). According to Greenberg (Reference Greenberg1990), research using simulated legal decisions found that verdicts were perceived as fairer and were better accepted when procedures offered the disputants process control than identical decisions denying process control. The concept of procedural justice was formerly introduced by Greenberg and Folger (Reference Greenberg and Folger1983) in the field of management. However, there exists a conceptual controversy on the relationship between procedural justice and distributive justice. Indeed, Mahony, Hums, Andrew, and Dittmore (Reference Mahony, Hums, Andrew and Dittmore2010) in their systematic review found that while some scholars provided evidence on high correlations between the two, others provided support for examining the two justices as separate dimensions. Although the conceptual debate exists in the relationship between the two justices, the vast majority of contemporary research on organizational justice conceptualizes distributive justice and procedural justice as distinctive dimensions (Mahony et al., Reference Mahony, Hums, Andrew and Dittmore2010).
Applying the concept of procedural justice to an athletic context, the athlete in the example of the starting position above would accept the fact that he/she was not selected in a starting position if the process or procedure adopted by the coach in selecting a starting position (i.e., team vote) was perceived as fair and objective.
Interactional Justice
The next dimension of organizational justice is interactional justice. Interactional justice is grounded on the interpersonal treatment by and quality of information an individual received from authority figures or decision makers (Bies & Moag, Reference Colbert and Kwon1986). In other words, interactional justice underscores the human side of organizational practice (Cohen-Charash & Spector, Reference Cohen-Charash and Spector2001). This form of justice is divided into two constructs: interpersonal and informational justice (Greenberg, Reference Greenberg1993). Interpersonal justice relates to the extent to which an individual is treated with respect, dignity, and politeness by authority figures during procedures (Brockner & Wiesenfeld, Reference Colquitt, Conlon, Wesson, Porter and Ng1996). If a coach treats athletes with dignity and respect, the athletes would be more likely to perceive fairness (Jordan, Gillentine, & Hunt Reference Jordan, Gillentine and Hunt2004). Informational justice refers to the extent to which information is honestly provided by decision makers and whether the information is communicated/explained in a thorough and timely fashion (Colquitt, Reference Colquitt2001; Li & Cropanzano, Reference Li and Cropanzano2009; Shao et al., Reference Shao, Rupp, Skarlicki and Jones2013). For example, even though the procedure of a particular decision made by a coach seems fair, the athlete may not perceive it as fair if the coach does not thoroughly and honestly explain the procedure.
ORGANIZATIONAL JUSTICE RESEARCH IN A SPORT SETTING
Organizational Justice in Intercollegiate Athletics
Research on organizational justice in a sport setting has largely been conducted in two settings: (a) intercollegiate (Hums & Chelladurai, Reference Hums and Chelladurai1994a, Reference Hums and Chelladurai1994b; Mahony, Hums, & Riemer, Reference Mahony, Hums and Riemer2002, Reference Mahony, Hums and Riemer2005; Patrick, Mahony, & Petrosko, Reference Patrick, Mahony and Petrosko2008) and (b) interscholastic athletics (Whisenant, Reference Whisenant2005; Whisenant & Jordan, Reference Whisenant and Jordan2006, Reference Whisenant and Jordan2008; Whisenant & Smucker, Reference Whisenant and Smucker2007). As previously mentioned, organizational justice consists of three dimensions: distributive, procedural, and interactional justice (divided into interpersonal and informational justice). For intercollegiate athletics, of the three dimensions, distributive justice has been the most widely examined with resource allocations in the athletic departments during the last 2 decades.
Using the conceptual framework of Tornblom and Josson (Reference Tornblom and Jonsson1985), Hums and Chelladurai (Reference Hums and Chelladurai1994a, Reference Hums and Chelladurai1994b) first developed three principles, based upon distributive justice, and applied them to the perceived fairness of resource allocations in collegiate athletic department. The three principles included equity (contribution), equality, and need. Under the equity principle, there are four sub-principles such as (a) effort, (b) ability, (c) productivity, and (d) spectator appeal. Later, an additional sub-principle of revenue generation was developed by Mahony, Hums, and Riemer (Reference Mahony, Hums and Riemer2002, Reference Mahony, Hums and Riemer2005). For example, an athletic team who has contributed more to an athletic department, in terms of productivity (e.g., win-loss record), effort, ability (e.g., skill level), spectator appeal (e.g., attracting more people), and revenue generation, should receive greater amounts of resource allocations than others.
With respect to the equality principle, there are three sub-principles including (a) equality of results, (b) equality of treatment, and (c) equality of opportunity (Hums & Chelladurai, Reference Hums and Chelladurai1994a). First, equality of results means that even though inequalities regarding resource allocation occur in the short term, the outcomes (results) are the same for every athletic team in the long term. Second, equality of treatment means that every athletic team in a given situation should receive the same amount of resources. Lastly, equality of opportunity suggests that every athletic team has the same chance to receive a certain amount of resources within the athletic department.
The last principle of distributive justice in an intercollegiate athletics setting is need. It refers to the notion that those who have less necessary resources should receive a greater amount of resources than others (Mahony, Hums, & Riemer, Reference Mahony, Hums and Riemer2002). The need principle was further defined as three sub-principles by Mahony and his colleagues: (a) need due to lack of resources, (b) need due to high costs, and (c) need to be competitively successful (Mahony, Hums, & Riemer, Reference Mahony, Hums and Riemer2005). These need-based sub-principles have recently been further examined in national sport governing bodies (Dittmore, Mahony, Andrew, & Hums, Reference Dittmore, Mahony, Andrew and Hums2009) as well as intercollegiate athletics (Kim, Andrew, Mahony, & Hums, Reference Kim, Andrew, Mahony and Hums2008; Patrick, Mahony, & Petrosko, Reference Patrick, Mahony and Petrosko2008). While all of the studies mentioned above focused primarily on distributive justice, Kim and Andrew (Reference Kim and Andrew2012) and Thorn (Reference Thorn2010) were the first to examine other dimensions of organizational justice in intercollegiate athletics (procedural and interactional justice) in order to fully understand the perceptions of fairness regarding resource allocation of the athletic department.
Even though these prior studies provided great insight into understanding resource allocations in a sport context, particularly in intercollegiate athletics, there are several limitations to be addressed. First, all of the works, with the exception of two studies (Thorn, Reference Thorn2010; Kim & Andrew, Reference Kim and Andrew2012), have solely centered on one dimension of organizational justice, distributive justice. Understanding other dimensions of organizational justice helps sport practitioners to better and more fully explain the perceptions of fairness regarding resource allocation. Second, Cropanzano, Bowen, and Gilliland (Reference Cropanzano, Bowen and Gilliland2007) pointed out that the rationale and importance to explore perceptions of organizational justice would be significantly decreased if empirical research keeps showing that organizational justice is not associated with attitudinal and behavioral outcome variables such as organizational commitment, job satisfaction, and OCB. However, while the aforementioned research in the intercollegiate setting has all focused on perceptions of organizational justice, little attention was paid to the outcome of organizational justice. Further, examining the effect of organizational justice on outcome variables may provide meaningful managerial implications for sport organizations. Lastly, the majority of the organizational justice studies in the intercollegiate athletics have employed hypothetical scenarios to measure the fairness perceptions of participants. It is uncertain whether hypothetical scenarios reflect actual perceptions of fairness of participants.
Organizational Justice in Interscholastic Athletics
While the research on organizational justice in intercollegiate athletics has primarily focused on resource allocations with distributive justice, the research focus in interscholastic athletics was on organizational behaviors in relation to the multiple-dimensions of organizational justice. Since Jordan , Gillentine, and Hunt (Reference Jordan, Gillentine and Hunt2004) proposed that the relationships between the construct of organizational justice and individuals’ attitudes and behaviors toward their organizations might be applicable in a team sport setting, there were several attempts that examined the relationships between multi-dimensions of organizational justice and organizational behaviors, such as commitment to sports participation (Whisenant, Reference Whisenant2005), job satisfaction (Whisenant & Smucker, Reference Whisenant and Smucker2007, Reference Whisenant and Smucker2009), team performance (Whisenant & Jordan, Reference Whisenant and Jordan2006), and enjoyment in sports participation (Whisenant & Jordan, Reference Whisenant and Jordan2008).
Whisenant (Reference Whisenant2005) initially attempted to examine the effect of three dimensions of organizational justice on levels of high school student-athletes’ commitment to participate in sports. The researcher had found that the student-athletes’ levels of commitment to sport participation were associated with their perceptions of fairness across each of the three justice dimensions (distributive, procedural, and interactional justice). Of the three dimensions, interactional justice had the greatest influence on commitment. In other words, if coaches treat student-athletes fairly with respect and dignity, the athletes are more likely to continue participating in sports. In a follow-up study, Whisenant et al. (Reference Whisenant and Smucker2007, Reference Whisenant and Smucker2009) have attempted to examine whether high school coaches’ perceptions of fairness could affect their job satisfaction and whether high school student-athletes’ perceptions of fairness toward their coaches were impacted by their team performance level (winning/losing record; Whisenant & Jordan, Reference Whisenant and Jordan2006). The former two studies found that there was a strong relationship between three distinctive dimensions of justice and job satisfaction. The latter study specifically compared the perceptions of fairness held by student-athletes who were in a winning season to the perceptions of fairness held by those who were in a losing season. Of the three dimensions, only procedural justice affected team performance. Finally, Whisenant and Jordan (Reference Whisenant and Jordan2008) sought to determine whether student-athletes’ perceptions of fairness in interscholastic athletics was significantly different between the sports athletes enjoyed the most and the sports they enjoyed the least. In their study, the authors found that there was a significant difference between the two groups in each of the three dimensions of justice.
GROUP COHESION AS AN ENHANCEMENT FOR AFFECTIVE COMMITMENT
As mentioned previously, organizational justice is found to be one of the significant predictors of affective commitment to an organization (Andrew et al., Reference Bies and Moag2008). Another factor influencing affective commitment is group cohesion. Indeed, group cohesion (also known as team cohesion) has frequently been examined for its impacts on affective commitment and found to have a positive relationship with it (Wech et al., Reference Wech, Mossholder, Steel and Bennett1998; Friedkin, Reference Friedkin2004; Pillai & Williams, Reference Pillai and Williams2004). Likewise, the group cohesion has been extensively studied and is generally acknowledged as a crucial factor influencing the affective commitment in sport settings (Greenberg, Mark, & Lehman Reference Greenberg, Mark and Lehman1985; Jordan, Gillentine, & Hunt Reference Jordan, Gillentine and Hunt2004). In this regard, affective commitment in a team sport setting can be strengthen through both group cohesion and organizational justice. For example, being surrounded by supportive and friendly teammates would make the team environment much more desirable and enjoyable (i.e., highly cohesive team). Particularly, when this occurs in a sport team where there are fair rules, procedures, outcomes, treatments, and stronger feelings of affective commitment can be developed. On the other hand, even though a sport team’s environment is relatively fair in terms of outcomes, procedures, and treatment, affective commitment may not be strengthen if the team’s environment is not highly cohesive. Thus, the link between organizational justice and affective commitment can be different depending on the level of group cohesion (i.e., moderating role of group cohesion).
Prior research outside of the sport contexts has already demonstrated the moderating role of group cohesion in the relationship between organizational justice and affective commitment (Andrew et al., Reference Bies and Moag2008). It is important to note that there are two sources of fairness/justice in an organization context: (a) overall organization and (b) authority figure (Koivisto, Lipponen, & Platow, Reference Koivisto, Lipponen and Platow2013). Somewhat surprisingly, however, research in general organization contexts has primarily focused on organization itself as the source of justice. In other words, while most research conducted in general organizations viewed organizational justice as an institutionalized part of fairness (i.e., the organization as the major source of justice), very little attention was paid authority figures as the source of justice (van Knippenberg, De Cremer, & van Knippenberg, Reference van Knippenberg, De Cremer and van Knippenberg2007). In this regard, the moderating role of group cohesion found in general organization settings may not be applicable to a team sport setting where athletes’ perception of fairness is influenced by both the team as a whole (i.e., team policies, regulations) and the team coach (i.e., authority figure). According to van Knippenberg, De Cremer, and van Knippenberg (Reference van Knippenberg, De Cremer and van Knippenberg2007) and Koivisto, Lipponen, and Platow (Reference Koivisto, Lipponen and Platow2013), distributive justice and procedural justice seem to be treated as more systemic and institutionalized part of fairness (i.e., organization as the source of outcomes and procedures), while interactional justice is regarded as an essential part of leader fairness. In the present study, therefore, we included the above all three justices in order to test the moderating effect of group cohesion.
Since Jordan, Gillentine, and Hunt (Reference Jordan, Gillentine and Hunt2004) conceptually proposed that student-athletes’ perceptions of fairness are affected by their coaches and can influence team cohesion and team commitment, several studies attempted to examine the impacts of organizational justice and group cohesion on affective commitment in a sport setting. However, there were no studies examining how affective commitment is influenced when the two antecedent variables are taken into account together in a team sport setting. Hence, this study aimed to determine the moderating effect of group cohesion in the relationships between all three forms of organizational justice and affective commitment in a team sport setting. A better understanding of the relationship among organizational justice, group cohesion, and affective commitment in an athletic context may provide meaningful information for coaches to develop appropriate strategies to effectively manage athletes, particularly during training and practice sessions.
METHOD
Participants and Procedure
Participants were 282 college student-athletes from two large NCAA Football Bowl Subdivision (FBS) institutions in the Southeastern United States using a convenience sample. Student-athletes for this study were recruited from only team sport programs at the two universities because this study focused on a team sport setting rather than individual sport setting. In addition, the selection of the former setting is deemed more appropriate than the latter setting, considering the equity principle of distributive justice (i.e., comparison one’s inputs and outputs with those of others). The researchers and research assistants first contacted and met with personnel working in the athletic departments and obtained the contact information of coaches for all team sport programs. Once permission was given by the coaches, the research team met with the student-athletes and made appointments in times and places the coaches preferred. The research team administered the survey and encouraged the respondents to answer honestly. Participants were also assured confidentiality and anonymity. During the process of filling out the survey questionnaires, coaches in each of athletic programs were not present because items on the survey were about coaches’ behaviors and attitudes toward their athletes.
Out of the 282 returned questionnaires, a total of 253 complete and usable questionnaires were included in the data analyses The sample comprised of 61.3% males (n=155) and 38.7% females (n=98) with a mean age of 19 years and 8 months. Participants’ grades were broken down as follows: freshman (n=80), sophomore (n=40), junior (n=64), and senior (n=69). For types of sports, 22.9% (n=58) participated in football, 21.7% (n=55) in baseball, 15.0% (n=38) in basketball, 14.6% (n=37) in soccer, and 25.8% (n=65) in others.
Instruments
Organizational justice scale
For the measurement of organizational justice, there has been a debate whether organizational justice should be measured by a three-factor model (distributive, procedural, and interactional justice) or a four-factor model (i.e., interactional justice should be divided into interpersonal justice and informational justice). While the three-factor model has been traditionally used, the four-factor model proposed by Greenberg (Reference Greenberg1993) has started to receive great attention. In fact, Colquitt (Reference Colquitt2001) attempted to contrast the three-factor model with the four-factor model and found that the four-factor model was a statically better model. However, it is important to note that there is still some controversy regarding the four-factor model (Cropanzano, Rupp, Mohler, & Schminke, Reference Cropanzano, Rupp, Mohler and Schminke2001), as evidenced by a staunch defense of the three-factor model (Bies, Reference Cohen-Charash and Spector2001) and support for a three-factor model from meta-analytic studies (e.g., Cohen-Charash & Spector, Reference Cohen-Charash and Spector2001). Thus, in the present study the three-dimensional measure of organizational justice was utilized.
Specifically, 11 items were adopted from Colquitt (Reference Colquitt2001) and modified to measure athletes’ perceptions of fairness toward their coaches’ behaviors and attitudes through a panel of experts. In addition, the items were changed to a descriptive style from an interrogative style. The modification also involved changing the response format from a 5-point Likert-type to a 7-point Likert-type scale from 1 (strongly disagree) to 7 (strongly agree). This instrument was comprised of distributive justice (four items), procedural justice (three items), and interactional justice (four items). The reliability estimates (Cronbach’s α) have been proved in a previous study (Colquitt, Reference Colquitt2001).
Organizational commitment
Meyer and Allen’s (Reference Meyer and Allen1991) Organizational Commitment Questionnaire with three components (affective, continuance, and normative commitment) is among the best well-known and respected scale with usage in many previous studies. Of the three components, this study only measured athletes’ affective commitment toward their teams using six items. The respondents were asked to indicate their levels of agreement with each of the six items on a 7-point Likert-type scale ranging from 1 (strongly disagree) to 7 (strongly agree). In this scale, the word team was substituted for organization. Previous study demonstrated the reliability of α=0.85 (Meyer & Allen, Reference Meyer and Allen1997).
Group cohesion
To measure group cohesion, the researchers adopted and modified eight items from a questionnaire developed by Dobbins and Zaccaro (Reference Dobbins and Zaccaro1986). The reliability estimate of the scale (Cronbach’s α) was 0.80 (Chen, Tang, & Wang, Reference Chen, Tang and Wang2009). The participants were required to respond to a 7-point Likert-type scale from 1 (strongly disagree) to 7 (strongly agree).
Data Analysis
A confirmatory factor analysis (CFA) was first performed to check the validity of the organizational justice scale. Then, to test the relationships between three dimensions of organizational justice and affective commitment and the moderating effect of group cohesion between the two variables, three separate hierarchical multiple regression analyses were performed. The three dimensions of organizational justice served as independent variables, while affective commitment served as a dependent variable in each regression analysis. We first entered the demographic variables, such as gender and age, as control variables due to their impacts on affective commitment (Colbert & Kwon, Reference Colbert and Kwon2000; Lok & Crawford, Reference Lok and Crawford2001). In the second step, each of the three organizational justices was entered into three separate regression analyses. Then, group cohesion, a moderator, was entered in the third step in each of the three analyses. Lastly, the cross-product term (interaction term) which was generated by multiplying the independent variables and the moderator (three dimensions of justice×group cohesion) was entered in the final step of the three regression analyses. In order to minimize multicollinearity problems, independent (three organizational justices) and moderating variables (group cohesion) were centered before putting them in the regression analyses. To further examine the form of interaction, the researchers plotted two slopes for the final equation: one at 1 standard deviation below the mean of group cohesion and the other at 1 standard deviation above the mean.
RESULTS
Reliability, Validity, and Descriptive Statistics
The CFA results of the organizational justice scale with three dimensions (procedural, distributive, and interactional justice) indicated that the data did adequately fit to the measurement model based upon several fit indices (χ2/df=192.926/41=4.706, RMSEA=0.07, TLI=0.938, CFI=0.954, SRMR=0.037). To check discriminant validity for the organizational justice scale, correlations between three dimensions of justice were used. The results of the measurement model indicated that all correlations among three dimensions of organizational justice were less than the cut-off of 0.85 (see Table 1; Kline, Reference Kline2005). To further check discriminant validity for the scale, we compared the average variance extracted (AVE) of each construct (i.e., justice dimension) with the shared variance between one construct and any others (i.e., squared correlation). The results indicated that all AVE values for the justice dimensions were greater than the squared correlations between one dimension and any others (see Table 2; Fornell & Larcker, Reference Fornell and Larcker1981).
Means, standard deviations, correlations, and Cronbach’s 
$$\ralpha$$
s
Note. N=253. All correlations were statistically significant (p<.01).
Average variance extracted (AVE) values and squared correlations of the organizational justice scale
Note. Values on the diagonal are AVE for each dimension. Values below the diagonal are squared.
Means, standard deviations, correlations and Cronbach’s αs for each of five variables were reported in Table 1. The overall means for each variable ranged from a low for procedural justice (M=3.72) to a high for group cohesion (M=4.45). Standard deviations ranged from 0.99 (group cohesion) to 1.61 (affective commitment). In general, participants of the current study revealed moderate levels of all three fairness perceptions toward their coaches’ behaviors as well as affective commitment and group cohesion. Cronbach’s alpha coefficients for each variable were computed in order to verify internal consistency. All Cronbach’s α values were above the traditional cut-off of 0.70 ranging from 0.92 (distributive justice) to 0.95 (affective commitment; Nunnally & Bernstein Reference Nunnally and Bernstein1994).
Assumption of Regression
Before performing the main regression analyses, several assumptions were checked. First, normality and homoscedasticity were checked by examining residuals. The normal probability plots in the three regression analyses were close to the straight diagonal, which indicated a normal distribution of the residuals. The partial regression plots showed random scatters of the residuals, which reflected no violation of homoscedasticity assumption. Second, influential data point assumption was checked by examining Cook’s D. The Cook’s D values across three regression analyses were <1.0 (ranged from 0.00 to 0.041). Lastly, multicollinearity assumption was checked by examining variance inflation factor (VIF). There were no VIF larger than 10 across three regressions (ranged from 1.02 to 3.64), which indicated no violation of the multicollinearity.
Relationship between Organizational Justice and Affective Commitment
The first purpose of the current study was to examine the relationship between three dimensions of organizational justice (distributive, procedural, and interactional justices) and affective commitment among collegiate student-athletes. Results for this purpose were provided in Step 2 of the three separate hierarchical regression analyses presented in Table 3. These results indicated that the relationship between the two variables after controlling for age and gender is positively significant for all dimensions of organizational justice: (a) procedural justice (β=0.66), (b) distributive justice (β=0.78), and (c) interactional justice (β=0.89). More specifically, beyond the variance contributed by gender and age, procedural, distributive, and interactional justices explained ~39% (∆R 2=0.39), 50% (∆R 2=0.50), and 50% (∆R 2=0.50) of variance in affective commitment, respectively.
Hierarchical multiple regression results for affective commitment
Note. Betas are standardized regression coefficients.
*p<.05, **p<.01, ***p<.001.
Moderating Effect of Group Cohesion
To examine the moderating effect of group cohesion on the relationship between organizational justice and affective commitment, we created three interaction terms using three dimensions of justice and group cohesion and entered each interaction term into Step 4 presented in Table 3. These results indicated that while interactions between distributive justice and group cohesion (β=0.06, p<.01), and between interactional justice and group cohesion (β=0.07, p<.01) were significantly and positively associated with affective commitment, the interaction effect between procedural justice and group cohesion was not significant (β=0.02, p>.05). More specifically, beyond the explained variance of gender, age, the two significant dimensions of justice, and group cohesion, the distributive justice×group cohesion (R 2=0.91, p<.01; ∆R 2=0.02) and the interactional justice×group cohesion interactions (R 2=0.90, p<.01; ∆R 2=0.02) explained significant incremental portions of variance in affective commitment. Even though the increment in R 2 reflects a very small amount of variance, it is within the range for moderator effects typically found in non-experimental studies (Champoux & Peters, Reference Cropanzano, Rupp, Mohler and Schminke1987).
To further examine whether the interaction effects exist, the researchers graphed them by plotting two slopes: one at 1 standard deviation below the mean and the other at 1 standard deviation above the mean. In addition to the two significant interaction graphs, the researchers plotted non-significant interaction graph for procedural justice, but did not interpret it. The graphs were displayed in Figure 1. The interaction graphs revealed that the positive relationships between distributive justice and affective commitment and between interactional justice and affective commitment were significantly stronger among student-athletes reporting high levels of group cohesion than among those reporting low levels of group cohesion.
Interaction graphs between three dimensions of organizational justice and group cohesion on affective commitment.
Note. AC=affective commitment; GC=group cohesion; DJ=distributive justice; PJ=procedural justice; IAJ=interactional justice
DISCUSSION
In an organizational context, employees typically perceive fairness based on the following three criteria: (a) outcomes they receive (distributive justice), (b) process or procedures used to determine the outcomes (procedural justice), and (c) the extent to which they are treated in good manner by authority figures and/or the extent to which information is honestly provided by authority figures (interactional justice; Cropanzano & Greenberg, Reference Cropanzano and Greenberg1997). Considering that sport teams have many similar characteristics to other types of organizations (Chelladurai, Reference Dittmore, Mahony, Andrew and Hums2001), it is plausible that athletes in a team sport context may use these criteria to make fair judgments toward their coaches. Athletes’ perceptions of fairness often affect their behaviors and attitudes in an athletic team, which would in turn influence organizational commitment and team unity (i.e., group cohesion; Jordan, Gillentine, & Hunt Reference Jordan, Gillentine and Hunt2004). Thus, this study attempted to identify the relationship among the three forms of organizational justice, group cohesion (moderator), and affective commitment.
When it comes to the overall levels of organizational justice held by collegiate student-athletes, the participants revealed that they reported the highest mean score on distributive justice (M=4.24) out of the three dimensions of justice. One plausible explanation of this result is that since winning or losing is often viewed as the most important end result (i.e., outcome) in an athletic context. Student-athletes’ wish to win is superior than their perceptions of fairness of how coaches make their decisions and how the coaches treat student-athletes (i.e., procedural, interactional justices; Whisenant, Reference Whisenant2005). Even when a coach’s decision regarding a certain outcome (e.g., the amount of playing time in competition) might be perceived as unfair by athletes, the athletes may care less about the fairness of their playing time as a reward than a team’s win or loss. This may account for the reasons why the level of distributive justice was relatively higher than the other two forms of justice.
With regard to the relationship between organizational justice and affective commitment, the results showed that all three dimensions of justice were positively and significantly related to affective commitment after controlling for age and gender. These results were consistent with two previous views found one outside the sport context (Andrew et al., Reference Bies and Moag2008) and the other within the sport context (Jordan, Gillentine, & Hunt, Reference Jordan, Gillentine and Hunt2004). As suggested by the former work, when individuals are fairly treated and rewarded by their supervisors, they are more likely to show positive attitudes toward the supervisors and then make stronger efforts to attach with the organization. Likewise, it is possible that athletes are more likely to be attached or committed to the team when they are fairly treated and rewarded by their coaches. This can be also supported by the essential tenet of social exchange theory which states that when organizations provide fair and just environment, individuals would be more likely to have affective commitment to their organizations in exchange for the fair environment (Rupp & Cropanzano, Reference Rupp and Cropanzano2002). The findings are further partially supported by recent meta-analytic reviews (Li & Cropanzano, Reference Li and Cropanzano2009; Shao et al., Reference Shao, Rupp, Skarlicki and Jones2013).
The above relationships between organizational justice and affective commitment in a team sport setting indicate coaches need to develop strategies intended to improve athletes’ perceptions of fairness. The strategies may be effective even when athletes are disappointed with the rewards/outcomes they receive. To be sure, athletes who do not receive the reward they desire would want more. However, they are often affectively committed to their team if the procedures used to come up with the desired rewards are perceived as fair (Cropanzano, Bowen, & Gilliland, Reference Cropanzano, Bowen and Gilliland2007). Indeed, McFarlin and Sweeney (Reference McFarlin and Sweeney1992) found that even though individuals were not fairly rewarded, they were more likely to have high levels of affective commitment to their organization as long as procedural justice was high. This can be applied to interactional justice because it is oftentimes considered a part of procedural justice (Jost & Kay, Reference Jost and Kay2009). As such, it is critical for coaches to develop proper managerial strategies regarding procedural and/or interactional justices in a team sport setting. For example, if an athlete who practices a lot fails to be selected as a starter, he/she would perceive unfairness regarding the outcome. However, the athlete may accept the unfair outcome if a coach thoroughly and honestly explains the processes and procedures of selecting starting members with dignity and respect.
Further managerial strategies can be developed pertaining to Leventhal’s six procedural rules (1981). The six rules are as follows: (a) consistency (e.g., coaches must be consistent in their application of team rules and policies across all athletes), (b) bias suppression (e.g., coaches must be neutral), (c) accuracy of information (e.g., coaches must use accurate and reliable information when making decisions), (d) correction (e.g., procedures used to correct wrong decisions must be viewed by athletes), (e) representation (e.g., coaches must consider the interests of all team members), and (f) ethicality (e.g., coaches’ decision-making process must follow the standards of ethics and morality). Therefore, coaches may need to make fair decisions by utilizing these six rules in order to enhance athletes’ affective commitment to a team. In turn, this may result in reducing athletes’ dropout rates.
With regard to the moderating effect of group cohesion on the relationships between the three dimensions of organizational justice and affective commitment, participants’ perception of group cohesion strengthened the distributive justice–affective commitment and the interactional justice–affective commitment relationships. The findings were partially different from previous work that showed group cohesion positively moderated not only the distributive- and interactional-, but also the procedural–affective commitment relationships (Andrew et al., Reference Bies and Moag2008). Specifically, the findings of the current study indicated that distributive justice and interactional justice had stronger relationships with affective commitment among student-athletes reporting high levels of group cohesion than their counterparts. With these findings, it is implied that coaches should emphasize not only fair treatment and rewarding for athletes’ efforts, performances, and athletic skills, but also team unity among team members, which in turn would enhance attachment/loyalty to the team and may eventually improve the team’s athletic performance. Further, social interactions between student-athletes may enable their distributive justice perceptions to be influenced as they learn of others’ treatment. In other words, the more cohesive an athletic team is perceived to be, the more likely athletes will develop shared perceptions of the justice climate as a result of these social interactions. Thus, coaches should also develop a strategy for encouraging social interactions among athletes in a team sport setting.
Although group cohesion was not a significant moderator of the procedural justice–affective relationship, the relationship between the procedural justice and affective commitment was positively significant, regardless of the levels of group cohesion. This finding showed that procedural justice may indeed be critically essential in a team sport setting. According to Greenberg (Reference Greenberg1993), procedural justice is often regarded as the most important influential dimension of justice on one’s behaviors and attitudes toward organization (e.g., affective commitment). As such, it may be difficult for any positive aspects of the team environment (e.g., group cohesion) to supplement for poor procedural justice.
The results of the current study offer a unique perspective on the relationships among organizational justice, group cohesion, and affective commitment in a team sport setting. In addition, while previous research on organizational justice in a sport context primarily focused on distributive justice, the current study measured the three dimensions of organizational justice which is the most widely used approach (Cropanzano et al., Reference Cropanzano, Rupp, Mohler and Schminke2001). In spite of the above strengths, some limitations should be addressed.
First, even though age and gender have been utilized as control variables, there may still be other important variables that can influence student-athletes’ perceptions of fairness or affective commitment. For example, athletes’ perceptions of fairness may differ depending on their levels of grade and gender of coaches (Whisenant, Reference Whisenant2005). Second, student-athletes’ perceptions of fairness were measured on the basis of coaches’ behaviors and actions. However, their perceptions of fairness are likely to be influenced by other authority figures such as the assistant coaches within the context of teams, and also athletic directors outside the teams. Third, the sample for the current study was recruited from only two FBS institutions, thereby limiting the generalizability of the results. Fourth, to measure the level of fairness perception among student-athletes, the study employed the organizational justice scale developed by Colquitt (Reference Colquitt2001) outside the sport context. Thus, future research may also need to consider developing an appropriate measurement tool for a team sport setting, given that structural and environmental differences between a general organization and a team sport setting may exist. Even though there are some organizational justice scales designed for a sport setting (Hums & Chelladurai, Reference Hums and Chelladurai1994a; Mahony, Hums, & Riemer, Reference Mahony, Hums and Riemer2005), these scales are fully designed to measure perceptions of fairness regarding resource allocation. Last but not least, sport is usually regarded as an outcome-obsessed context so it is quite plausible that athletes’ perceptions of fairness may not be an important component in a team sport setting. Thus, there may be a difference in the organizational justice-group cohesion–affective commitment relationships based on team’s success factor. Considering the limitations listed above, more studies should be conducted with regards to this line of research in the future.
In conclusion, this study extended the organizational justice literature by empirically identifying the relationships among organizational justice, group cohesion (moderator), and affective commitment among collegiate student-athletes in a team sport setting. All three dimensions of justice were positively and significantly related to affective commitment. Group cohesion moderated the distributive justice–affective commitment and the interactional justice–affective commitment relationships out of the three dimensions of justice. Specifically, the above two relationships were significantly stronger among student-athletes reporting high levels of group cohesion than those who reported low levels of group cohesion. Managerial strategies by coaches to improve athletes’ perceptions of fairness may increase their affective commitment.
Acknowledgment
The authors would like to thank to two anonymous reviewers for their meaningful critiques and suggestions on an earlier form of this article. This was supported by a research grant from the University of Arkansas at Little Rock, USA.
