Seeking Sustainable Development in Teams: Towards Improving Team Commitment through Person-Group Fit

: This study explores the sustainability of work groups in the industry by enhancing team commitment via person-group ﬁt and its boundary conditions. Speciﬁcally, we examine the relationship between person-group (PG ﬁt) and team commitment, introducing person-organization (PO) ﬁt and person-supervisor (PS) ﬁt as moderators. Using data collected from a Korean company, this study reports that a three-way interaction between PG ﬁt, PS ﬁt, and PO ﬁt predicts an employee’s commitment to their team. The results highlight that PS ﬁt can compensate for PG ﬁt when the level of PG ﬁt is low by stimulating team members to commit to the work team. Likewise, PG ﬁt also plays a compensatory role for inducing team commitment when confronting poor PS ﬁt. The results reveal that as long as PG ﬁt is high, PO ﬁt is more important than PS ﬁt in enhancing team commitment. However, when PG ﬁt is low, high PS ﬁt with low PO ﬁt produces a higher level of commitment to the team than low PS ﬁt with high PO ﬁt.


Introduction
The increasing adoption of team-based structures by modern organizations enhances the influence of team members on fellow employees, as employees spend more time and interact more frequently with their co-workers than with their supervisors [1,2]. Researchers have noticed the importance of co-workers in the workplace and have begun to pay attention to the role of co-workers as the social environment that influences workers' attitudes and behaviors. Despite recent research on person-group (PG) fit in the workplace and its outcomes at both group and individual levels [3][4][5], PG fit is still one of the most underdeveloped areas in person-environment (PE) fit research.
Although PG fit mainly functions in a team context, the relationship of employees to other targets or actors (e.g., their supervisor and the organization) is also critical for enhancing individual outcomes such as work satisfaction, organizational citizenship behavior, and task performance [3,6]. Team dynamics, such as high-quality person-supervisor (PS) fit, often lessens differentiation or segmentation in a team and facilitates integration between team members [7]. PS fit or leader-follower fit has been extensively discussed in the leadership literature, particularly from leader-member exchange (LMX) perspectives [8][9][10][11][12]. Also, the level of PO fit makes team members either commit themselves to their job on the team, or conversely, engenders deviant behavior that might damage both individual and group outcomes. In this respect, the effect of PG fit on individual outcomes, such as team commitment and task performance, seems to be moderated by PO fit and PS fit [13,14].
However, the majority of earlier fit studies have focused on a single type of PE fit, and investigated its effect on individual or group outcomes [3], even though the concept of PE fit has been differentiated into separate aspects [5,15,16], including supplementary fit and complementary fit [17][18][19]. To fill this gap, we extend the prior studies by examining the various types of PE fit. Previous research that dealt with multiple types of fit reported that each type influences individual outcomes such as job satisfaction, organizational commitment, and turnover intention independently and uniquely [6,20,21]. Although various types of PE fit have been found to influence individual and workgroup outcomes, little research has been conducted to explain how strongly and under what circumstances each type affects the outcomes. It was not until the early 2000s that a comprehensive investigation of the multiple types of PE fit and how they affect work-related attitudes and behaviors simultaneously was attempted [14,[22][23][24][25][26]. Researchers have tried to examine whether the effects of multiple types are simply additive or whether there is a multiplicative or interactive configuration in addition to their main effects [13,14].
In the literature on fit-outcome relationships, each type of PE fit has been shown to have variable effects on individual outcomes [14]. For example, a high level of PG fit may not necessarily contribute to the achievement of organizational goals and values by raising PO fit, or it may even have the opposite effect, so that a strong commitment to the group weakens organizational commitment [27]. Also, the relative impact of the relationship between PG fit and PO fit on workgroup outcomes may vary according to the level of PS fit. There may even be conflicts between various types of PE fit due to the different targets of commitment: the group, the supervisor, and the organization. Each type of PE fit may reinforce or undermine the others in producing work-related outcomes.
To respond to this question, we need to integrate PG fit as a subdimension of PE fit with PO fit and PS fit as contextual factors in a single conceptual framework. In this study, we examine the effects of PG fit, PO fit, and PS fit on team commitment simultaneously. We also extend the fit literature by investigating the effect of fit interaction, specifically a three-way interaction between subtypes of PE fit, on individual outcomes.
This study contributes to the existing knowledge on the relationship between PE fit and organizational outcomes, showing ways in which a fit dimension either complements the positive effect of another fit dimension or compensates for its negative effect (i.e., a lower level of fit) on team commitment. In previous research, the additive effects of PE fit dimensions have been widely proposed and verified to a considerable extent [13,23], but the compensatory effect of PE fit for a misfit or poor fit in its subdimension has rarely been discussed and thus, we still do not know much about it. Salthouse [28] (p. 22) stated that "as deficits appear, changes may occur in the nature of activities in which one engages, or in the level of expectation in the same activities, in order to minimize the consequences of those deficits." We presume that employees who feel a lower fit in one dimension may seek psychological compensation in another dimension of PE fit [28,29]. Thus, we attempt to fill the gap left in existing multi-dimensional fit research.

Person-Environment (PE) Fit and Multiple Targets of Commitment
PE fit has been conceptualized in various ways. The most frequently used dimensions in organizations are person-job (PJ) fit, person-group (PG) fit, and person-organization (PO) fit, which correspond to the individual, group, and organizational levels of PE fit, respectively. Furthermore, in a dyadic relationship, person-supervisor (PS) fit, which has been extensively researched, emphasizes the match between supervisors and subordinates [3,30]. PO fit is defined as the "compatibility between individuals and the organization that occurs when (a) at least one party provides what the other needs, and/or (b) both share similar fundamental characteristics" [27] (pp. [4][5]. PG fit refers to "the interpersonal compatibility between individuals and their work group" [27] (p. 7). Among the various dimensions of PE fit, we focus on PG fit, PO fit, and PS fit, which refer to the similarities between the values, goals, and/or personalities of employees and those of their group, organization, and supervisor. We focus on the organization (PO fit) and the co-workers (PG fit) and supervisors (PS fit) that comprise the organization.
Research has shown that PE fit enhances both individual and group-level outcomes such as job satisfaction, organizational commitment, and turnover intention. Each element of PE fit is differentially associated with individual attitudes and behaviors in group and organizational settings. For example, PO fit is more likely to be associated with organizational commitment, PS fit with satisfaction with supervisors, and employees with a high level of PG fit should be satisfied with their coworkers [3]. Furthermore, in the commitment literature, multiple commitments that employees may make have been extensively discussed [31][32][33]. It has been shown that commitment to one target (or referent) is predictive of the outcomes specifically focused on that particular target. For example, organizational commitment explains outcomes at the organization level such as intent to quit and the commitment to a supervisor predicts supervisor-related outcomes such as citizenship behavior towards the supervisor [34].
Individual employees are expected to commit to the workgroup or team, supervisors, and the organization they work for. However, commitment to different targets sometimes does not go together. In this vein, Reichers (1985) addressed the need to re-conceptualize "organizational commitment" to reflect the multifaceted characteristics of the organization, rather than regarding it as "a monolithic, undifferentiated entity" [33]. He argued that organizations should be viewed "as composites of coalitions and constituencies, each of which espouses a unique set of goals and values that may be in conflict with the goals and values of other organizational groups" [33] (pp. 469-470). Hunt and Morgan (1994) found that commitment to work group influences organizational outcomes independently from organizational commitment [32]. In this study, we examine the interaction effect of the three types of PE fit (PG, PO, and PS fit) on team commitment.

PG Fit and Team Commitment
PG fit occurs when individuals feel a sense of compatibility with other members of their workgroup [30,35]. In most fit studies, the criteria of interest are individual-level consequences such as satisfaction, adjustment, stress reduction, turnover, or performance [25]. The three most commonly studied outcomes of PG fit are job satisfaction, organizational commitment, and intention to quit.
However, various conceptualizations of fit differentially predict particular dependent variables. Bretz and his colleagues [36,37] provide a good example of investigating multiple conceptions of fit, but their work could be extended to disentangle the effects of each type of fit on the dependent variables [27]. In particular, PG fit on values and goals may be expected to have a strong influence on affective outcomes because they both involve attitudes, but it may have a smaller effect on individual performance. The opposite might be the case for complementary fit on knowledge, skills, and abilities (KSAs), which might have a strong influence on on-the-job performance.
According to similarity-attraction and social identity theory [38,39], individuals are attracted to other individuals or organizations that have similar values, goals, and characteristics. Thus, when employees are congruent with their team in terms of values and KSAs, they feel more committed to their team. Researchers have found positive effects of PG fit on the attitudes and behaviors of individual workers [3,24]. For example, a meta-analysis by Kristof-Brown et al. (2005) showed that PG fit was significantly related to job satisfaction, organizational commitment, and intention to quit. PG fit has a strong relationship with group cohesion [3]. Seong and Kristof-Brown (2012) examined the differential effect of PG fit on team commitment, vocal behavior, and knowledge sharing [5]. They found a differential effect of PG fit on its corresponding outcomes. PG fit research suggests that when employees' values are congruent with those of the team, greater satisfaction and commitment can result as the team fulfills the needs or desires of the individual [5,27,40].

Interactive Effects of the Multiple Types of Fit on Work-Related Outcomes
Some scholars have introduced perceived fits as boundary conditions [41]. Even though each type of perceived fit is positively associated with work-related outcomes, different types of fit may not be additive but interactive in affecting the outcomes. Sometimes, they go in opposite directions. A deficiency in one type of fit may be compensated by another type of fit, and a high level of fit in one dimension may not contribute to outcomes when perceived fit in another dimension is also high [26]. Vogel and Feldman (2009) found that there are significant interactive effects of PG fit with PJ fit on performance and organizational citizenship behavior (OCB), but in groups with high PG fit, PJ fit is negatively related to citizenship behaviors, and hardly affects performance [26]. Their explanation for the negative relationship between PJ fit and OCB when PG fit is high is that there may be a less urgent need for team members to be involved in citizenship behaviors in high-compared to low-PG-fit groups. In the same vein, Resick et al. (2007) suggested that two types of person-job (PJ) fit such as demands-abilities (DA) fit and needs-supplies (NS) fit are likely to moderate relationships between PO fit and work-related attitudes [41]. Resick et al. (2007) reported that the relationship between PO fit and satisfaction is stronger when NS job fit is low and weaker when NS job fit is high, and that PO fit is also more strongly related to job choice decisions when DA job fit is low than when DA job fit is high [41].

PO Fit as a Moderator
Researchers have found positive effects of PE fit, especially PO fit and PS fit, on attitudes and behaviors [3,24]. For example, Kristof-Brown et al. (2005) showed that both PO fit and PS fit are significantly related to organizational commitment [3]. Most researchers have focused on the effects of PO fit [27,42,43]. PO fit is different from PG fit [27,44]. Support for the distinction between PO fit and PG fit is found in the literature, which suggests that sub-organizational units, such as groups, may have different norms and values from the organization in which they are contained [45]. Thus, "the degree of fit between an individual and group may differ radically from the fit between the person and the organization" [27] (p. 8).
Although there is little research examining direct and interactive relationships between PO fit and PG fit, we can deduce these relationships from social identity theory [46,47]. When employees feel committed to the organization, they will be more committed to their team because employees regard both organization and team as one target. There has also been a view that regards organizational commitment as a more comprehensive construct than other types of commitment to the various individuals and groups with whom an employee is associated in the organization [32]. Hunt and Morgan (1994) reported that "global organizational commitment" plays a key mediating role between "constituency-specific commitments" and important organizational outcomes [32]. However, organizational commitment cannot adequately explain an employee's commitment to the workplace since commitment is multidimensional and complex [33,48]. The organization and the team are collapsed into one broad target such as "our group" or "our organization." PO fit research suggests that when employees' values are congruent with those of the organization, greater satisfaction and commitment may result as the organization fulfills the needs or desires of the individual [27,[49][50][51].

PS Fit as a Moderator
Over the years, many studies of PE fit have examined the positive relationship between PS fit and individual outcomes [52][53][54][55][56]. As the team context and work itself are more complicated, we may deduce that there will be a compound effect of PS fit and PG fit on group outcomes. This is an area that has not previously been fully examined. However, a similar logic to that used in relation to the relationship between PO fit and PG fit may also apply to the PS fit/PG fit relationship. The leader-follower relationship and the interaction between superiors and subordinates determine the effectiveness and efficiency of the employees' task performance [8,56]. Meglino, Ravlin, and Adkins (1989) found that value congruence between employees and their supervisors is positively related to satisfaction and commitment on the part of subordinates [53]. Other studies have shown a positive relationship between personality congruence of the leader and the follower and the follower's performance [52], as well as a relationship between the value and attitude congruence of the leader and the follower and the follower's positive outcomes [54,55].
Since the findings from studies on the effect of subordinate-supervisor similarity in work-related values have shown fairly consistent results, it can be deduced that employees who are congruent with their supervisor with respect to values, goals, and personalities will be more committed to their team than those who are not. As noted above, both similarity-attraction theory and social identity theory reinforce the fact that individuals who have similar values, goals, and characteristics are attracted to other individuals on the team or in the organization. Employees whose values, goals, and personalities are congruent with those of their supervisor are more likely to commit themselves to the supervisor and trust him/her. Thus, we expect positive effects of PS fit on team commitment.
LMX theory has focused on the reciprocal relationship between the leader and followers within the group, on the assumption that the leader does not treat each follower the same way and interacts differently with followers [8,57]. Thus, the follower's work attitudes are dependent upon the leader's manner with each subordinate. "As leaders initiate social exchanges by bestowing favorable treatment upon certain members, members in turn feel obliged to work harder to benefit the leader as a means of reciprocation" [58] (p. 1097). In this respect, stronger relationships between employees and their supervisor lead to higher levels of team commitment.
However, employees confronted with highly abusive supervision (e.g., PS misfit) may have a stronger cohesion with their work group (e.g., PG fit) when organizational identification (e.g., PO fit) is high [59], and PG fit eventually enhances team commitment [3,5]. In the leadership literature, it has been widely discussed that negative (e.g., abusive or destructive) leadership adversely affects positive outcomes such as organizational commitment, job satisfaction, and performance [60]. However, destructive leadership (i.e., the lack of PS fit) may constitute a work environment where employees attempt to find an exit solution through integration with their co-workers (i.e., a higher level of PG fit). The harsh group environment caused by team leaders may induce solidarity between team members that raises team commitment. As psychological compensation theory addresses [28,29], employees who confront deficits in a fit dimension may seek compensation in another fit dimension to reduce the deficits incurred.

Three-Way Interactive Effects of PG Fit, PO Fit, and PS Fit on Team Commitment
The considerations set out above suggest that PG fit interacts with PS fit to affect team commitment most strongly when PO fit is high. We propose that team commitment will be most strongly related to PG fit when PO fit is high and PS fit is high. In these circumstances, team members are likely to focus on team values and goals as a source of team identification and are also likely to be oriented toward integrating the best of the team's and the organization's goals. In situations of high PO fit, team commitment is more likely to be facilitated by a sense of unity of direction, so that similar activities in the organization can move toward a common target, either as a team or as an organization. In the same way, we expect similar outcomes in situations of high PS fit. Therefore, PG fit, when combined with PS fit, may bring about positive outcomes such as commitment to the team and other team members.
We proposed that when both PO fit and PS fit are high, PG fit will enhance team commitment most strongly. Alternatively, we may ask in what way PG fit affects team commitment in different conditions of PO fit and PS fit; for example, "high PO fit and low PS fit" or "low PO fit and high PS fit". As discussed above, PO fit constitutes the basis for a pleasant environment in which PG fit operates most effectively. A high degree of PO fit appears to be a critical factor for employees' commitment to the organization and to their group. Thus, we present the following hypotheses: Hypothesis 1 (H1). PG fit, PO fit, and PS fit will interact to affect employees' team commitment such that PG fit has a stronger positive relationship with team commitment when both PO fit and PS fit are high than when PO fit is high but PS fit is low.
Hypothesis 2 (H2). PG fit, PO fit, and PS fit will interact to affect employees' team commitment such that PG fit has a stronger positive relationship with team commitment when PO fit is high but PS fit is low than when PO fit is low but PS fit is high.

Data and Sample
To test a series of working hypotheses, we collected data from an electrical company in South Korea as part of a larger project conducted for both policy and research purposes. The company is a team-based organization with 29 teams grouped in eight sections: general affairs/personnel, strategy/planning, finance/accounting, sales, production, R&D, purchasing, and quality control. For this survey, employees who work in the main office located in Seoul participated as respondents.
Questionnaires were distributed to 200 employees and 163 employees responded (response rate of 81.5%). Of the 163 employees, five were eventually excluded from analysis due to the absence of relevant individual-level data. Consequently, the useful sample was reduced to 158. The sample included both full-time and part-time employees. The questionnaires were sent to employees via email and returned to one of the stewards or the head of the planning section in sealed envelopes, and then delivered directly to the researchers. Eighty-five percent of the respondents were male, and 30% of the respondents were in their twenties, 54% in their thirties, and 16% were aged 40 or older. The respondents' average length of service in the organization was 2.8 years (SD = 1.5 years).

Dependent Variable
Team commitment. We constructed a four-item scale adapted from the team commitment measure of Van der Vegt and Bunderson [61]. The sample item included the following: "I feel emotionally attached to my team." Each item was measured on a five-point scale from 1 ("strongly disagree") to 5 ("strongly agree").

Independent Variables
Perceived person-environment fit. Perceived person-organization fit was measured using a four-item scale developed by Saks and Ashforth [21]. The items were explicitly designed to measure perceptions of fit and to capture specific aspects of PO fit such as goals and values. The sample item for the PO fit scale is the following: "To what extent are the values of the organization similar to your own values?" The same set of questions used for measuring the person-organization (PO) fit was asked to capture the person-supervisor (PS) fit and the person-co-worker (PG) fit by simply changing the word "organization" to "supervisor" and "team member," respectively. An example item for the perceived PS fit is: "To what extent are your supervisor's values similar to your own values?" An example item for the perceived PG fit was: "To what extent are your co-workers' values similar to your own values?"

Control Variables
As in other studies of PE fit and commitment [20,54,62], gender, age, rank, and organizational tenure were used as control variables in all analyses. Gender was coded 1 for male and 0 for female. Age was measured on a four-point scale from 1 ("20-29 years") to 4 ("50-59 years"). Rank was coded from 1 ("rank-and-file employee") to 6 ("executive"). Organizational tenure was measured from 1 ("less than 1 year") to 5 ("more than 10 years"). The correlation matrix for all the variables used in this study is shown in Table 2.

Confirmatory Factor Analysis
We first performed confirmatory factor analyses (CFAs) to examine the distinctiveness of the scales for PG fit, PO fit, PS fit, and team commitment using SPSS Amos Version 23 (see Table 1). Overall, the results show that the hypothesized four-factor model provides substantially improved fit over the relevant alternative models.  Table 2 presents the means, standard deviations, and correlations among the study variables. The three types of PE fit variables were strongly correlated, although they were empirically distinguishable, as was demonstrated by the CFA results.

Examination of Common Method Bias
Cross-sectional research often raises the question of common method variance (CMV) in the sense that correlations between constructs can be inflated. To check for the presence of CMV that might seriously undermine causal inferences, we followed Lindell and Whitney's (2001) suggestions to use a partial correlation analysis [63]. To this end, we introduced a scale theoretically unrelated to at least one of the other variables as the method variance (MV) marker variable. The marker variable needs to be a highly reliable multi-item scale. In this paper, we chose organizational culture as the MV marker variable that is theoretically unrelated to team commitment as the criterion variable. Quinn and Rohrbaugh (1983) proposed a competing value framework (CVF) measured with four dimensions: the management of employees, organizational glue, strategic emphasis, and the criterion of success [64]. The respondents were asked to give scores (from 0 to 100) to each of the four statements in each dimension describing the current team climate. The sum of scores assigned to each statement should add up to 100. One of the four statements indicates a human-oriented culture, as opposed to a competitive or goal-oriented culture. The alpha reliability coefficient of the organizational scale was 0.80. This scale was used as the MV for organizational culture.
Since the correlation between the marker and criterion variables is not statistically significant (r = 0.11), the marker can be used as the estimate of CMV, as shown in Table 3. CVF, as the theoretically irrelevant predictor, is not correlated with the predictor (r = 0.04, p = n.s.), while PG fit as the theoretically relevant predictor has a high correlation with team commitment as the criterion variable (r = 0.65, p < 0.001). After controlling for CMV, the correlation of the predictor and the criterion variable is still high (r = 0.67, p < 0.001), which implies that the CMV did not seriously distort the true correlation.

Hypothesis Testing
Model 4 in Table 4 shows that the three-way interaction between the three types of PE fit was related to team commitment (β = −0.17, p < 0.05), which, overall, supports Hypotheses 1 and 2. We plotted the three-way interaction in Figure 1, following the widely used procedure specified by Aiken and West [65]. As Figure 1 shows, PG fit had the highest level of positive relationship with team commitment when both PO fit and PS fit were high. To investigate this further, we calculated the unbiased beta weights for each slope along with the t-test for each pairwise comparison [66]. The results are reported in Table 5. The slope difference between "high PO fit and high PS fit" (3) and "high PO fit and low PS fit" (4) (Hypothesis 1) is statistically significant at the 1% level (t = 2.66, p < 0.01). However, the difference between "high PO fit and low PS fit" (4) and "low PO fit and high PS fit" (1) is statistically not significant at the 5% level even though the slope of (4) is steeper than (1) (t = 1.61, p = n.s., for two-tailed test), not supporting Hypothesis 2. Taken together, these results imply that high PG fit brings about a stronger level of team commitment when PO fit is high but PS fit is low than when PS fit is high but PO fit is low. As shown in Figure 1, however, when PG fit is low, the pattern turns out to be the reverse. High PO fit with low PS fit produces a lower level of team commitment than low PO fit with high PS fit.     (4) 1.61 (2) and (3) 1.80 + (2) and (4) 1.61 (3) and (4) 2.66 ** + p < 0.10, * p < 0.05, ** p < 0.01, *** p < 0.001.

Discussion
The objective of our research was to investigate the relationship between perceived PG fit and team commitment. To explain this causal relationship, we introduced other types of PE fit, namely PO fit and PS fit, as moderators, placing PG fit as the main category of PE fit. The three types of PE fit interacted to affect team commitment in a significant way.
The results consistently show that the employees who perceive high levels of PG fit, PO fit, and PS fit are very committed to the team, whereas those who perceive low levels of PG fit, PO fit, and PS fit exhibit the lowest level of team commitment. As long as PG fit is high, PO fit is more important than PS fit in enhancing team commitment. However, when PG fit is low, high PS fit with low PO fit produces a higher level of commitment to the team than low PS fit with high PO fit. Stated in this way, PS fit appears to be complementary to PG fit, which implies that when fit between team members is not appropriate, PS fit may play a compensatory role in encouraging them to commit to the workgroup. Conversely, PG fit becomes more important for heightening team commitment when PO fit is high but PS fit is low, than when PO fit is low but PS fit is high. Thus, a high level of PG fit can also compensate for a low level of PS fit.

Theoretical Implications
This study strengthened the existing literature of fit by elaborating a multidimensional theory of PE fit [23]. Little research has examined the interaction, specifically the three-way interaction, between the effects of the three components of PE fit on individual outcomes such as team commitment. By incorporating PG fit, PS fit, and PO fit in terms of values, personality and goal similarity, the findings of this study explain group outcomes, demonstrating the three-way interaction between the dimensions of PE fit.
In general, previous research has suggested a significant association between a "good fit" and positive outcomes for the individual and the organization [3,67]. However, one dimension of PE fit may not go together with another in the same way, and sometimes they even go in the opposite direction, showing a good fit in one dimension, but a poor fit in another [23]. Thus, this study discovered that the effects of fit with multiple targets are more likely to be interactive beyond simple additive effects in predicting organizational outcomes [25].
This study also combines social identity theory and LMX theory in the workplace, and shows that PS fit is related to positive outcomes for leaders, team members, and the organization as a whole. When PG fit is high, high levels of PS fit will lessen the turnover intention of team members and other deviant behavior through team commitment and also enhance the participation of employees on the team and job involvement. In this respect, this study makes full use of social identity theory and strengthens LMX theory by examining the impact of the three types of PE fit on team commitment.
This study highlighted the salience of PG fit for team commitment. Congruence and integration among team members become important for team performance, no matter when and how they get recruited, and their commitment to the team and fellow workers will be heightened as long as they are well fitted to each other with respect to the values they pursue. This eventually contributes to the enhancement of team performance and better organizational outcomes. Such a seminal idea goes back to the early stage of the human relations school that originated with the Hawthorne studies led by Elton Mayo and Wallace Donham [68][69][70]. The salience of group norms and leadership styles works for worker behavior.
Since the Hawthorne studies, the function and dysfunction of informal structure in a formal organization have long been a critical issue in organization studies [71]. Informal groups in formal organizations are formed naturally and voluntarily to meet the social needs and common interests of group members that are not easily satisfied by the formal organization. More recently, Soda and Zaheer (2012) argued that the interplay of formal and informal elements in the organization influences the performance of individual employees, and more specifically, the coherence of an informal social network and formal structures is strongly associated with higher performance [72]. However, strong ties among in-group members may not lead to positive outcomes for the organization. Oh, Chung, and Labianca (2004) discussed the role of informal socializing ties in affecting the association between "group social capital" and group effectiveness [73]. Group social capital indicates "the configuration of group members' social relationships within a group" [73].
PG fit, which refers to the interpersonal compatibility between team members with respect to their values, goals, and/or personality, is more likely to occur within the framework of an informal structure, while PS fit and PO fit reflect the goals and needs of a formal organization. Therefore, our research on fit interactions can expand the scope of fit studies to broader organization theories. Our research also suggests that a high level of PS fit can be compensatory or complementary to the lower level of PG fit at both high and low levels of PO fit for heightening team commitment. LMX theory strengthens the positive association between PS fit and team commitment in that a high quality of LMX facilitates cooperation between team members [74].

Practical Implications
The results of our research indicate that PS fit can compensate for PG fit when the level of PG fit is low by stimulating team members to commit to the work team. Likewise, PG fit also plays a compensatory role for inducing team commitment when confronting poor PS fit. By linking PG fit, PO fit, and PS fit, this study identifies a mechanism that explains how promoting the relationship between PG fit and team commitment can be mutually reinforcing. By understanding the relationship from individual and organizational perspectives, this study explains how managers can simultaneously promote both PS fit and PO fit in work environments. Balancing PS and PO fit while focusing on PG fit is a good example of strengthening the overall PE fit in a company and a team.

Limitations and Future Research Directions
Our findings have to be interpreted subject to some limitations. First, given the cross-sectional design, we cannot infer causality. Therefore, future research should include longitudinal measurement of teams' commitment so that the impact of employees' perceptions of commitment can be studied across time. Another limitation may be a potential risk of common method bias. The data were collected using a single questionnaire. However, common method bias for this particular study is not very likely to be a matter of concern given our findings of significant interaction effects [75]. Siemsen, Roth, and Oliveira [76] (p. 472) showed that "quadratic and interaction effects cannot be artificially created through CMV . . . A finding of significant quadratic or interaction effects connotes that researchers can be confident that they are not the result of CMV." Nonetheless, future research is needed to collect data from different sources. Also, a mixed-method study is worth pursuing to augment the quantitative analysis, including qualitative approaches such as in-depth interviews or ethnographic observations [77].
Second, the measurement of fit should be reexamined. For example, Seong and Kristof-Brown (2012) examined whether the fit measurement is formative (i.e., composed of various independent types of fit) or reflective (i.e., one overarching construct that influences perceptions of various types of fit). In order to advance, more attention must be paid to underexplored areas of fit [3].
Third, this study posited PG fit as an independent variable explaining team commitment, introducing PO fit and PS fit as moderators. However, there can be alternative possibilities with PG fit as a moderating variable in the relationship between either PO fit or PJ fit and team commitment [26].
Fourth, we should think about reverse causation. Prior research on fit has posited unidirectional causal relations from fit to outcomes, and in that sense, this study is no exception. However, it is also probable that team commitment enhances perceived PG fit. By being loyal to teamwork and fully committing to group performance, employees may feel congruent with their team. Astakhova (2016) suggested that commitment precedes fit to the work team, the supervisor, and the organization.
Fifth, fit between targets other than individual employees, such as group-organization (GO), supervisor-organization (SO), and group-supervisor (GS) fit beyond PE fit, is also worth pursuing [23].
As to GO fit, for example, a sample question should read as follows: "To what extent are the values of your team members (or the team) similar to the values of the organization (e.g., company)?" For SO fit and GS fit, "your supervisor" should be replaced by either the organization or the team. This kind of work has to be carried out at a group level or multi-level [4,78].
Finally, this study has cross-cultural implications. It is generally believed that Koreans are more collectivism-oriented than Westerners. In general, collectivist culture places more importance on interdependence among people, while individualist culture emphasizes the independent roles and functions of individual human beings [79]. Our study found significant interaction effects of PG, PO, and PS fit on team commitment. The relationship with the group, the organization, and the supervisor is more likely an important factor in determining team commitment in Eastern societies characterized by Confucian values than is the case with their Western counterparts. It is suggested that comparative research encompassing other cultures is needed to test the tentative conclusion drawn in this study. Stated in this way, we can conclude that the interactive effects of PG, PO, PS fit on team commitment might be quite different among employees in other countries.

Conclusions
Despite its limitations, this study enriches our understanding of PE fit and its effects on attitudes. We investigated the mechanism by which PE fit influences team commitment by examining the three-way interaction of PE fit and team commitment. Although a significant amount of research on PE fit has been conducted, we still do not fully understand the mechanism. In particular, we do not know much about the compounding effects of PS fit and PG fit. Hence, additional research in this area seems not only to be warranted, but also critical to advancing our knowledge of PE fit.
Our findings suggest that PG fit is an essential component of PE fit in enhancing the sustainable development in teams since the relationship with coworkers has become a critical source of social influence. To extend the importance of the current research, we need to investigate the mechanism of how PE fit influences behaviors and performance by examining the mediating and moderating role of team commitment at multi-levels.