GROUP ENGAGEMENT BASED ON SOCIAL EXCHANGE THEORY: ANTECEDENTS AND CONSEQUENCES

Danang Sunyoto1, Heru Kurnianto Tjahjono2, Zaenal Mustafa El Qodri3, Wisnu Prajogo4, and Syamsul Hadi5* 1 Department of Management, Faculty of Economics, Universitas Janabadra, Yogyakarta, 55231, Indonesia 2 Department of Management, Faculty of Economics and Business, Universitas Muhammadiyah Yogyakarta, Yogyakarta, 55183, Indonesia 3 Department of Management, Faculty of Business and Economics, Universitas Islam Indonesia, Yogyakarta, 55283, Indonesia 4 Department of Management, Faculty of Economics, STIE YKPN, Yogyakarta, 55281, Indonesia 5 Department of Management, Faculty of Economics, Universitas Sarjanawiyata Tamansiswa, Yogyakarta, 55165, Indonesia


Introduction
Work engagement is an affective-motivation and positive psychological state related to work characterized by enthusiasm, dedication, and absorption (Schaufeli et al., 2002). Despite its relevance in work settings, most scholars have focused on work engagement at the individual level, and ignored the role of groups (Richardson & West, 2010). It is even more incredible if we consider that groups play an important role in the employees' health and well-being (Wilson, et al., 2004), and productivity (Salanova et al., 2003).
This study analyzes the mediating role of group engagement including the spirit of group work, group dedication, group absorption, between the perceived organizational support including superiors' support, rewards, and working conditions, and the distributive justice including contribution, hard work, compensation, with the group performance including work harmony, open communication, ways to get work done, work effectiveness, and work efficiency. Group engagement plays a mediating role between perceived organizational support and distributive justice at the level of the group and group performance (Torrente et al., 2012). The previous study shows different results for the same antecedents and consequences on work engagement variables at the individual level. According to the studies by Ram & Prabhakar (2011), Dajani (2015 they show different results. On the other hand, effective group performance comes from several fundamental characteristics. First, group members must successfully integrate their actions individually. Second, groups are increasingly demanded to perform in complex and dynamic environments. Third, group leadership represents an effective group performance. Most groups contain certain individuals who are primarily responsible for setting group goals, developing, and organizing groups in achieving this mission. According to Marks et al., (2001), effective groups integrate four processes based on cognition, motivation, affection, and coordination.
Conceptually, group performance is a set of interrelated cognitions, attitudes, and behaviors that contribute to dynamic performance processes. A group cognition or a group-level macro cognition is an example of the type of interrelation between processes and has been the focus of many recent studies.
Group performance in research is defined as measurable results or group members' self-assessment of their own group's performance (Belbin, 2010;Senior, 1997;Partington & Harris, 1999). In Belbin's study (Belbin, 2010;Dulewicz, 1995), Group performance is measured in terms of wins and losses. Partington & Harris (1999) define group performance as a measure of group output and how to meet their quality standards. Group performance is also defined as social cohesiveness, group learning, personal well-being, and satisfaction expressed by group members with group experience and achievements, for example, the value obtained on the project. Because of the importance of groups and group work in organizations, many studies have focused on how to achieve highperformance groups (Abadi & Perkasa, 2020;Putri et al., 2020;Tjahjono et al., 2020). One of the most well-known and widely used group-role frameworks in the development of group composition and management development is built by Belbin (2010). Belbin's study reveals that the difference between successful and unsuccessful groups does not depend on individuals such as intelligence and experience, but mainly on behavior, the way group members make decisions, interact with each other, and apply their abilities to achieve group results. Belbin observes that individuals in groups tend to perceive different "group roles" which are defined as "inclinations to behave, contribute and interact with others in certain ways." Belbin found that successful groups were marked by the compatibility of the roles played by their members while working. Groups that fail become targets of role conflicts (personality clashes, conflicts, lack of cooperation) from surplus roles/mismatches and or loss of the required "behavioral" roles. That is why Belbin argues that the method of group role composition that is identified to form groups is proven to determine the success of the group and therefore one of the most important individuals who influence group performance. The role of groups as used in this study is based on Belbin's team role theory (2010).
Finally, group effectiveness is an evaluation of the results of relative group performance processes to some group criteria (Hackman, 1987). It means that performance is an activity carried out while completing a task, and effectiveness involves evaluating the results of that activity (Fitts & Posner, 1967). Through this foundation, further research will be carried out at the group level.

Literature Review and Hypothesis
According to various opinions of sociologists, social exchange theory is a theory in social science which states that in social relations there are elements of reward, sacrifice, and profit influencing each other. This theory explains how humans perceive our relationship with other people according to their self-perception towards 1) The balance between what is given into a relationship and what is excluded from that relationship, 2) The type of relationship that is carried out, 3) How people relate to society (between a group of people), the smallest is the household and the community, and 4) The opportunity to have a better relationship with other people. Social exchange theory includes the basic theory of rationality, more integrative exchange theory, network theory, network exchange theory, and rational choice theory.

Perceived Organizational Support, Group Engagement, and Group Performance
The perceived organizational support (POS) is based on general beliefs where organizations pay attention to their contributions and are concerned with their well-being (Rhoades & Eisenberger, 2002). Perceived organizational support creates an obligation for employees to care about the welfare of the organization and to help the organization achieve its goals (Rhoades et al., 2001). The perceived organizational support is also to determine the organization's readiness to appreciate the increased work effort and to meet social-emotional needs, the employees develop global trust about how far the organizations appreciate their contribution and are concerned with their well-being (Eisenberger et al., 1986;Shore & Shore 1995;Ram & Prabhakar, 2011). Perceived organizational support is valued as a guarantee that the organization will assist when needed to carry out one's work effectively and to deal with stressful situations (George, 1993). In other words, employees with higher perceived organizational support may become more engaged to their work and organization as part of the norms to help the organization achieve its goals (Rhoades et al., 2001). When employees believe that their organization cares about them and cares about their wellbeing, they tend to respond by trying to fulfill their obligations to the organization by becoming more involved (Rhoades & Eisenberger, 2002). This is supported by Social Exchange Theory/SET that the employees who are given a more challenging workload will feel obliged to respond with a higher level of employee engagement. In social exchange theory, obligations are generated through a series of interactions between parties who are in mutual interdependence. SET has a basic principle that relationships develop over time, commitments of mutual trust, loyalty, and reciprocity as long as the parties obey to certain exchange rules.
Perceived organizational support by the employees that also plays an important role in the employees' group engagement has a definition of employees' general beliefs about how far the organization appreciates contributions and cares about the employees' welfare (Eisenberger et al., 1986). This organizational support felt by the employees is considered as a certainty of the availability of assistance from the organization when the assistance is needed to support the implementation of employees' duties to be able to run effectively and to deal with stressful situations (Rhoades & Eisenberger, 2002). Organizational support can also be seen as an organizational commitment to the employees. If the organization, in general, appreciates the dedication and loyalty of the employees as a form of employee' commitment to the organization, then the employees also generally pay attention to how the organization's commitment towards them.
Appreciation given by the organization can be considered to provide benefits for the employees, such as the feeling of being accepted and acknowledged, obtaining salaries and promotions, getting access to information, and other forms of assistance needed by the employees to be able to carry out their work effectively. The existence of norms of reciprocity causes the employees and the organizations need to pay attention to each other's goals in the employment relationship (Rhoades & Eisenberger, 2002). Eisenberger et al., (1986) put forward two aspects to determine the conditions of organizational support felt by employees. The two aspects are the organizational appreciation of employees' contributions and the organizational attention to employees' welfare (Hadi & Supardi, 2020;Hadi, Tjahjono, El Qadri, et al., 2020;Hadi, Tjahjono, & Palupi, 2020;supardi & Hadi, 2020).
Organizational trust towards the policies of the employees in completing their work will also increase organizational support felt by the employees (Eisenberger et al., 1999). Role stressors such as overwork/work demands that are impossible to complete in a limited time, the vagueness of roles including the absence of clear information about individual responsibilities, and role conflict including conflicting work responsibilities can reduce organizational support felt by employees (Rhoades & Eisenberger, 2002). The employees' perceived procedural fairness influence their perception that the organization respects them. In turn, this impression encourages the employees to reciprocate the organization by showing organizational citizenship behavior. This is where social exchange with organizations takes place. The employees will tend to feel "indebted" to the organizations that have supported them, as a result, they will make more contributions to the organization in the form of doing work beyond what is voluntary. The employees will view exchanges in the organization as an injustice if the organization does not provide any support. The emergence of injustice in the organization will automatically lead to a feeling of lack of support from the organization, as a result, later it will have an impact on decreasing the frequency of organizational citizenship behavior. The employees are not motivated to contribute more to the organization. They will only show in-role work behavior that may also experience a decline due to a lack of justice and organizational support. Hypothesis 1 (H1) Perceived organizational support influences group engagement positively. Hypothesis 3 (H3) Perceived organizational support influences group performance positively Hypothesis 6 (H6) Perceived organizational support positively influences group performance with group attachment as a mediator variable.

2.2.Distributive
Justice, Group Engagement, and Group Performance Distributive justice relates to the goals achieved (what the decision is) or the content of justice and predicts satisfaction with the results (Sweeney & McFarlin, 1993;Ram & Prabhakar, 2011). On the other hand, distributive justice is related to one's perceived justice of the decision outcomes (Colquitt, 2001;Rhoades et al., 2001). A review of organizational justice research found that perceived fairness were related to organizational outcomes such as job satisfaction, organizational commitment, organizational citizenship behavior, withdrawal, and performance (Colquitt et al., 2001). When the employees have a high perception of justice in their organization, they are more likely to feel obliged to also be fair in the way they carry out their roles by giving more through greater levels of attachment. Conversely, low perceptions of justice tend to cause employees to withdraw and escape from their job roles. A lack of justice can worsen fatigue while a positive perception of justice can increase attachment (Maslach et al., 2001). Hypothesis 2 (H2) Distributive justice positively influences group attachment. Hypothesis 4 (H4) Distributive justice positively influences group performance. Hypothesis 7 (H7) Distributive justice positively influences group performance with group engagement as a mediator variable.

2.3.Group Attachment and Group Performance
Group attachment of the employees related to group performance will positively contribute to the results achieved by an organization.
Here we define job performance as an aggregate of financial or non-financial added value by employees in contributing to fulfillment, both directly and indirectly, to the organization's targets. Research has shown that employees' engagement has a positive influence on organizational performance indicators such as employees' satisfaction, productivity, turnover, organizational commitment, and security (Dajani, 2015).
Implicitly, the relationship between engagement and job performance has been described (Kahn, 1990;Bruce & Jeffrey, 2010). In general, the employees who have a high attachment to their work, apparently not only focus themselves on achievements related to roles but also are cognitively alert and emotionally connected to the hard work. Therefore, the employees who are weak in their work roles hold their physical energy, cognition, and emotions, and this is reflected in passive and separated task activities (Hochschild, 1983;Kahn, 1990;Ashforth & Humphrey, 1995;Bruce & Jeffrey, 2010).
Some research results indicate that there are three boosters of the employees' engagement with their performance, namely: 1) the spirit of taking part in work has to do with organizational goals, because the high organizational behavior values will last longer time (Kahn, 1990;1992;Bruce & Jeffrey, 2010). When many employees play a role in their work, it reflects the behavior of other employees as a positive contributor to their organization (Katz & Kahn, 1978;Bruce & Jeffrey, 2010). This makes it clear that the employees who work harder have a higher level of job performance (Brown & Leigh, 1996), 2) Cognitive energy investments in work roles contribute to organizational goals because they promote more alert, attentive and focused behavior (Kahn, 1990). Attention decreases due to a reduction in cognitive energy investment, performance decreases due to failure to see, or to pay attention to one's work role, and, 3) investments in emotional energy into work roles contribute to organizational goals in some related ways (Kahn, 1990). The employees invest emotional energy into their performance roles through the promotion of increased relationships between colleagues in achieving organizational goals (Ashforth & Humphrey, 1995). Emotional energy investment also helps individuals meet the emotional demands of their role in producing more complete and authentic performance (Kahn, 1990;1992). In another part of the previous research, it was mentioned that social resources are not exclusively antecedents of group work engagement. Also, Torrente et al.'s, (2012) model fails to integrate what we already know about group processes and group effectiveness, and is a homologous transposition (Kozlowski & Klein, 2000) model of individual-level engagement, therefore it ignores possible important differences between levels.
Overall, previous researches on work engagement in groups have several limitations. The studies have not presented clear definitions of constructs or theoretical models for group work engagement that contribute variables that are exclusively relevant in group contexts. Even when considering relevant group variables and group member interactions, the researches on group work engagement have not been integrated into the specific literature on the group. See Figure 1. As Research Framework Model.

3.1.Research Population
According to Arikunto (2013), "Population is the whole of the research subjects. The population is a very important data source because, without the presence of population, research will not be meaningful and is impossible to carry out". This research using a specific population is usually conducted by means of the census method, which involves the overall data used as research objects. The population in this study were officials and teaching staff of social sciences study programs at private universities in Yogyakarta, comprising 113 study programs. Based on the population, this study used the census method, which is a way of collecting data if the entire population is examined one by one. The data obtained as the result of census processing is referred to as actual data (Supranto, 2008). In this study, from 113 study program groups (population) those qualified as data are 105 samples for the social sciences study programs of private universities in Yogyakarta, consisting of the head and secretaries of the study programs under the research.

Measurement
The data collection used a questionnaire that was applied after it met valid and reliable requirements. The measurement of research variables was based on respondents' alternative answers using a Likert scale of 1 (strongly disagree), 2 (disagree), 3 (neutral), 4 (agree), and 5 (strongly agree). A high answer score indicates a high level, and a low answer score indicates a low level.

4.1.Test Model Goodness of Fit
After the SEM assumptions were fulfilled in Figure 2., the next step was testing by using some suitability indexes to measure the "truth" of the proposed model. The test is known as the "goodness of fit" test. The following results of the goodness of fit test can be seen in Table 1 Table 2, 3, and 4 below:

4.2.Perceived Organizational Support (POS) dan Group Engagement (GE)
From table 2 it is obtained the standardized regression weight coefficient between perceived organizational support (X1) and group engagement (Y1) of -0.296. To test the effect between the two variables it shows a C.R. value of -2.119 with a probability value of 0.034. Based on these results, the first hypothesis is accepted, in which the statement is based on a probability value of 0.034 that meets the <0.05 requirement and a C.R. value of -2.119 that meets the >+ 1.96 requirement. Then, it can be concluded that perceived organizational support (X1) has a significant effect on group engagement (Y1).

Distributive Justice (DJ) and Group Engagement (GE)
From table 2, the standardized regression weight coefficient between distributive justice (X2) and group engagement (Y1) is obtained at 0.368. To test the effect between the two variables it shows a C.R. value of 2.521 with a probability value of 0.012. Based on these results, the second hypothesis is accepted, in which the statement is based on a probability value of 0.012 that meets the <0.05 requirement and a C.R. value of 2.521 that meets the >+ 1.96 requirement. It can be concluded that distributive justice (X2) has a positive and significant effect on group engagement (Y1).

Perceived Organizational Support (POS) and Group Performance (GE)
From table 2 the standardized regression weight coefficient between perceived organizational support (X1) and group performance (Y2) is obtained at -0.333. To test the effect between the two variables it shows a C.R. value of -2.121 with a probability value of 0.034. Based on these results, the third hypothesis is accepted, in which the statement is based on a probability value of 0.034 that meets the <0.05 requirement and a C.R. value of -2.121 that meets the > + 1.96 requirement. Then, it can be concluded that perceived organizational support (X1) has a significant effect on group performance (Y2).

Distributive Justice (DJ) and Group Performance (GP)
From table 2 we get the standardized regression weight coefficient between distributive justice (X2) and group performance (Y2) of 0.101. To test the effect between the two variables it shows a C.R. value of 0.711 with a probability value of 0.477. Based on these results, the fourth hypothesis is rejected, in which the statement is based on a probability value of 0.477 that does not meet the < 0.05 requirement and C.R. value of 0.711 does not meet the >+ 1.96 requirement. Then, it can be concluded that distributive justice (X2) does not have a significant effect on group performance (Y2).

Group Engagement (GE) and Group Performance (GP)
From table 2, the coefficient of standardized regression weight between group engagement (Y1) and group performance (Y2) is 0.375. To test the effect between the two variables it shows a C.R. value of 2.076 with a probability value of 0.038. Based on these results the fifth hypothesis is accepted, in which the statement is based on a probability value below 0.038 that meets the < 0.05 requirement and C.R. value of 2.076 that has fulfilled the > + 1.96 requirement. Then, it can be concluded that group engagement (Y1) has a positive and significant effect on group performance (Y2).

Perceived Organizational Support (POS), Group Engagement (GE), and Group Performance
The sixth hypothesis aims to examine the effect of perceived organizational support (X1) on group performance (Y2) mediated by group engagement (Y1). From table 3 and  table 4

Distributive Justice (DJ,) Group Engagement (GE), and Group Performance
The seventh hypothesis aims to examine the effect of distributive justice (X2) on group performance (Y2) mediated by group engagement (Y1). From table 3 and table 4

Discussion
Perception of organizational support is influenced by the experiences possessed by individuals, as well as observations about the daily lives of organizations in treating someone individually and in groups. In this case, the organization's attitude to ideas raised by employees, the response to employees who experience problems, and the company's attention to employees' welfare and health are three factors that are the main concern of employees. Waileruny (2014) states that the perception of organizational support is the level at which the employees believe the organization appreciates their contribution and cares about their well-being. The perception of organizational support is an employee's perception that he/she is appreciated and getting cared by the organization or company where he/she works. If the organization pays attention to and appreciates the efforts made by individuals to achieve the company goals, individuals will perceive that the organization provides support for them (Sunyoto, Tjahjono, El Qodri, 2019). According to SET theory, employee contributions to distributive justice are rewarding employees based on the contributions they make to the company and also based on workload under their job descriptions. Because distributive justice is a resource justice and appreciation of rewards, it reflects the perceived fairness of how resources and rewards are distributed and allocated (Kreitner & Kinicki, 2003). According to Tjahjono (2008;2011), distributive justice is transactional between the organization and the employees. The employees are motivated to get long-term welfare, therefore fair distribution is very important for the employees. Distributive justice is related to the results obtained by the employees of the organization (Folger & Konovsky, 1989) and distributive justice is related to the distribution of circumstances and goods that will affect individual welfare (Deutsch, 1975).
According to SET theory, employee group attachment related to group performance will positively contribute to the results achieved by an organization. Job performance as an aggregate of financial or non-financial value added by the employees in contributing to fulfillment both directly and indirectly to the target organization is targeted. Research has shown that employee engagement has a positive effect on organizational performance with indicators such as employees' satisfaction, productivity, employees' turnover, organizational commitment, and security (Dajani, 2015).

Conclusion
Based on the results of this research data analysis, the conclusions are as follows: -Perceived organizational support has a significant effect on group engagement. -Distributive justice has a positive and significant effect on group engagement. -Perceived organizational support has a significant effect on group performance. -Distributive justice has an insignificant influence on group performance. -Group engagement has a positive and significant effect on group performance. -Perceived organizational support has a significant effect on group performance with group engagement as a mediator variable. In other words, group attachment has a significant role as a mediator.
-In the effect of perceived organizational support on group performance. -Distributive justice has a significant effect on group performance with group engagement as a mediator variable. In other words, group engagement has a significant role as a mediator in the effect of distributive justice on group performance.

Research Limitations and Recommendations
The important things in this research are about the limitations of the study and recommendations for further research, they are as follows: -This study uses research subjects in the social sciences study program of private institutions in Yogyakarta, Indonesia. Therefore, for further research, it can be recommended to use research subjects from other study programs at private and state institutions/universities with a wider quantity and scope both in Indonesia and outside Indonesia. In another term, further research can also be conducted using the subject of industrial companies or other organizational institutions. -The antecedent variables in this study are only two (perceived organizational support, distributive justice), therefore, for further research, it is recommended to use the antecedent variables of other group attachments, such as human resource practices, social resources, organizational justice and leadership, and employee communication.
-The consequence variable in this study is only one, that is, group performance, as a result, for further research, it is recommended to use the variable group engagement consequences except for group performance, for example, group satisfaction. -The mediator variable in this study is only group engagement, therefore for further research, it is recommended to use another mediator variable, that is, work team engagement, with the consequent variable that is work team satisfaction.