Organizational Commitment as a Mediator of Organizational Support and Emotional Intelligence in Shaping Organizational Citizenship Behavior

Research aims : This study aims to analyze the effect of organizational support and emotional intelligence on organizational citizenship behavior (OCB) through employee organizational commitment. Design/Methodology/Approach : This research was conducted in the Central Bureau of Statistics in Medan City with a population of 100 permanent partner employees. Using non-probability sampling, the entire population became the research sample. Meanwhile, the analysis technique employed path analysis. Research findings : The results proved that organizational support and emotional intelligence had a significant effect directly on organizational citizenship behavior (OCB) and indirectly through employees' organizational commitment. In other words, increased organizational support and emotional intelligence of employees will intensify the level of employee organizational commitment, raising the level of organizational citizenship behavior (OCB) and vice versa. In this case, organizational support strongly predicts employees’ high org anizational commitment. Theoretical contribution/Originality : Further research that explores other predictors needs to be carried out with more significant respondents to generalize the findings better. Practitioner/Policy implication : It is essential to provide more intensive organizational support to employees since this variable is a strong organizational predictor of generating a superior level of employee organizational commitment. Organizations also need to conduct in-depth personality tests to detect the level of individual emotional intelligence of employees who work in the field and interact with the community for better work effectiveness. Research limitation/Implication : The limited number of population and samples in the agency is a limitation of the study to generalize the findings better.


Introduction
Various aspects can determine the company's success, mainly from its employees' work behavior. As such, employees are required to complete the tasks and responsibilities that become their job descriptions. However, in addition to that, employees are often willing to voluntarily expend tremendous effort for the organization by doing various tasks outside the Another individual predictor that contributes significantly to the emergence of employee OCB is organizational commitment. The previous research results have proven that organizational commitment is a strong predictor of the emergence of OCB in employees (Aisyah & Wartini, 2016;Wijaya & Yuniawan, 2017;Pigome et al., 2019;Junita, 2017;Junita et al., 2019;Busharmaidi, 2020). In this case, commitment is an affective aspect that influences an individual's work behavior (McShane & Von Glinow, 2010). When employees identify themselves with the organization, they will feel bound as members of the organization. Aisyah and Wartini (2016) and Junita et al. (2019) even confirmed that organizational commitment is a critical mediating variable that connects various organizational predictors to raise an employee's OCB.
Moreover, organizational support provided by the Central Bureau of Statistics (BPS) in Medan City can be perceived differently by each employee, impacting organizational commitment and OCB. Similarly, employees' emotional intelligence is considered relevant to be analyzed in their work activities outside the office and far from the direct supervision of their superiors. The respondents also interacted with the public to collect various field survey data needed by organizations which, in the interaction process, were required to have qualified emotional intelligence to support their work to run well. Therefore, this

Junita, Hafas, & Nasution
Organizational Commitment as a Mediator of Organizational Support … Jurnal Manajemen Bisnis, 2022 | 237 research emphasizes organizational antecedents in the form of organizational support, which varies in character, and individual antecedents in emotional intelligence, which is rarely observed empirically in previous research. For this reason, analyzing the impact of organizational (organizational support) and individual (emotional intelligence) predictors in an integrative manner on the emergence of employee OCB through organizational commitment as a mediator becomes this study's significance and research objective. It also completes the theoretical and empirical gap from previous research, which only partially analyzed organizational or individual predictors in generating OCB and organizational commitment.

Organizational Citizenship Behavior
Organizational citizenship behavior (henceforth called OCB) is individual behavior that is independent (discretionary), does not directly receive rewards from the formal reward system, and encourages the effectiveness of organizational functions (Organ, 1990;Greenberg & Baron, 2003). Organ (1988) argues that the OCB concept differs from organizational commitment, where OCB is tangible as work behavior, while commitment is an attitude character (Mowday et al., 1979). Van Dyne et al. (1995) specifically mentioned OCB as a form of extra-role behavior, i.e., behavior that benefits the organization beyond the obligations of the individual role. Borman and Motowidlo (1997) refer to it as the contextual performance term.
Moreover, OCB in employees manifests itself in several dimensions (Organ, 1988;Organ et al., 2006): 1. Altruism is the behavior of employees willing to help colleagues experiencing task or personal problems. 2. Conscientiousness is the behavior of doing things that exceed the company's expectations and not only completing the task. 3. Sportsmanship is a tolerant work behavior of less-than-ideal conditions in the organization, so it tends to create a pleasant work environment. 4. Courtesy is the behavior of maintaining and fostering good relations with colleagues. 5. Civic virtue is a work behavior full of initiative and responsible for improving the quality of the field of work occupied.  Griffin (2004) and Mowday et al. (1979) define organizational commitment as an individual's relative strength to identify with the organization and be deeply involved, including trust, support for the goals and values of the organization, an ardent desire to work hard for, and a strong will to maintain membership in the organization.

Junita, Hafas, & Nasution
In addition, organizational commitment is one of the attitude variables that significantly affect various positive employee behaviors, including OCB. As an attitude variable, it allows employees to have a variety of positive and negative assessments reflected in their level of organizational commitment. Variations in the level of employee commitment are also determined by numerous factors, both organizational and individual factors (Junita, 2017;Bodjrenou et al., 2019;Bodjrenou et al., 2019;Parimita et al., 2020). The previous research findings support this, stating that organizational commitment could significantly contribute to employee OCB (Ardi & Sudarma, 2015;Fazio, 2015;Devece et al., 2016;Pigome et al., 2019;Siregar et al., 2019;Wilkanandya & Sudarma., 2020) but also not significant (Novianti, 2021).
According to Meyer and Allen (1991) dan Mowday et al. (1982), components of organizational commitment include: 1. Affective commitment relates to employees' emotional identification and involvement in an organization. 2. Continuance commitment is based on employees' perceptions of the losses they will face if they leave the organization. 3. Normative commitment is based on a sense of employee compliance with the organization to repay all the benefits received from the organization.
Each individual can have all three components within him at once. However, the diversity of individual characters in organizations can provide diverse levels of organizational commitment. The way individuals perceive organizational situations also contributes greatly to employee organizational commitment.
The previous research results have verified a substantial contribution of organizational commitment in making employees' OCB more dominant. Thus, this phenomenon justifies this research to place organizational commitment as a mediating variable between organizational antecedents, namely organizational support and individual antecedents, i.e., emotional intelligence, on employee OCB. It aligns with Junita's (2017) results, who concluded that organizational commitment is one of the employee attitude variables appropriately positioned as a mediating variable.

Organizational Support
Organizational support is the perception and belief of employees that the organization where they work cares, appreciates the contribution of workers to the organization, and always conditions various forms of support and assistance for its employees (Blau, 1964;Eisenberger et al., 1986;Eisenberger & Stinglhamber, 2011;Erdogan & Enders, 2007).

Junita, Hafas, & Nasution Organizational Commitment as a Mediator of Organizational Support …
Jurnal Manajemen Bisnis, 2022 | 239 Employees' perceptions of organizational support also largely determine whether the organization can meet employee expectations regarding things that should be received from the organization.
In essence, organizational support includes three principles: reciprocity (a mutually beneficial exchange between the interests of the organization and employees), meeting the social needs of employees at work, and a fair reward system (Rhoades & Eisenberger, 2002). Some dimensions of organizational support, according to Eisenberger et al. (1986), Rhoades & Eisenberger (2002), and Robbins and Judge (2018), are procedural justice that arises from the process used to determine the distribution of rewards, working conditions, and rewards as the treatment of organizations that consider workers as substantial investments (assets) for the organization; thus, attention is paid to the implementation of various practices that can increase the value of human assets. It comprises training, promotions (Rhoades & Eisenberger, 2002), job autonomy (Allen et al., 2003), career development (Lew, 2009), distributive justice (McShane & Von Glinow, 2010;Putranto et al., 2019;Fachriansyah et al.., 2021), supervisory support (Susskind et al., 2007;Mustika et al., 2020;Fachriansyah et al., 2021), work structure, personal development (Ginduz, 2014), rewards, working conditions (Mustika et al., 2020;Fachriansyah et al., 2021), and welfare (Ansori & Wulansari, 2021).

Emotional Intelligence
McShane and Glinow (2010), Ivancevich (2011), and Daud (2012) define emotional intelligence as a set of abilities to understand and express emotions, assimilate emotions in mind, understand and reason emotions, and regulate emotions in oneself and others. Besides, Goleman (1998) defines emotional intelligence as a person's understanding of the feelings of others, especially in the individual decision-making process. Salkojani et al. (2012) also elaborated on emotional intelligence as a non-cognitive competence that makes individuals strong in facing various internal and external pressures. Meanwhile, the FLICC (2011) model operationally defines emotional intelligence as competence to perform conflict management, external awareness, flexibility, negotiation, interpersonal skills, self-management, teamwork, and collaboration.

Research Framework
Based on theoretical and empirical studies relevant to the conceptual research framework (Figure 1), the research hypotheses were formulated as follows: H 1 : Organizational support and emotional intelligence significantly affect employee organizational commitment.
H 2 : Organizational support and emotional intelligence significantly affect organizational citizenship behavior mediated by employee organizational commitment.

Research Methods
This causality research used quantitative/statistical analysis to assess research hypotheses (Sugiyono, 2013). Testing and analyzing the direct influence of organizational support and emotional intelligence in generating employee organizational citizenship behavior, as well as indirectly through organizational commitment, was conducted with the help of inferential and descriptive statistics. In addition, the path analysis technique employed inferential statistics to analyze the relationship between research variables. Besides, descriptive statistics were also utilized to help deepen the discussion of the inferential statistical test results (path analysis). All were based on a quantitative approach. The research variables included: 1. Organizational support is employees' general belief in the organization's care and contribution to their welfare (Rhoades & Eisenberger, 2002). Organizational support in this study was measured by four dimensions proposed by Eisenberger et al. (1986): rewards, superior support, working conditions, and employee welfare. 2. Emotional intelligence is recognizing and controlling feelings to help emotional and intellectual development (Salovey & Mayer, 1990). The emotional intelligence variable in this research was assessed by five dimensions of Goleman (1998): selfawareness, self-regulation, self-motivation, empathy, and social skills. 3. Organizational commitment is an attitude that reflects the individual attachment to the organization (Griffin, 2004). It was evaluated by five dimensions of Griffin (2004), including feeling happy to spend the rest of their career in the organization, feeling worried if stopping working in the company, feeling unethical to move to another organization, loyalty, and lack of intention to leave the organization despite getting a better offer in another organization. 4. Organizational citizenship behavior (OCB) is the voluntary behavior of a worker to want to do a task or job outside his responsibilities or obligations for the advancement or benefit of his organization (Organ et al., 2006). The OCB variable in this study was measured by five dimensions used by Organ et al. (2006), including altruism, sportsmanship, courtesy, civic virtue, and conscientiousness behavior.
Further, the population and samples in this study were all employees (permanent partners) who worked at the Central Bureau of Statistics in Medan City, totaling 100 people. The sampling technique was then referred to as non-probability or saturated sampling (Sugiyono, 2013). The path analysis technique was used to examine the direct effect of exogenous variables on endogenous variables (Sarwono, 2012). Path analysis is usually used by first fulfilling various required tests, including instrument validity and reliability, normality, multicollinearity, and heteroscedasticity (Ghozali 2017).

Respondents' Characteristics
Research respondents can be described based on education level, length of work, age, and gender ( Figure 2). Most respondents attained a bachelor's level of education (86%), had worked at BPS Medan City for 3-6 years (90%), were aged > 31 years (74%), and were female (56%).

Validity and Reliability Instruments Test Results
Based on the validity test results of eight question items for the organizational support variable, ten question items for the emotional intelligence variable, and eight question items for the organizational commitment variable, the overall results of the items in each variable were declared valid since they had an r-value > r- and a significance level of 5%. As for the OCB variable, a second item was found invalid of the ten question items tested for validity, so it had to be removed and retested for validity.

Test Results of Substructure 1 Path Analysis Model (Hypothesis 1)
The model test in this study using path analysis included the analysis of two substructures representing the problem areas and research hypotheses. Substructure 1 examined the effect of organizational support and emotional intelligence on employees' organizational commitment. Based on the normality test results conducted with the Kolmogorov-Smirnov Test (KS Test), the KS Test value was obtained at 0.078 > 0.050 (a), so it was concluded that the data in the research model for hypothesis 1 were normally distributed (Ghozali, 2017). Based on the multicollinearity test results, there was no significant correlation between the study's independent variables. The VIF value of the organizational support and emotional intelligence variables was 1.625 < 10, and the tolerance value was 0.615 > 0.100. Thus, there was no heteroscedasticity problem in the first hypothesis model since the data were randomly distributed.
According to Ferdinand (2014), the F-test is used to assess the model feasibility in explaining the relationship between variables in the observed phenomenon or the research model's suitability with the analyzed phenomenon. Based on the F-test results of hypothesis 1, the calculated F-value was 0.000 < 0.050, and the coefficient of multiple determination (R2) value was 71.5%. It means that organizational support (X1) and emotional intelligence (X2) were feasible to explain the organizational commitment (Y1) of 71.5%, while the remaining 28.5% were influenced by various other variables not observed in this study. The next is to find out the contribution of organizational support (X1) and organizational intelligence (X2) variables to organizational commitment (Y1) partially (Table 2). Based on the t-test results, the significance value of the organizational support variable (X1) was 0.000 < 0.050 (a), and the emotional intelligence variable (X2) was 0.013 < 0.050 (a). It means that the increased organizational support and individual emotional intelligence of employees will increase the level of organizational commitment of employees and vice versa. In addition, organizational support significantly affected organizational commitment, with a path coefficient value (p1) of 0.731, and emotional intelligence, with a path coefficient value (p2) of 0.172. Meanwhile, e1 shows the variance of the research model, calculated by the formula: e1:√1-R2. Then, the t-test results became the basis for describing the path analysis model of substructure 1, as shown in Figure 3.

Test Results of Substructure 2 Path Analysis Model (Hypothesis 2)
Substructure 2 path analysis model tested hypothesis 2: the effect of organizational support and emotional intelligence on organizational citizenship behavior (OCB) through organizational commitment. To evaluate the impact of the research variables, it is necessary to test the data normality, multicollinearity, and heteroscedasticity. Based on the normality test results conducted with the Kolmogorov-Smirnov Test (KS Test), the KS Test value was 0.062 > 0.050 (a), so it was concluded that the data in the research model of hypothesis 1 were normally distributed (Ghozali, 2017). Based on the multicollinearity test results, the study's independent variables showed no significant correlation between the VIF value of the organizational support of 3.539, emotional intelligence of 1.731, and organizational commitment of 3.580. Besides, the

Junita, Hafas, & Nasution Organizational Commitment as a Mediator of Organizational Support …
Jurnal Manajemen Bisnis, 2022 | 245 overall value of the independent variable VIF in the second hypothesis model was < 10. The tolerance value of the organizational support was 0.283, emotional intelligence was 0.578, and organizational commitment was 0.279. Meanwhile, the overall value of the tolerance of the independent variables in the second hypothesis model was > 0.100. Therefore, the heteroscedasticity test results concluded that there was no heteroscedasticity problem in the first hypothesis model since the data were randomly distributed.

Figure 4 Model of Path Analysis Substructure 2
Based on the F-test results of hypothesis 2, the calculated F-value was 0.000 < 0.050, and the coefficient of multiple determination (R2) was 52.7%. It indicates that organizational support (X1), emotional intelligence (X2), and organizational commitment (Y1) were feasible to explain organizational citizenship behavior (OCB) (Y2) by 52.7%, while the remaining 47.3% were influenced by various other variables not observed in this study. To find out the contribution of organizational support (X1) and organizational intelligence (X2) variables to organizational citizenship behavior (OCB) (Y2) through organizational commitment (Y1) partially, it can be seen from the t-test results as shown in Table 3. Based on the t-test results in Table 3, the significance value of the organizational support variable (X1) was 0.001 < 0.050 (a), emotional intelligence variable (X2) was 0.012 < 0.050 (a), and the organizational commitment variable (Y1) was 0.023 < 0.050 (a). Based on the t-test results of the second hypothesis model, the path coefficient value of the influence of organizational support on organizational citizenship behavior (OCB) was 0.444 (p4). Then, the path coefficient value of the emotional intelligence effect on organizational citizenship behavior was 0.327 (p5), while the path coefficient value of the organizational commitment variable effect on organizational citizenship behavior was 0.047 (p6). These  The path analysis test results of substructure 1 (Figure 3) and sub-structure 2 (Figure 4) then became the basis for describing the overall path analysis model, including paths of substructures 1 and 2, as shown in Figure 5. It appears that the organizational support variable (X1) and emotional intelligence (X2) had a significant effect on organizational citizenship behavior (OCB) (Y2) directly or indirectly through organizational commitment (Y1).

Discussion
Based on the path analysis test results, this study has proven the truth of hypotheses 1 and 2. Thus, organizational predictors in the form of organizational support and individual predictors in the form of emotional intelligence could simultaneously form employees' organizational citizenship behavior. The organizational support provided by the Central Bureau of Statistics to employees in the form of rewards, working conditions, welfare, and full support (Eisenberger et al., 1986) was perceived positively by employees. Theoretically, employees' perceptions of organizational support are determined mainly by whether the organization can meet employee expectations regarding things that should be received from the organization. The principle of reciprocity applies in this case (Rhoades & Eisenberger, 2002), i.e., there is a mutually beneficial exchange between the organization's and employees' interests. Hence, it is concluded that the obtained results confirm the current state of knowledge that organizational support is a principal factor determining an employee's organizational citizenship behavior (Frenkel & Bednall, 2016;Saputra & Supartha;Ansori & Wulansari, 2021).
Employees also had a relatively proficient level of emotional intelligence in terms of understanding their condition, managing emotions, motivating themselves to perform well, empathizing with others, and having social skills in interacting with various parties (Goleman, 1998). Theoretically, employees' emotional intelligence level is determined by one work experience (Al-Fawaeer & AlKhatib, 2020). Based on the characteristics of the research respondents, the respondents had a long working period of 3-5 years to adapt to their work environment, which had implications for good emotional intelligence. Good emotional intelligence will then manifest in good organizational citizenship behavior as well. Thus, it is concluded that the obtained results verify the current state of knowledge (Sahafi et al., 2011;Ibrahim, 2013;Naghdi & Shatalebi, 2013).
Between the two independent variables, the organizational support variable was a stronger organizational predictor in shaping employee organizational commitment than individual predictors of emotional intelligence. The relatively high level of employee commitment could be seen from feeling happy to spend the rest of the career in the organization, loyal to last a long time working in the organization, and having no intention to move to another company even though they get a better offer in another organization (Griffin, 2004). Thus, this study's results align with the findings of Celep and Yilmazturk (2012) and Khan et al. (2014), stating a significant positive relationship between organizational support and emotional intelligence on organizational commitment.
Based on the second hypothesis test results, it was concluded that organizational support and emotional intelligence variables significantly affected organizational citizenship behavior mediated by the organizational commitment of employees at the Central Bureau of Statistics Medan City (Table 3). The organizational commitment variable in this research model was positioned as a partial mediation variable. A partial mediation variable means that the variable can be placed as a mediating variable and an independent variable in the research model (Baron & Kenny, 1986;Suliyanto, 2011).
Moreover, Griffin (2004) and Mowday et al. (1979) define organizational commitment as an individual's relative strength to identify with the organization and be deeply involved, including trust, support for the goals and values of the organization, a powerful desire to work hard for, and a strong will to maintain membership in the organization. Organizational commitment is also an attitude variable that significantly affects various positive employee behaviors, including OCB. However, the diversity of individual characters in organizations can provide distinct levels of organizational commitment. The way individuals perceive organizational situations also contributes significantly to employee organizational commitment. The previous research results (Ardi & Sudarma, 2015;Fazio, 2015;Devece et al., 2016;Pigome et al., 2019;Siregar et al., 2019;Wilkanandya & Sudarma., 2020) have also proven that there was a substantial contribution of organizational commitment in making employees' OCB more dominant. Hence, this phenomenon justifies this research to place organizational commitment as a mediating variable between organizational antecedents, i.e., organizational support and individual antecedents, namely emotional intelligence, on employee OCB. It is consistent with Junita (2017), who concluded that the organizational commitment variable is one of the employee attitude variables appropriately positioned as a mediating variable. The placement of the organizational commitment variable as a mediating variable is also in line with the research findings of Aisyah and Wartini (2016), Junita (2017), and Junita et al. (2019), concluding that the organizational commitment variable is one of the employee attitude variables appropriately positioned as a variable mediation. Thus, the obtained results confirm the current state of knowledge.

Conclusion
Organizational citizenship behavior of employees is formed by various organizational and individual factors. In this case, organizational support represented organizational factors, and emotional intelligence signified individual factors. Directly, organizational support and emotional intelligence variables contributed significantly to the emergence of superior levels of employee OCB at the Central Bureau of Statistics in Medan City. In addition, the organizational support variable was a stronger organizational predictor in shaping employee organizational commitment than emotional intelligence as an individual predictor.
Indirectly, organizational support and emotional intelligence variables significantly affected organizational citizenship behavior (OCB) through organizational commitment. Compared with the enormous contribution of influence, the indirect effect of organizational support and emotional intelligence on OCB through organizational commitment contributed more than the direct effect without going through organizational commitment as a mediating variable.
Further, theoretical implications provide that organizational aspects in the form of support for employees and individual aspects related to emotional intelligence are essential in conditioned organizations to generate levels of organizational commitment and organizational citizenship behavior (OCB), which contribute positively to the organization. Therefore, further research exploring other predictors needs to be conducted, among others, related to the diversity of individual employee characteristics. Further research with a larger sample size also needs to be done. It is related to the limitations of this research in the number of respondents. Thus, more significant respondents will allow the generalization of research findings and more accurate scientific justification.
Practically, the leader of the Central Bureau of Statistics in Medan City should try to increase the positive attitude of employees, especially organizational commitment, because it plays a significant role in generating high employee OCB, among others, by providing more intensive organizational support as a strong organizational predictor to generate a high level of employee organizational commitment. Organizations also need to conduct in-depth personality tests to detect the level of individual emotional intelligence of employees who work in the field and interact with the community for better work effectiveness.