The role of religiosity in moderating the influence of servant leadership on job satisfaction Abdul Jamala, Budiyantoa and Agustedia*

Article history: Received: September 25, 2020 Received in revised format: September 3


Introduction
The results of the empirical research by several researchers, namely McCann et al. (2014), Girei (2015), Agustine and Muslimah (2016), Harwiki (2016), and Koesmono (2014) concluded that servant leadership influences the employees' performances, but on the other side there is a different result which conclude that servant leadership not influence the employees' performances, which is the result from , Lisbijanto and Budiyanto (2014), Darto et al. (2015), and Fahrura (2016). The result of the research from Judge et al., (2001), Hua (2008), Ghozali (2002), Biancha et al. (2013), Perera et al. (2014), Murgianto and Suhermin (2016) and Nurnaningsih and Wahyono (2017) concluded that job satisfaction influences the employees' performances. But there is a different result from the empirical research from Darmanegara (2016), Pamungkas (2016), Subekti and Setiadi (2016), Kristine (2017), Can and Yasri (2015), and Handoko (2010), who stated that job satisfaction not influence the employees' performances. Thus, the researchers can conclude that there are no consistent results from the previous study and this research gap about the variables of servant leadership on job satisfaction needs further research. The selection of variables in this study is based on the previous literature review which indicates the research gap and as a solution from the research gap, the researchers used a moderating approach. It is based on empirical and theoretical studies and is supported by the phenomena in the field, the researchers suspect that there is another variable which is an exclusive moment and can fill the research gap, which is religiosity variable (Fitrio et al., 2020). The researchers hope that the variable of religiosity as a moderation could strengthen the influence of servant leadership on the employees' performances that is an upgrading in this research.

Literature Review and Hypotheses Development
Previous research about the Influence of Servant Leadership and Job Satisfaction on the Employee's Performance can be seen on the Table 1:

The Influence of Servant Leadership on the Employees' Performances
Basic concept of servant leadership starts from a natural feeling that the leader wants to serve first. Even by a complex definition of servant leadership, the basic idea of servant leadership placed the interests of participants above their own interests and emphasized the participants' development. Servant leadership is sometimes treated as a trait, but in the context of the research servant leadership is seen as a behavior. Several researches which related with this research like the result from McCann et al. (2014), Awan et al. (2012), Girei (2015), Agustine and Muslimah (2016), Harwiki (2016), and Koesmono (2014) concluded that servant leadership influenced on the employee's performance. Linden et al. (2008), stated that servant leadership can be measured with indicators of (1) recover emotions, (2) create some values for community, (3) forming a concept, (4) empowering, (5) help the participants grow and success, (6) prioritizing the participant, and (7) behave ethically.
H 1 : Servant leadership has a significant effect on the employees' performances.

The influence of Job Satisfaction on the Employees' Performances
Job satisfaction is an employees' positive feeling on the result of work from the evaluation of the various characteristics that have been determined. An employee job satisfaction is the sum of several different elements in accordance with the existing line of work in the organization. Some of the results show that employees are generally more satisfied with their work as a whole than with their colleagues, salaries, and promotion opportunities. The result from Robbins and Judge (2015: 49-50), claim that employee's satisfaction does not depends on the salaries and promotion opportunities, because according to Davis and Newstrom (1985) satisfaction involves many dimensions, in general it involves two aspects, which are satisfaction on the job itself and satisfaction on the work environment (colleagues, work condition, supervisor and organization). Robbins and Judge (2015:50) said that job satisfaction can be measured with some indicators of (1) suitability of work, (2) supervision, (3) salary/allowance, (4) promotion, and (5) colleagues.
H 2 : Job satisfaction has a significant effect on the employees' performances. Ancok and Suroso (2008:124) stated that, in the aspect of life and job, religion or religiosity has a several functions: 1) Religion as a source of knowledge and ethics, 2) as a justification tools and hypotheses to prove the truth, 3) Religion as a motivator, and 4) Religion drives a social surveillance function (guardian of morals) so that each individual responsible in activities. Ghozali (2002:2) said that religiosity leads to the quality of appreciation and someone's life attitude based on the religious values that are believed with an emphasis on the substance of noble values also tend to turn away from the religious formalism. Zubairu and Sakariyau (2016) stated that there is no significant correlation between religiosity and accountant students' academic performances at International Islamic University Malaysia (IIUM). The result concluded that each individual does his/her best and does not guarantee to get the highest score, because he/she has different talents given by God in various areas of life. The researchers have not found the variable of religiosity as a moderate which plays a role to influence servant leadership on the performance. In accordance with the role of religiosity moderation, when servant leadership plays a good role and the employee's understanding on the religiosity is higher, then it can be predicted that religiosity will play a role as a moderation which can strengthen the influence of servant leadership on the performance, or vice versa the religiosity weakens the influence of servant leadership on the performance if the role of leaderships is weak and the employees' understanding of religiosity is low.

The moderating effect of religiosity on relationship between servant leadership and employee performance
H 3 : Religiosity moderates the effect of servant leadership on the employees' performances.

The moderating effect of religiosity on the relationship between Job Satisfaction and Employee performance
Satisfaction involves many dimensions, but in general it involves two aspects, which are satisfaction on the job itself and satisfaction on the work environment. The meaning of job satisfaction most widely used is created by Spector (1997) which explains that most of the job satisfaction is centered on the individual feeling about a whole work which concerns on the level where people love or hate their jobs. Until now, the researchers have not found that religiosity is a variable of moderation that influences job satisfaction on the performance. In reality, job satisfaction is related to the role which is performed by employees while working. If the employees did their job well and the company gave rewards according to the level of work which is done, then the employees will be satisfied to accept that. Or, the employees have worked well but the company did not give welldeserved rewards, so the employees will not feel satisfied. As the role of religiosity moderation, if the job satisfaction is achieved and the employees' understanding on religiosity is high, then it can be suspected that religiosity will play a role as a moderation which can strengthen the influence of job satisfaction on the performance, or vice versa the religiosity plays a role to weaken the influence of job satisfaction on the performance If there is no job satisfaction and the employees' understanding of religiosity is low. Religiosity indicator according to Glock and Stark (1965), Ancok and Suroso (2011) can be measured through (1) ideology/beliefs, (2) religious knowledge / intellectual, (3) religious practices/rituals, (4) experiences, and (5) consequences of religion.

968
H 4 : Religiosity moderates the relationship between job satisfaction and the employees' performances.

Research Questions
Research questions of the research are as follows: a. Does the servant leadership have any significant effect on the Pekanbaru Education Office employees' performances? b. Does job satisfaction have any significant effect on the Pekanbaru Educational Office employees' performances? c. Does religiosity moderates the influence of servant leadership on the Pekanbaru Educational Office employees' performances? d. Does religiosity moderates the influence of job satisfaction on the Pekanbaru Educational Office employees' performances?

Instruments Testing
The research instruments testing can be done by doing validity and reliability tests. The instruments can be valid if the correlation coefficient (r count) > 0,3 while if the correlation coefficient (r count) < 0,3 then can be concluded that the statement item is not valid (Sugiyono, 2016:183). The result test of all the statement items has a correlation coefficient score > 0.3 then it can be concluded that all of the statement items are valid. Next, for the research reliability variables can be said reliable if the score of alpha Cronbach coefficient > 0,60 it means the level of reliability for 0.60 is indicating the reliability of a construct. (Sugiyono 2016: 184). The test result of all the statements with Cronbach alpha score > 0,6 shows the statements items from each research variable could create the construct of variables. Therefore, the list of whole statements of the questionnaire is valid and reliable to measure each variable.

Analysis of Measurement Model (Outer Model) Validity Test
The following are the results of the validity test obtained in Fig. 2.

Fig. 2. Validity Testing
The leading factor score is 0.7 or more considered very satisfying (Henseler et al., 2009;Gotz, 2010) meanwhile the indicator which have the leading factor score same with 0.5 can be considered accepted, while the indicator with the leading factor score smaller from 0.5 must be issued (Chin, 1995;Hair et.al., 2010). From Figure 2 can be seen that all of the leading factor scores are bigger than 0.7 it means all of the variable indicators of the research are valid.

Reliability Test
The following are the result test of reliability which obtained: Reliability tests in PLS can use two methods, which are Cronbach's alpha and composite reliability. Cronbach's alpha measures the lower limit of reliability score whereas composite reliability measures the real reliability score of a construct (Chin, 1995). Composite reliability is considered to be better in estimating internal consistency of a construct (Werst et al, 1974). Rule of thumb of Cronbach's alpha score and composite reliability should be bigger than 0.70 although 0.60 can be accepted (Hair et al, 2010). From Table 2 we can see that Cronbach's alpha and composite reliability from the variable of servant leadership, job satisfaction, religiosity, and employees' performance are bigger than 0.70, which means the construct of the research model is reliable. Table 3 shows the results of analysis of determinacy coefficient or R-Square. The results of R 2 are 0.67, 0.33 and 0.19 for endogen variables in the structural model indicating the "good", "moderate" and "weak" model (Ghozali, 2012). Goodness of 970 fit on the PLS model can be known from the R 2 score, the higher R 2 score then the model can be said to be more fit. The calculation result of R 2 from the employees' performances variable shows the score of 0.912 is higher than 0.67, which means the model is already fit, the great diversity of the research data can be explained by the structural model and has a relevant good score which is almost close to one.  Table 4 shows the results of hypotheses testing in this study. Ghozali (2012) stated that coefficient path or inner model score which been showed by the score of T statistic must be above 1.96 to test the hypotheses on 5 percent alpha.

H 1 : Servant Leadership Has a Significant Effect on the Pekanbaru Educational Office Employees' Performance.
The influence magnitude of servant leadership on the employees' performances is 0.601 with the t-statistics is 14,204 and the score of the t-statistics is bigger than 1.96 and P=Value 0,000, then it can be concluded that servant leadership has a significant effect on the performances. This number shows that the better the implementation of servant leadership which is carried out by the leader of Pekanbaru Education Office then the employees' performances will be better. Therefore, the first hypothesis which stated that servant leadership has a significant effect on the Pekanbaru Education Office employees' performances can be accepted (H 1 accepted).

H 2 : Job Satisfaction Has a Significant Effect on the Pekanbaru Education Office Employees' Performance.
The influence magnitude of job satisfaction on the employees' performances is 0.401 with t-statistics 6,102 and the score of tstatistics is bigger than 1.96 and P-Value 0,000, then it can be concluded that job satisfaction has a significant effect on the employees' performances. This number shows that the higher level of job satisfaction felt by employees then the employees' performances will get higher. Therefore, the second hypothesis which stated that job satisfaction has a significant effect on the employees' performances who work at Pekanbaru Education Office can be accepted (H 2 accepted).

H 3 : Religiosity Moderates the Influence of Servant Leadership on the Pekanbaru Education Office Employees' Performance.
The influence magnitude of religiosity and servant leadership interaction (Z×X 2 ) on the employees' performances is-0.093 with the score of t-statistics 2,702, where the score of t-statistics is smaller than 1.96 and the score of P-Value 0,007, then it can be concluded that the variable of religiosity moderates the influence of competence variable on the employees' performances. Therefore, it can be concluded that the role of religiosity as a moderation could affect (lowered) servant leadership on the employees' performances (H 3 accepted). From the result of the previous direct effect test, noted that servant leadership affect the employees' performances, however when the moderating variable for religiosity was added, then there is a (negative) effect which lowered the performances, which means the implementation of a good servant leadership and the high level of religiosity will decrease the employees' performances. Or vice versa in any other words, by decreasing religiosity then the employees' performances will increase.

H 4 : Religiosity Moderates the Influence of Job Satisfaction on the Pekanbaru Education Office Employees' Performances.
The influence magnitude of religiosity and job satisfaction interaction (Z×X 2 ) on the employees' performances is 0.016 with tstatistics 0,382 and the score of t-statistics is smaller than 1.96 and the P-Value 0,702, then it can be concluded that religiosity is not moderating the influence of job satisfaction on the employees' performances (H 4 rejected). This number shows that the role of religiosity as a moderation is not giving any influences of job satisfaction on the employees' performances. Job satisfaction directly could influence the employees' performances, both for the employees who have a high level of religiosity and for the employees who have low levels of religiosity.

Discussion
The result of research hypothesis testing shows that servant leadership has a significant effect on the employees' performances, which means the servant leadership gives a meaningful effect in increasing the employees' performances who work at Pekanbaru Education Office. Or, in other words, the higher level of servant leadership carried out by the leader, then the employees' performances will be higher. To get the higher employees' performances, then the organization needs to pay attention to the implementation of servant leadership, especially related to the indicators of restoring employees' emotions by listening and helping them to solve the problem. The result of hypothesis testing is consistent with the research which has been done by Meuser et al. (2011), Northouse (2013, Hussain and Ali (2012), Harwiki (2013), Setiawan (2015), andLiden et al. (2008). According to Meuser et al. (2011) servant leadership has a favorable impact on follower performance in their role, namely the way followers perform their assigned work, and followers become more effective at completing their work. Servant leadership is a new leadership style with a more orientation towards service to others by synergizing with subordinates, a sense of togetherness and sharing in making organizational decisions (Spears, 2005). Northouse (2013) suggests that leadership that serves a focus and is more sensitive to subordinates' problems, there is a sense of empathy and development for the better. The results of this study state that indicators that reflect the leadership of the Pekanbaru City Education office are helping employees grow and succeed. The results of hypothesis testing on the job satisfaction variable indicate that job satisfaction has a significant effect on employee performance of the Pekanbaru City Education Office, meaning that job satisfaction has a significant effect in improving employee performance. Or in other words, the more job satisfaction increases, the employee's performance will also increase. Thus, to produce good employee performance, the Pekanbaru City Education Office needs to pay attention to job satisfaction, especially those related to supervision indicators with the hope that the leadership will supervise and pay attention to the progress of work carried out by employees. The results of testing this hypothesis are consistent with research conducted by Kreitner and Kinicki (2014: 273) that employee job satisfaction is based on employee assessments of work and what the organization provides to employees, or based on their work performance. Gibson (2000) states that satisfaction is one of the criteria for effectiveness and is a measure of the success of an organization in meeting employee needs. There is a strong relationship between job satisfaction and employee performance (Robbins and Judge, 2001). Several other researchers, Ghozali (2002), Hua (2008) Murgianto et al. (2016), Javed, et al. (2014 , and Shahab and Nisa (2014) also stated the same result. This means that the higher the level of job satisfaction an employee has, the performance will also increase. From the results of this study it can be stated that the five indicators of job satisfaction, namely 1) reward system, 2) job suitability, 3) job challenges, 4) coworkers, and 5) work environment, support research conducted by Robbins and Judge (2015). Likewise, the results of research conducted by Ariffin and Said (2016) state that the dimensions of job satisfaction consisting of the job itself, salary, promotion opportunities, supervision, and co-workers have a significant effect on employee performance. The indicator that reflects the job satisfaction of employees at the Pekanbaru City Education Office is supervision. Gibson et al. (2000: 102) state that the task of supervising a leader is to provide support, provide and receive suggestions, and give praise to outstanding employees so that employee and organizational performance can improve. The results of the hypothesis test state that religiosity acts as a moderation that affects the leadership servant and the performance of the Pekanbaru City Education Office employees. After testing the path coefficient, it turns out that the moderation variable for religiosity is negative (-0.093), meaning that the role of moderation for religiosity is negative and significant. As previously explained, leaders who implement servant leadership can improve employee performance. But with a negative path coefficient, the role of religiosity will weaken the influence of servant leadership on employee performance. Thus, even though the path coefficient is negative, the third hypothesis which states that religiosity moderates the influence of servant leadership on the performance of the Pekanbaru City Education Office employees is acceptable. Based on the interaction relationship between the moderator variable and the predictor variable, the moderator variable of religiosity is a quasi moderator variable. Quasi moderator (religiosity) is a variable that moderates the relationship between exogenous variables and endogenous variables which are also exogenous variables. According to Ancok and Suroso (2011: 124) religiosity has several functions, including carrying out the function of social supervision (guarding morale). Thus, the lowest value perceived by employees on the religiosity variable is the consequence of religion with statements not to embarrass colleagues and not to seize work / job opportunities from other employees to support social / moral monitoring functions and according to researchers it is correct. Furthermore, the role of servant leadership that employees expect is how to recover emotions and leaders are willing to listen to, recognize and overcome problems faced by employees. The results of interviews by researchers with several leaders in the Pekanbaru city education office stated that most employee problems were how to complete the assigned tasks in accordance with predetermined main tasks and functions, including some personal problems. In completing work, some employees do not want to share experiences and do not want to help other employees with problems so that it affects the achievement of job targets and ultimately affects the employee performance benefits. The researcher suspects that in this section there will be pros and cons in the thinking and in the hearts of employees whether they will carry out their duties according to the prevailing duties and functions and carry out religious moral messages (religiosity) which are also an inseparable part of the work in their institution. From the research results, although the role of moderation of religiosity is 972 significant but with a negative path coefficient it can be ascertained that employees prioritize completing tasks by asking other employees who are willing to help rather than carrying out the moral message mandated by religiosity. The social / moral monitoring function of religiosity can be carried out inside or outside the organization. Thus, the results of this study can be stated, that the role of religiosity as moderation will weaken the influence of servant leadership on the performance of the Pekanbaru City Education Office employees. To improve employee performance requires a strong servant leadership role supported by a high level of individual religiosity. The level of religiosity in question is not just visible symbols, it is possible that some people with this religious symbol show that the individual is good but in everyday life it does not necessarily reflect the individual's good behavior. Researchers suspect that there are elements of culture and customs that influence religiosity and become a challenge for future researchers to research. Finally, from the results of the hypothesis test conducted, it shows that religiosity does not moderate the effect of job satisfaction on the performance of the Pekanbaru City education office employees. Thus, the fourth hypothesis, namely the role of religiosity in moderating the effect of job satisfaction on employee performance, is rejected. To produce good employee performance, the Pekanbaru City Education Office only needs to pay attention to employee job satisfaction without the need for a full role of religiosity, because it can be obtained from outside the education office such as attending recitation and religious lectures that are routinely carried out by religious institutions or mosques around the office. or around the employee's residence. To produce employee performance in accordance with agency targets, the Pekanbaru City Education Office needs to meet employee job satisfaction levels, especially indicators that are considered low, namely colleagues, especially in creating harmonious relationships and promotion opportunities so that employee careers can increase. The results of Ghozali's (2002: 10) research state that job satisfaction has a positive effect on employee productivity, and there is an indirect effect of religiosity (dimensions of trust, commitment, and behavior) and organizational commitment and job involvement on employee job satisfaction.

Conclusions and Implications
Based on the results of the research and discussion in the previous chapter, the several conclusions can be drawn to answer the research questions as follows: a. Servicing leadership has a positive and significant effect on the performance of employees who work at the Pekanbaru City Education Office. The first hypothesis can be accepted, meaning that the leader or head of the service who runs the servant leadership is able to improve the performance of his employees. b. Job satisfaction has a positive and significant effect on the performance of employees who work at the Pekanbaru City Education Office. The second hypothesis can be accepted, it means that an increase in employee job satisfaction can improve their performance. who work at the Pekanbaru City Education Office. c. Religiosity does not play a role as moderation affecting job satisfaction with the performance of the Pekanbaru Ci d. Religiosity acts as a moderation (negative and significant) to weaken the influence of the servant leadership on the performance of the Pekanbaru City Education Office employees. The third hypothesis can be accepted, which means that the role of religiosity has not been able to strengthen the influence of servant leadership on the performance of employees by Education Office employees. The fourth hypothesis is rejected, because the role of religiosity is not able to increase the effect of job satisfaction on the performance of employees who work at the Pekanbaru City Education Office, although directly job satisfaction can improve employee performance.
The implication of the results of this study can identify that the variable of religiosity has not been able to strengthen the influence of the leadership that serves the improvement of the performance of the Pekanbaru City education office employees, so that educational institutions / offices need employees who carry out high religiosity so that it is truly reflected in the implementation of daily tasks at the same time. as an implementation of competitive advantage according to Resources Based View Theory.