THE ROLE OF EMPLOYEE DEMOGRAPHICS, WORK STRESS, AND REWARDS ON JOB SATISFACTION AND EMPLOYEE PERFORMANCE

: This research was conducted at a private university in East Java (University X). The purpose of this study was to examine demographic factors as moderating variables on the relationship between job stress and job satisfaction and the relationship between rewards and job satisfaction and measure the effect of job satisfaction on job performance at the university. 106 administrative staff and 101 lecturers are the samples in this study. The sampling technique was employed using simple random sampling. Structural Equation Model used to analyze the multivariate data. Results showed that employee demographics of both respondents do not act as moderators on the relationship between job stress and job satisfaction and the relationship between rewards and job satisfaction. Likewise, the job satisfaction variable in this study does not provide a mediating effect on the relationship between job stress and employee performance and the relationship between rewards and employee performance. Significant results for administrative staff are shown in the relationship between rewards and job satisfaction and the relationship between rewards and employee performance. As for lecturers, significant results are shown in the relationship between job stress and performance and the relationship between reward and job satisfaction. Further research suggested conducting workload analysis for each position, developing an appropriate reward system, further research to determine the factors that can predict performance, and research continued by using unstructured demographic data, such as income levels or workloads.

The success of a university in a disruptive and competitive era to be the best and achieving its goals is strongly influenced by many aspects. Human resource per for-mance is one of those that influence success in the university. Universities that are not optimal in stimulating the performance of vulnerable employees experience a quality slowdown. This research takes a setting at a University in East Java (University X). As one of the young private universities, University X progressively seeks to compete with other universities. The intense competition between universities makes the quality improvement and quality of services from the University of X carried out by lecturers and administrative staff a fundamental aspect. Lecturers cause this. Administrative staff and university leaders are the main drivers of creating innovative academic and non-academic works. Collaboration and synergy of all these elements is a determining factor that accumulates the university's overall performance (Bear and Woolley, 2011). Understanding the factors that may affect employee performance will provide great benefits for universities, especially in terms of more effective employee management.
The role of human resource management (HRM) in a university is very important to monitor the performance and quality of employees. As humans, employees have thoughts, feelings, desires, status, and backgrounds brought into the organization and influenced the evaluation results of their work. The results of this evaluation are very important in determining how to work achievements, dedication, and love in work charged to employees. When an employee's evaluation r esults are achieved, it will be more opportunities for performing, providing dedication and loyalty to the organization. The results of the employee's evaluation are known as employment satisfaction (Kampkötter, 2014;Locke, 1969).
Job satisfaction is one of the psychological factors that can affect the performance of an employee. Many studies argue that job satisfaction determines the performance of an employee (Ayundasari et al., 2017;Mulia, 2013;Tentama and Kusuma, 2017;Wright and Cropanzano, 2000). If individual employees feel satisfied in the work and life, it will have a positive effect on the employee's performance. Conversely, if the employee is not satisfied, it will negatively influence the work of the employee. The link between job satisfaction and performance is a relationship that is not yet conclusive. Previous studies stated that employee satisfaction on employees has a positive effect on employee performance (Ahmad et al., 2017;Bakan et al., 2014;Perera and Weerakkody, 2016;Syaharuddin, 2016), but other studies state that job satisfaction and the performance of employees have pseudo relations, or there is no influence between job satisfaction and employee performance (Bowling, 2007;Meirina, 2013). That becomes interesting for further investigation, whether job satisfaction in University X employees plays a role in improving employee performance.
One factor affecting employee job satisfaction is work stress (Jehangir et al., 2011;Nanjundeswaraswamy, 2019). Based on experts' explanations, work stress can have a good impact (Eustress), and the adverse impact on employees and organizations (Le Fevre et al., 2006). Eustress occurs when the workload makes someone challenged, conscious, and grow. But excessive work stress tends to make employee performance unproductive, ineffective, and inefficient. So, a balance between work and life is needed. The university as an organization always strives to create a variety of supportive working conditions to suppress the work stress of employees, one of which is through rewarding.
Through giving rewards to lecturers and administrative staff, the university recognizes that the work performance possessed by university employees is part of the university's achievements. Research that examines the role of awarding in improving performance has been carried out (Jiang et al., 2009;Minarsih, 2009;Mustapha, 2013). However, research carried out in the university setting has not been done much. Knowing the dynamics of job satisfaction and the performance of university employees through the aspect of giving awards is very useful to see the university's gait as a humanist organization.
Job satisfaction is not obtained in a vacuum space. That is, there is a role of socio-cultural factors that do it. That is adopted by individuals and demographic factors such as age, gender, marital status, education, and life level (Sule, 2002). University employees have heterogeneous backgrounds. Based on field data, researchers noted a variety of socio-cultural profiles at universities in Indonesia. That shows that university employees have different dependent and psychological pressure profiles even though they have a definite role.
Heterogeneous demographics can cause a difference in response to everything in the university. Each individual has a different level of satisfaction by following the system values that apply to him (Badriyah, 2015). On this basis, this research also considers the involvement of demographic aspects in predicting performance.
Various phenomena and previous research results show that performance is a complex dependent variable and must be approached by considering various dimensions. This study is important because the authors seek to explain performance through a combination of psychological variables (job stress and job satisfaction), organizational variables (rewards), and social variables (demography) in a research model consisting of a combination of moderating and mediating variables. The researcher also wants to explain how the influence of these variables on two types of respondents who have different functions and characteristics of their work in a university (lecturers and administrative staff). The main purpose of this study was to examine the effect of employee demographics, job stress, rewards, and job satisfaction and their effect on employee performance. Expected after learning the effect of these factors, universities can establish appropriate employee management policies related to employee performance and employee job satisfaction, especially in the context of universities.

HYPOTHESIS DEVELOPMENT
Job stress can occur due to too many tasks, time constraints, role ambiguity, differences in values within the company, frustration, and family environment (Pujiastuti, 2013). Excessive stress can threaten the ability of employees to deal with it, which results in disruption of the implementation of tasks and work performance. Job stress also affects the level of employee job satisfaction. Untreated stress on employees can decrease employee job satisfaction (Afrizal et al., 2014;Jehangir et al., 2011;Khan et al., 2014).

Figure 1. Research Concept Framework
Job stress does not always have a negative effect but also has a positive influence on the company. At a certain level of stress, stress is expected to spur employees to complete work as well as possible. Employees who are in conditions of work stress will show changes in behavior. At low-stress levels, employees may not be challenged or engaged enough to perform at their best. At the optimal level of stress can make employees give their best performance. More than the optimal level of stress, performance begins to deteriorate. Excessive levels of stress can cause employees to be too nervous or feel threatened so that they cannot perform well (Certo, 2003). That leads to two current research hypotheses: H1 : Job stress affects the job satisfaction H2 : Job stress affects employee performance One of the factors that have a significant role in influencing job satisfaction is salary and rewards. Bustamam et al. (2014) found that financial rewards significantly impact frontline job satisfaction at four and five-star hotels in the Klang Valley, Malaysia. In their research, Ahmad et al. (2017) also stated that awards and recognition play an important role in improving employee performance and satisfaction and gaining a competitive advantage in global business.
Employee rewards and performance have a positive influence relationship (Jiang et al., 2009). There are two sides of interest in rewarding, namely from the employee and the organization. Employees motivated by the reward will work harder, which is very beneficial for the organization and the workers themselves in the long run. For organizations, the provision of rewards can lead to an increase in organizational performance due to the enthusiasm and enthusiasm of employees (Markova and Ford, 2011). That leads to two current research hypotheses: H3 : Rewards affect job satisfaction H4 : Rewards affect employee performance High job satisfaction on employees will support the level of employee performance. Employees will be motivated to work hard and achieve higher per-formance when they get job satisfaction. On the other hand, employees who do not get job satisfaction will have a negative impact on the employee's performance and can even have a bad impact on the environment in which the employee is assigned (Syaharuddin, 2016;Perera and Weerakkody, 2016). This research supports previous research by Bakan et al. (2014), which states that employee job satisfaction positively affects employee performance and organizational commitment. That leads to one current research hypothesis: H5: Job satisfaction affects employee performance Job stress and employee job satisfaction partially and simultaneously can affect employee performance. Adawiyah and Siswanto (2015) found that job stress indirectly affected performance through employee job satisfaction. Employees will continue to provide high performance if it is supported by high job satisfaction, even though work stress factors influence it. When employees feel satisfied at work, they will be able to work together with coworkers and appreciate the work done by their coworkers (Adawiyah and Siswanto, 2015;Sanjaya, 2012).
Rewards can be one of the determinants of employee job satisfaction. In addition, giving rewards to employees is to encourage employees to achieve high levels of performance. When the company gives an award, it can motivate employees to perform even better (Aini and Frianto, 2020); Aktar et al., 2013); Wasiati, 2018). That leads to two current research hypotheses: H6: Job stress affects employee performance by mediating job satisfaction H7: Rewards affect employee performance by mediating job satisfaction Some studies have shown that employee demographics can influence the relationship between job stress and job satisfaction and the relationship between rewards and job satisfaction. Kim et al. (2009) found that age, gender, tenure, and employee status strengthen the negative influence between job stress and employee satisfaction. Rehman et al. (2010) stated that employee age moderates rewards and employee job satisfaction. Tausif (2012) also explains that the difference in employee age affects the relationship between rewards and employee job satisfaction. Employee job satisfaction increases with the increasing age of employees. This leads to two current research hypotheses: H8 : Employee demographics moderate the relationship between job stress and job satisfaction H9 : Employee demographics moderate the relationship between rewards and job satisfaction

METHOD
The type of research used is multivariate quantitative research, consisting of four types of variables, namely independent variables (job stress and reward), dependent variables (employee performance), mediator variables (job satisfaction), and moderator variables (employee demographics). The population in this study were all full-time employees of University X and had a score of the Performance Appraisal with a total population of 263 people consisting of 127 administrative staff and 136 lecturers. The samples used in this study were obtained using simple random sampling techniques, and the Slovin formula calculated the sample size: (1) Note: n = the number of samples needed; N = population size; E = 5% sampling error Obtained sample size 96 education persons and 101 educators. The sample in this study was distinguished between lecturers and administrative staff because they showed different work characteristics. Researchers indicated differences in work stress and views related to the reward and job satisfaction experienced by the two types of employees. Data collection sources and techniques are carried out as follows primary data and secondary data. Primary data consists of demographic characteristics, job stress, reward, and job satisfaction from the questionnaire distributed to respondents directly online in December 2020. Secondary data used are employee performance variables obtained from University X data.
This study used the Partial Least Square (PLS) analysis method using SmartPLS 3.0 program. In Partial Least Square (PLS), there are two measurement models: the outer and inner models. The outer model analysis consists of validity and reliability tests. Validity was measured using convergent validity (loading factor value >0.7 and Average Variance Extracted (AVE) value >0.5) and discriminant validity (cross-loading value >0.7 in one variable). Reliability was measured using composite reliability. Good results are indicated by the value of composite reliability >0.7 and Cronbach's alpha >0.6 (Ghozali, 2015). The inner model analysis consists of an R-square test that shows the strength of the independent variable in influencing the dependent variable (R-Square value is 0.7, it is declared strong, 0.33 is moderate, and 0.19 is weak). The T-value test indicates the significance level of the hypothesis test (T-statistic value must be above 1.96 for the two-tailed hypothesis). Intervening variable analysis is using estimation of direct-indirect effect and VAF (Variance Extracted For) method for advanced mediation testing.

Respondent Demographic Profiles
Respondents in this study were dominated by female employees, namely 63.21% for lecturers and 50.50% for administrative staff. Based on age, employees were the most likely to be aged 30-40 years. While in the second place, there is a difference between administrative staff who are dominated by the age <30 years and lecturer with an age range of 41-50 years. This could be due to the different needs for the two types of workforce. The minimum requirement to become a lecturer is to have a master's level of education. That will affect employees' age because they also need a minimum of two years to continue their master's studies. While the requirements to become an administrative staff are sufficient for the undergraduate level. Based on the tenure, the employee demographics are dominated by 5-10 years of service. Then for the tenure is below five years is in the next order. As for the tenure of >10 years, it ranks last. That is due to the relatively young age of the university, so there are not many employees who have more than ten years of service. In addition, based on marital status, employees at University X are dominated by married employees, namely 61.32% for administrative staff and 72.28% for lecturers. The estimation results of cross loading in Table  2 show that the loading value of each indicator on the latent variable is greater than the cross-loading value. It can be concluded that all latent constructs or variables have good validity.

Composite Reliability
The reliability of measuring instruments for educational staff and teaching staff respondents in the table above has a good value. Shown by a Cronbach's Alpha score > 0.6 and Composite Reli-ability> 0.7.    The hypothesis testing table above represents Hypothesis 6 and Hypothesis 7 on the teaching staff was rejected because the t-value results were below the standard in the two-tailed analysis (<1.96).

Variable
The next test is analyzing intervening variables by comparing T-statistics on a direct relationship with an indirect one. Based on the test results, the following results were obtained in Table 8.
In the results of this study, the intervening analysis showed insignificant results and there was no mediating effect of job satisfaction. Therefore, this study did not conduct a VAF analysis.

DISCUSSION Job Stress, Job Satisfaction, and Job Performance
The analysis results on examining the relationship between job stress-job satisfaction and job stress-performance show that job stress is not a predictor of job satisfaction for both administrative staff and lecturers. However, job stress can be a good predictor of an increase or decrease in the performance of lecturers. That is thought to be related to the difference in the complexity of the work between the administrative staff and the lecturer, the structural positions currently being held in the university, other supporting elements being worked on, and so on. That is supported by several previous studies which explain that work stress has implications for performance (Ahmad et al., 2017;Akbar et al., 2016;Jehangir et al., 2011), especially in jobs that are oriented towards interpersonal services (LeBlanc, 2009), such as the profession of health workers, lecturer, customer service, and so on. Job stress affects performance for types of work that have a higher nature of social and intellectual interaction, such as lecturers (Taris et al., 2001). The high administrative demands of lecturers, such as performance reporting through various DIKTI platforms, can also trigger work stress on lecturers. With these administrative demands, lecturers are required to know technology quickly (Akbar et al., 2016).
Based on the hypothesis testing table for the educational staff above, the results show that only Hypothesis 3 and Hypothesis 4 are accepted.
Based on the hypothesis testing table for the educational staff above, the results show that only Hypothesis 2 and Hypothesis 3 are accepted.

Intervening Variable Analysis Variable
Hypothesis

Reward, Job Satisfaction, and Job Performance
The analysis results on testing the reward-job satisfaction and reward-performance relationship show significant results for the administrative staff. The higher t-value on the relationship between reward and job satisfaction on the administrative staff shows that their work-life is influenced by two factors which are necessities based on Herzberg's Two Factor theory, namely:

Hygiene factors
This factor is related to basic human needs and continuously. The absence of this factor can trigger an employee's dissatisfaction.

Motivation factors
Motivation factors are related to employee psychology. Job satisfaction will increase motivation which can result in good employee performance. The absence of this factor will result in increased job dissatisfaction. Based on this theory, it can be concluded that University X has provided a big motivating factor for administrative staff. They feel motivated at work, but they still have complaints about their basic needs or working conditions, so perceived job satisfaction could not improve employee performance. Institutions focus more on improving motivational factors but have not maximally improved hygiene factors.
Employee rewards can only predict job satisfaction but are not predictors for a lecturer's performance (Harvey-Beavis, 2003). Even though University X has given awards to each of its employees, the rewards do not affect the performance of lecturers. That is possible because the University X award system has not met the lecturers' expectations (Djoemadi and Noermijati, 2014). Previous research states that high performance-based extrinsic rewards for lecturers can increase dislike of management and reduce employee loyalty. That can result in decreased productivity or performance (Ramirez, 2001).
This study also shows that job satisfaction affects employee performance, both for administrative staff and lecturers. That contradicts previous research, which states that job satisfaction affects employee performance (Ayundasari et al., 2017;Bakan et al., 2014;Kawiana et al., 2020;Perera and Weerakkody, 2016;Rafsanjani et al., 2019;Syaharuddin, 2016). These results indicate that job satisfaction is not a predictor for employee performance in University X. Satisfied employees do not necessarily perform well. On the other hand, dissatisfied employees do not necessarily give poor performance. According to Suzzana (2007) in Hidayat et al. (2012), job satisfaction only motivates employees to work but not improves their performance. Good employee performance is not always influenced by job satisfaction. Apart from job satisfaction, other factors impact strengthening or weakening employee productivity or performance. These factors include individual characteristics of their work, locus of control, and trust in management (Bowling, 2007;Rich, 1997).

Moderating Effects of Employee Demographics
Employee demographics that act as moderating variables in this study are gender, age, tenure, and marital status. The results showed that these variables did not moderate job stress-job satisfaction and reward-job satisfaction. Previous results suggest that individual characteristics have inconsistent effects in individual studies. Several studies have shown a positive effect of individual characteristics on employee job satisfaction (Kim et al., 2009;Tausif, 2012;Vadivu, 2017;Rehman et al., 2011). Another research group states that individual characteristics do not affect employee job satisfaction (Farla WK, 2016;Hidayah, 2012).
The characteristics of work in education tend not to differentiate between male and female employees, especially for educators. They have the same roles, duties, and responsibilities that do not require gender discrimination. For administrative staff, gender differentiation only occurs for certain work types, such as building maintenance work. Whereas for other types of work for educational personnel, there is no gender distinction. For example, for the position of secretary, the stigma that develops in society shows that the position of secretary will be better if female employees do it. However, at University X, a male secretary plays a role in the organization's operations. That is in line with the opinion of previous research, which states that there are no consistent differences between men and women in the ability to solve problems, analytical skills, competitive drive, motivation, social skills, or learning abilities (Farla WK, 2016).
Some researchers state that age influences job satisfaction (Tausif, 2012;Vadivu, 2017;Rehman et al., 2011). Middle-aged people tend to be more satisfied with their work because they already know the organization they work for. That creates a strong psychological bond with the organization and its coworkers. Another factor that causes older employees to be more loyal is to start a career again in a place that is not easy for them anymore. The findings in this study contradict the opinion of previous studies, where age has not affect the level of job satisfaction. That is possible because of the friendly working atmosphere at University X. Every elderly and younger employee has the same opportunity for self-development. Employees with elderly people are not reluctant to associate with employees with young age, which causes younger employees to feel free to interact with them.
Employee age is not necessarily linear with employee tenure, which means elderly employees are not necessarily senior employees at University X. On the other hand, younger employees are not necessarily junior employees at University X. Previous studies have shown that tenure has a relationship with employee job satisfaction (Vadivu, 2017). The analysis results in this study contradict previous studies, where tenure did not moderate the relationship between the independent and dependent variables. This result is possible because there is no culture of seniority at X University. Both senior and junior employees have the same duties and responsibilities to achieve organizational goals and have the same opportunity to develop.
Previous studies have shown that an employee's marital status influences job satisfaction (Farla WK, 2016;Khan et al., 2014;Vadivu, 2017). Employees with unmarried status have higher job satisfaction than married employees (Sutanto and Anggraini, 2011). Still, there are also studies that state that married employees have a higher level of satisfaction than employees with unmarried status (Jain and Nair, 2019). That is possible because there is a factor in the importance of a job in a person's career, apart from their marital status, which results in employees developing positive feelings about their work by ignoring their negative aspects. In addition, even though someone is not married, it does not mean that they do not have a family burden (Azim et al., 2013). As a collectivist society, the culture in Indonesia shows that an employee, even though he is not married, has a sense of obligation to support his family (father, mother, or siblings).
The absence of the influence of employee demographics in increasing job satisfaction can occur due to organizational culture factors. Organizational culture is the values adopted by the institution and its respective employees, which form the basis of individual behavior. University X itself has strong values shared by all employees, both male and female, young or old, senior and junior employees. One of these values is professionalism. Based on the results of the research, the job satisfaction is not influenced by demographic factors can represent that there is high professionalism for each employee following the culture of University X. This is following statement by Wani (2017), which states that employees are motivated well because they live according to the organizational culture of the company regardless of the demographic factors inherent in each individual employee.

CONCLUSIONS
Based on the results of the previous discussion, it is known that the job satisfaction of each employee does not determine the performance of X University employees. There are differences in the predictors of determining employee performance between administrative staff and lecturers. For administrative employees, the predictor variable for employee performance is a reward, while for lecturers, the predictor variable for employee performance is job stress. Job stress in each type of respondent is not a predictor of employee job satisfaction. Still, the reward can be a predictor of job satisfaction at University X. In addition. Employee characteristics factors consist of gender, age, years of service, and marital status cannot predict University X employee's job satisfaction.

IMPLICATIONS Implications for lecturers
Job stress on lecturers can affect their performance. That is due to the complexity of the work held by the lecturers, the structural position being held, the supporting elements being worked on, and so on. In addition, work stress will affect performance for types of work with a high social and intellectual nature of the interaction, such as lecturers. In other words, a lecturer's performance is determined from the intrinsic conditions of the main task and psychological conditions and personality tendencies, such as how susceptible an individual is to stress. Therefore, universities can continue to apply innovation through technology to create work programs that are more effective and efficient in assisting the main tasks of lecturers. Rewards only affect job satisfaction (psychological effects) but not on the lecturers' performance. That means that the reward received by the lecturer confirms that the university appreciates the hard work that has been done. Rewards in this study do not affect performance, so they cannot be used as extrinsic motivation to improve lecturer performance. In lecturers, job satisfaction is not a determinant to provide the best performance. That can be caused by other aspects that can affect their performance, including work stress, intrinsic motivation such as love for work, individual employee factors, organizational policies, and others.

Implications for administrative staff
Job stress is not a factor that determines the satisfaction and performance of administrative staff. Based on the results of this study, the reward factor has a positive effect on the performance and job satisfaction of administrative employees. That shows that by giving rewards, the organization will increase job satisfaction and the performance of education personnel. Giving rewards can mean that the university appreciates the hard work of administrative staff and recognizes that they are an important part of the progress of a university. In contrast, job satisfaction is not a determinant for administration staff to provide the best performance. That is due to other aspects outside this research that can affect performance, such as the nature or character of a person towards his work, locus of control, or trust in management.

Demographic factors
Heterogeneous employee demographics do not affect job satisfaction felt by all employees. The existence of a cultural climate at University X that does not differentiate between male and female employees, young or old, senior, and junior employees, also affects the results of this study. The marital status factor also does not influence which can be caused by the existence of essential career factors in each individual and a culture of feeling to support family members. In addition, the organizational culture adopted by University X can also underlie the individual attitudes of employees in acting on each task and responsibility assigned to them.

LIMITATIONS
The sample size used in this study was not large (± 100 respondents), so it became one factor that caused the t-value test results to be insignificant. In addition, this study uses SmartPLS software in the data processing. One of the weaknesses of this software is that it is only used to see the effect between variables without testing the theory used. The independent variable used in this study was taken from the lowest job satisfaction score at University X. Therefore, this study only measures work stress and rewards on job satisfaction and employee performance, without exploring other variables that can be predictors of job satisfaction and employee performance. In addition, the demographic variables used in this study only cover nominal and interval data such as age, gender, education level. But it has not accommodated unstructured demographics in predicting stress and performance, such as income level or workload.

RECOMMENDATIONS
Perform workload analysis for each job title within the department at X University. This workload analysis can provide an overview and become a reference for institutions to optimize properly. So that with the right work stress can spur employees to give better performance. Develop an appropriate reward system by reviewing the awards types and measurement parameters for each award given, conducting Focus Group Discussions (FGD) with representatives of several departments or study programs to formulate the suitable types of awards. Further research can be carried out to determine what factors are capable of being predictors for the performance of each employee in the education and teaching staff, one of which is by conducting longitudinal research. Further research can also use other demographic factors, unstructured demographic data, such as income levels or workloads.