THE ROLE OF EMPLOYEE ENGAGEMENT ON JOB SATISFACTION AND ITS EFFECT ON ORGANIZATIONAL PERFORMANCE

The purpose of this research was to analyze the role of employee engagement from the perspective of job engagement and organization engagement on job satisfaction and its effect on organizational performance. This research was conducted at a multinational manufacturing chemicals company located in Tangerang and Cilegon. The population was approximately 121 employees consisting of three job levels: Manager, White Collar, and Blue Collar. A target sample determined approximately 93 employees. Sampling using a non-probability sampling approach with a quota sampling method. The questionnaire was distributed to the population, but only 86 respondents filled out and returned the questionnaire. Method of hypothesis testing using Partial Least Square of Structural Equation Modeling (PLS-SEM) approach. The results suggested that job engagement has a positive and significant effect on job satisfaction but does not significantly affect organizational performance. Next, organization engagement has a positive and significant effect on job satisfaction but does not significantly affect organizational performance. Furthermore, job satisfaction has a positive effect and significantly influences organizational performance. Future research is advisable to examine the relations of other variables such as workload, work-life balance, and implementation of an integrated management system, which, believed, can provide a comprehensive view of employee engagement, job satisfaction, and organizational performance.

of organization members' selves to their work roles; in engagement, people employ and express themselves physically, cognitively, and emotionally during role performances.
Employee engagement in this research is viewed from two perspectives, job, and organization engagement. What underlies engagement as a multidimensional construct is because this concept carries a multi-focus meaning, recognizing that employees are simultaneously engaged in two different objects: their job and their organization (Saks, 2006;Suhartanto and Brien, 2018). The concept of employee engagement is generally viewed as managing the discretionary effort. When employees have choices, they will act in a way that furthers their organization's interest (Bauer and Erdogan, 2012). When an employee is engaged, he is aware of his responsibility in the business goals and motivates his colleagues alongside the success of organizational goals; engaged employees go beyond the call of duty to perform their role in excellence (Anitha, 2014).
Engaged employees tend to be satisfied. People want to do meaningful work and contribute to organizational success. In addition, the interventions that create the conditions for meaningful work and enhance the capacity of employees to engage can have a positive effect on satisfaction as well (Macey et al., 2009). The employee's positive attitude toward his workplace and its value system is otherwise called the positive emotional connection of an employee towards his/her work (Anitha, 2014).
As part of managing human resources operations, it is essential to identify matters related to employee engagement that affect job satisfaction as the basis for formulating policies related to human resources, which can improve organizational performance. Several studies discussed the relationship between employee engagement and job satisfaction (Saks, 2006;Al-dalahmeh, 2018;Suhartanto and Brien, 2018;Saks, 2019), employee engagement and organizational performance (Tahir, 2013;Farndale et al., 2014;Otieno et al., 2015;Kazimoto, 2016;Al-dalahmeh, 2018;Arfeen and Shafi, 2018), and job satisfaction and organizational performance (Brahmasari and Suprayetno, 2008;Al-dalahmeh, 2018).
From the initial discussion with the company's representatives, it is noted that the employee engagement survey did not go as expected during the six years since the company was founded. It is difficult for management to identify matters related to employee engagement from the perspective of job and organization engagement, which affects job satisfaction as the basis for formulating policies related to human resources, which in turn can improve organizational performance to achieve the company's goals. Another interesting point is that very few studies have discussed the relationship between job and organization engagement on job satisfaction and its effect on organizational performance.
This research was conducted to assist the company in identifying and solving these problems and contribute to answering the research gap related to the role of employee engagement on job satisfaction and its effect on organizational performance. The objectives of this research can be described as follows: 1. Analyzing the effect of job engagement on job satisfaction. 2. Analyzing the effect of job engagement on organizational performance. 3. Analyzing the effect of organization engagement on job satisfaction. 4. Analyzing the effect of organization engagement on organizational performance. 5. Analyzing the effect of job satisfaction on organizational performance. Schaufeli et al. (2002) propose operationalization by describing engagement as a positive, fulfilling, work-related state of mind characterized by vigor, dedication, and absorption. Vigor is characterized by high levels of energy and mental resilience while working, the willingness to invest effort in one's work, and persistence even in the face of difficulties. Dedication is characterized by a sense of significance, enthusiasm, inspiration, pride, and challenge. Absorption is characterized by being fully concentrated and deeply engrossed in one's work, whereby time passes quickly, and one has difficulties with detaching oneself from work (Schaufeli et al., 2002).

LITERATURE REVIEW
One way for individuals to repay their organization is through their level of engagement. Employees will choose to engage themselves to various degrees and respond to the resources they receive from their organization (Saks, 2006). Observing Saks' (2006) operationalization of organization engagement, Farndale et al. (2014) argue that organization engagement can be defined along similar lines to work engagement but with an organization-level referent. When engaged, people are attached to their work role and are absorbed by enacting it. They invest a lot of their energy into performing the role, as it is an important part of their identity. They have internalized the goals and aspirations of the organization as their own (Fleck and Inceoglu, 2010).
Since Kahn's (1990) concept of engagement, most theory and application have emphasized consequences for employees and organizations (Akingbola and van den Berg, 2016). The research held by Akingbola and van den Berg (2016) in nonprofit organizations reveals that engaged employees are more likely to experience job satisfaction, demonstrate behaviors consistent with organizational commitment, organizational citizenship behavior, and less likely to have the intention to quit. Job satisfaction is an area of inquiry concerned with both sorts of influence: the influence of work organizations on people and the influence of people on work organizations (Rowden, 2002).
The energy and focus inherent in work engagement allow employees to bring their full potential to the job, enhancing the quality of their core work responsibilities (Anitha, 2014). According to Caplan (2013) in Al-dalahmeh et al. (2018), employee engagement increases productivity, and overall performance creates a better and more productive work environment and reduces non-attendance and turnover. Engaged employees want good communication with their superiors, work that has meaning for them and motivates them, and a safe working place. When these conditions are present, employees be-come engaged and, as a result, produce better financial results, are proud of their organizations, and demonstrate enthusiasm (Al-dalahmeh et al., 2018). One approach to measuring the performance of an organization is the Balanced Scorecard (BSC) developed by Kaplan and Norton in 1996 through four perspectives: financial, customer, internal business process, and learning and growth (Uniati, 2014). The four perspectives of the scorecard permit a balance between short-term and long-term objectives, between desired outcomes and the performance drivers of those outcomes, and between hard objective measures and softer, more subjective measures (Kaplan and Norton, 1996).

HYPOTHESIS DEVELOPMENT
Research conducted by Saks (2006) shows that job engagement mediates the relationship between the antecedents of employee engagement and job satisfaction. Align with Saks (2006), Suhartanto and Brien (2018) confirm that job satisfaction mediates the relationship between job engagement and organizational performance. Saks (2019) reviewed research by Saks (2006) confirmed that job engagement predicts job satisfaction. Based on those results, we propose the first hypothesis as follows. H1: Job engagement affects the job satisfaction The relationship between the employee and the business organization he works for is reciprocal (Suhartanto and Brien, 2018). The results of Suhartanto and Brien's (2018) research show that job engagement and affecting job satisfaction also affect store performance and employee performance. Hence, the second hypothesis formulated as follows. H2: Job engagement affects organizational performance Research conducted by Saks (2006) shows that organization engagement mediates the relationship between the antecedents of employee engagement and job satisfaction. Furthermore, Suhartanto and Brien (2018) reveal that job satisfaction mediates the relationship between organization engagement and organizational performance. Saks (2019) reviewed Saks (2006) and concluded that organization engagement predicts job satisfaction. Based on those results, we propose the third hypothesis as follows. H3: Organization engagement affects the job satisfaction According to Suhartanto and Brien (2018), an organization engaged employees might not necessarily engage in his or her job. For instance, an employee may engage in the business organization due to its favorable reputation. Still, he or she might not engage in his/her job due to the job position being not what he or she intended. Research conducted by Farndale et al. (2014) shows a positive relationship between organization engagement and organizational performance. We thus propose the fourth hypothesis below.
H4: Organization engagement affects organizational performance Maarif and Kartika (2013) describe the performance due to the company's real value, e.g., total sales, total production, quality level, cost efficiency, profit, etc. Research conducted by Brahmasari and Suprayetno (2008) and Al-dalahmeh et al. (2018) show that job satisfaction positively and significantly affects company performance. Hence, the fifth hypothesis formulated as follows. H5: Job satisfaction affects organizational performance In sum, the model in Figure 1 is proposed:

METHOD
This research was conducted at a multinational manufacturing chemicals company engaged in the organic base chemical industry producing specialty chemicals located in Tangerang and Cilegon. Data collection, processing, and analysis were carried out during the period June to July 2020. The data used in this research are primary and secondary. The primary data obtained directly from employees by filling out a closed questionnaire, and the secondary data obtained through literature studies relevant to this research.
After collecting respondents' answers to the closed questionnaire, data analysis was carried out in descriptive statistics and inferential statistics. Descriptive statistics are a type of data analysis used to analyze data by describing or describing the data that has been collected as it is (Sugiyono, 2017). On the other hand, inferential statistics (often referred to as inductive statistics or probability statistics) are data analyses used to analyze sample data. The results are applied to the population (Sugiyono, 2017).
The population in this research was 121 employees consisting of three job levels: Manager, White Collar, and Blue Collar. For data analysis and hypothesis testing, a target sample of 93 people was determined. Sampling was carried out using a nonprobability sampling approach with a quota sampling method. Quota sampling is a technique for determining a sample of a population that has certain characteristics to the desired quota (Sugiyono, 2017). The quota amount representing the Manager level is determined in the range of 20% -21%, the White Collar level is determined in the range of 31% -33%, and the Blue Collar level is determined in the range of 44% -46%.
The sampling was distributed to the entire population by using online questionnaires google form. The use of google form carried out considering the COVID-19 pandemic and the company's health protocols to minimize interactions between employees when this research was taking place. The questionnaires were distributed to the population, but only 86 respondents filled out and returned the questionnaires.
The scale used in this research was a Likert scale. The Likert scale is appropriate to test how strongly respondents agree with the statements in the questionnaire. Interval-shaped scales classify individuals according to certain categories and arrange the order of the groups and measure the magnitude of differences in preferences between individuals (Sekaran and Bougie, 2013). The Likert scale used in this research has six choice categories, ranging from 1 = strongly disagree to 6 = strongly agree, which is a modification of the 7-point score by eliminating 'neutral' or 'doubtful' points. The 6-point score category will stir up respondents to determine the direction of their perceptions more and avoid the tendency of neutral perceptions in each response. In terms of correlation between questions, the overall correlation of the questions was low for a scale of two, three, or four response categories. It would increase the number of response categories to about six (Preston and Colman, 2000).
For the significance and hypotheses testing with a significance level of  = .05, degree of freedom (df) = 82, and a two-tailed approach, then the ttable value is 1.989 (1.99). After collecting respondents' answers to the questionnaire, data analysis was carried out in descriptive statistics and inferential statistics using the Partial Least Square of Structural Equation Modeling (PLS-SEM) approach.
PLS-SEM is used to test the proposed research
In this research, the characteristics of the respondents consist of gender, age, education, years of service, and job level. These aspects play an important role in obtaining information related to perceptions of job and organization engagement, job satisfaction, and organizational performance. The respondents of this research consist of sixty-nine male respondents (80%) and seventeen female respondents (20%).
In the age characteristics data, the majority of respondents were in the age between 41-47 years old, approximately twenty-two respondents (26%), followed in sequence with age between 34-40 years old approximately twenty respondents (23%), the age between 48-54 years approximately nineteen respondents (22%), the age between 27-33 years hypothesis, and on that basis, a conclusion is drawn (Santosa, 2018). The number of respondents who filled out the questionnaire was 86 employees from 121 employees. This number is less than the target sample. Still, based on the rule of thumb (Hair et al., 2011), the number of respondents has met the minimum sample requirements for PLS-SEM analysis, so it is valid for use in PLS-SEM analysis and hypothesis testing.
old approximately nine respondents (10%), the age between 20-26 years old approximately six respondents (7%), and lastly the age >55 years old approximately three respondents (4%). Some respondents did not answer the age data approximately seven respondents (8%). For the description analysis, seven people who did not answer the age question were categorized as "Others".
The education characteristics data divided into five levels: middle school, high school, diploma (I-III), bachelor, and graduate. Most respondents have a bachelor degree approximately thirty-two respondents (37%), followed in sequence with a high school approximately twenty-nine respondents (34%), a diploma degree approximately nineteen respondents (22%), a graduate degree approximately five respondents (6%), and lastly a middle school approximately one respondent (1%).
Data on the year of service shows that most of the respondents have worked for more than five years, approximately forty-six respondents (53%), followed by respondents who have worked for one to five years approximately thirty-five respondents (41%), and new employees with a service period less than one year approximately five respondents (6%).
The last characteristic data of respondents is the job level consist of Manager, White Collar, and Blue Collar. The majority of respondents were Blue Collar, approximately 38 respondents (44%), followed by White Collar approximately 28 respondents (33%), and Manager, approximately 20 respondents (23%). Table 1 shows that the initial indicator loading of JE3 is .668, which means less than the threshold according to the criteria of .708 (Hair et al., 2019). The JE3 indicator was then removed from the structure and the path coefficient recalculated. After JE3 is removed, the adjusted indicator loadings of all reflective indicators show all values are greater than .708, which means the reliability is met. The formative indicators OP1, OP2, OP3, and OP4 were not analyzed using loading factors. However, it will be analyzed using collinearity, significance, and outer weight tests (Sarwono and Narimawati, 2015;Santosa, 2018).

Measurement Model Reflective Indicators
The next step is to analyze the internal consistency reliability through Cronbach's alpha and composite reliability values and analyzing convergent validity through average variance extracted (AVE) value. Cronbach's alpha value shall greater than .70, the composite reliability value shall greater than .60; and the AVE value shall greater than .50 (Santosa, 2018;Hair et al., 2019). Table 2 shows that Cronbach's alpha and composite reliability values of reflective indicators are higher than .70 and .60, respectively. The AVE values show all values are greater than .50. That means that internal consistency and convergent validity values have met the requirements.
The last step in the measurement model reflective indicators is to analyze discriminant validity at indicators level (cross-loading) and variables level (Fornell-Larcker criterion) (Santosa, 2018;Hair et al., 2019). Table 3 shows that all numbers written in bold have the greatest value compared to other numbers on the same row, which means that discriminant validity at the indicator level is met. Furthermore, Table 4 shows that the square root of AVE written in bold has the greatest value compared to the value of correlation to other variables. That indicates that discriminant validity is meant for each sample group.

Measurement Model Formative Indicators
The organizational performance (OP) variable in this research is a variable with a formative indicator. In the formative measurement model, a variable is defined as a combination of its indicators (Santosa, 2018). Analysis of the formative measurement model evaluated through multicollinearity, significance, and weight (Ghozali and Latan, 2015). Several things need to be considered before eliminating or maintaining formative indicators that do not meet the significance and weight relevance criteria. In contrast to reflective measurement models, formative indicators are not interchangeable, and removing even a single indicator can, therefore, reduce the measurement model's content validity (Diamantopoulos andWinklhofer, 2001 in Hair et al., 2019). Based on the literature, measurements and BSC have been widely applied in various industrial sectors to measure company performance. All formative indicators will be used for the structural model analysis.

Structural Model
The first step in conducting a structural model analysis is to measure the coefficient of determination (R 2 ), which shows the variance of the endogenous variables caused by all the exogenous vari-  (Santosa, 2018). Figure 2 shows the coefficient of determination (R 2 ) of job satisfaction and organizational performance. The R 2 value of the job satisfaction variable is .743; this indicates that 74.3% of the variance of job satisfaction caused by job and organization engagement, while other things not examined in this research cause 25.7%. The R 2 value of the organizational performance variable is .668; this indicates that 66.8% of the variance of organizational performance is caused by job engagement, organization engagement, and job satisfaction, while other things not examined in this research cause 33.2%. The second step in analyzing the structural model is to analyze the effect size ƒ 2 on R 2 . The effect size is measured by comparing the R 2 value when all exogenous variables are involved against the R 2 value when one of the exogenous variables is omitted in the path model analysis. That aims to deter- mine the effect of an exogenous variable on endogenous variables that are interconnected (Santosa 2018). According to Cohen (1988) in Santosa (2018), the size of the effect when an exogenous variable is removed from the path model is indicated by 0.02 (small), 0.15 (moderate), and 0.35 (large). Table 6 shows the effect size of ƒ 2 on R 2 . It is revealed that: (1) the effect size of job engagement on job satisfaction is 0.17 (moderate); (2) the effect size of job engagement on organizational performance is 0.00 (small); (3) the effect size of organization engagement on job satisfaction is 0.84 (large); (4) the effect size organization engagement on organizational performance is 0.00 (small); and (5) the effect size of job satisfaction on organizational performance is 0.41 (large).
The third and final step of structural model analysis through the combination of the path coefficient (b) and the significance of the relationship (ttest) between variables to determine whether the proposed hypothesis is accepted or rejected (Santosa, 2018). Table 7 shows the results of the path coefficient test and the significance of the relationship as the basis for testing the hypothesis. The path coefficient (b) is positive and t-test > 1.99 for each path H1, H3, and H5, which means positive and significant. Thus, H 1 accepted and H 0 rejected.
The path coefficient (b) is positive and t-test < 1.99 for each path H2 and H4, which means positive but not significant. Thus, H 1 was rejected and H 0 accepted.

DISCUSSION
Hypothesis 1 (H1) in this research states that job engagement affects job satisfaction based on the research conducted by Saks (2006), Suhartanto and Brien (2018), and Saks (2019). Research conducted by Saks (2006) shows that job engagement mediates the relationship between the antecedents of employee engagement and job satisfaction. Align with Saks (2006), Suhartanto and Brien (2018) confirm that job satisfaction mediates the relationship between job engagement and organizational performance. Saks (2019) reviewed research by Saks (2006) and confirmed that job engagement predicts job satisfaction. The results of the first hypothesis (H1) analysis show that H 1 accepted and H 0 rejected. That means that job engagement has a positive and significant effect on job satisfaction. This result is in line with the research of Saks (2006), Suhartanto and Brien (2018), and Saks (2019). This result shows that the greater the employee's job engagement, which is indicated by a positive state of mind regarding which they are responsible, will affect the greater their job satisfaction. In this research, hypothesis 2 (H2) states that job engagement affects organizational performance with the reference r esearch conducted by (Suhartanto and Brien, 2018). This reveals that the relationship between the employee and the business organization he works for is reciprocal. The results of Suhartanto and Brien's (2018) research show that job engagement and affecting job satisfaction also affect store performance and employee performance. The results of the second hypothesis (H2) analysis show that H 1 was rejected and H 0 accepted. That means that even though job engagement has a unidirectional relationship, it does not affect organizational performance. This research provides empirical evidence that different from previous research conducted by Suhartanto and Brien (2018). A possible explanation if many employees came from previous organizations through acquisitions would impact responding to the survey questions regarding the working conditions of each employee. That is in line with Sparrow and Balain (2010) revealed that different employees respond differently to the same work context and conditions. It will depend on many other characteristics, such as tenure, age, gender, hours of work and pay patterns, what country they work in, whether they work for a core or a more peripheral organizational unit (for example, one that is outsourced), or whether they come from a particular organizational constituency.
Hypothesis 3 (H3) in this research states that organization engagement affects job satisfaction based on the research conducted by Saks (2006), Suhartanto and Brien (2018), and Saks (2019). Research conducted by Saks (2006) shows that organization engagement mediates the relationship between the antecedents of employee engagement and job satisfaction. Furthermore, Suhartanto and Brien (2018) reveal that job satisfaction mediates the relationship between organization engagement and organizational performance. Saks (2019) reviewed research by Saks (2006) that concluded that organization engagement predicts job satisfaction. The results of the third hypothesis (H3) analysis show that H 1 accepted and H 0 rejected. This means that organizational engagement has a positive and significant effect on job satisfaction. This result is in line with the research of Saks (2006), Suhartanto andBrien (2018), andSaks (2019). This result shows that the more aligned and higher the company's emotional bond to the company, the more their job satisfaction will be.
Hypothesis 4 (H4) in this research states that organization engagement affects organizational performance with the reference research conducted by Farndale et al. (2014) that shows a positive and significant relationship between organization engagement and organizational performance. The results of the fourth hypothesis (H4) analysis show that H 1 was rejected and H 0 accepted. That means that even though organization engagement has a unidirectional relationship, it does not affect organizational performance. This research provides empirical evidence that different from previous research conducted by Farndale et al. (2014). A possible explanation is that organization engagement is seen from employees' perspective towards the organization down to the smallest scale of the unit in which they work. When the unit where they work is in a not comfortable condition, it will cause negative perceptions to employees that can affect organizational performance. According to Sparrow and Balain (2010), sometimes being in a good performing unit makes employees engaged, not the other way round. Sometimes performance effects only begin at extreme levels of engagement and this level of "sensitivity" may change over time.
Hypothesis 5 (H5) in this research states that job satisfaction affects organizational performance based on the research conducted by Brahmasari andSuprayetno (2008), andAl-dalahmeh et al. (2018). Research conducted by Brahmasari and Suprayetno (2008) proves that employee job satisfaction has a positive and significant effect on company performance, meaning that, in general, high employee job satisfaction will be able to improve company performance. Al-dalahmeh et al. (2018) concluded that job satisfaction significantly and positively affected organizational performance. The results of the fifth hypothesis (H5) analysis show that H 1 accepted and H 0 rejected. That means that job satisfaction has a positive and significant effect on organizational performance. This result is in line with the research of Brahmasari and Suprayetno (2008) and Al-dalahmeh et al. (2018). That indicates that employees who have high levels of job satisfaction like their jobs and have high trust in the company. Employees are willing to go beyond the call of duty by devoting personal time to their work activities and becoming more committed to the company.

CONCLUSIONS
This research answers the questions of how employee engagement is viewed from two perspectives; job and organization engagement directly and significantly affect job satisfaction, which in turn can improve organizational performance to achieve company goals. This research provides empirical evidence for the existence of a multidimensional attachment model that implies that employees are simultaneously engaged in two different objects, namely their work and their organizations. This research examines the construction and relationship between variables in the path model using the PLS-SEM approach.
The results suggested that job engagement has a positive and significant effect on job satisfaction. The effect size indicated that job engagement has a moderate effect on job satisfaction. That indicates that the more employees feel engaged with their work, which is characterized by enthusiasm, dedication, and full concentration at work, the more satisfied they are with their work.
Although job engagement has a positive relationship, it does not have a significant effect on organizational performance. The effect size indicated that job engagement has a small effect on organizational performance. A possible explanation for this is that the respondent's characteristics, especially if many employees come from the previous organization through the acquisition, will have a major role in responding to the survey questions regarding the working conditions that each employee lived.
Next, the results suggested that organizational engagement has a positive and significant effect on job satisfaction. The effect size indicated that organization engagement has a large effect on job sat-isfaction. That shows that when engaged, employees will immerse themselves in their job roles and devote a lot of personal energy to carrying out the role. Employees internalize the goals and aspirations of the organization as their own.
Although organization engagement has a positive relationship, it does not have a significant effect on organizational performance. The effect size indicated that organization engagement has a small effect on organizational performance. A possible explanation for this is that organization engagement is viewed from employees' perspective toward the organization down to the smallest scale of the unit in which they work. When the unit where they work is in a not comfortable condition, it will cause negative perceptions that can affect organizational performance.
Furthermore, the results suggested that job satisfaction has a positive and significant effect on organizational performance. The effect size indicated that job satisfaction has a large effect on organizational performance. When job satisfaction factors are fulfilled, it will affect and improve organizational performance. Employees who have a high level of job satisfaction like their jobs and have high trust in the company. Employees are willing to go beyond the call of duty by devoting personal time to their work activities and becoming more committed to the company.
The scores were good on average in a sequence related to organizational performance: (1) Customers, (2) Finance, (3) Internal Business Processes, and (4) Learning and Growth. These results indicated that the company performs well in retaining customers and managing to attract new customers. The company periodically sets sales and revenue targets. The company carries out Research and Development (R & R&D) activities. There is a rewards program as an appreciation to achievers.

IMPLICATIONS
This research contributes to the conceptual development related to the role of employee engagement viewed from two perspectives: job and organization engagement on job satisfaction and its effect on organizational performance. The result in