THE INFLUENCE OF ORGANIZATIONAL CULTURE AND PERCEIVED ORGANIZATIONAL SUPPORT ON EMPLOYEE ENGAGEMENT WITH ORGANIZATIONAL COMMITMENT AS A MEDIATOR

The focus of this research is how to improve employee engagement at the PT.X head office by examining the relationship between organizational culture and organizational perceptions of employee engagement with organizational commitment as a mediator. This research uses quantitative approach, which is a study that aims to explain a cause of the occurrence of social phenomena, which the phenomenon will be explained by conducting an objective measurement. Primary data collection through the dissemination of questionnaires to 74 respondents. Data analysis using the partial least square equation model or can be known as SEM-PLS. Based on the results of research that has been found, it can be concluded that organizational culture variable does not give a significant direct positive influence on employee engagement, whereas when viewed on variables perceived organizational support and organizational commitment, these two variables have a significant positive direct affect on employee engagement. If the direct influence relationship between organizational culture and employee engagement does not have a significant influence, but when the variable is mediated by organizational commitment, then the results state that there is a significant positive influence relationship between the two variables. This shows that the organization's commitment to succeed has a significant role in mediating the relationship between organizational culture and employee engagement.


A. INTRODUCTION
Indonesia has good bilateral relations with various countries in the world. This resulted in the emergence of various companies that invest in Indonesia or even not a few companies in Indonesia that cooperate with foreign companies with the intention of developing business. One of the countries that is often active in investing in Indonesia or establishing good cooperation with Indonesia is Japan.
So much cooperation, investment, and expatriate in various countries, it makes Japan always try to transfer and apply their characteristics in doing business and running the organization. Not infrequently it is widely adopted and adapted by various countries. The important thing that includes the culture of Japanese organizations such as teamwork, openness, and so on to create efficiency and competitive advantage.
Culture is a set of values, guidance, beliefs, understanding, norms, philosophy, ethics, and ways of thinking. The culture that exists in an environment has a huge influence on the personal formation that is in that environment. Apart from the socialization side, the workmanship of a job as well as the cultural role of the Japanese people greatly influenced their work. Therefore, many Japanese companies or Indonesian multinational companies that cooperate with Japan apply the existing culture in working, to achieve the same goals and through the same process.
Moving straight with the improvement of cooperation and globalization, Indonesia is obliged to improve the quality and competence of the existing workforce. The improvement of competence and quality received by the Indonesian workforce can not be separated from the intervention of the organization or related companies to increase the percentage of engagement owned by employees. Employee engagement is very important for the organization and it can be determining factor behind the high business performance of a company (Triple Creek Associates, 2007). In general, employee engagement can be defined as the level of commitment and attachment of employees to the organization and the value applied in the organization. Employees who feel attached are those who feel really engaged and have an enthusiasm for their work and organization.
According to research conducted, about eighty-seven percent (87%) of respondents believe this issue related to the culture and attachment of employees as an important thing with fifty percent (50%) cited the issue as a very important matter (Delloitte, 2015).
There are a variety of scales that can be measured to know the engagement that employees have in an organization, one of them is the scale of tardiness. Paycor (2020) states that tardiness can be the first sign of decreased attachment, burnout, or an indication that employees are struggling to balance work and family commitments. Turnover in an organization is one indication of low employee attachment. High turnover or more than 10% annually will increase the costs that organizations have to spend. When viewed from the data above, in 2017, 11 people left the organization or about 7.4% of the total employees. While in 2018 16 employees left the organization or about 11% of the total employees and experienced a decrease in percentage in 2019, which is about 8.4% of the total employees. Through pre-research interviews that have been conducted to several respondents, several factors cause turnover intentions from employees, such as lack of appreciation, further response to what employees say or, there is no clear career path in the company. One of the variables that has something to do with the reasons of the respondents above is the organizational culture and perception of organizational support.

B. LITERATURE REVIEW Organizational Culture
Organizational culture defines the right way to behave within an organization. This culture consists of shared beliefs and values established by leaders and then communicated and strengthened through various methods, which in turn build the employees' perceptions, behaviors, and understandings. Many definitions of organizational culture have been proposed in the literature, but the common themes underlying these definitions are based on the values underlying their shared organization, beliefs, and philosophies (Barney, 1986;Corbett, 1986;Deal and Kennedy, 1982;Ouchi 1981;Pettigrew, 1979;Schein, 1990;1996;Shockley-Zalabak andMorley, 1989 in Tang, 2000). Tang et al (2000) found that the quality of circle members will develop a variety of work-related behaviors and performances due to the different types of business environments, values, and cultures within the organization.

Japanese Organizational Culture
Japanese society is very concerned with the process at work rather than the results they get. Because the Japanese people believe that what they do through a good process will get a good result. Therefore, it is not only in their daily lives to apply the culture they adhere to but in various things that fill their daily activities. Culture is a set of values, guidance, beliefs, understanding, norms, philosophy, ethics, and ways of thinking. The culture that exists in an environment has a huge influence on the personal formation that is in that environment. Therefore, many Japanese companies or multinational companies that cooperate with Japan in working that apply the existing culture in order to achieve the same goals and go through the same process.
Japan applies some basic principles of their organization. Such as trust in employees, building employee loyalty to the company, treating employees as resources, decentralizing decisions, and using consensus systems for decision making (Cole;1980;Harper 1988;Johnson and Ouchi 1974). Tang (2000) presents four (4) dimensions and fifteen (15) indicators contained in the culture of the Japanese organization, namely; a. Family orientation and loyalty, b. openness communication and decision making, c. team approach, and d. manager knowledge.

Perceived Organizational Support
Erdogan and Enders (2007) in Nezhad et al (2012) state that perceived organizational support refers to the extent to which a person believes that the organization cares for him, appreciates, and provides assistance and support. In addition, the appropriate reaction if an employee makes a mistake in his performance, the employee advises the company to take a better approach and communicate again with them Mitchell, Gagné, Beaudry, &Dyer (2012). Perceived organizational support is learned as something that is experienced by employees and will lead to organizational commitment owned by employees.
Employees tend to see pleasant or unpleasant treatment as a reflection or an indication that the organization is happy or not with the employee. Rewards given for pleasant work such as high compensation, promotion of positions, enrichment of employment, as well as the influence of organizational regulations will contribute positively to perceived organizational support only if employees believe that the whole thing is the result of the organization's willingness without any coercion (Eisenberger, 1986).
The theory of perceived organizational support (Eisenberger, 1986) categorizes three (3) dimensions of perceived organizational support. This is a common treatment received or felt by employees of the organization. In the form of fairness, supervisory support, and organizational awards, and working conditions (rewards and job conditions).

Organizational Commitment
Organizational Commitment, proposed by Whyte (1956), refers to individual concern and loyalty to the organization. Organizational commitment is defined as "the relative strength of individual identification and attachment in a particular respect and can be characterized by a strong belief in and acceptance of the objectives and values of the organization, a willingness to exert considerable effort on behalf of the organization and a strong desire to maintain membership of the organization" (Mowday, Porter, and steer, 1982, p., 27). Meyer and Allen (1991) have classified the organization's commitment into three categories and they emphasize three different themes in the definition of work commitment terms. This model proposes that organizational commitments are experienced by employees as three simultaneous mindsets that include affective organizational commitments, normative, and sustainable. Cook and Wall (1980) developed a scaled version of nine (9) shorter items and adapted it from a longer organizational commitment questionnaire (OCQ) designed for the working population. This scale has 3 dimensions that have an interrelationship with each other, namely: a. acceptance of organizational values (identification); b. willingness to exert efforts on behalf of the organization (involvement); c. the desire to remain an employee in the organization (loyalty). Saks (2006) defines employee attachment as the extent to which an individual is attentive and absorbed in the performance of his role. Attachment is a positive, fulfilling, work-related state of mind characterized by strength, dedication, and absorption (Schaufeli and Bakker, 2004).

Employee Engagement
Schaufeli, Bekker classifies attachment into three dimensions: vigor, dedication, and absorption. Vigor refers to energy, mental resilience, determination, and investing consistent effort in work (Rayton and Yalabik, 2014). Vigor is one aspect of attachment that implies a high level of energy and mental endurance while working. Dedication is identified by enthusiasm, pride, and also challenge. Both vigor and dedication are positives as opposed to work fatigue levels. Absorption is characterized by a full concentration of work, a happy heart, and a feeling of fun at work. In addition to this dimension when working employees feel a very fast time passed and employees feel immersed in their work.
The measurement scale issued by Schaufeli, Bekker is the 17 items Utrecht Work Engagement Scale (UWES) which is then shortened to 9 items (Schaufeli et al., 2006). Schaufeli and his colleagues argue that rather than momentary and specific circumstances, engagement refers to a persistent and pervasive affective cognitive state that does not focus on a particular object, event, individual, or behavior (Schaufeli et al., 2002).

C. METHOD Research Hypothesis
Based on theory in literature review above, researchers built a research hypothesis between the variables, namely the organizational culture, the perceived of organizational support, organizational commitment, and employee attachment.
Here are seven (7) hypotheses built by researchers in this study.
1. Ho1: Organizational culture has no effect on employee engagement. Ha1: Organizational culture has a positive relationship and influence on employee engagement. 2. Ho2: Organizational culture has no effect on employee organizational commitment.
Ha2: Organizational culture has a positive relationship and influence on employee organizational commitment. 3. Ho3: Perceived of organizational support has no effect on employee engagement.
Ha3: Perceived of organizational support has a positive relationship and influence on employee engagement. 4. Ho4: Perceived of organizational support has no effect on employee organizational commitment Ha4: Perceived of organizational support has a positive relationship and influence on employee organizational commitment. 5. Ho5: Organizational commitment has no positive relationship and influence with employee engagement Ha5: organizational commitment has a positive relationship and influence on employee engagement 6. Ho6: organizational commitment cannot mediate the influence of organizational culture on employee engagement Ha6: Organizational commitment can mediate the influence of organizational culture on employee engagement. 7. Ha7: Organizational commitment cannot mediate the effect of perceived organizational support on employee engagement. Ho7: Organizational commitment can mediate the effect of perceived organizational support on employee engagement.

Analysis Model
In this research, there were 4 variables studied, namely two (2) free variables (exogenous variables) and bound variables (endogenous variables).
1. Exogenous Variables Exogenous variables are variables that exist in research and affect other variables, but are not influenced by other variables. In this study which is an exogenous variable is the JOrganizational Culture (X1) and the perceived organizational support (X2).

Endogenous Variables
Endogenous variables or can be known as dependent variables are variables that are bound or influenced by other variables. In this study, employee engagement (Y2) is a dependent variable, and organizational commitment as a mediating variable (Y1).

Figure 1 Analysis Model Processed by Researchers, 2020
In this research, the researcher used a quantitative approach with an explanatory type. Quantitative research intends to describe a phenomenon objectively and explain it by collecting focused numerical data (Muhammad, 2014). The quantitative approach also aims to explain a cause of social phenomena, in which these phenomena will be explained by taking an objective measurement and making numerical analysis (Muhammad, 2014). Explanatory or causal research is a study aimed at testing a hypothesis related to cause and effect in its implementation (Kotler, 2006).
In the process of data collection, researchers used two types of data sources, namely primary data and secondary data. Primary data is taken from the division of questionnaires consisting of forty-one (41) question items arranged based on indicators of each variable. Malhotra (2007) said that the questionnaire is a series of formal question items that will be disseminated to respondents, which aims to obtain an information.
The scale used in this questionnaire is an interval scale in which the respondent can put a checklist (√) in one of the five interval classes. Five classes on the interval scale used in the study were strongly disagreed, disagree, somewhat agree, agree, and strongly agree. In this research, researchers will use a questionnaire which consists of 5 (five) parts, namely; the identity of the respondent, the screening question that determines whether the employee can fill out further questionnaires, if so, then continue with the main questions related to the variables in this research.
Secondary data in this study is in the form of personnel department documents or Human Resource Department of PT X. In addition, researchers will also use scientific books, the internet, research journals that have been conducted before where the source has a good relationship in terms of themes and others with this study.
Before researching all data objects, researchers collected pre-test analysis to thirty (30) respondents in the locus of the study that aims to test the validity and reliability of the indicators in the questionnaire in the form of google form consisting of four (4) variables and fourty one (41) core question items derived from the research instrument also equipped with screening questions. Data obtained by researchers from questionnaires and then processed using SPSS version 20.
Pre-test analysis is conducted to test the validity level and reliability level of an instrument. Freidenberg (1955) in his book Psychological Testing: Design, Analysis, and Use, said that a correlation coefficient value used to be used as a measurement of validity is ≥ 0.3. The technique used to test the validity of the statement items in this study is Pearson Product Moment. If the correlation coefficient value of the item of the statement item being tested is greater than rcritical by 0.3, it can be concluded that the statement item is a valid construct.
After conducting a pre-test analysis and it is stated that the instrument is valid and reliable, the existing instruments will be used in further analysis, namely distributing questionnaires to the target population. Ismiyanto (2003) stated that the population is the totality or object of research that can be an object, person, or something which it can provide information or research data. In this study, the population used consisted of seventy-four (74) non-managerial employees and directors in the main office of PT X.
This study uses analytical techniques with a partial least square equation model or can be known as SEM-PLS with Smart PLS 3.0 tools. PLS analysis can be used if there are many independent variables or there is multicoserity between existing variables.

D. RESULT AND DISCUSSION Pre Test Analysis
Researchers have conducted pre-test tests that include validity tests and reliability tests on research instruments before collecting primary data through questionnaires that have been compiled based on selected variables. The results of the questionnaire validity test for the variables studied are presented in the following table: Furthermore, reliability tests are conducted when the instrument has been confirmed valid. Reliability has the meaning of a trustworthy thing, Sugiharto and Situnjak (2006) said that reliability refers to an understanding where the instrument used in a study to obtain information can be trusted as a data collection tool and can reveal information in the field. A research instrument can be said to have adequate reliability if it has a coefficient of Cronbach's alpha greater than or equal to 0.70 (Now, 2006). Based on table 7 on reliability tests of the four research variables. The reliability test results of the four variables have good reliability because they have a reliability coefficient greater than the critical value (0,7) as shown in the table above.
The instruments contained in the questionnaire are entirely valid and reliable, so all statements on the questionnaire in this study are eligible to be included in further analysis, namely the spread of questionnaires in the total population.

Descriptive Analysis of Research Data
The data description of the research results can be used to enrich the discussion, through the description of the respondent's response data can be known how the respondent's response to each variable being studied. The principle of categorization of the number of respondents' response scores in this study used the average value of the calculation as a reference to determine the classifies of the assessment category. Average calculations are used to see the tendency of answers from statement items in order to describe the condition of each of the variables studied. The classification of assessment categories in this study is determined based on the number of measurement scales used, which is as many as five classifications. In this study, the maximum rating value was five and the minimum value of the study was one.

OUTER MODEL TEST RESULTS Convergent Validity Test
The first stage assesses the criteria of convergent validity. An indicator can be said to have good validity if the indicator has a loading factor value greater than 0.70. While loading factor 0.50 to 0.60 can still be maintained for models that are still in the development stage (Ghozali, 2014:39). Based on the results of the estimation by using the help of smartpls 3 program application. obtained output as follows : Figure 2 Outer Model Evaluation Loading Factor Value Chart Figure 2 above shows the loading factor value for each construct of each variable. Based on the figure can be seen that the entire loading factor is worth more than 0.7. So it can be concluded that based on each construct in the study has a good validity.
Furthermore, it will be conducted testing average variance extracted (AVE) to strengthen the results of convergent validity with criteria if the AVE value > 0.5, then the construct used in the study is valid. The following are presented the results of the average variance extracted test using SmartPLS 3.0 program: Based on Table 10 convergent validity results can be found based on average variance extracted values. The results showed that all latent variables had an AVE value of more than 0.5. This indicates that the indicators that make up the latent construct have a good convergent validity when viewed from the average variance extracted value.

Discriminant Validity Test
Discriminant Validity can be seen from the cross-loading value. The correlation value of the indicator to its construct should be greater than the correlation value between the indicator and other constructs. And it can also be seen from the comparison between the square root of AVE with the correlation between latent constructs. If the square root value of AVE is greater than the correlation between latent constructs indicates that the latent construct has a good discriminant validity in the model (Fornell and Larcker, 1981). Based on the table above, it can be known that all indicators have a high correlation to the construct compared to other constructs. So it can be concluded that the research model has good discriminant validity.

Reliability Test
The next stage assesses the criteria of Cronbach's Alpha and Composite Reliability. Each construct is said to be reliable if it has Cronbach's Alpha and Composite Reliability greater than 0.70 (Ghozali, 2014:40). The following are presented reliability test results using Smart PLS 3.0 program. Based on Table it can be known that there is a latent construct has a value of Cronbach's alpha more than 0.7, it indicates that latent constructs have good reliability. In addition, the composite reliability value of all latent constructs also has a value greater than 0.70, it indicates that the latent construct has good reliability.

INNER MODEL TEST RESULTS WITH SEM PLS
The inner model analysis test in this study through four (4) approaches, namely calculating the values of R-square, F-Square, Q Square Predictive Relevance, and Godness of Fit (GoF). According to Chin (1998) Table 4.37, R-Square can be seen for organizational commitment variable of 0.748 which means that Organizational Culture and Perceived Organizational Support contributed 0.747 or 74.7% influence to Organizational Commitment. While the remaining 25.3% is the influence of other factors that are not observed.

R-Square Test Results
R-Square for Employee Engagement variables is 0.864 which means that Organizational Culture and Perceived Organizational Support contributed 0.864 or 86.4% of Employee Engagement with mediated Organizational Commitment. While the remaining 13.6% is the influence of other factors that are not observed.

F-Square Test Results
The F Square value of 0.02 indicates a small rating, Effect Size 0.15 indicates a medium rating and Effect Size 0.35 indicates a large rating. Based on the test results with SmartPLS 3, F Square results were obtained as follows. Based on Table 4.19 shows that on variables Organizational Culture has an influence with a small category in influencing Employee Engagement with mediated Organizational Commitment, Perceived Organizational Support has an influence with a small category in influencing Employee Engagement with mediated Organizational Commitment, and Organizational Commitment has an influence with a large category in influencing Employee Engagement.

Q Square Predictive Relevance Test Results
Q-square testing is used to measure how well the observation value is generated by the model as well as the estimated parameters. Q-square value obtained by using the R Square value in the table above, obtained the calculation result as follows: Based on the calculation results above it is known that the value of Q square is greater than 0, this means that the observed values have been well reconstructed so that the model has predictive relevance. This Q square value can also be used to see the relative influence of structural models on observational measurements for endogenous latent variables. This means that there is a 0.965 or 96.5% relative influence of structural models on observational measurements for endogenous latent variables, and as much as 3.5% is a model error.

Goodness of Fit (GoF) Test Results.
This index is used to evaluate measurement models and structural models and in addition provides simple measurements for the entirety of the model prediction. The Value of GoF 0.10 is concluded that GoF belongs to the category of small, 0.250 medium and 0.36 belongs to the large category. For this reason the GoF index is calculated from the square root of the average AVE value and the Rsquare average as follows: Based on the calculation above, it is known that the value of GoF obtained is 0.714, it is known that the GoF value belongs to the large category.

RESEARCH HYPOTHESIS TEST RESULTS
Hypothesis testing in this study was carried out using the path coefficient, tvalue, and p-value. According to Abdillah & Hartono (2015: 211), assessing predictions and significance in hypothesis testing can be seen from t-value and pvalue. The t-table value can be seen in the following table. According to Abdillah &Hartono (2015: 211), with a confidence level of 95% (alpha 5%), two tailed, obtained the value of t-table as follows: 1. If the t-statistical value > 1.96 (used for direct influence), then H0 is rejected and H1 is accepted. 2. If the t-statistical value < 1.96 (used for direct influence), then H0 is accepted and H1 is rejected. The magnitude of the significance value between the variables tested is presented in the form of the value contained in the arrow that connects one of the variables to the variable being the goal. From the results of Table 18 above obtained the Original Sample (O) value of 0.425 indicates that the direction of organizational culture's influence on Organizational Commitment is positive or in the direction. The influence of Organizational Culture on Organizational Commitment is significant, with a tstatistical value of 3,511 greater than t table or 3,511 > 1.96, and p value of 0.000 smaller than alpha 5% (0.05). So H1 is accepted meaning organizational culture affects Organizational Commitment. From the results of Table 19 above obtained the Original Sample (O) value of 0.469 indicates that the direction of influence of Perceived Organizational Support on Organizational Commitment is positive or in the direction. The influence of Perceived Organizational Support on Organizational Commitment is significant, with a t-statistical value of 4,542 greater than t table or 3,698 > 1.96, and a p value of 0.000 smaller than alpha 5% (0.05). Thus, H1 is accepted meaning Perceived Organizational Support influence on Organizational Commitment. From the results of Table 20 above obtained the Original Sample (O) value of 0.173 indicates that the direction of Organizational Culture influence on Employee Engagement is positive or in the direction. Organizational Culture's influence on Employee Engagement is significant, with a t-statistical value of 1,653 smaller than t table or 1,653 < 1.96, and a p value of 0.099 greater than alpha 5% (0.05). So it can be stated that Ho1 is accepted and H3 is rejected.

The Influence of Organizational Culture on Employee Engagement
The results of this study are contrary to previous research conducted by Widyaswendra, 2020 which stated that Organizational Culture has a significant effect on Employee Engagement. But these results support research conducted by Akbar, 2020 which states that Organizational Culture has no direct effect on Employee Engagement. From the results of Table 21 above obtained the original sample value (O) of 0.259 indicates that the direction of influence of Perceived Organizational Support on Employee Engagement is positive or unidirectional, meaning that if Perceived Organizational Support increases or gets better then Employee Engagement will increase anyway or will be better. The effect of Perceived Organizational Support on Employee Engagement is significant, with a tstatistical value of 2,491 greater than t table or 2,491 > 1.96, and a p value of 0.013 smaller than alpha 5% (0.05). So H1 is rejected meaning that Perceived Organizational Support affects Employee Engagement. From the results of Table 22 above obtained the Original Sample (O) value of 0.545 indicates that the direction of Organizational Commitment influence on Employee Engagement is positive or in it's direction, meaning that if Organizational Commitment increases or gets better then Employee Engagement will increase as well or will be better. The effect of Organizational Commitment on Employee Engagement is signifikan, with a t-statistical value of 4,936 greater than t table or 4,936 > 1.96, and p value of 0.000 smaller than alpha 5% (0.05). So H1 is accepted meaning Organizational Commitment influence on Employee Engagement.  From the result of Table 24 above obtained the original sample value (O) which is positive is 0.255. Organizational Commitment mediates the influence of Perceived Organizational Support on Employee Engagement significantly, where the t-statistical value of 2,753 is greater than the table or 2,753 > 1.96, and the p value of 0.006 is smaller than alpha 5% (0.05). Thus, H1 is accepted meaning that Organizational Commitment can mediate the influence of Perceived Organizational Support on Employee Engagement.

E. CONCLUSION
Based on descriptive analysis that has been done by researchers on each variable through an average score (mean score) on the respondent's response to each variable. the value obtained by the researcher is in the range of 3.40 to 4.19, it can be concluded that in general employees in the main office of PT. X has a good perceive of every variable that exists, the variables in question are Organizational Culture, Perceived Organizational Support, Organizational Commitment and Employee Engagement.
Through the evaluation of outer models both through the test of convergent validity, discriminant validity and also the reliability of all items in this study have been qualified. In addition, in the evaluation of the inner model this study has a large or high level of conformity in Goodness of Fit (GoF) with a point gain of 0.714.
Hypothetical test results show that organizational culture variables do not have a significant direct positive influence on employee engagement, whereas when viewed on variables perceived organizational support and organizational commitment, these two variables have a significant positive direct influence on employee engagement.
If the direct influence relationship between organizational culture and employee engagement does not have a significant influence, but when the variable is mediated by organizational commitment, then the results state that there is a significant positive influence relationship between the two variables. This shows that the organization's commitment to succeed has a significant role in mediating the relationship between organizational culture and employee engagement.
The relationship that organizational commitment to employee engagement has the highest influence when compared to other relationships. This can be proven by the presence of t statistic value in this research. In addition to having the highest influence, organizational commitment succeeds in becoming a mediation variable that has a significant role. Then it can be used by PT. X to make organizational commitment owned by employees as the main focus in order to create high engagement and impact on improving performance, improved discipline as well as decreased voluntary turnover rate in the company. This is without ignore the presence of other variables.

SUGGESTION
The suggestion from researcher for academics who will conduct deeper research: 1. Research on a wider range of respondents, not only in the scope of the head office, but can conduct research to branch offices or even other private companies. 2. Further research should use not only using primary data through questionnaires, but can use other instruments to collect data. 3. Conduct further research on the variables in this study using the latest theory and scale of measurements.