The Effect of Toxic Workplace Environment on Employee Performance Mediated by Employee Engagement and Work Stress Among F&B Employees in Jakarta

Employee engagement and high work stress often become obstacles for MSMEs to produce good employee performance. One of the problems in human resources that often occur is a toxic workplace environment. The purpose of this study was to determine the effect of a toxic workplace environment on employee performance mediated by work stress and employee engagement among F&B MSMEs employees in Jakarta. This research uses quantitative research methods. Questionnaires were distributed to F&B MSMEs employees in Jakarta and the data obtained was processed using SmartPLS. This study has a respondent of 400 respondents. The results showed that there was a positive effect between a toxic workplace environment on work stress, a negative effect between a toxic workplace environment on employee engagement, a negative effect between work stress and employee performance, a positive effect between employee engagement and employee performance, a non-significant negative effect of a toxic workplace environment on employee performance, there is a significant negative effect of a toxic workplace environment mediated by work stress on employee performance, and there is a significant negative effect of a toxic workplace environment mediated by employee engagement on employee performance.


INTRODUCTION
In the daily practice of organizations or companies, human resource problems such as low employee engagement and high work stress often become obstacles for companies to produce good employee performance. One factor in human resource problems that often occurs is a toxic workplace environment. According to (Rasool, Wang, Tang, Saeed, & Iqbal, 2021) a toxic workplace environment is an explanation of the relationship between employees and their workplace, which is caused by narcissistic behavior, aggressive leadership, harassment, exclusion, and intimidation from managers and co-workers. (Masliani, 2021) stated that a Breather HR study from England in 2021 found that a third of British employees left their jobs because of a toxic workplace environment culture. In addition, based on (Wirotama, 2022), research conducted by MIT Sloan in 2022 stated that more than 40% of employees thought of leaving their jobs in early 2021, this was apparently influenced by several problem factors such as compensation, toxic culture, to the lack of self-development opportunities, MIT Sloan states that toxic cultural issues are 10.4 times more influential than compensation issues.
The influence and impact of a toxic workplace is very damaging, it will harm the organization or company in the near and long term. According to (Rasool, Wang, Tang, Saeed, & Iqbal, 2021) a toxic workplace environment has a negative effect on employee engagement. (Teo, Bentley, & Nguyen, 2020) stated that workplace bullying or toxic workplaces can undermine organizational efforts to increase employee engagement and commitment. Problems regarding human resources with low engagement will directly impact the company's employee performance. According to Scullion and Collings (Hoque, Awang, Siddiqui, & Sabiu, 2018), employee engagement is related to several important organizational outcomes such as employee performance. This is also supported by Mone and London (Mughal, 2020) who state that the main determinant that drives employee performance to increase is employee engagement.
Apart from causing low employee engagement, the negative effects of a toxic workplace environment can also increase work stress. Work stress is one of the most common problems for employees. This can interfere with employees in carrying out and completing the tasks they have at work. In addition, stress can affect the ability of employees to make decisions and think creatively. Directly, work stress will also have a negative effect on employee performance. (Wang, Zaman, Rasool, Zaman, & Amin, 2020) states that a direct increase in a toxic workplace will also increase work stress. Of course this is a bad thing for companies and organizations, because it can affect the resulting performance. (Rasool, Wang, Zhang, & Samma, 2020), states that work stress is an important element in reducing employee performance. This was also explained in the research by (Yu, Park, & Hyun, 2021) where their research found that increased work stress would reduce overall employee performance.
MSMEs have a contribution of 61 percent to national GDP, therefore, the development of MSMEs is needed to encourage economic growth according to the Coordinating Ministry for the Economic Affairs (Limanseto, 2022). According to (Bayu, 2021) research results, 63 percent of MSMEs are pessimistic that they will grow in the first quarter of 2021. The workforce is one of the six reasons why MSMEs limit or close business operations (Pusparisa, 2020). According to (Haryanti, 2018), the MSME F&B sector in Indonesia is the second largest sector after the wholesale and retail trade sector. Therefore, food and beverage MSMEs are an interesting field to study by looking at the performance of employees who contribute to the success of food and beverage MSMEs. Especially during the current economic recovery, the contribution and performance of employees is very important so that the business can survive, run and develop even more.
Based on the discussion of the problems and a brief understanding of the toxic workplace environment, employee engagement, work stress, and employee performance, this research focuses on F&B MSME employees in Jakarta with the aim of finding out (1) is there a significant effect of toxic workplace environment on work stress, (2) is there a significant effect of a toxic workplace environment on employee engagement, (3) is there a significant effect of work stress on employee performance, (4) is there a significant effect of employee engagement on employees performance, (5) is there a significant effect of a toxic workplace environment on employee performance, (6) is there a significant effect of a toxic workplace environment on employee performance mediated by work stress, (7) is there a significant effect of a toxic workplace environment on employee performance mediated by employee engagement.

METHODS
The research method used is a quantitative research and associative method with individual analysis units, namely F&B MSME industry employees in Jakarta. Cross sectional is the time horizon used in this study. F&B MSME employees in Jakarta have a population of 365,191 people. To get the minimum number of samples studied, the researchers used the following formulas, the minimum number of samples obtained using the formula based on Yamane (Uakarn, Chaokromthong, & Sintao, 2021) was 400, then the minimum number of samples obtained using the Slovin formula (Rizki, Arhami, & Huzeni, 2021) also numbered 400, while the minimum number of samples obtained using the Krejcie and Morgan formula (Uakarn, Chaokromthong, & Sintao, 2021) was 384. Therefore, based on the calculation results of the three formulas, in this study researchers will take the results of the highest sample size of 400 samples to answer all three formulas required samples.
The data were collected using questionnaire with 4 indicators of toxic workplace environment, 11 indicators of work stress, 9 indicators of employee engagement, and 9 indicators of employee performance. The indicators of toxic workplace environment consist of Ostracism, Incivility, Harassment and Bullying. The indicators of work stress consist of Workload, Physical Relationship, Role management, Interpersonal relationship, Organizational style, Career development, Work-family balance, Leadership, Competence, Open information, Burnout. The indicators of employee engagement consist of Vigor, Dedication, Absorption, Word recognition, Empowerment, Supportive Feedback, Partnership, Attention, and Trust. The indicators of employee performance consist of Working time, Cooperation, Quality, Quantity, Responsibility, Job Knowledge, Independence, Initiative, and Adaptability.

The analytical method in this study uses the Structural Equation Model (SEM) PLS -Smart PLS where
there are three stages of testing, namely the outer model test, the goodness of fit model test, and the inner model test (Muhson, 2022).

RESULT AND DISCUSSIONS
The results of this test aim to determine the validity and reliability of the indicators and variables used in the study. There are two tests used in the validity test, namely the convergent validity test and the discriminant validity test. Both of these tests aim to find out whether the indicators and variables used are valid. While the composite reliability test and Cronbach's Alpha test aim to find out whether the indicators and variables are reliable. The following are the results of the validity and reliability tests obtained using the SmartPLS 3.0 application.

Convergent Validity Test
The convergent validity test is carried out by looking at the loading factor value which must be greater than 0.7 and the AVE value which must be greater than 0.5. After testing with the SmartPLS 3.0 application, in the table of loading factor values it can be seen that the loading factor values of all indicators are more than 0.7 and declared as valid, shown in the Table 1.

AVE Value
Parameter Validity

Discriminant ValidityTest
After retesting seven times on the the SmartPLS 3.0 application, with the removal of indicators TWE2, WS11, EE1, EE2, EE4, EE5, EE6, EE7, EP1, EP3, EP4, EP5, EP6, and EP7. In the image of the Fornell-Larcker criterion results, it can be seen that the correlation value of the TWE variable with the TWE variable itself is 0.880, which is greater than the correlation value of the WS variable with WS of 0.823. then the correlation value of the EE variable with the EE variable itself is 0.873 greater than the correlation value of the EE variable with the EP variable of 0.869. So it can be concluded that all the root values of the AVE variables TWE, WS, EE, and EP are valid, shown in the Figure 1. Then after testing with the SmartPLS 3.0 application, in the table of cross loading values results it can be seen that in cross loadings AVE root value or the correlation value of the variable indicator to the variable itself, everything is greater than the correlation value of the variable indicator to other variables. So it can be concluded that all indicators and constructs are valid, shown in the Table 3. After all the indicators and constructs have met the requirements in the validity test, it can be concluded that the indicators used in the study are valid or have been able to measure the variables, and can proceed to the reliability test.

Composite ReliabilityTest
The composite reliability test is carried out by looking at the composite reliability value which must be greater than 0.7. After testing with the SmartPLS 3.0 application, in the composite reliability values results table it can be seen that the composite reliability value is greater than 0.7. So it can be concluded that all variables are reliable and have passed the composite reliability test, shown in the Table 4.

Cronbach's AlphaTest
The Cronbach's alpha test is done by looking at the Cronbach's alpha value which must be greater than 0.7. After testing with the SmartPLS 3.0 application, the Cronbach's alpha value results table shows that the Cronbach's alpha value is greater than 0.7. So it can be concluded that all variables have passed the Cronbach's alpha test and all variables are reliable, shown in the Table 5. Next is the goodness of fit model testing stage where it is carried out by conducting blind folding tests and model fit tests. Testing this stage aims to determine the predictive power and feasibility of the structural model.

Blind Folding Test
The blindfolding test is carried out by looking at the predictive relevance or Q 2 value which must be greater than 0 so that the predictive power of the model is good, but if the predictive relevance or Q 2 value is less than 0 the predictive power of the model is not good. After testing with the SmartPLS3.0 application, in the image of the results of predictive relevance values it can be seen that all predictive relevance values or Q 2 are greater than 0. So it can be concluded that all the predictive power of the structural model is good, shown in the Figure 2.

Model FitTest
The model fit test is carried out by looking at the SRMR value which must be less than 0.1. After testing with the SmartPLS 3.0 application, in the image of the results of the SRMR value it can be seen that the SRMR value of 0.062 is smaller than 0.1. So it can be concluded that the model is feasible to test the influence, shown in the Figure 3. The next step is the inner model testing phase where it is carried out by carrying out a significance test, path coefficient test, and R square test. Testing this stage aims to determine the significance and magnitude of the influence of exogenous variables on endogenous variables.

Significance Test Results
The significance test is carried out by looking at the T statistics value which must be greater than 1.96 in order to have a significant effect. After testing with the SmartPLS 3.0 application, in the results table for the T statistics value it can be seen that X(TWE) has a significant influence on Y1(WS), X(TWE) has a significant influence on Y2(EE), Y1(WS) has a significant effect on Z(EP), Y2(EE) has a significant influence on Z(EP), X(TWE) has no significant effect on Z(EP), X(TWE) mediated by Y1(WS) has a significant effect on Z(EP), and X(TWE) which is mediated by Y2(EE) has a significant effect on Z(EP), shown in the Table 6 and Figure 4.

Path Coefficient Test Results
The path coefficient test is carried out by looking at the path coefficients value which must be greater than 0 in order to have a positive influence, but if the path coefficients value is less than 0 it means it has a negative influence. After testing with the SmartPLS 3.0 application, in the results table for path coefficients it can be seen that X(TWE) has a positive influence on Y1(WS), X(TWE) has a negative effect on Y2(EE), Y1(WS) has a negative effect on Z(EP), Y2(EE) has a positive influence on Z(EP), X(TWE) has a negative influence on Z(EP), X(TWE) which is mediated by Y1(WS) has a negative effect on Z(EP), and X(TWE) which is mediated by Y2(EE) has a negative effect on Z(EP), shown in the Table 7.

Path Coefficients Value
Parameter Influence

R SquareTest Results
The Rsquare test is done by looking at the R square value to find out how much influence it has. After testing the SmartPLS application, in Figure 5 it can be seen that Y1(WS) has an R square value of 0.678. So it was concluded that Y1(WS) was influenced by 67.8% by X(TWE). Then Y2(EE) has an R square value of 0.570. it is concluded that Y2(EE) is influenced by 57% by X(TWE). And finally Z(EP) has an R square value of 0.789. it can be concluded that Z(EP) is influenced by 78.9% by X (TWE), Y1(WS) and Y2(EE), shown in the Figure 5.

Discussion
From the results of the above analysis, it was found that there was a significant positive effect of the toxic workplace environment on work stress on F&B MSME employees in Jakarta. The results of this study support previous research conducted by (Rasool, Wang, Zhang, & Samma, 2020) where it was found that the toxic workplace environment became one of the causes of work stress. The other finding is it was found that is a significant negative effect of a toxic workplace environment on employee engagement among F&B MSME employees in Jakarta. The results of this study are in line with previous research conducted by (Rasool, Wang, Tang, Saeed, & Iqbal, 2021), whose research shows that a toxic workplace environment has a negative effect on employee engagement,where this can reduce worker productivity.
There is a significant negative effect of work stress on employee performance in F&B MSME employees in Jakarta, this research is in line with previous research conducted by (Yu, Park, & Hyun, 2021) which found that work stress experienced by hotel employees under study can reduce employee performance in general. There is a significant positive effect of employee engagement on employee performance in F&B MSME employees in Jakarta, this research is in line with previous research conducted by (Hoque, Awang, Siddiqui, & Sabiu, 2018) in his research which stated that employee engagement is an important factor in companies in an effort to create satisfaction, competitive advantage and achievement of employee performance targets.
Regarding the impact towards employee performance the finding is there is a non-significant negative effect of a toxic workplace environment on employee performance in F&B MSME employees in Jakarta. The results of this study are supported by research conducted by (Haeruddin, Akbar, Dipoatmodjo, Kurniawan, & Abadi, 2022) where their research found that a negative or toxic workplace environment negatively affect their level of performance. Furthermore, toxic workplace environment mediated by work stress has a negative and significant effect on employee performance, work stress plays a role as a mediating variable between the effect of a toxic workplace environment on employee performance. The results of this study have similarities from previous studies that have conducted by (Rasool, Wang, Zhang, & Samma, 2020) where the results of his research found that there was a significant effect of a toxic workplace environment on employee performance which was mediated by work stress. Toxic workplace environment mediated by employee engagement has a negative and significant effect on employee performance, employee engagement plays a role as a mediating variable between the effect of a toxic workplace environment on employee performance, in line with research by (Wan, Li, Zhou, & Shang, 2018) research results found that there is a significant negative effect of the work environment on turnover intention which is mediated by employee engagement.

CONCLUSIONS
Toxic workplace environment has a significant and positive effect on work stress directly received by employees. Therefore MSME owners must be agile to see this toxic workplace environment if it occurs in their business. Because a toxic workplace environment will spread negative feelings towards employees, these negative feelings can be detrimental and cause stress. A toxic workplace environment has a significant and negative effect on employee engagement directly. Of course this is not good for MSMEs because in carrying out business strategies, cooperation and collaboration between MSME owners and employees must be carried out so that the business can compete with the competitors and also grow. The impact that a toxic workplace environment can have is not very good for employees as the main drivers of business and business development in the long term, because this can reduce enthusiasm, sincerity and full attention at work, because work stress has a significant and negative effect on employee performance directly. Therefore this will be very detrimental not only for employees, but also for MSMEs. In particular, this will burden employees in a psychological sense, difficulty in allocating time, so that employees will easily feel tired at work and lead to decreased performance. Employee engagement has a significant and positive effect on employee performance directly. MSME owners must be sensitive to the positive influence provided by employee engagement on employee performance, because this will be very beneficial for the business value they have. By taking advantage of the advantages of employee engagement, MSME owners will be able to increase the enthusiasm, sincerity and attention that employees have in their work. This will have a long-term positive influence on the sustainability and development of MSMEs in achieving goals. Toxic workplace environment has no significant and negative effect on employee performance directly. Even though the effect that a toxic workplace environment has on employee performance is not significant, there is still a negative influence that is exerted and will cause a decrease in employee performance. In the long term, things such as a slow decline in the quality of performance, difficulties in achieving sales quantity targets, and can reduce employees accuracy. Of course this will have a detrimental impact on the organization, as well as hinder business development and competition. Work stress plays a full role as a mediating variable in influencing a toxic workplace environment indirectly and significantly and negatively on employee performance. Therefore MSME actors must realize that a toxic workplace environment will directly give negative feelings to employees, as well as cause work stress. Then the work stress experienced by employees will burden them psychologically, easily feel tired at work, resulting in decreased employee performance. Employee engagement plays a full role as a mediating variable in the influence of a toxic workplace environment that is both significant and negative on employee performance indirectly. Therefore, MSME owners must also realize that a toxic workplace environment will have a negative impact on employee engagement which can disrupt business processes and organizations such as reducing enthusiasm, sincerity, and focusing on work. Then it will also have a negative effect on employee performance, such as low quality and quantity of work. MSME actors should see a toxic workplace environment as a big problem for MSMEs. Paying attention to, preventing, and providing solutions are some of the efforts that can be made by MSME actors to eliminate toxic workplace environments in organizations. Work stress received by employees through a toxic workplace environment or other factors needs to be resolved quickly, because this will reduce employee performance in MSMEs which hinders MSMEs in running their business. Employee engagement must be maintained by paying attention to factors that disturb employee engagement or provide encouragement in increasing employee engagement, this will make employees give their best for the organization. So it can be concluded, by preventing negative factors that can hinder business continuity such as a toxic workplace environment and work stress, while simultaneously increasing positive factors such as employee engagement in running a business, it will greatly help MSMEs grow bigger and compete with competitors in a competitive market. tight. This is an important foundation that should not be forgotten by all MSME actors. This study has limitations that can be developed in further research, the researcher suggests adding unresearched variables that might influence or be affected by a toxic workplace environment such as organizational support, organizational culture, job satisfaction, and workplace spirituality. Future research is expected to consider demographic factors to see how the same variables are in different demographics and examine in different regions or countries to provide a broader picture regarding the results of this study.