Correspondence to : sitinuraziz@yahoo.com Received: 7 August 2021 Accepted: 18 August 2021 Published: 30 September 2021 JURNAL AKUNTANSI, MANAJEMEN DAN EKONOMI Vol.23, No. 3, 2021, pp. 1 - 9 Published online in http://jos.unsoed.ac.id/index.php/jame ISSN: 1410-9336 / E-ISSN: 2620-8482 INTRODUCTION Front line employees are very important resources because of their important role in the overall operation of the business. In the previous literature, more research discussed phenomena that appear under different names, such as resource management human resources sustainability, green human resources management, social management, and ethical human resources management (Ehnert, 2009). Underlying the aforementioned studies is to develop long-term employment orientation and to the indifference of the employees, which is largely determined by the trust, loyalty, commitment, and equity in employment relations. Therefore, it is important to further investigate the determinants in the context of organizational interpersonal relationships, such as the relationship between superiors and subordinates. Various studies have explored interpersonal trust not only in the context of a sustainable relationship but also to improve the desired outcomes of employees such as job satisfaction, organizational commitment, organizational citizenship behavior, cooperation, and performance (Stankevciut and Savaneviciene, 2018). Despite the importance of interpersonal trust in organizations for both sustainable work relationships and the results of positive employee attitudes and behavior, researchers in the past paid little attention to the determinants of interpersonal trust felt by frontline employees towards supervisors. Hence, there is a need to investigate interpersonal trust, particularly trust-in- supervisors in front-line employees. To address the academic gap mentioned above, a study was conducted on the interpersonal trust of frontline employees in trust-in-supervisors. The study of antecedents of the trust-in-supervisors in the context of frontline employees is still rarely explored. This research explores the abusive supervisor as negative antecedents of trust in - supervisor and supervisor support and communication quality as positive antecedents of trust-in- supervisor. Furthermore, the moderating effect of relationship quality is examined on the effect of abusive supervisor and supervisor support on trust-in-supervisor The findings of this study are expected to contribute to theory and practice in several ways. The first examines the unique positions of frontline employees in the workplace, who are influenced by their immediate supervisor. As owners of power in the organization, leaders have a great influence on both the staff and the organization. Organizations can run effectively if there is an emphasis on the best leadership style or one best way (Rukaiyah and Muliana, 2018). Second, this study can contribute to the literature on sustainable employment relationships by empirically investigating some of the positive and negative Trust In Supervisor for Frontline Employee: Quality of Relationship as Moderator Siti Nur Azizah Management Department, Economy and Business, University of Putra Bangsa, Indonesia Abstract This study investigates the antecedents of a frontline employee’s trust-in-supervisor with relation quality as moderator variable.This research is a causal descriptive study to investigate the relationship between variables. A personally administered survey was conducted to collect data from 100 frontline employees in Kebumen District, Cenral Java, Indonesia. Covariance- based structural equation modeling was applied using PLS3 to explore the proposed relationships. The results show that abusive supervision is negatively related to fro ntline employee’s trust-in-supervisor, whereas supervisor support and communication quality have significant positive effects on frontline employee’s trust-in-supervisor. Finally, the results show that relationship quality moderates (strengthens) the supportive influence of supervisors on trust in supervisors, but the quality of the relationship does not moderate the effect of abusive supervisors on trust in supervisors Keywords Trust-In-Supervisor; Abusive Supervisor; Supportive Supervisor; Qualuty of Communication antecedents of frontline employees with trust-in- supervisors. Third, the findings of this study can contribute to the literature on employee cooperation behavior with trust-in-supervisors Research conducted by Barton, et al. (2012) explore abusive supervisor affect the relationship between the regulatory and supervisory stress rude. The results showed that the increase in the stress level of the supervisor was related to the increase in the experience of rough supervision that was rated by employees. Tepper (2000) describes in his research as “rough supervision” and this has led to a decline in the trust of his subordinates. In another study, abusive behavior of the leadership of the abusive acts as a mediator in full relationship Machiavellianism with counter-productive behavior (Amir and Malik, 2016) The research also shows that there is support from superiors who can increase the trust of subordinates, which then affects, among others, employee performance. The supervisor's role is very important to ensure that employees are competently capable of performing their jobs (Rowold, 2008). The ability, virtue, and integrity of supervisors can increase or decrease employees' trust in them ( Mayer and Gavin, 2005). Conversely, leaders will gain the trust of their subordinates if their behavior is by the principles of openness, fairness, and impartiality (De Carlo, et al. 2020) so that superior communication will also increase subordinates' trust. Trust is very important for cooperation. This is the most direct, economical, and effective way for leaders to increase organizational effectiveness (Kramer and Tyler, 1996). Increasing the confidence of subordinates in their superiors will help form harmonious relationships in the organization that result in organizational citizenship behavior and increase job performance, loyalty, and willingness of subordinates to comply with superior decisions and organizational regulations, as well as reduce uncertainty, lower transaction costs, increase internal resources for use more rational, thereby increasing organizational effectiveness (Barney and Hansen, 1994). The quality of the relationship is a determining factor in increasing trust if the support from the supervisor is carried out properly. This research will take in several organizations in Kebumen, Central Java and the subject of this research are front-line employees. The framework of this research uses social exchange theory, where subordinates will provide remuneration in the form of a positive contribution because superiors give positive attention or vice versa. Social exchange theory is one of the main theories for interpreting social exchange relations between parties in the regulation of human interaction. In particular, this theory is frequently discussed in the literature to explain social exchanges between superiors and subordinates. Cropanzano and Mitchell (2005) describe that social exchange occurs when employers treat their employees and the employees have responded with the attitude and positive work behaviors. In other words, if one party feels that the other is fulfilling a social obligation, then the exchange process continues; However, if the party views it negatively, then social exchange does not continue, and as a result, negative outcomes are likely. Based on the social exchange theory, this study investigates the social interactions of frontline employees and their supervisors. Abusive Supervisor (X1) and Trust -in- Supervisor (Y) Mayer et al. (1995) define trust as the willingness of a party to be vulnerable to the actions of others, based on the expectation that the other party will take certain important actions, regardless of the ability to monitor or control the other party. To be more specific, trust in supervisors in frontline employees refers to the willingness of front-line employees to rely on the actions, promises, words, or intentions of their immediate supervisors. Confidence in superiors ( trust in the supervisor ) is a trust in the integrity, fairness, and reliability of a person against a supervisor (Dizgah, Farahbod & Khoeni, 2011). As the holders of power in the organization, leaders have a great influence on both the staff and the organization. Researchers have paid attention to many positive things about leadership such as transformational leadership, charismatic leadership, and concern for increasing leadership effectiveness (Mitcel and Ambrose, 2007) such as trust in superiors which will impact commitment and performance. However, both the leader and the leadership have their dark side. Therefore, studying the "dark" side of leaders contributed to amendments to the theoretical leadership system. Increased productivity work environment, the sustainability of the employment relationship, and build trust play a very important role. therefore, it is important to explore the factors that increase or decrease the perception of trust-in-supervisors. one factor that has not been explored that may affect the subordinate is the abusive supervisor. abusive e supervisor refers to the perception of subordinates on the extent to which their supervisors are involved in display behavior hostile, verbal and non-verbal sustainability exclude physical contact (tepper, 2000). Researchers have made a list of examples of abusive supervisors, such as intimidation by threat of firing, aggressive eye contact, silence, and embarrass or ridicule someone in front of others. Studies of the results of abusive supervisors on the attitudes and behavioral outcomes of employees in the workplace have been conducted. For example, researchers have shown that abusive supervisors significantly improve employee morale release of employee stress and intention to move, avoidance interaction, and dehumanization of the organization. The researcher proposes a hypothesis: H1 Abusive supervisors ber influence negatively trust-in- supervisor Supervisor Support (X2) and Trust-in-Supervisor (Y) Supervisor support in the workplace is defined as the belief offered by the supervisor (Susskind, 2007) as related to the work to help the performance of their jobs. In other words, supervisor support is the general perception of employees that their supervisors acknowledge their contribution and pay significant attention to their well-being. Supervisor support reduces feelings of misery and increases the safety of an employee mindset, which is an important component of confidence. More specifically, according to social exchange theory, a positive social exchange between superiors and subordinates fosters a reciprocal norm (Gouldner, 1960). In other words, supervisory support can increase the interpersonal trust frontline employees feel in their supervisors. Based on the findings from the studies mentioned above, for this study, we conclude that supervisor support can help to increase the confidence of frontline employees. This is because, frontline employees generally perform a challenging task and many, therefore, help supervisors in learning and successfully handle the responsibility is an important indicator, support oversight tends to increase confidence in their superiors. Therefore, the second hypothesis is : H2 Supervisor support ber influence positively on the trust-in-supervisor. Communication Quality (X3) and Trust-in- Supervisor (Y) Communication quality refers to the sharing of information that is timely and accurate, both formal and informal (Morgan, et al., 1994). Ouedraogo and Ouakouak (2018) have studied the effects of communication as the natural process of change management. Their study revealed that good supervisor communication leads to an effective commitment from employees. Conversely, several studies have also emphasized that failure in proper communication leads to decreased trust and more uncertainty. Thus, in the context of a service setting, supervisors, co-workers, and frontline employees work together to deliver high-quality customer service. According to social exchange theory, positive social exchanges between superiors and subordinates foster reciprocal norms (Gouldner, 1960). In other words, the quality of communication from supervisors can increase the interpersonal perception of frontline employees' trust. Therefore, the researchers found the quality of communication in the workplace will improve the perception of trust-in-supervisor of frontline employees because of a piece of sharing information in a timely and accurate increase their trust in the supervisor/supervisor. Therefore, the third hypothesis in this study is: H3. The quality of the air communication positive impact on the trust-in-supervisor Relationship Quality (Z) as moderator If supervision is cruel frequent or prolonged, it always causes employees antipathy and then affects mistrust of subordinates (Xiaqi, 2012). Relating to the relationship factor with the leader/group. Tosi et al. (1990) said that the relationship between a leader and his subordinates will vary, and the variation in this relationship will be reflected in the quality of the relationship between superiors and subordinates. Therefore, the quality of the superior-subordinate relationship affects the formation of an employee's commitment to the company. The quality of the relationship between a superior and a subordinate is a level that shows the extent to which the closeness of the work and social relationship that is established between a superior and a subordinate in a dyad, as a result of the process of forming a role carried out by a subordinate. Furthermore, Landy (1989) states that if the superior- subordinate relationship occurs it is of high quality. then a boss will often discuss with his subordinates personal and work problems, and the boss will be very interested in the fun difficulties faced by subordinates. This indicates a personal and social bond between superiors and subordinates. Personal attachment and social happens this will prevent subordinates from the race isolation in the company. This research proposes a hypothesis H4. The quality of the relationship me moderating influence of abusive supervisor to t rust-in-supervisor H5. The quality of the relationship moderates the influence of supervisor support on t rust-in- supervisor Based on the above hypothesis, the thinking framework for this study is: Figure 1. Research Framework Variable measurement items were taken and adapted from previous research in the literature. Respondents use a five-point scale where 1 means "strongly disagree" and 5 "strongly agree" to respond to the questionnaire abusive Abusive SUperviso r Supervisor support Trust in supervisor Kualitas hubungan Kualitas komunikasi supervisor, supervisor support, the quality of communication, and the quality of relationships. Abusive supervisor refers to the perception of subordinates on the extent to which supervisors (supervisors) involved in displaying hostile behavior, verbal and non-verbal sustainability does not include physical contact (Tepper, 2002) was measured with five items. Supervisor support or employee support in the workplace is defined as the individual beliefs offered by supervisors ( Susskind, 2007) related to work to help their job performance. This variable is measured by 3 items. Communication quality refers to the timely and accurate sharing of information both formal and informal (Morgan, et al. 1994). This variable is measured by 2 items. Trust in a supervisor is a belief in the integrity, fairness, and reliability of a person towards superiors (Dizgah, Farahbod, and Khoeni, 2011). This variable was measured by an indicator according to Heider et al. (2015 ) measured at 4 items. Landy (1989) states the quality of the relationship as a level that shows the extent to which the closeness of the work and social relationship that is established between a superior and a subordinate in a dyad, as a result of the role formation process carried out by a subordinate. The quality of the relationship is measured by 2 items RESEARCH METHODS Referring to the problem formulation that has been set in the previous chapter, this study uses quantitative methods to test and prove the hypotheses that have been made through various tests and data processing. This is mentioned by Schiffman and Kanuk (2000: 19), quantitative research methods are related to methods for collecting data, sample design, and construction of data collection instruments. Malhotra (2006: 161) reveals that the quantitative approach is a research methodology that seeks to quantify data and usually applies certain statistical analyzes. Based on the dominant type of data processed in the form of numbers, this research is a quantitative research category (Sekaran, 2007). The tool used in this study is a questionnaire distributed to a sample of a predetermined population Population According to Sekaran (2011) population is an entire group of people, events, or things that researchers want to investigate. The area of generalization consisting of subjects who have certain qualities and characteristics determined by the researcher to study and then draw the conclusion is also called the population (Anshori and Iswati, 2009: 92). The target population is addressed clearly concerning the sampling unit, element, extent or scope, and time. The population used in this study is front-line employees in Kebumen, Central Java, Indonesia. Sample The sample is part of the population consisting of several members selected from the population (Sekaran, 2006: 123). The sample in this study was determined by the non-probability sampling method, which is a sampling technique that does not provide equal opportunities or opportunities for each member of the population to become a sample, by using the purposive sampling technique. The process of distributing research questionnaires was carried out through field surveys. The number of samples in this study is the first 100 samples. In this study, the sample is front-line employees who have worked at least 1 year to experience perceptions of leadership. Testing the research hypothesis was carried out by using the Structural Equation Model (SEM) approach based on Partial Least Square (PLS). Structural Equation Model (SEM) is a field of statistical studies that can test a series of relationships that are relatively difficult to measure simultaneously. According to Santoso (2014), SEM is a multivariate analysis technique which is a combination of factor analysis and regression analysis (correlation), which aims to examine the relationship between variables in a model, be it between indicators and constructs, or relationships between constructs. Validity and Reliability This study uses a questionnaire in collecting research data. To determine the level of validity and reliability of the questionnaire, the researcher used the SmartPLS 3 .0 program. The validity testing procedure is convergent validity by correlating the item score with the construct score which then produces the loading factor value. The rated loading factor is said to be high if the component or indicator correlates more than 0.7 0 the construct to be measured. However, for research in the early stages of development, a loading factor of 0.5 to 0.6 is considered sufficient (Ghozali, 2008). Reliability states the extent to which the results of measurements can be trusted or reliable and provide relatively consistent measurement results after several measurements are made. To measure the reliability level of the research variables, the coefficient alpha or Cronbachs alpha and composite reliability were used. Measurement items are said to be reliable if they have an alpha coefficient value greater than 0.6 (Malhotra, 1996). Inner Model The purpose of the structural model test is to see the correlation between the measured constructs which is the t-test of the partial least square itself. The structural or inner model can be measured by looking at the R-Square model value which shows how much influence between variables in the model. Then the next step is to estimate the path coefficient which is the estimated value for the path relationship in the structural model obtained by the bootstrapping procedure with a value that is considered significant if the statistical t value is greater than 1.96 (significance level 5%) This research included descriptive research. The study aims to determine the effect of abusive supervisors, supportive supervisors, and the communication quality of the trust in supervisor with the quality of the relationship as a moderating. The dependent variable used in this study is trust in the supervisor, while the independent variable is abusive supervisor, supportive supervisor, and communication quality, while relationship quality is the moderator. RESULTS AND DISCUSSION Respondent Characteristics The results of the analysis of the characteristics of the respondents are shown in table 1: Table 1 Descriptive Total Percentage Gender Male 34 0.34 Female 66 0.66 Age (years) 17-25 37 0.37 26-35 36 0.36 36-40 19 0.19 > 41 8 0.08 Education High school 78 0.78 Associate Degree 15 0.15 Bachelor 7 0.07 Length of working 1-3 53 0.53 (years) 4-6 37 0.37 > 6 10 0.10 Source: primary data processed, 2021 Based on table 1 the respondents can be divided into two categories: male and female. Of the 100 respondents who obtained, the composition of the respondents by sex is 66 respondents or 66% gender to female, and the remaining 34 or 34% were males. Table 1 shows the highest number of respondents in the range 17-25 years (37%), and at least> 41 years as many as 8 people (8%). Based on the education of the respondents, 78% are high school graduates, 15 respondents (15%) are associate degree graduates, and 7 respondents (7%) are bachelor degree. Based on the data of respondents in this study, 53 respondents (53%) with 1-3 years length of work, 37 respondents (37%) with 4-6 years length of work, and 10 respondents (10%) with > 6 years length of work. Outer Model Testing (Measurement Model) This research model will be analyzed using the Partial Least Square (PLS) method and assisted by the SmartPLS 3.0 software. PLS is an alternative method of Structural Equation Modeling (SEM) that can be done to solve problems in the relationship between variables which are very complex but the sample size is small (30- 100 samples) and has non-parametric assumptions, meaning that the data does not refer to either certain distribution (Yamin and Kurniawan, 2009). Convergent Validity Convergent validity is done by looking at the item reliability which is indicated by the loading factor value. A loading factor is a number that shows the correlation between the score of a question item with the score of the indicator construct that measures the construct. The loading factor value is greater than 0.7 which is said to be valid. However, according to Hair et al. (2010) for the initial examination of the loading factor matrix is approximately 0.3 considered to have met the minimum level, and for a loading factor of approximately 0.4 is considered better, and for a loading factor greater than 0.5 is generally considered significant. In this study, the loading factor limit used was 0.7. The results of the loading factor can be shown as in Table 2 : Table 2 Load ing Factor Value Variable loading Factor X1.1 0.783 X1.2 0.936 X1.3 0.820 X1.4 0.773 X2.1 0.780 X2.2 0.851 X3.1 0.958 X3.2 0.946 X3.3 0.926 Y1.1 0.880 Y1.2 0.926 Y1.3 0.908 Z1.1 0.935 Z1.2 0.999 Source: primary data processed, 2021 From the results of data processing with SmartP LS shown in table 2, that the majority of the indicators on each variable in this study have a loading factor value greater than 0.70 and are said to be valid (Ghozali, 2008). Discriminant Validity The next evaluation is to see and compare discriminant validity and the square root of average variance extracted (AVE). The measurement model is assessed based on the measurement of cross-loading with the construct. If the correlation between the constructs with each indicator is greater than the size of the other constructs, then the latent construct predicts the indicator better than the other constructs. According to Ghozali (2008), if the value is higher than the correlation value between the constructs, good discriminant validity is achieved (if AVE> 0.5). The following are the measurement results with AVE for each indicato Table 3 AVE Variable AVE X1 0.583 X2 0890 X3 0.592 Y 0813 Z 0.710 Source: primary data processed, 2021 From the measurement results in table 4 above, it can be seen that five variables meet the criteria Composite Reliability To determine composite reliability, if the composite reliability value is > 0.8, it can be said that the construct has high or reliable reliability and> 0.6 is said to be quite reliable (Ghozali, 2008). The results of the reliability test in this study are: Table 5 Composite Reliability Variable Composite Reliability X1 0.873 X2 0.961 X3 0812 Y 0.946 Z 0.873 Source: primary data processed in 2021 The SmartPLS output results in Table 4, it shows that the composite reliability value for all constructs is above a value of 0.70. With the resulting value, all constructs have good reliability following the required value limits. Inner Model Testing The goodness of fit model is measured using the R-square of the dependent latent variable with the same interpretation as the regression. R-square predictive relevance for structural models measures how well the observed value is generated by the model and also its parameter estimates. The value of R-square> 0 indicates the inner model measurement model with PLS is as follows: Table 5 R Square R Square Adjusted R Square Y 0.4 0. 2 2 Source: primary data processed, 2021 Results of the study showed variability in the variables X1, X2, and X3 is 40% while the remaining 6 0% explained by other variables. The adjusted R square value has a value with an interval between 0 and 1. If the Adjusted R Square value is getting closer to 1, it shows that the independent variable (X) explains the variation of the dependent variable (Y) is getting better. In the present study, can not you the value of adjusted R square 0.22 or 22 %. So it can be concluded that the 22 % variation that occurs in variable Y can be explained by the independent variable, while the rest can be explained by other variables. Hypothesis test Hypothesis testing is carried out based on the results of the inner model which includes the r-square output, parameter coefficient, and t-statistic. To see whether a hypothesis can be accepted or rejected, among others, by paying attention to the significance value between the constructs, t-statistics, and p-values. This research hypothesis testing was carried out with the help of SmartPLS (Partial Least Square) 3.0 software. The rules of thumb used in this study are t- statistic> 1.96 with a significance level of p-value 0.05 (5%) and the beta coefficient is positive. The value of testing the hypothesis of this study can be shown in table 5 and the results of this research model can be described as shown in Figure 2 : In testing the hypothesis, the value analyzed is the value that is in the t-statistic generated from the PLS output by comparing it with the t-table value. The PLS output is an estimation of the latent variable which is the aggregate linear of the indicator. Hypothesis Testing Based on the statistical results, the hypothesis test can be explained in the following table: Table 6 Hypothesis Test Origi nal sam ple (O) Sam ple mea n (M) Standar d deviatio n (STAT DEV) T statistic (O / STATD EV) P- val ue X1 → Y 0.58 3 0.56 8 0.192 3,033 0.0 33 X2 → Y 0.89 0 0.84 0 0.137 6.502 0.0 00 X3 → Y1 0.59 2 0.52 6 0.116 5.088 0.0 00 Modera ting effect 1 X1 → Z 0.01 8 0.12 2 0.172 1.281 0.2 01 Modera ting effect 2 X2 → Z 0.22 0 0.85 5 0.038 5.825 0.0 00 Source: primary data processed, 2021 The first hypothesis which tests the negative effect of abusive supervisors on trust in supervisors in table 6 showed that the original sample value is 0.583 and the t statistic is 3.033 with a p-value of 0.0 33, so hypothesis 1 is accepted. Figure 2 shows the regression coefficient of -0323 indicates this effect is negative (H1 accepted) Hypotheses 2 that tested the effect of positive supervisors' support to trust in supervisors in Table 6 showed the value of the original sample was 0. 890 and statistical t 6.502 with p-value 0.000 then hypothesis 2 is accepted. Figure 2 showed the regression coefficient of 0. 1 indicates this effect was positive (H2 accepted) The third hypothesis showed the positive influence ofcommunication quality of subordinates' superiors on trust in supervisors. Table 5 showed that the original sample value is 0. 592 and t statistic 5. 0 88 with a p-value of 0.0 00, then hypothesis 3 is accepted. Figure 2 showed the regression coefficient of 0. 158, it indicates that this effect was positive (H3 accepted). Based on statistical calculations, it can be concluded that a quality relationship is not moderating abusive supervisor to trust in the supervisor. It can be seen that the t-statistic is 1.281 which is less than the t- table 1.96 (H4 is rejected) Based on statistical calculations, it can be concluded that relationship quality moderating supportive supervisors to trust in the supervisor It can be seen that the t-statistic is 5,825 greater than the t- table 1.96 (H5 accepted). CONCLUSION Front employees as an important resource of the company play an important role in business operations. Therefore, researchers and practitioners place great importance on the determinants of front-line employees' ongoing relationships with their superiors. In this case, taking the social exchange theory, this study was conducted to test the trust-in- supervisor antecedents within one framework. Following social exchange theory, the research findings reveal that hypothesis 1 is that abusive supervision is significantly proven to reduce trust in frontline employee supervisors. This finding is in line with previous research, which showed that abusive supervision led to negative outcomes such as low job satisfaction, low affective commitment, low organizational identification, employee stress, and intention to leave and enter. The results of this study support the research of Ji & Jan (2020) that abusive supervisors harm trust in supervisors. So if the supervisor performs abusive behavior, the subordinates' trust in their superiors will decrease. Hypothesis 2 showed that supervisor support has a positive effect on trust in supervisors. This supports the research of Ji & Jan (2020) that supervisor support has a positive effect on trust in supervisors. So if supervisors show supportive behavior, then the subordinates' trust in their superiors will increase. Consistent with previous studies, the study also shows that the communication quality has a positive effect on trust in supervisor in hypothesis 3. This research supports previous studies that do Ji and Jan (2020) that the communication quality has a positive effect on trust in the supervisor. Based on statistical calculations, it can be concluded that a quality relationship is not moderating abusive supervisor to trust in the supervisor. The results of this study support Tepper (2000) that good relationship quality can strengthen supportive behavior towards trust in supervisors. The results showed support for the social exchange theory which states that subordinates will provide remuneration in the form of a positive contribution because the boss gives positive attention or vice versa. But, this study does not prove the moderating effect of relationship quality on the abusive effect of supervisors on trust in supervisors in the context of frontline employees. Managers must educate supervisors about the negative outcomes of abusive supervisors. The findings of this study reveal that poor supervision in the workplace can exacerbate perceptions of trust-in- supervisors, which in turn affects the likelihood of collaboration between frontline employees and supervisors. Hence, in this regard, there is a need to formulate an organizational policy based on an organization-wide zero tolerance of abusive behavior, leading to a sustainable working relationship. Also, managers are advised to improve the selection and recruitment process for supervisors to reduce abuse of supervision in the workplace. Managers must educate and train supervisors to be more supportive of front-line employees. To increase supervisory support in the workplace, supervisors need to be encouraged to provide recognition and appreciation to front line employees, providing support in the form of assistance in carrying out tasks, and providing useful feedback. Supportive supervision will increase the level of trust of front-line employees in their superiors, which will trigger their cooperative behavior. The research findings reveal that the supervisor's communication quality and the quality of relationships that leads to trust-in-supervisors. Managers should place more emphasis on the communication skills of the applicant during the hiring process. Besides, managers must offer extensive training and on-the-job development workshops. REFERENCES Amir, T. S., & Malik, A. A. (2016). Machiavellianism and counterproductive work behavior at workplace. Journal of Educational & Social Sciences, (4)1, 14-29. Doi 10.20547/jess0411604102 Barney, Jay B., Hansen, & Mark, H. (1994). Trustworthiness as a Source of Competitive Advantage. Strategic Management Journal, 15, 175-190 Burton, James, P., Hoobler, Jenny M., Scheuer, Melinda D. (2012). Supervisor Workplace Stress and Abusive Supervision: The Buffering Effect of Exercise, Journal of Business and Psychology, 27(3), 271-279 De Carlo A., Dal, Laura, Carluccio, Francesca, Colledani, Daiana, & Falco Alessandra. (2020). Positive Supervisor Behaviors and Employee Performance: The Serial Mediation of Workplace Spirituality and Work Engagement. Front Psychol, 11, 1- 8. Doi: 10.3389/fpsyg.2020.01834 Dizgah, Morad, Rezaie, Farahbod, Farzin & Khoeni, Bahare. (2011). Relationship between perceptions of organizational justice and trust in Guilan Tax Affair Organization, Interdisciplinary Journal of Contemporary Research in Business, 3(7), 341-351. Ehnert, I. (2009). A Conceptual and Exploratory Analysis from a Paradox Perspective. Springer: Heidelberg. Germany Ji, Seonggo & Jan, Ihsan Ullah. (2020). Antecedents and Consequences of Frontline Employee’s Trust-in- Supervisor and Trust-in-Coworker, Sustainability, 12, 2-17 Ghozali, Imam. (2008). Structural Equation Modelling, Edisi II. Universitas Diponegoro. Semarang. Gouldner, A.W. (1960). Norm of Reprocity: A Preliminary Statements. American Sociological Review, 25, 161-178 Landy, J.F. (1989). Psychology at Work Behavior, California: Books. Cole Publishing Company. Malhotra, N. K. (1996). Marketing Research An Applied Oriented. Second Edition. Prentice Hall, Inc.New Jersey Mayer, Roger J., Gavin, Mark B. (2005). Trust in Management and Performance: Who Minds the Shop While the Employees Watch the Boss?. The Academy of Management Journal. 48(5), 874-888 Mitchell, M. S., & Ambrose, M. L. (2007). Abusive Supervision and Workplace Deviance and The Moderating Effects of Negative Reciprocity Beliefs. Journal of Applied Psychology, 92(4), 1159–1168. Hair, Jr et.al. (2010). Multivariate Data Analysis (7th ed). United States: Pearson https://dx.doi.org/10.3389%2Ffpsyg.2020.01834 https://www.researchgate.net/journal/The-Academy-of-Management-Journal-0001-4273 Rukaiyah, ST. & Muliana. (2018). Penguatan Hubungan Antara Leadership Stategy, Knowledge Management dan Orientasi Bisnis Terhadap Kinerja UKM. Jurnal Manajemen Bisnis, 5(2), 32-40 Rapisarda S., Mora D., Ometto I. (2020). Three uses of Virtual Reality in work and organizational psychology interventions - A dialogue between Virtual Reality and organizational well-being: relaxation techniques, personal resources, and anxiety/depression treatments. TPM Test. Psychom. Methodol. Appl. Psychol, 27, 129– 143. DOI 10.4473/TPM27.1.8 Rowold, Jens. (2008). Effects of Transactional and Transformational Leadership of Pastors. Pastoral Psychology, 56(4), 403-411 Santoso, Singgih. (2014). Statistik Parametrik Edisi Revisi. Jakarta: Elex Media Komputindo Sekaran, Uma. 2006. Research Methods For Business. Edisi 4. Buku 1. Jakarta: Salemba Empat Schiffman and Lazar Kanuk. 2000. Costumer Behavior, Internasional Edition, Prentice Hall Stankevciute, Ž. dan Savanevcien, A. 2018. Designing sustainable HRM: The core characteristics of emerging field. Sustainability. 10, 4798. Xiaqi, Ding, Kun, Tian, Chongsen, Yang. 2012. Abusive supervision and LMX: Leaders' emotional intelligence as antecedent variable and trust as consequence variable, Chinese Management Studies, 6(2):257-27 Tepper, B.J. Consequences of abusive supervision. 2000. Acad. Manag. J. 43, 178–190. Tosi, H.L., Rizzo, ].R. & Carrol, S.I. 1990. Managing Organizational Behavior. (2nd Edition). New York: Harper Collins Publisher Yamin Sofyan, Kurniawan Heri. 2009. SPSS Complete: Tekhnik Analisis Statistik Terlengkap dengan software SPSS. Jakarta: Salemba Infotek https://www.researchgate.net/journal/Pastoral-Psychology-1573-6679 https://www.researchgate.net/journal/Chinese-Management-Studies-1750-614X https://www.researchgate.net/journal/Chinese-Management-Studies-1750-614X