The role of trust as an informal mechanism in the management control system of performance effectiveness

Performance is a variable explaining the effectiveness of internal control systems that managed based on management accounting information and functions. There are some aspects of behavior that remain unexplained in understanding the effectiveness of the performance of organizations and individuals affected by the data and accounting functions. In addition, social mechanisms such as culture, honesty, group identification and trust play an important role in mitigating agency problems. This study aims to investigate what and how the role of trust in organizational performance is. It develops and tests a conceptual relation model between trust and performance through organizational commitment and job satisfaction. It also examines contextual factors in the form of cognitive orientation of the conceptual model that has been built. The data were collected by using questionnaire-based survey method for the employees based on the criteria of having direct supervisors and directly responsible for them, and a minimum work period of one year. They were analyzed using SEM PLS. It shows that organizational commitment mediates the positive relationship between trust and job satisfaction. The positive relationship between trust and performance is mediated by organizational commitment and job satisfaction. It also shows that cognitive orientation influences the relationship between job satisfaction and performance. This study contributes to the management accounting literature, that is, a conceptual relationship model of trust to performance in the realm of empirical research.


A B S T R A C T
Performance is a variable explaining the effectiveness of internal control systems that managed based on management accounting information and functions.There are some aspects of behavior that remain unexplained in understanding the effectiveness of the performance of organizations and individuals affected by the data and accounting functions.In addition, social mechanisms such as culture, honesty, group identification and trust play an important role in mitigating agency problems.This study aims to investigate what and how the role of trust in organizational performance is.It develops and tests a conceptual relation model between trust and performance through organizational commitment and job satisfaction.It also examines contextual factors in the form of cognitive orientation of the conceptual model that has been built.The data were collected by using questionnaire-based survey method for the employees based on the criteria of having direct supervisors and directly responsible for them, and a minimum work period of one year.They were analyzed using SEM PLS.It shows that organizational commitment mediates the positive relationship between trust and job satisfaction.The positive relationship between trust and performance is mediated by organizational commitment and job satisfaction.It also shows that cognitive orientation influences the relationship between job satisfaction and performance.This study contributes to the management accounting literature, that is, a conceptual relationship model of trust to performance in the realm of empirical research.

INTRODUCTION
In today's business environment, organizational performance is dominantly influenced by various factors, regardless of the choice of strategy and the level of environmental uncertainty encountered.In this case, performance is a variable capable of explaining the effectiveness of a control system, based on management accounting information and functions.Among the influential factors, performance is the most frequently variable in management accounting research.However, there are still aspects of behavior that are unexplained in understanding the effectiveness of the performance of organizations and individuals affected by data and accounting functions.
Management control system, as a process conducted by managers to ensure the effectiveness and efficiency of resource use (Anthony 1965), is an important factor that contributes to the success and achievement of performance.Some of the literature on management control defines control systems into formal and informal forms (e.g.Morand 1995, Rowe 2004, Sitkin & George 2005).Formal control refers to the ability of an organization to create incentive mechanisms (e.g.performance evaluation systems, compensation, etc.), while informal control refers to the ability of an organization to influence/establish a social control mechanism, in which both are expected to motivate members' behavior in order to achieve the objective alignment (Rowe 2004).
So far, there has been some empirical evidence explaining the importance of formal control in organizations (e.g.Lau & Sholihin 2005, Lau et al. 2008, Sholihin & Pike 2010).Yet, the role of informal control systems seems to be in question.The development of literature puts attention to tradeoffs between formal control mechanisms inclined to incentive orientation and informal control mechanisms indicating that social mechanisms such as culture, honesty, group and trust identification play an important role in mitigating agency problems (Evans et al. 2001, Towry 2003).Meanwhile, Rowe (2004) also finds that trust is an important informal control mechanism to reduce free-rider problems in cross-functional teams.Specifically, Coletti et al. (2005) explains that trust becomes important when the members of the organization are in situations and conditions of work that require collaboration.
Although trust, as an informal control mechanism, is important, it is not generally accepted but is considered insufficient for superiors to control the behavior of their subordinates (Hartman & Slapnicar 2009).Therefore, bosses within an organization tend to keep using formal controls, such as performance evaluation systems (Malhotra & Munighan 2002).Hartman & Slapnicar (2009) conducted a study by establishing and testing the model of the relationship between formal control and trust to investigate whether the formal control system also has an effect on trust.The result is that the formal performance evaluation system has a positive effect on trust through the quality of feedback and procedural fairness.In addition, Sholihin & Pike (2010) also developed and tested a model of the relationship between financial and non-financial performance measurement and organizational commitment.The result indicates that trust mediates a positive relationship between non-financial performance measurement and organizational commitment.The findings explain that trust as an informal control mechanism is also established through a formal control system.
It is clear about the importance of the role of trust as the element of informal control that explains the relationship between the control system and the behavior of organizational members.For that reason, this study aims to investigate what and how the role of trust in increasing the performance of the organization.Such is important to investigate because the relationship between trust and performance is not clearly understood.If formal control increases the informal control mechanism (e.g., trust), is trust as an informal control mechanism important enough to improve organizational performance?Several previous findings, such as Li & Tan (2013), provide evidence that trust to superiors positively influence the performance of subordinates through psychological perceptions.De Jong & Elfring (2010) also conducted a study and found that trusts positively affected the team's performance.Meanwhile, Sholihin & Pike (2009) found that trust was negatively related to performance.In general, the findings conclude that trust has an effect on performance, but still raises the question on how the relationship patterns which are formed between trust and performance, and whether differences in cognitive orientation affect the relationship pattern.
This research tries to answer the research question by developing and testing a model of relationship between trust and performance.Some earlier studies that tested the antecedents of trust (Hartman & Slapnicar 2009) and tested trusts as mediating variable that linked performance measurement and organizational behavior define trust as "subordinate"s trust to superior."In order to facilitate the contradictory findings of previous research, this study attempts to define trust into broader terms and concepts.Trust in this study is defined as the overall trust formed on individuals within the organization, that is, interpersonal trusts consisting of "subordinate"s trust to superior" and " trust between co-workers" (Cho & Park 2011, Cook & Wall 1980, Costigan et al. 1998, Errol & Bruce 2005) and institutional trust consisting trust between individuals and the organizations or organizational units in which they work (Baek & Jung 2015).In other words, the trust measured in this study is a latent construct formed of three dimensions ("trust between superiors and subordinates", "trust between co-workers" and "institutional trust").
Furthermore, this research tries to explore the path that connects trust to performance, that is, through organizational commitment perception and job satisfaction perception.The reason for the researchers to test these two variables as mediating relationships between trust and performance is because some empirical evidence from previous studies indicates the linkages, but the findings are also contradictory.Through multivariate testing, Lau et al. (2008) found a significant positive relationship between trust and job satisfaction, but also found no significant relationship between trust and organizational commitment.While Sholihin & Pike (2009) through a similar study but on different objects found the opposite relationship, i.e. there is a significant positive relationship between trust and organizational commitment, but found no significant relationship between trust and job satisfaction.In order to facilitate the second contradictory findings, the researchers also aim to examine whether the model or pattern of relationships formed among trust, organizational commitment, job satisfaction and performance is moderated by the individual's cognitive orientation within the organization.

THEORETICAL FRAMEWORK AND HYPO-THESIS Trust, Organizational Commitment, and Performance
Trust has been defined and conceptualized by many experts.Hosmer (1995) defines trust as an expectation of individual decisions and actions based on ethical principles.Meanwhile, many studies of organizational behavior (Mayer & Schoorman 1998, Colquit et al. 2007, McEvily & Tortorielo 2011) tend to adopt the definition of Rousseau et al. (1998), where trust is defined as a psychological state in which the individual accepts the risk of loss that may be obtained because it has positive expec-tations on the intentions or behavior of others.
From the domain of accounting research, researchers tend to use a trust definition as part of the interpersonal trust, i.e. subordinate"s trust to the superior.The reason is that the performance evaluation system is the process by which the superior/boos evaluates the performance of his subordinates based on a set of targets that have been set.However, based on the hierarchy level of trust in the organization, there are two types of interpersonal trust, namely subordinate"s trust to superiors and trust among colleagues (Cho & Park 2011, Cool & Wall 1980, Costigan et al. 1998and Errol & Bruce 2005).However, interpersonal trust has not adequately explained all the trusts that are formed within the organization.Cook & Wall (1980) uses the size of trust to management to predict organizational commitment.Meanwhile, Baek & Jung (2015) state that there are two general types of trust relationships within organizations, namely interpersonal trust and institutional trust.Institutional trust is trust formed between individuals and organizations or the organizational units where they work (Baek & Jung 2015).In order to facilitate the contradictory findings of the previous findings, this study uses these three types of trust as the dimensions that form the construct latent of trust generally contained within the organization.
Sholihin & Pike (2010) conclude that trust culture between subordinate and superior in police organizations is expected to increase organizational commitment.While in other studies, Sholihin & Pike (2009) also find a significant positive relationship between trust and organizational commitment.Rousseau et al. (1998) previously also explains that individuals with a high degree of trust toward colleagues or organizations will have higher confidence and will form a more positive picture of the object that he believed.Therefore, when an individual has a high trust to colleagues or organization as an object that he believes, the positive image formed on the object will create a sense of comfort and calm for the individual to be in the situation and environment being faced.When comfortable feelings and work situations have arisen, it is possible for the individual to be loyal to the organization where he works.
Loyalty to the organization is a reflection of organizational commitment that is formed at the level of each individual.Therefore, researchers predict that trust is positively related to organizational commitment.H1a: Trust as an informal control mechanism is positively related to organizational commitment.
There are only few studies, related to accounting literature, that examine the relationship between organizational commitment and performance, such as Nouri & Parker (1998), Chong & Eggleton (2007) and Sholihin & Pike (2009).However, they find that organizational commitment is associated with performance.Porter et al. (1974) explains that two important characteristics of organizational commitment are strong beliefs and acceptance of organizational goals and values as well as a desire to provide hard effort for the benefit of the organization.If organizational commitment creates an individual desire to strive to achieve organizational goals, it is most likely that performance improvement is influenced by the level of organizational commitment.Through this empirical evidence, this study predicts the relationship between organizational commitment and performance.H1b: Organizational commitment is positively related to performance.
When hypothesis 1a and hypothesis 1b are supported (there is positive relationship between trust -organizational commitment and organizational commitment -performance), it shows that there is an indirect relationship between trust and performance through organizational commitment.H1c: Trust as an informal control mechanism has an indirect positive relationship with performance through organizational commitment.

Trust, Job Satisfaction, and Performance
The relationship between trust and performance which is described through organizational commitment may not reflect the overall model/relationship pattern.Some empirical studies have shown that job satisfaction is also influenced by trust.Lau & Sholihin (2005) assert that trust can be used to explain why the use of the metrics of specific performance measure can improve individual job satisfaction.Read (1962) previously explains that trust affects job satisfaction through increasing subordinate confidence in his boss.When the individual believes that his or her boss or partner will act without harming his interests (interests within certain limits), it will create a harmonious work relationship within the organization.
A harmonious work environment will increase the amount of information exchange among individuals within the organization, create communication and collaboration, and help each other in every difficulty encountered in every job, which ultimately leads to satisfaction in work.But on the contrary, if in the working environment only creates a low trust in the organization, the individuals are reluc-tant to be open to colleagues, superiors and subordinates (Lau et al. 2008).The result is that the environment and harmonious working condition will not be created.So, with such condition, it will be difficult to get job satisfaction.Based on the description, this study predicts that trust is positively related to job satisfaction.H2a: Trust as an informal control mechanism is positively related to job satisfaction.
When trust can increase the exchange of information, the intensity of communication, and the desire to help each other in every difficulty to create job satisfaction, the job satisfaction will stimulate the spirit to work that ultimately affect the productivity of performance.Therefore, the resulting performance will also be influenced by job satisfaction formed on each individual.The next hypothesis predicts that job satisfaction affects performance.H2b: Job satisfaction is positively related to performance.
When hypothesis 2a and hypothesis 2b are supported (positive relationship between trust-job satisfaction and job satisfaction-performance), it shows that there is an indirect relationship between trust and performance through job satisfaction.H2c: Trust as an informal control mechanism has an indirect positive relationship with performance through job satisfaction.

Organizational Commitment and Job Satisfaction
Some previous studies agree that employees who are committed to the organization may already have high job satisfaction experience (Sterr 1977, Bateman & Strasser 1984, DeCotiss & Summers 1987, in Lau et al. 2008).Individuals who are highly committed to the organization will find that the organization's goals are important, so they tend to complete their tasks non-coercively (Lau et al. 2008).In such a condition, job satisfaction will be higher because the individual motivation will appear to complete tasks more quickly and appropriately in order to create efficiency achievement of organizational goals.Therefore, this study predicts that high organizational commitment will be associated with high job satisfaction.H3: Organizational commitment is positively related to job satisfaction.

Trust and Cognitive Orientation
The theory of individualism-collectivism distinguishes two main types of behavior or individual inclinations (Chow et al. 2001).Individualism is a condition in which the individual interests take precedence over group needs, while collectivism is a condition in which the individual feels that he or she is part of the team (Wagner 1995).When an individual has a tendency to cognitive orientation of individualism, the trust formed as an informal control mechanism within the organization will be less useful because of the decreased individual desire to cooperate with each other.Conversely, when an individual has a tendency to cognitive orientation of collectivism, the trust formed as an informal control mechanism within the organization will be more beneficial because of the increased desire of individuals to work together.
Based on the above descriptions, this study suggests that the individual condition with the cognitive orientation of individualism will weaken the positive relationship among trust, job satisfaction, organizational commitment and performance.Conversely, the individual condition with cognitive orientation of collectivism will strengthen the positive relationship among trust, job satisfaction, organizational commitment and performance.H4: A model of the relationship among the variables of trust, job satisfaction, organizational commitment and performance will show a stronger positive relationship in individuals with cognitive orientation of collectivism.

Research Model
The pattern of relationships among the variables of the overall hypotheses predicted in this study is illustrated in Figure 1.

RESEARCH METHOD Research Sample and Design
The collection of research data for the purpose of hypothesis analysis and testing is done using sur-vey method.Survey method is chosen because this method is considered the most appropriate and most feasible method to capture the phenomenon and analyze the path of relationship among the variables that have been built.The survey is conducted using a measurement tool in the form of questionnaires.The form of questionnaires that will be presented to the respondent is in the form of printed and online questionnaires.The selection of printed questionnaires is intended to make it easier for respondents who do not have online access, while the selection of online questionnaires is to make it easier for respondents to fill out questionnaires through their gadgets.
Population of this research is employees who work at organizations in DI Yogyakarta.The main criteria of employees to be selected by respondents in this study are those who have direct supervisors and directly responsible to them.The second criterion is the employees who have working period of more than one year, because within a minimum of one year, it is assumed that the employees have had enough experience to assess their employer, colleague and work environment.

Variables Measurement 1. Trust
Trust in this study is a latent construct formed from three dimensions that reflect the phenomenon of trust that occurs in the organization, namely subordinates trust to superiors, trust among coworkers, and institutional trust (intra-individuals trust to organizations or organizational units).Conceptual level differences will become an issue and are important to consider when estimating the influence of trust on behavioral measures such as organizational commitment (Baek & Jung 2015).Based on the hierarchy of the levels of trust in the organization, there are two types of trust relationships between individuals, that is, subordinate trust to superiors and trust among coworkers, in which both of them are part of the interpersonal trust (Cho & Park 2011, Cool & Wall 1980, Costigan et al. 1998and Errol & Bruce 2005).
The subordinate trust to superior which in this study is referred to as trust to superiors is measured using a questionnaire instrument containing 4 items of statement adopted from Posdakoff et al. (1990).Respondents will be asked to complete a questionnaire containing statements about: 1) "I feel pretty sure that my boss will always try to treat me fairly" 2) "My boss will never try to make a profit by cheating employees" 3) I trust and believe in the integrity of my boss" 4)" I will support my boss in almost any emergency."The all responses are measured using Likert scale, 1 = strongly disagree, 2 = disagree, 3 = neutral, 4 = agree, 5 = strongly agree.
Trust among coworkers in this study is defined as trust between individuals with coworkers who are at one functional level or the same division within the organization.This variable is measured using an instrument consisting of 4 items of statements developed by Cook and Wall (1980).The responses measured in this study are statements about: 1) "If I get trouble with my job, I know that my colleague will try to give help" 2) "Most of my coworkers can be trusted.When they say that they will do something, it will be done" 3) "I feel very confident with my coworkers" ability" 4) "Most of my coworkers will keep doing their work even though no boss who oversees."All responses are measured using Likert scale, 1 = strongly disagree, 2 = disagree, 3 = neutral, 4 = agree, 5 = strongly agree.
Institutional trust is defined as trust between individuals and organizational units or within the context of this research is the department within the organization.The instrument used is adopted from Cook and Wall (1980) containing 4 items of statement measured using Likert scale, 1 = strongly disagree, 2 = disagree, 3 = neutral, 4 = agree, 5 = strongly agree.The items of statement are about: 1) "My department is seriously striving to carry out the advice and opinions I express" 2) "My department can be trusted to make the right decisions for the future of the company" 3) "My department is seen doing a job efficiently "4) "I feel pretty sure that my department will always try to treat me fairly."

Job satisfaction
If several researchers such as Brownel (1982), Chenhall & Brownel (1988), Frucot & Shearon (1991), Harrison (1992) and Lau et al. (2008) used the short version of the Minesota Satisfaction Questionnaire instrument containing 20 statement items developed by Weiss et al. (1967), this study prefers to use the questionnaire instrument developed by Rusbult and Farrel (1983) which is also used by Sholihin and Pike (2009).This instrument is chosen because it is considered more understandable and the characteristics of the questions are quite concrete compared to 20 Minnesota statement items.
The instrument used by Rusbult and Farrel (1983) contains 6 items of questions: 1) "Considering all things, how satisfied are you with the current job?" 1 = very dissatisfied, 5 = very satisfied.2) "In general, how much do you love your job?" 1 = really dislike, 5 = really like 3) "Aware of what you know now, if you have a chance to go back to the past time, what do you do about this work?" 1 = I certainly will not take this job, 5 = I will take this job without hesitations.4) "If your good friend says that he is interested in the work you are doing, what would you say to him", 1 = strongly not recommend, 5 = strongly recommend.5) "How good is the present job when compared to the ideal job you think", 1 = very far from ideal, 5 = very close to ideal.6) "How good is your present job when compared to the job you want", 1 = very unsuitable as the job I want, 5 = very suitable as the job I want.

Organizational Commitment
In this study, organizational commitment is measured using a short version of OCQ (organizational commitment questionnaire) of Porter et al. (1974) and is also used by Mowday et al. (1979).This instrument contains 9 statement points with 7 likert scores (1 = strongly disagree, 7 = strongly agree).This instrument is chosen because of the reliability that has been tested through some previous research in management accounting related to organizational commitment (Magner et al. 1995, Nouri & Parker 1998, Lau et al. 2008, Sholihin & Pike 2009).Some examples of statement points in this questionnaire are: 1) "I will accept almost all types of work assigned in order to be able to stay in this organization" 2) "I am proud to tell others that I am part of this organization" 3) "I do care for the fate of this organization "4)" I want to give a tougher effort than I should give for the purpose of helping the organization achieve its success."

Cognitive Orientation
The cognitive orientation in this study is defined as the type of behavior or individual tendency that is more likely to be individualist (preferably selfemployed) or collectivist (prefer to cooperate).This variable is measured using an instrument adapted from Triandis & Gelfand (1998).Respondents will be asked to complete a questionnaire containing statements such as: 1) "When other people work better than me, I feel encouraged."2) "I feel comfortable when working with others."3) "I prefer to depend on myself rather than others."4) "It is important for me to finish the job better than others."The questionnaire is measured using Likert scale, 1 = strongly disagree, 9 = strongly agree.Respondents will be asked to answer questions in the questionnaire based on their actual experience and actions on the condition, not on what should be done.

Hypothesis Testing
Hypothesis testing is done using Structural Equation Modeling (SEM) with Partial Least Square (PLS) method approach.PLS is used in order to maximize the value of variance explained by the predictor variable to the dependent variable through R-Square as a goodness-of-fit measure (Chin & Newsted 1999).But before that, the researchers assess the quality of the measurement model using item reliability and construct reliability test as well as convergence validity and discriminant validity test for reflective construct (Bagozzi 1994).Hypothesis 1a is tested by analyzing the coefficient score of path between the variable of trust -job satisfaction.Hypothesis 1b is tested by analyzing coefficient score of path between the variable of job satisfaction -performance.Hypothesis 1c is tested by analyzing the coefficient score of path among the variables of trust -job satisfaction -performance.Hypothesis 2a is tested by analyzing the coefficient score of path between the variables of trust -organizational commitment.Hypothesis 2b is tested by analyzing coefficient score of path between the variables of organizational commitmentperformance.Hypothesis 2c is tested by analyzing coefficient score of path among the variables of trust -organizational commitment-performance.Hypothesis 3 is tested by analyzing coefficient score of path between the variables of organizational commitment -job satisfaction.Hypothesis 4 is tested by analyzing the sub-group.The group of respondents with high individualistic levels will be tested sepa-rately from the group of respondents who have high collectivism levels.

DATA ANALYSIS AND DISCUSSION Respondent
The respondents in this study are 97 employees who work in organizations or institutions in DI Yogyakarta.Respondents are selected based on the criteria determined that they have direct supervisors and have worked for more than one year.The demographic characteristics of the respondents are divided into some of the information presented in Table 1.
Based on Table 1, it can be seen that the 97 respondents are divided into several characteristics, namely gender, age, education level, and length of work.The above information provides a percentage difference in demographic characteristics of respondents.Such information helps researchers in the after-data conclusion.

Descriptive Statistics Analysis
Descriptive statistics analysis is performed to de- scribe data between variables.Descriptive statistics help the reader to know in detail about the spread of the value of the numbers given by respondents in the instrument.The results of descriptive statistics analysis are presented in Table 2. Based on Table 2, it can be known the information about the number of respondents, minimum and maximum values for the theoretical score and actual score, mean value, and standard deviation value for each variable.Each variable has a different scale.The results of the above analysis show the distribution of data from theoretical scores and actual scores filled by respondents.One of the information provided is with a sample of 97 people, the variable of trust has minimum value of 2.25, maximum value of 5.00, mean value of 3.8081, and standard deviation value of 0.53730 (the smallest standard deviation value of all variables).Thus, the information helps researchers see the data distribution of the respondents' perceptions.

Validity Test
Validity test is conducted before performing hypo-thesis test.Validity test is conducted to determine the ability of the instrument in measuring the object to be measured.The validity of the instrument is determined by testing the validity of convergence and the validity of discriminant.The validity of convergent is determined using some estimation in which outer loading should be greater than 0.7 (> 0.7), communality should be greater than 0.5 (> 0.5), and Average Variance Extracted (AVE) should be greater than 0.5 (> 0.5).The validity of discriminant is determined using the square root of AVE whose value should be higher than the correlation between latent variables in the same column.The instrument can be said to be valid if the value of cross loading is greater than 0.7 (> 0.7).Test results of convergence validity and discriminant validity can be seen in Table 3 and 4.
Table 3 shows the information on convergent validity.The data show that latent variable has a loading value above 0.7 (> 0.7) and AVE value above 0.5 (> 0.5).This means that the assumption of convergent validity is met.The table also shows the result that the convergent validity for reflective con- Table 4 shows the information on discriminant validity.The data show that the research instrument meets the discriminant validity assumption.
The assumption is met when the loading value to another construct (cross loading) is worth less than that to the construct.Discriminant validity for the trust construct has been fulfilled because the AVE root value is 0.752 which is greater than 0.289; 0.518; and 0.419, and so have organizational commitment, job satisfaction, and performance.

Reliability Test
Reliability test was done to see the accuracy and reliability of the measuring tool to determine the consistency of the results of the measurement (Hartono, 2008).The instrument of reliability is determined using Cronbach's alpha and composite reliability values.The Cronbach's alpha value should be greater than 0.7 (> 0.7) and the composite reliability value should also be greater than 0.7 (> 0.7) so that the indicator can be said to be reliable.The results of the research of reliability instrument test can be seen in Table 5.
Table 5 provides information on reliability test results.The data show that the research instrument meets the assumption of reliability because Cronbach's alpha and composite reliability values are greater than 0.7 (> 0.7).This indicates that the research instrument has accuracy and reliability to measure each variable.

Structural Model
Structural model test is performed to test the hypothesis.The structural model in SEM-PLS is analyzed using the value of R2 for the dependent construct and looks at the value of path coefficient or p-value to see the significance in hypothesis testing.The value of R2 gives information about the level of

PFM (R)1i
Note: OC = Organizational Commitment JO = Job Satisfaction PFM = Performance variation of the change of the independent variable to the dependent variable.The higher the R2 value, the better the prediction of the proposed research model.

Direct Effect Test
The first thing to do to test the relationship between variables is to test the direct relationship between independent variable and dependent variable.Table 6 shows the results of the direct relationship between variable of trust and variable of performance.
Table 6 shows the information that there is a positive relationship between the independent variable of trust and the dependent variable of performance.The test result shows that the path coefficient value is 0.35 and significant with p-value <0.01 with R2 value of 0.12.This means that there is a positive relationship between trust and performance.The higher the trust, the higher the performance of individuals within the organization.

Mediation Effect Test
After performing the direct test, the next is to test the mediation relationship.Hypothesis testing procedure is done by estimating the indirect effect simultaneously among independent variables, mediation variables, and dependent variable.Figure 2 shows the test results simultaneously.
Figure 2 shows the results of simultaneous testing between independent variables, mediation variables, and dependent variable.Based on the picture above it can be drawn hypothesis conclusion.The relationship between variables can be seen in Table 7.
Table 7 shows the results of hypothesis test of the variables of trust, organizational commitment, job satisfaction, and performance.The first result indicates that trust and organizational commitment have a significant positive relationship, with a coefficient value of 0.53 (<0.01) and R2 value of 0.28.The result indicates that H1a, which states that trust as an informal control mechanism has positive relationship with organizational commitment, is supported significantly.The second result shows that organizational commitment and performance have a significant positive relationship, with a coefficient value of 0.33 (<0.01) and R2 value of 0.49.This result indicates that H1b, which states that organizational commitment has positive relationship with performance, is supported significantly.The third result shows that trust and job satisfaction have no significant positive relationship, with a coefficient value of 0.12 (0.10) and R2 value of 0.43.The result cannot support Hypothesis 2a, but shows that the positive relationship between trust and job satisfaction is fully mediated by organizational commitment.This is evidenced by additional testing of the direct relationship between trust variable and job satisfaction variable.And the result  shows a coefficient value of 0.43 (<0.01) and R2 value of 0.19.When organizational commitment is incorporated into the model, the relationship between trust and job satisfaction becomes insignificant.This means that there is full mediation.The fourth result shows that job satisfaction and performance have a significant positive relationship, with a coefficient value of 0.43 (<0.01) and R2 value of 0.49.This result indicates that H2b, which states that job satisfaction has positive relationship with performance, is supported significantly.The fifth result shows that organizational commitment and job satisfaction have a significant positive relationship, with a coefficient value of 0.58 (<0.01) and R2 value of 0.43.This result indicates that H3, which states that organizational commitment has positive relationship with job satisfaction, is supported significantly.The R2 value of each relationship shows the amount of variation change of variable X to variable Y.
Figure 3 also shows the mediation relationship between variables.There are two mediation variables: organizational commitment and job satis-faction.The results of the mediation test can be seen in the following Table 8.
Table 8 shows the mediation relationship among trust, organizational commitment, and performance.The data show that the path coefficient of trust to organizational commitment has a value of 0.53 and significant to (p <0.01), the path coefficient of organizational commitment to performance has a value of 0.33 and significant to (p <0.01).The data also show that the path coefficient value of trust to the performance variable decreases to -0.01 and not significant to (p = 0.48) when organizational commitment variable is incorporated into the model.This result supports hypothesis 1c which states that trust as an informal control mechanism has an indirect positive relationship with performance through organizational commitment, and the type of mediation is full mediation.
Hypothesis 2c, which states that trust as an informal control mechanism has an indirect positive relationship with performance through job satisfaction, is not supported significantly.This is because there is no significant positive relationship between  trust and job satisfaction (β = 0.12; p = 0.10).That means that job satisfaction does not mediate the relationship between trust and performance.However, the significant indirect relationship between trust and performance can occur through organizational commitment and job satisfaction.Given that the relationship between trust and job satisfaction is mediated by organizational commitment.See Figure 3.

Moderation Effect Test
The moderation test is performed using sub-group analysis.The sub-group analysis is performed by comparing the results of SEM-PLS analysis of two groups of individuals who have cognitive orientation of individualism and collectivism.The first step is to divide the respondents into two groups of cognitive orientation.The division of the group is based on the greatest score on each indicator that measures the cognitive orientation of individualism and collectivism.A higher score on one of these orientations becomes the basis for the grouping of respondents.Based on the analysis, the respondents who have cognitive orientation of individualism are 43 people and the respondents who have cognitive orientation collectivism are 54 people.Furthermore, the data of each group is tested using SEM-PLS.The results of the moderation test can be seen in the comparison of the two figures, Figure 3 and Figure 4.
Based on Figures 3 and 4, it can be seen that there are differences in coefficient and R-square values of each relationship.Individuals with cognitive orientation of collectivism exhibit higher coefficient and R-square values than individuals with cognitive orientation of individualism.The comparison of the two groups can be seen in Table 9.
Table 9 shows the value of path coefficients for each relationship between individualism group and collectivism group.Based on these data, it can be seen that the coefficient value of the whole collectivism sample group is larger than that of the whole individualism sample group, except for the path of organizational commitment -performance.The entire R-Square value for the collectivism sample group is also greater than that of the individualism sample group.Different level of significance are shown only on the Job Satisfaction -Performance path, in which collectivism sample group shows β = 0.56; p <0.01, whereas the individualism sample group shows β = 0.32 p = 0.16.This means that individuals in the collectivism sample group show a significantly stronger positive relationship on the Performance-Performance path, but the results have not been able to give full support to hypothesis 4.

CONCLUSION, IMPLICATION, SUGGES-TION, AND LIMITATIONS
This study examines the relationship among trust, organizational commitment, job satisfaction, and individual performance.In order to facilitate contradictory findings of previous research, this research attempts to define trust into broader term and concept.The trust measured in this study is a latent construct formed from three dimensions ("trust between superiors and subordinates", "trust among co-workers" and "institutional trust").The result shows that trust has a positive relationship with performance.This means that the better the level of individual confidence in the work environment, the better the performance they provide.Rousseau et al. (1998) explains that individuals with a high level of trust to coworkers or organizations will have higher confidence and will form a more positive picture of the object that they believe.The more positive the individual's belief in his organization, the more comfortable the individual performs his task, thus affecting his better performance.
In more detail, this study provides evidence that trust is positively related to organizational commitment.This result is in line with the results of the research conducted by Sholihin & Pike (2009), Sholihin &Pike (2010), andRousseau et al. (1998) that trust between subordinates and superiors within the police organization can increase organizational commitment.Rousseau et al (1998) defines trust as a psychological state in which the individual accepts the risk of loss that may be obtained because he has positive expectations on the intentions or behavior of others.Positive expectations encourage the emergence of individual positive behaviors so that individual commitment will increase.
In addition, there are interesting finding that trust is positively related to job satisfaction through organizational commitment.This finding indicates that trust is not the only factor that leads to job satisfaction.However, the relationship between trust and job satisfaction is fully mediated by organizational commitment.Lau et al: (2008) states that individuals who are highly committed to the organization will regard the organization's goals as an important factor.Therefore, they will complete the task/work without coercion.In such a condition, the individual"s job satisfaction will increase because of the emergence of motivation to complete tasks and obligations quickly and accurately.This result supports previous research conducted by Sterr (1977), Bateman &Strasser (1984), andDeCotiss &Summers (1987) that commitment to the organization causes individuals to have high job satisfaction experience.
This study also shows a significant relationship between organizational commitment and performance as well as job satisfaction and performance.Organizational commitment variable has a positive relationship with performance.These results support previous studies conducted by Nouri & Parker (1998), Chong & Eggleton (2007), and Sholihin & Pike (2009).There are two important characteristics of organizational commitment: 1) strong trust, acceptance of purpose, and organizational value; 2) a desire to give hard effort for the benefit of the organization (Porter et al., 1974).Organizational commitment creates an individual desire to strive to achieve organizational goals, thereby impacting on performance improvement.In addition, the results of this study also indicate that job satisfaction has a significant positive relationship with performance.Job satisfaction can stimulate individual work morale.Individuals who are satisfied with their work environment will have positive emotions.It is this positive emotion that encourages individuals to feel enthusiasm in performing their duties which ultimately affecting the productivity of their performance.This result indicates that the positive relationship between trust and performance must go through the path of organizational commitment and job satisfaction first.
The result of mediation testing successfully proves that the positive relationship between trust and performance is fully mediated by organizational commitment and job satisfaction.Porter et al (1974) explains that strong beliefs will encourage individuals to have a high commitment to the organization.This happens because of the height of individual confidence in organizational behavior.Porter also explained that the acceptance of goals, organizational values, and the desire to provide hard effort for the benefit of the organization will have an impact on better performance.Therefore, trust creates organizational commitment and organizational commitment creates an individual's desire to strive to achieve organizational goals, thereby leading to high individual organizational performance.In addition to through organizational commitment, the positive relationship between trust and performance must also be through job satisfaction in advance.When the trust is able to increase the commitment of the organization, there will be an exchange of information and the intensity of good communication in each job, so as to create job satisfaction.Increased job satisfaction will ultimately stimulate the spirit of work that leads to increased productivity performance.Therefore, the resulting performance will also be influenced by job satisfaction formed on each individual The result of moderation testing show that a stronger positive relationship model in individuals with cognitive orientation of collectivism only occurs in the path of job satisfaction -performance.This means that the difference in cognitive orientation only causes differences in individual behavior changes caused by job satisfaction.Individuals with cognitive orientation of collectivism will react more positively to work when gaining satisfaction in work.This finding indicates that cognitive orientation is not the only variable that leads to differences in trust and performance relationship models in previous studies.However, the results of this study are able to show that cognitive orientation is an influential factor and paves the way for further research to explore other variables that explain the relationship model.This research contributes practically, theoretically, and methodologically.Practically, the results of this study provide important information that the most important thing to do to improve individual performance is to create individual confidence in the organization.This can be a reference for leaders in the organization to behave appropriately so that all members of the organization have a high trust.Theoretically, the results of this study contribute to the management accounting literature in the form of a conceptual relationship model of trust as a mechanism of informal control systems to performance that is influenced by accounting functions and information.There is new evidence that differences in cognitive orientation affect the pattern of relationships between job satisfaction and individual performance, and the relationship between trust and performance must go through the path of organizational commitment and job satisfaction first.Methodologically, the results of this research show that testing the whole dimension of trust into a construct is worth doing when many previous studies only tested the trust dimensions between superiors and subordinates.
This study is one of the studies, examining trust in a broader construct.Yet, this study still has some limitations.The main limitation is that the researchers could not prevent the possibility that some employees refer to the same boss for the response in the survey.Although biases are possible, the researchers believe that the results of this study cannot be explained.
Further research is expected to reexamine this model to objects and samples with different characteristics.After this research is able to show cognitive orientation as one variable that influences one of the paths, further research is expected to be able to explore other variables that potentially influence the relationship model of trust and performance.
Figure 1 Research Model Figure 2 Mediation Effect of Research Model Figure 3 Individualism Group Test Result Figure 4 Collectivism Group Test Result

Table 8 Indirect Effect Test Path Path Coefficient Total effect Type of Mediation
*** p<0.01