Abstract: Although numerous studies have investigated the effect of trust on job performance in organizations, there has been little research directed at the relationship between individual trust for different referents and task performance. Trust is a special dyadic relationship between trustees and trustors. Many previous researches revealed that the trustworthiness of different referents (direct leader, co-workers, or top management) would influence employee’s psychological conditions which, in turn, leads to different job performance. In this study, we are motivated by the idea to study whether the trustworthiness of different referents have its unique and mutually complementary impact on individual performance. Hence, we simultaneously adopted three trust referents (direct leader, co-workers, and top management) as independent variables, and empirically examined their relationships with task performance. The interactive effect of different referents is also explored, because we believe it would help present a clearer picture to understanding how trust in workplace facilitates superior individual performance.
Data was obtained from full-time employees of a large manufacturing corporation in Northern China. We combined the data from three representative departments, an engineering center (64), a manufacturing plant (157) and a financial department (58). In order to avoid the common method bias, we collected the employee’s psychological data and their corresponding performance information separately. For the purpose of diminishing the effect of social desirability, we employed Aryee’s sampling procedure (2002) that is regarded as an effective solution to eliminate negative outcome of social desirability in trust researches. The statistical package of LISRAL 8.45 was used to examine the hypotheses.
In the first stage of data analysis, we examined whether the effect of trust in different referents on individual task performance is mutually exclusive. In order to test our hypotheses, we established six alternative nested models. By comparing the alternative nested models with the hypothesized model, the results indicated that trustworthiness of each referents played a unique role on job performance. In the second stage of data analysis, we evaluated our hypotheses by assessing the conditions for mediation. Specifically, we examined the hypothesis by comparing hypothesized model (partially mediated) with alternative nested model (fully mediated), and the result supported our hypothesis that trust in top management is partially mediated by trust in direct leader.
The results have several theoretical implications. First, it showed that the positive effect of any trust foci on individual behaviors could not be superseded by other trust foci, which hints researchers to pay more attention to the unique value of the trustworthiness of different referents. In addition, this study also starts a new trend of investigating the interactive effect among different trust dimensions. From a practical perspective, the result of this study provided specific evidence for manager to believe that building a trustworthy organizational culture is a good investment which will enhance the competitiveness of the organization

Key words: organizational trust, trust in top management, trust in direct leader, trust in co-workers, task performance