Investigating follower felt trust from a social cognitive perspective

ABSTRACT

Previous organizational research on trust has focused more on subordinates’ trust in their leaders than on their experience of felt trust from the leader, even though the latter is also an important component of trust relationships. Our paper addresses a recent call for more theoretical explanations of the mechanism through which followers’ felt trust influences their in-role and extra-role performance. Based on social cognitive theory, we proposed that occupational self-efficacy (OSE) mediates the felt trust-performance relationship in workplace settings, and tested these relationships in two empirical studies. Study 1 was a cross-sectional pre-study (N = 189) investigating only the mediating effects of OSE. For the main study, i.e., Study 2 (N = 500), we collected data at three different measurement occasions to minimize response bias. Study 2 investigated the mediation of the felt trust-performance relationship not only by OSE, but also by an additional mediator variable (organization-based self-esteem) that had been identified in previous studies, in order to determine whether the OSE effects remained significant. In both studies, structural equation modelling results supported the proposed mediating effects of OSE on the three performance outcomes for the reliance component of felt trust, but not for the disclosure component of felt trust.

KEYWORDS:

• Felt trust
• social cognitive theory
• self-efficacy
• performance

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. There was some ambiguity around the inference of statistical significance for this particular result. Specifically, the bootstrapped confidence interval for the standardized indirect effect for OCBI had a lower limit of 0, however, the CI for the unstandardized effect (which is more replicable than the standardized effect and more typically used for statistical inference) did indicate a statistically significant result using conventional limits. Also, both components of the mediating paths were statistically significant at conventional levels, thus we considered this effect interpretable. Supporting our decision, we note that this result did indeed replicate in the main study.