Abstract: With the advent of the knowledge economy era, the highly efficient knowledge sharing is one of the key steps to improve the performance of organizations. Despite the established need for knowledge sharing, knowledge hiding is prevalent in many organizations and impairs knowledge transfer. Compared with the negative influence of knowledge hiding on organizations and employees, the antecedents of knowledge hiding have not been extensively examined. Extant studies which mostly based on social exchange theory have tested the relationship between workplace ostracism and employees′ knowledge hiding behavior. This study seeks to identify a new interpersonal antecedent of knowledge hiding, specifically workplace negative gossip, and mediating role of emotional exhaustion and the moderating role of negative reciprocity. Thus, we propose a moderated mediation model wherein workplace negative gossip predicts employees′ knowledge hiding behavior via emotional exhaustion, and negative reciprocity as the first stage moderator. Several hypotheses are proposed: (1) workplace negative gossip is positively related to knowledge hiding; (2) emotional exhaustion mediates the relationship between workplace negative gossip and knowledge hiding; (3) negative reciprocity moderates the relationship between workplace negative gossip and emotional exhaustion; and (4) negative reciprocity moderates the indirect effect of workplace negative gossip on knowledge hiding through emotional exhaustion.The participants of this study were 416 full-time employees from five high-tech enterprises in Shandong Province. In order to reduce the potential common method bias, the data was collected in three waves with two-month intervals. In the first-wave survey (T1), participants completed measures of predictor variable (perceived workplace negative gossip) and demographic variables (gender, age, and organizational tenure). In the second-wave survey (T2), two months later, participants were surveyed again and were required to rate mediating variable (emotional exhaustion) and moderating variable (negative reciprocity). In the third wave survey (T3), participants rated dependent variable (knowledge hiding behavior). Following the back- translation procedure, we created a Chinese version of all measures. With the assistance of human resource managers, we first sent 580 questionnaires to participants at Time 1, 507 responded the survey. About two months later (T2), the questionnaires on emotional exhaustion and negative reciprocity were completed by participants who finished the survey at Time 1, 452 participants finished the survey. About two months later (T3), the questionnaires on knowledge hiding were completed by those who finished the second-wave survey, 416 participants finished the survey. Finally, we got 416 valid samples, giving a response rate of 71.7%. Regarding demographics, of the 416 participating employees, 54.6 percent were male, most of the participants are under 35 years old (73.1 %)，and most of the participants′ organizational tenure is less than 10 years (78.4%).Among the major measures, Chandra and Robinson′s (2010) 3-item scale of workplace negative gossip, Peng′s (2012) 3-item scale of knowledge hiding,Perugini et al.′s 9-item scale of negative reciprocity, Maslach and Jackson′s (1986) 9-item emotional exhaustion scale were adopted. We conducted a Harman′s single-factor test to address potential common method bias. The analysis returned four factors with eigenvalues greater than one, with the first factor explained less than 50% of the variance (23.95% of 68.04%), the findings provided no indications of common method variance. We used hierarchical regression and bootstrapping method to test the hypotheses. Hypothesis testing was conducted by examining two nested models, mediation analysis and first stage moderated mediation analysis. Analyses were conducted using SPSS 21.0 and PROCESS version 2.16 for SPSS, bootstrapping was set to 5000 resamples. The research reported in this article used a time-lagged design to explore the impact mechanism of workplace negative gossip on knowledge hiding behaviors. As hypothesized, results showed that: (1) workplace negative gossip was positively related to employees′ knowledge hiding behavior; (2) emotional exhaustion mediated the relation between workplace negative gossip and knowledge hiding; and (3) negative reciprocity not only moderated the relation between workplace negative gossip and emotional exhaustion, but also moderated the mediated path though emotional exhaustion.Drawing from conservation of resources theory, this study offers following major contributions. First, although scholars have previously studied the antecedents of knowledge hiding from different perspectives, the interpersonal antecedents of knowledge hiding have not been extensively examined. In employees′ daily work, interpersonal interaction cues not only provide social context information to employees, but also influence their way of adapting and dealing with information. Therefore, it is necessary to explore antecedents of knowledge hiding from interpersonal interaction perspective. As an unfavorable kind of interpersonal treatment, workplace negative gossip can affect the employees′ behaviors, attitudes, and performance. Considering this, we expect workplace negative gossip may be an interpersonal antecedent of knowledge hiding. As hypothesized, this study fills in the gap of the research on the relation between workplace negative gossip and knowledge hiding, and enriches knowledge hiding literature by identifying a new interpersonal antecedent of knowledge hiding. Second, in addition to exploring the main effect of workplace negative gossip on knowledge hiding, the present study predicted the mediating role of emotional exhaustion and the moderating roles of negative reciprocity on the above association. As such, the second contribution of this study is to provide a better understanding of how and when gossiped employees hide knowledge. As predicted, negative reciprocity strengthened the positive direct effect of workplace negative gossip on emotional exhaustion and the indirect effect of workplace negative gossip on knowledge hiding through emotional exhaustion. For gossiped employees with high negative reciprocity, the impact of workplace negative gossip on knowledge hiding was more positive. For gossiped employees with low negative reciprocity beliefs, the influence was less positive. Thus, this study contributes to understanding the black box between workplace negative gossip and knowledge hiding.

Key words: workplace negative gossip, knowledge hiding, emotional exhaustion, negative reciprocity