Leader–member Exchange Dimensions and Knowledge Hiding: The Moderating Role of Prosocial Motivation in IT Sector in Sri Lanka

Sri Lanka’s Information Technology (IT) sector has become a key driver of national economic growth, characterized by a rapidly expanding workforce and increasing integration into global markets. The sector relies heavily on teamwork, innovation, and effective knowledge sharing to maintain competitiveness. However, the collaborative nature of IT work also heightens the influence of interpersonal relationships, which may either enable or constrain knowledge flows. Drawing on Social Exchange Theory (SET), this study investigates the effects of Social Leader–Member Exchange (SLMX) and Economic Leader–Member Exchange (ELMX) on knowledge hiding behavior. It further examines prosocial motivation as a moderating factor, proposing that higher prosocial motivation strengthens the negative association between SLMX and knowledge hiding, while ELMX is expected to promote knowledge withholding through transactional and self-interested exchanges. Using a deductive quantitative design and convenience sampling, data were collected from 280 IT professionals across seven organizations located in Colombo. Correlation analysis, regression techniques, and conditional effects modeling were employed to test the hypothesized relationships. The results indicate that ELMX does not have a significant effect on knowledge hiding (β = .08, p > .05), whereas SLMX shows a significant negative relationship with knowledge hiding (β = −.33, p < .001). Moreover, prosocial motivation significantly moderates the SLMX–knowledge hiding relationship, with stronger negative effects observed at higher levels of prosocial motivation (β = −.46, p < .001). Theoretically, this study advances understanding of how distinct social and economic dimensions of Leader–Member Exchange differentially influence knowledge-related behaviors in the workplace. Nevertheless, the use of a cross-sectional design limits causal interpretations, suggesting that future research should adopt longitudinal or experimental approaches to strengthen causal inference.