how today’s shocks predict tomorrow’s leaving

This research explores “shocking events” as part of the unfolding model of turnover, extending our understanding of the influence of various types of shocks on future voluntary employee separations. 1536 new hires at a large financial institution reported shocks monthly during their first 8 months at work as well as their job satisfaction and perceptions of marketability. we used event history to estimate the basic distributional properties of the shocks and cox proportional hazards models to examine the effects of shocks on job satisfaction and turnover over the subsequent year as reported by the organization. organizational shocks generally occur earlier than personal shocks. further, unexpected shocks have a stronger impact than expected shocks on subsequent turnover. finally, the effects of organizational shocks on turnover are mediated by job satisfaction, whereas personal shocks have direct effects on turnover. our findings offer evidence for the utility of the shock construct in the unfolding model of turnover and speak to the importance of encouraging managers to monitor shocks on an ongoing basis in order to predict when different types of shocks will occur and their likely influence on turnover. ours is the first study to examine shocks as they occur. this is a contrast to prior studies that relied on retrospective accounts. thus, we are able to test new hypotheses (e.g., direct effects vs. mediation) that expand the unfolding model of turnover.