Empowered to Perform: A Multilevel Investigation of the Influence of Empowerment on Performance in Hospital Units

Abstract

Psychological empowerment has been studied extensively over the past few decades in a variety of contexts and appears to be especially salient within dynamic and complex environments such as healthcare. However, a recent meta-analysis found that psychological empowerment relationships vary significantly across studies, and there is still a rather limited understanding of how empowerment operates across levels. Accordingly, we advance and test a multi-level model of empowerment which seeks to better understand the unique and synergistic effects between unit and individual empowerment in hospital units. Analysis of data involving 544 individuals in 78 units, collected from multiple sources over three different time periods, revealed that unit empowerment evidenced a synergistic interaction with individual-level psychological empowerment as related to individuals' job performance, as well as an indirect effect on performance via individual empowerment, while controlling for previous performance levels.Notably, these effects were significant at relatively high, but not at relatively low levels of unit empowerment. Furthermore, we found that unit voice climate increased unit empowerment and thereby enhanced individual psychological empowerment. These findings suggest that, in complex and dynamic environments, empowering work units is an important means by which leaders can enhance individuals' performance. ) is a Professor and the Friar Chair in Leadership and Teams at the University of Connecticut. His interests include models of team and multi-team effectiveness, and cross-level models of organizational behavior. He has worked with several Fortune 500 companies, the armed services, federal, state, and public organizations. M. Travis Maynard

Does Servant Leadership’s People Focus Facilitate or Constrain Its Positive Impact on Performance? An Examination of Servant Leadership’s Direct, Indirect, and Total Effects on Branch Financial Performance

“…Given branch support climate’s low ICC(2) value, the tests of its relationships with servant leadership and branch OCBs were somewhat conservative yet still significant. Second, branch support climate’s ICC(2) value is consistent with extant research that also reported significant effects with constructs that have similar or smaller ICC(2) values and were accompanied by ICC(1) and r wg values that supported aggregation (e.g., Chen et al, 2009; D’Innocenzo et al, 2016; Lanaj & Hollenbeck, 2015; Mathieu et al, 2007; Porck et al, 2019). Third, we theoretically defined servant leadership and branch support and goal achievement climate as group-level constructs (reflecting shared perceptions of the leader’s behaviors and of the unit’s policies, procedures and practices); our ICC(1) and r wg values provided support for modeling the constructs at the group level.…”

Understanding “It Depends” in Organizational Research

“…They further suggested that "higher levels of unit empowerment create a condition or context within which psychologically empowered employees are liberated to take actions to improve their performance. In contrast, one's degree of psychological empowerment would matter less in contexts that do not afford them with opportunities to exercise that empowerment" (D'Innocenzo et al, 2016(D'Innocenzo et al, , p. 1294. In other words, they anticipated that the X-Y relationship would be significant at relatively high, but not at relatively low, levels of the moderator.…”

Inspired to perform: A multilevel investigation of antecedents and consequences of thriving at work

“…Specifically, multilevel structural equation modeling partitions variance and creates latent group‐level variables, which allows for testing the within‐ and between‐unit influences simultaneously. Moreover, because thriving at work and affective commitment were hypothesized at both the individual and group levels, we centered them around group means, so that the group mean centered portions of the two variables are used to model the individual‐level relationships and the means are used to model the group‐level relationships (see also D'innocenzo, Luciano, Mathieu, Maynard, & Chen, ).…”