Moving Engagement Research To A Higher Level: The Impact Of Unit-Level Engagement On Business Metric Outcomes

Although engagement has been slow to take hold within the academic literature, it has quickly become a hot topic within the applied and business environments. Because of the rapid growth within these areas, there has been a great deal of conceptual confusion and mystery surrounding the engagement construct. Recent research within the literature has sought to define engagement, differentiate it from existing job attitude constructs, and link it with both personal and organizational outcomes. To date, a majority of the research demonstrating the impact of engagement has been conducted at the individual-level. While individual-level outcomes are of use to organizations, the success of a company is usually measured at higher levels of analysis. The purpose of the current study was to explore the factor structure of a new engagement index, investigate its discriminant validity with two common job attitude measures (job satisfaction and organizational commitment), look at various composition models for aggregating individual-level engagement to the unit level, and determine the relationship between work unit engagement and business metric outcomes (voluntary turnover, operating costs, and earnings). Results of the current study offer support for the three-factor model of engagement (vigor, dedication, and absorption), as well as the discriminant validity of engagement from other job attitude measures in terms of factor structure. Findings also indicated that alternative compositions models of engagement, such as the lowest and highest scores within the unit, are useful in predicting organizational outcomes. For example, the lowest unit engagement score was negatively related to voluntary turnover and the highest unit engagement score was positively related to earnings. Lastly, while the results offered some evidence to suggest that engagement may uniquely contribute to the prediction of earnings, all three job attitude variables exhibited a similar pattern of correlations with the outcome variables. Implications and future research directions are discussed