Abstract
With hybrid work becoming the norm in many organizations, new challenges arise in fostering employee inclusion. While prior research has begun to examine inclusion in hybrid teams, large-scale quantitative evidence on factors that foster inclusion in these contexts remains limited. Drawing on a large employee sample from a German automotive company (N > 1,300) and a broad set of predictors, this study applies machine-learning techniques to examine the relative importance, non-linear effects, and interactions of factors shaping inclusion in hybrid work settings. We conceptualize inclusion as comprising two distinct dimensions – belongingness and authenticity – and analyse them separately. Results identify novel predictors (e.g., rules for hybrid work), highlight the heightened importance of established factors such as intrateam trust, and reveal non-linear and interacting effects. By leveraging machine learning, this study extends inclusion models to contemporary hybrid work contexts and and generates new insights and hypotheses about what matters for employee inclusion.