The Curious Case of Cross-Level Effects: Refining Our Understanding to Match Our Methods

New methodological advances and guidance in estimating ‘cross-level effects’ in multilevel models have led to a proliferation of published research examining purportedly cross-level direct relationships, mediations, and interactions. However, our review of published ‘cross- level’ papers in top journals reveals that in many cases, scholars argue for or interpret relationships that are meaningfully different from the ones actually being tested. In this paper, we attempt to build an accessible explanation for what ‘cross-level effects’ actually are and how they operate in practice, detailing with sample datasets, results, and examples from the literature how they are frequently misunderstood. We conclude that to date the management literature has not yet developed and refined theory accordingly to match recent helpful methodological innovations. We attempt to address this issue by providing specific guidance on how to argue and interpret the various types of ‘cross-level’ effects, and spotlighting exemplar papers in the literature that accomplish this accurately.