Transparency and Replicability of Multilevel Modeling Applications: A Guideline for Improved Reporting Practices

Multilevel modeling (MLM) is a statistical technique for analyzing clustered data. Given the complexity of multilevel models, it is crucial for researchers to provide complete and transparent descriptions of the data, statistical analyses, and results. A recent systematic review of the reporting practices in MLM applications in education and psychology showed that the reporting practices still lack clarity and completeness in some areas, including reliability and validity of multilevel measures, model specifications, description of missing data mechanisms, power analyses, assumption checking, model comparisons, and effect sizes (Luo et al, 2021). In this chapter, we aim to provide a guideline for improved reporting practices in the identified areas to enhance the transparency and replicability of MLM applications. We will offer suggestions for what and how to report MLM results in those areas, use examples from real life research to illustrate the principles and guidelines, and provide readers with a checklist to describe the main points that should be thoroughly checked and clearly conveyed in reports when applying MLM.