Abstract Communication has long been recognized as a “variable field,” with researchers investigating media and communication phenomena at multiple levels of analysis. Despite early discussions in the 1990s and 2000s emphasizing the importance of multilevel analysis for theory development, little is known about the adoption and evolution of multilevel modeling (MLM) within the discipline. This study addresses this gap by analyzing 287 articles employing MLM, published in communication journals between 1990 and 2024. The findings reveal a steady increase in MLM studies over time, accompanied by diversification across various subfields. Most studies feature two levels of analysis, and multilevel regression is the most commonly used statistical approach. Moreover, there has been a shift in recent years toward the use of the open-source R package for analysis instead of proprietary software tools. However, the scholarship is predominantly produced by authors affiliated with institutions in the United States and Europe, further highlighting persistent inequities in the field.
Chan et al. (Tue,) studied this question.