Guidelines for the Development of Three-Level Models: Bridging Levels of Analysis and Integrating Contextual Influences in IS Research

Abstract

The use of multilevel analysis has steadily increased in information systems (IS) research. Many studies are doing an admirable job of integrating two-level models into their examination of IS phenomena. However, two-level models are limited in how well they enable researchers to (1) more explicitly incorporate context into theory development and testing and (2) bridge the existing gap between micro- and macrolevel research by focusing on intervening mechanisms that link hierarchically distal levels of analysis. Three-level models have emerged as a potential way to address these limitations. While the literature has clearly outlined the mechanics of how to estimate three-level models, there is very little, if any, guidance on when and how to integrate the use of such models with theory development. Consequently, IS researchers have little guidance to inform their decisions about integrating the use of three-level models with their theory development and testing. In this article, we identify the circumstances under which IS researchers should consider the use of three-level models, develop guidelines about how to map the use of three-level model estimation to the theoretical objectives, and provide an illustration of how to implement the guidelines.