Reviewing Radicalization Research Using a Network Approach

In an effort to discern determinants of political radicalization, scholars have discussed and investigated a considerable number of personal or contextual constructs. Yet the existing literature reviews on this topic have mainly focused on specific data sources and research approaches (e.g., survey research), whereas an integrative overview is still missing. This study provides a systematic review of 57 published studies while particularly focusing on differences in the prevalence of considered determinants across research approaches (i.e., survey approaches, experimental approaches, and digital trace data approaches). As an innovative approach to systematic review, we apply a network approach for analyzing the most prevalent constructs and related hypotheses in the literature. Network analysis is particularly useful in this context because, it allows the visualization of the structure of constructs and hypotheses proposed in the field as well as the identification of crucial concepts. The review reveals differences across empirical approaches and closes with a discussion of over- and underresearched constructs, their generalizability across research approaches, and potentials for future research. We conclude by recommending a stronger integration of constructs and perspectives as well as a more rigid consideration of causal inference.