Research Design: Testing the Hypotheses

In the previous chapter, I have derived six hypotheses from my conceptual framework. Throughout the remainder of this book, I am now bringing quantitative and experimental evidence to bear on these hypotheses. This chapter discusses the rationale behind my multi-method approach, the data sources and experimental design, and the measures and variables I use. The general idea is to combine different methods in such a way that the results generated by them are both generalizable and lend themselves to a causal interpretation. To ensure generalizability, I estimate a series of multilevel models using data from 14 Western European countries, and to establish causality, I focus on a subset of those countries and conduct two original survey experiments as well as panel data analysis. In addition, consistent with a design-based approach to causal inference, I discuss a number of robustness checks and auxiliary analyses which will be used to validate the results of the multilevel models, including falsification tests and pattern specificity.