ResIN: A new method to analyze socio-political attitude systems

Investigating attitude systems is of key interest to many social scientists. However, “classic” approaches like linear scaling or dimensionality reduction methods risk to hide useful information and may artificially narrow the scope of explanatory theories. As an alternative, scholars have turned to Belief Network Analysis (BNA), which formalizes mass attitudes as a set of interconnected nodes in a network. BNA, however, inherits many of the assumptions typical of latent variable methods which can preclude explorations of more complex phenomena, such as inter-group asymmetries and ideological non-linearities.This chapter discusses Response Item Networks (ResIN)—an extension of BNA which is detached from the classic assumptions inherent in latent variables models. We demonstrate how ResIN can detect attitudinal asymmetries without “forcing” specific (linear or monotonic) dependencies onto source data. In ResIN, structural patterns instead emerge from co-endorsement patterns within the data. We use simulations and investigate real-world survey data to highlight how ResIN provides parsimonious, yet realistic reproductions of diverse attitude structures in cases that are difficult to handle both for latent variable models and BNA. We further discuss unitary, bi-nary, and trinary attitude structures based on survey responses across three European countries as well as non-linearities within the US liberal-conservative spectrum.