Affective Polarization and Democratic Attitudes

Introduction

We currently witness democratic backsliding around the world where political rights, civil liberties, and free and fair elections are under assault. At the same time, politics in many countries are increasingly characterized by hostility and distrust across partisan lines; a phenomenon called affective polarization.

Research Question

Could there be a causal link between the two? Despite several theoretical conjectures about how affective polarization erodes democracy through its negative effects on citizens’ democratic attitudes, we lack the evidence to support them. We also do not have reliable interventions to reduce affective polarization that are generalizable to different contexts.

Objectives of DEPOLARIZE

DEPOLARIZE addresses these puzzles around affective polarization with two objectives:

1. Identify reliable and generalizable experimental interventions to reduce affective polarization in multiple contexts.
2. Establish any causal relationship between affective polarization and changes in democratic attitudes through novel empirical approaches.

Methodology

DEPOLARIZE will achieve these goals by producing and analysing high-quality observational and experimental data from multiple waves of surveys in Hungary, Poland, Turkey, and the US. These countries are selected for analyses because they have recently experienced democratic backsliding to different degrees, and they are also examples of high levels of affective polarization in society.

Importance of the Study

The stakes for identifying factors contributing to democratic backsliding could not be any higher. DEPOLARIZE will contribute to this effort by establishing whether affective polarization is a factor driving democratic backsliding using state-of-the-art causal inference methods.

Innovative Approach

DEPOLARIZE is also one of the first projects in political science to employ comprehensive measures for cumulative learning: coordinated and comparable randomized controlled trials in multiple contexts with pre-registration of experimental interventions.