close× Call Us
Title Dividing the American Voter: What Conjoint Analysis Tells Us About Behavioral Voting and American’s Preference Utilities on Contentious Policy Issues
Post date 03/13/2019
C1 Background and Explanation of Rationale This study explores the contours and impetus of polarization in contemporary American politics, drawing upon both primary research and a broad social psychology literature review. It is an attempt to unpack the subconscious and conscious motivations directing voter behavior, and how this ‘psychology of voting’ translates individual behavior to group behavior in the form of policy choices. Primary research is built upon a novel conjoint analysis, through a nationally representative sample of 1,200 registered US voters. Literature will be reviewed and surveyed from both political science and social psychology in order to ascertain the cognitive processes and motivations that could influence behavior in voting.
C2 What are the hypotheses to be tested?

(1) Does priming shape voter behavior on the issue of climate change in the United States? What can social psychology tell us about why voters have become increasingly polarized – and on this issue specifically?

(2) What are the confounding effects driving behavioral voting and how can conjoint analysis help us better understand how and why voters make choices on polarizing policy issues like climate change?

C3 How will these hypotheses be tested? *

In conjoint analysis, respondents score a set of alternatives, where each has randomly varied attributes. This approach estimates the effects of multiple treatment components in terms of a single behavioral outcome and, therefore, allows me to evaluate the relative explanatory power of how voters make issue-based decisions. By testing multiple hypotheses simultaneously, I will be able to identify the correct explanation of why divergence exists on contentious policy issues. The dependent variable will be the preferences of respondents, while the independent variables will be policy issue attributes, plus additional variables for interaction terms of the individuals (e.g., education level or geographic region).

The respondents will be asked to complete several questions, with surveys lasting no more than 10 minutes. It uses a model analogous to Hainmueller et. al 2013. The first question in the survey will ask respondents to identify their preferred policy bundles, which will be randomly allocated as treatments. Climate change will always be given as a policy issue, and will be compared to the following issues: healthcare, immigration and taxes. Because conjoint analysis is a summation of partial utility preferences, the total utility is found by adding all partial utilities together, yielding the importance/priority respondents give to climate / environmental policy relative to other contentious policy issues. The study will go one step further however, in assessing social psychology by building on Eidelman, S. (2009). A second treatment, “Priming,” will be randomly allocated to respondents to see if there is a shift in relative importance of environmental policy after respondents treated with both negative and positive primes. There are two outcomes I expect through priming: 1) relative to the baseline of the airplane plots, positive primes should raise the level of importance given by respondents to environmental policy; 2) negative primes will be more powerful than positive. In this way, the analytical framework is linking my literature review in social psychology and political science to revealed preference voting behavior. It is comparing the literature from these fields to the data observed through primary research.

C4 Country United States
C5 Scale (# of Units) N = 1200
C6 Was a power analysis conducted prior to data collection? Yes
C7 Has this research received Insitutional Review Board (IRB) or ethics committee approval? Was accepted by London School of Economics Ethics Committee review
C8 IRB Number not provided by authors
C9 Date of IRB Approval not provided by authors
C10 Will the intervention be implemented by the researcher or a third party? YouGov US
C11 Did any of the research team receive remuneration from the implementing agency for taking part in this research? Yes
C12 If relevant, is there an advance agreement with the implementation group that all results can be published? Yes
C13 JEL Classification(s) not provided by authors