Immigration and Political Selection

How to cite

Jensen, Katarina. 2019. "Immigration and Political Selection." AEA RCT Registry. AEARCTR-0004466. https://doi.org/10.1257/rct.4466-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Katarina Jensen*

* Primary investigator

Registry ID
AEARCTR-0004466
DOI
https://doi.org/10.1257/rct.4466-1.0
First registered
Last updated
Status
In development
Trial dates
2019-08-08 – 2019-08-23
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/4466

Abstract

In this project I study how immigration and economic information shocks affect political preferences. After viewing a projection of either future immigration or job automation, subjects are asked to choose between fictional political candidates. The candidates' social background, stances on immigration and redistribution are cross-randomized. This allows me to study whether (1) immigration and economic shocks change subjects' choice over political candidates, and (2) whether the change can be attributed to social background or preferred policy.

Intervention

The study has three treatment arms. In each of the treatment arms, subjects are first informed that they are about the view a projection conducted by a panel of experts, and that they will be asked follow-up questions about the information.

After the first page, subjects view a 30 second animation that shows information about immigration, automation, or population growth in Denmark. In the immigration treatment arm, subjects view a projection of immigration in Denmark until 2050 from Statistics Denmark. In the automation treatment arm, subjects view a projection of the number of current jobs that will be automated by 2040 in Denmark based on estimates from the automation literature. Finally, in the control arm subjects view a projection of the Danish population until 2050 from Statistics Denmark.

In all three treatment arms, the animation page is followed by a page where subjects are asked follow-up questions. This is to check if they can recall the information shown to them.

Experimental design

First, we collect basic covariates, such as income, urbanization, and party ID. Respondents are then randomized, in equal proportion, into three treatment arms. These treatment arms are described under "intervention". Then we show the fictional candidates described under "primary outcomes", ask subjects to rank the three policies that they care most about. We then ask questions about beliefs about e.g. Denmark's future economy and number of immigrants. Then we ask the questions described under "secondary outcomes". Finally, we ask which proportion of different groups the responds believe that immigrants make up (e.g. financial aid recipients and cash benefit recipients).

Design details

Randomization method
Randomization is done by a computer, using YouGov's algorithm.
Randomization unit
Individual randomization
Clusters
2000 individuals
Observations
2000 individuals
Arms
About 640 individuals in control and each of the two treatment arms. 70 individuals answer a supplementary survey, and do not participate in the experiment.
Primary outcomes
Preferences over political candidates; specifically, whether subjects prefer low SES, anti-immigration, and/or anti-redistribution candidates.
Secondary outcomes
Policy preferences over immigration and redistribution, and trust in politicians.

Cited by

No citation data available for this registration yet.

Keywords