Right-wing populism in Ecuador

The study aims to analyze the effect of a populist message on anti-immigration attitudes by applying a survey experiment in Ecuador. Besides that, the research measures the populist vote, and considers a prosocial preference like altruism as a variable linked to the populist message in terms of altruistic punishment. As mechanisms, the survey uses emotions —anger and fear — to evaluate the effect of the manipulation. Moreover, another set of variables will be controlled to explain the model proposed.

This research will be performed in Ecuador, a country that has adequate dynamics to test the hypothesis proposed, such as inequality, delinquency, institutional instability, right-wing tradition, immigration, and corruption. Considering that there are no studies related to this in Latin America to the best of my knowledge, the results will contribute to understanding the politics in the region and how right-wing messages advocates for exclusionist attitudes. The study incorporates different aspects like behavior, preferences, and emotions interacting with a populist message.

2023-06-19

2023-08-15

Primary outcome 1: Anti-immigration attitudes
Primary outcome 2: Right-wing populist vote
Primary outcome 3: Altruism

- Ecuador is a worse or better country when foreigners come to live here? [scale from 1 to 5 from much worse to much better - recoded being 5 much worse] [Right-wing populist message increases anti-immigration attitudes]
- The safety of Ecuadorians is generally undermined or affected by immigrants? ? [scale from 1 to 5 from very affected to nothing affected - recoded being 5 very affected] [Right-wing populist message increases anti-immigration attitudes]
- Is it generally bad or good for the Ecuadorian economy that foreigners come to live here? ? [scale from 1 to 5 from very bad to very good - recoded being 5 very bad] [Right-wing populist message increases anti-immigration attitudes]
- In the next presidencial election, who would you prefer to vote? A candidate that implement restrictions for immigrants with criminal records (coded 1). A candidate that focus on redistributive policies (coded 2). None (coded 3). [The right-wing populist message will increase the support for right-wing populist candidate]
- Would you donate to an NGO that helps immigrants, ONG that helps national, or you keep the money? 3 euros can be split for those three options. [The right-wing populist message will increase altruism toward nationals through and NGO that donates to nationals]

Secondary outcome 1: Anger
Secondary outcome 2: Fear

- How much anger does the video make you feel? (scale of 1 to 5 from nothing annoyed to extremely annoyed) [A manipulation externalising blame for delinquency to immigrants would generate higher levels of anger compared to the control group]
- How much fear does the video make you feel? (scale of 1 to 5 from nothing scared to extremely scared) [A manipulation externalising blame for delinquency to immigrants would generate higher levels of fear compared to the control group]

This research will be conducted through a face-to-face survey experiment in the streets of Quito-Ecuador, specifically outside malls to preserve surveyors’ security and integrity. This methodology will allow me to test how a populist message may influence the political attitudes of the participants. Thus, the experiment will have a control and a treatment arm to compare the effect of the populist message versus the non-populist one. Under this scheme, the eligible participants are Ecuadorians between 20- and 65-year-old that have the right to vote and have voted in the last presidential election. The survey will be developed using Qualtrics, so surveyors will use a devise to conduct the surveys and answers will automatically upload to the system once the survey is finished. Hence, the surveyors will not have access to the collected information. Also, Qualtrics allows me to include filters of security for the data collected like not recording the IP or dropping incomplete observations. Surveyors will be students from universities in the capital of Ecuador who study related areas to administration (e.g., students from Central University of Ecuador).

The randomized sequence will be done before hand (computer system), so each person that is interviewed will be assigned to the treatment or control group regarding the mentioned sequence. The control group will have a neutral message of delinquency, while the treatment group will receive a populist message about delinquency, the causes, and information blaming out-groups (immigrants in this case). This latter might induce people to support directly or indirectly right-wing populism and its policies.

Nonresponse and Attrition from the Sample

The study uses a survey experiment in the streets of Ecuador and some people would not agree to participate on it. One reason could be because they feel unsafe talking to people on the streets due to delinquency. Here, it will be mandatory to use a distinction with special clothes that make people feel safe. Moreover, the survey will be conducted outside malls where people could feel safer. Another reason could be because they are reluctant to provide information. For that, they will be informed that the survey is anonymous, and their answers will not be tracked to them. Besides that, I would remark that the survey is for academic purpose and to make it interesting I would inform them that there is a game where they would like to participate. Finally, there will be people that do not want to participate because they are in rush and, in this case, I could not convince them to be part of the survey.

It is also possible that people do not want to continue with the survey, either because of rush or they do not feel comfortable providing further personal or political information. Hence, these observations should be dropped from the sample and be replaced for the next person surveyed under the same randomization criteria. With the purpose to avoid attrition and people do not get exhausted of the questionnaire, the survey will not last more than 10 minutes.

Pilot

A pilot of 200 surveys was conducted with the purpose to test the hypotheses (5 to 9 of June 2023). There are difference in response between the treatment and the control group, and the results suggest that the manipulation has an effect on the primary and secondary outcomes. Few changes were made to the survey after the pilot, among them, in the scale of response it was included an option for the participants to say "I don't know" for not to force an answer. In the final paper the results of the pilot will be presented as a short analysis and compared it with the results of real experiment.

The randomisation is made before hand in the computer, using excel randomization formula.

The randomization is at individual level. People will be contacted in the streets of the capital of Ecuador.

No

1100 individual surveys.

1100 individual surveys.

550 control and 550 treatment.

Sample size was determined based on the power calculation from previous elections of Ecuador (control 54.77% and treatment 63.22%). with command pwr.t.test (sig.level =0.05, power =0.8 ). The resulting sample size is 1100 individual surveys.