Assessing the Effectiveness of Correcting COVID-19 Political Misinformation in Peru

Abstract

What is the effect of corrective information on support for candidates promoting COVID-19 misinformation campaigns? Since the COVID-19 pandemic began in early 2020, misinformation about the use of unproven drugs to prevent and/or treat COVID-19 has spread widely on social media. Due to the popularity of this misinformation, some politicians have included this discourse in their campaign platforms. In Peru, congressional candidates for the April 2021 elections are promoting the benefits of chlorine dioxide and the dangers of COVID-19 vaccines as part of their electoral campaigns. Despite the danger of these treatments, the government and health authorities have provided confusing and limited corrective information about them. In this study, I propose a survey experiment that tests if voters reduce their support for these candidates when provided with corrective information that refutes their misinformation campaigns.

Background

In contexts where corrective information is scarce and party loyalties are weak, voters are more likely to punish politicians who use misinformation as part of their campaign strategy when faced with clear and targeted corrective information. I test my theory using an online survey experiment in the context of the Peruvian Presidential elections of 2021, where candidates have employed misinformation about COVID-19 treatment options and vaccines as part of their campaign strategy.

Research Design

I propose a survey experiment to test the effectiveness of clear and targeted corrective information during the Peruvian general election in 2021. Using a real campaign statement by a presidential candidate and a targeted and clear correction statement, I will conduct a Facebook Messenger survey experiment to measure three outcomes: belief in misinformation, disposition to receive COVID-19 vaccine, and support for politician using misinformation. As a result, the study aims to identify the negative effect of the pandemics on elections, in terms of the types of candidates who are running for office and voters’ evaluation of candidates, and propose potential strategies for mitigating these effects.

I will implement the survey from April to June, during the campaign for the Presidential run-off election. For this experiment, I focus on misinformation messages around COVID-19 and draw stimuli from real information. In the last three months of the electoral campaigns, populist candidates have used misinformation around COVID-19 treatment options and vaccines to attract new voters. My survey experiment is designed to test the effectiveness of corrections. Through the survey experiment, I expect to find that citizens, when provided with clear corrective information, (1) increase their doubts about the misinformation presented, (2) increase their disposition to receive the COVID-19 vaccine, and finally, (3) reduce their support for the politician promoting the misinformation campaign.

Hypotheses

H1: Participants who receive corrective information are more likely to report higher levels of trust on the COVID-19 vaccine compared to those who don’t receive any correction.

H2: Participants who receive corrective information are more likely to increase their willingness to receive the COVID-19 vaccine compared to those who don’t receive any correction.

H3: Participants who receive corrective information are more likely to report lower levels of trust for the politician promoting the misinformation campaign compared to those who don’t receive any correction.

H4: Participants who receive corrective information are more likely to reduce their support for the politician promoting the misinformation campaign compared to those who don’t receive any correction.

H5: Participants who receive corrective information are more likely to report lower levels of satisfaction with the political class compared to those who don’t receive any correction.

H6: Participants who receive corrective information are more likely to seek additional information compared to those who don’t receive any correction.

H7: On average, participants who are part of the candidates’ party are more likely to support the candidate compared to participants who do not report any party affiliation.

H8: On average, participants who are part of the candidates’ party are more likely to support the candidate compared to participants who report a different party affiliation.

H9: Participants who only receive misinformation and no correction are more likely to express populist attitudes than those who received misinformation and correction.

H10: Participants who only receive misinformation and no correction are more likely to express dissatisfaction with the government compared to those who received misinformation and correction.

H11: Highly religious participants are less likely to be affected by the corrective information.

Findings

• My results show that corrections to scientific misinformation work. Respondents’ misperceptions about the Sinopharm vaccine were reduced and they were more willing to get vaccinated after receiving corrective information. This is evidence that in low information contexts, providing clear and targeted corrective statement about complex information such as the COVID-19 vaccine can steer citizens away from belief in misinformation.
• While some candidates used misinformation explicitly as part of their campaign platform, others used proxies to reap the electoral benefits of spreading misinformation while avoiding tarnishing their image. Future research should explore how direct vs indirect uses of misinformation affect voters’ political attitudes.
• My findings cast doubt on the theory that party identities in a polarized setting explain the limited effect of corrections on political attitudes. I find that even in context of low partisanship, candidate’s use of misinformation goes unpunished. Corrective information does not affect respondents’ political attitudes.
• Instead of partisanship, voters who indicated highest and lowest levels of likeliness for the candidate are the ones registering changes in their political attitudes. Although the correction updated their factual misperceptions about the vaccine, strong supporters of the candidate indicated an increased willingness to vote for their candidate. In contrast, strong detractors of the candidate indicated both an update of the factual misperceptions and decreased willingness to vote for the candidate.

Watch Verónica Hurtado present the findings of her project: