Access to quality professional journalism considerably reduces the chance of a voter believing in fake news, according to an unprecedented academic research project conducted in São Paulo in November and December.
The work was done by political scientists from the University of North Carolina - Charlotte (USA), Federal University of Minas Gerais (UFMG), and the Federal University of Pernambuco (UFPE), in partnership with Folha and Quaest, a research consultant group.
The project tested different ways of contacting people with communication outlets. The general finding was that readers with access to outlets like Folha tended to believe less in false information.
In one of the analyses, two groups were selected, with 500 voters each reflecting the São Paulo electorate.
The groups were interviewed through an online panel between the 19th and 24th of November. Half of the interviewees (experimental group) received a free subscription to Folha for three months. They also received a study that explained the information checking process.
The second group, the control group, received neither the subscription nor the study on fact-checking.
The two groups of voters, who were randomly drawn, were practically identical in terms of gender, age, social class, and religion. They were interviewed in a second panel, between the 8th and 16th of December.
The study showed both groups texts whose content was classified as false by fact-checking agencies (but this classification was not given to respondents).
In the group that did not receive the subscription, 65% of respondents considered at least one of the texts with false content to be true in the second round of interviews. In the group that received the subscription, the percentage of those who believed in at least one fake news item was lower--46% a statistically significant difference.
Some of the false information shown claimed that PT President Gleisi Hoffmann instructed party activists to refuse government aid; that the CNN news channel reported that ex-judge Sergio Moro had received a bribe to benefit currency-exchangers; and that YouTuber Felipe Neto made a case for sexual violence against children.
Another false text said that Rede Globo belongs to three Arab countries.
The study showed this last fake news item in the two rounds of interviews. In the first, in November, the percentage of people who classified the false information as true was similar between the two groups (close to 20%).
In the second interview, about 20 days later, the results between the groups widened. Among those who received the subscription to Folha, this fake news belief fell from 20% to 12% of respondents. The belief in false information regarding Globo rose from 21% to 39% in the other group.
In the researchers' assessment, this increase in the control group is because respondents saw the text for the second time, which reinforced the message. The opposite happened in the experimental group.
For this analysis, a technique called difference in difference was used, which, among other elements, controls factors other than the one analyzed as the main one, to prevent the variation from occurring for any reason other than the object of the analysis (in this case, access to the content of Folha ).
The academics who conducted the research were Frederico Batista (University of North Carolina - Charlotte), Felipe Nunes (UFMG), and Nara Pavão (UFPE). They intend to publish a paper (scientific work) with the results.
Calculations show that the experimental group was 25% less likely to believe the false information. It is a result compatible with research conducted by Professor Andrew Guess, from Princeton University, in the United States and India.
The researcher followed the campaign's impact that provided tips on how to check information in several countries. In the United States, people who received this content were 27% less likely to believe a fake text; in India, 18% (considering a public with more years of schooling).
Experts see the deliberate dissemination of false news as a great risk to the democratic process.
Researcher Ricardo Ribeiro Ferreira, from the Faculty of Letters of the University of Coimbra, found that the engagement of false texts was up to three times greater than the news from the professional press during the 2018 Brazilian presidential election.
For lawyer and professor Marco Antonio da Costa Sabino, coordinator of the Center for Media and Internet Research at Ibmec-SP, the study carried out in São Paulo showed that the higher the notion people had on the subject, the lower the risk of contamination.
Sabino, who holds a doctorate in law from USP, said that although access to professional journalism can be one of the ways to combat disinformation, the solution must go through a previous stage, which he calls basic education in media content.
The Brazilian survey of the São Paulo electorate also measured the impact of different media outlets - the news consumption profiles of the 731 interviewees who participated in the two rounds of study was examined.
They were asked from which of eight vehicles they consumed information at least four times a week.
Regular readers of Folha tended to believe 17 percentage points lower in fake news than non-regular readers (that is, those who read the newspaper regularly believe less in false information).
The consumption of information from UOL (15 points lower in the belief in fake news) and Rede Globo (10 points) also had a positive effect.
In the researchers' assessment, "only a massive awareness campaign can help different audiences better distinguish between false and true news." They say that not only should every newspaper or channel mobilize, but governments, TSE (supreme courts) and political parties "interested in keeping democracy alive."
The authors defend the idea that the public be exposed to alerts "in various professional media, all at the same time, in large quantities, with tips, examples and everything else that is necessary."
In the same vein, Sabino says that awareness must begin at school and that the debate must not lose sight of freedom of expression. "Removal and silencing are, in theory, the simplest solution [for fake news]. Care is needed because any regulation can dangerously flirt with censorship," says the expert.