After more than a year of raising suspicions about Brazil’s electronic voting system, which has worked flawlessly on the 2nd, as it has since its introduction in the 1990s, Jair Bolsonaro is now turning against the polls. Arthur Lira, Speaker of the Chamber of Deputies and an ally of the President, has decided to speed up the vote on a project that would criminalize election polls that fail to produce results beyond the margin of error – according to the text, those responsible for the polls and statisticians risk prison terms of up to 10 years.
“In my opinion, the research agency, which is paid for doing research, should compensate, be fined. You can’t miss 20, 15, 10 points, that’s not a mistake, that’s targeting. It’s disservice that incites voters in good faith,” said Lira.
In question are the polls of Datafolha and IPeC, the two main institutions in Brazil, which predicted Lula da Silva with 50% and 51% respectively and the current president with 36% and 37% respectively on the eve of the election. With the PT candidate, they missed the two-point limit, or close in the case of the IPeC, the margin of error – Lula had 48% – but they got far below the vote for the PL candidate, who had 43% .
Lawyers and the press warn of the “dangers” of the bill and academic Marcelo Crespo, coordinator of the Law course at Escola Superior de Propaganda e Marketing de São Paulo and a specialist in digital law, tells DN that the idea “isn’t right.” “.
“The attempt by the Chamber of Deputies to criminalize research institutes that fail beyond the margin of error, including mathematicians and statisticians, is futile, criminal law must not be used in vain, it cannot be used by mistake, otherwise rulers who make mistakes in predictions, such as Bolsonaro regarding those killed by covid should also be criminalized”.
“I don’t believe this bill will be passed by either house, including the Senate, or, if passed, it will be passed by the Federal Supreme Court, let alone in time for the second round. For me it is nothing more than pressure”.
According to Crespo, “Since 2018, like Donald Trump in the US, Jair Bolsonaro has tried to discredit the entire electoral system, first the electronic voting machines, which presented no problems, and now the polls, that is, he replaces the one smoke screen through the other.”
“It is not up to me to put my hand in the fire for all the research institutes, but of course it is much more likely that there was just a mistake than a fraud, many of these institutes use the census to implement the cuts the interviewees and the census in Brazil has not been done for years and therefore it may be outdated”.
For the academic: “the polls actually showed different results than the ballot box, but we must remember that polls are portraits, not predictions”. On the contrary, it seems that Lula’s call in the last stretch for the helpful vote of Ciro Gomes and Simone Tebet voters has led voters who don’t want him as president to vote against him.
Shirlei Camargo, Master in Marketing from the Federal University of Paraná, agrees. “I spent many years working in a large company in the cosmetics industry and making sales forecasts. When there were discrepancies between what was planned and what was done, I reminded the critics on duty that our work was forecasting and not sales assurance. Polls”.
“Like a sales forecast, the researcher uses all possible techniques to get as close to reality as possible, but there are always ‘uncontrollable’ variables that can change the outcome. In this case, the ‘uncontrollable’ variable was the voter himself,” he wrote Camargo.
“Often during an investigation, in order to fit in, to be accepted or even to be nice, the person gives an answer that he thinks is the most socially correct, but that does not correspond to what he really thinks. In this context the telephone surveys have a slight advantage, as not being face to face with the interviewer, or even being a “robot” on the other end of the line, makes the person feel more comfortable and gives his honest opinion. opinion”.
“Unlike the conspiracy theorists, it makes no sense for big institutions to manipulate their research because it is their image, their brand, that is at stake. The closer the research gets to reality, the better for this one.” companies,” he concludes. .
For Alberto Carlos Almeida, professor of political science, “the poll failed, yes.”
“The mistake of the main polls, whose method is face-to-face, was that they didn’t reach Bolsonaro’s voters, period not for the right reasons, but simply by chance: they shot what they saw and hit what they didn’t,” he tells DN.
How can a political analyst interpret the polls then and now? “What we have concretely for this second round is the effective result of the ballot box in the first round, it is with this number, rigid, that does not change easily for the second round, that we political scientists now have to work”.
“And this result left Lula behind 1.6% of the win, meaning he meets all the conditions to win on the 30th, the polling stations now have to correct the numbers based on the first ballot, whoever does that has a good chance to do it right, if you don’t you will get hurt”.
In the latest polls released before the end of this report, Lula led 54% to 46% on the 24th, over Bolsonaro, in Ipec, and by 52% to 48%, in Datafolha, on the 20th.
Source: DN
