Edição #28
Latitude Ocasional é uma newsletter escrita ao longo da semana, enviada aos sábados. Se tiver interesse, subscreva-a gratuitamente, através do botão abaixo; se já é assinante, ajude a comunidade a crescer, partilhando-a com que poderá gostar de ler.
Latitude Global - Questão fundamental: Estão as sondagens adequadas à nossa conjuntura?
So, I wonder: Are polls fundamentally flawed or is it the interpretation of them that’s out of kilter?
A questão acima surge depois do excerto que abaixo copio, do artigo US election shock: why we need to rethink polling:
Should we blame “the polls”? The media certainly thinks so. Here’s Freddy Gray in The Spectator:
“Can we just agree never to pay any attention to polls from now on?...The big pollsters all assured us that they had learned the lesson of 2016*…Some projections even gave Biden a 97% chance of winning. Everyone who isn’t delusional knew that was absurd.”
This got me thinking: what does “wrong” mean anyway?
As sondagens não são em si um problema, mas estamos a chegar a um momento em que a conjuntura em que vivemos apresenta obstáculos à fiabilidade das mesmas. Esses obstáculos não se esgotam no facto de fazerem as sondagens falhar: no limite, podem fazer a democracia claudicar.
Entre cada um dos limites, há muitas nuances que invariavelmente danam a representatividade do povo nos vários parlamentos. Assim, não é necessária uma questão de extremos para nos preocuparmos com o efeito nefasto das sondagens; e, consequentemente, apurar o que se pode fazer para que os seus falhanços sucessivos em várias partes do mundo deixem de acontecer.
O tema da fiabilidade das sondagens tem literatura muito mais acessível nos EUA do que na Europa e, consequentemente, em Portugal - mas também há; contudo, antes de abordar as sondagens mais recentes a propósito das legislativas de 30 de janeiro, passemos rapidamente pelos Estados Unidos.
Há pouco tempo, registou-se mais um falhanço das empresas de sondagens, nomeadamente nas eleições para Governador da Virginia e de Nova Jersey. Na newsletter Tangle, temos o seguinte:
Speaking of the polls, they are still off. Whatever skewing pollsters are doing to account for conservatives seems to once again be failing. In the final days it looked like the Virginia race came into focus (with RealClearPolitics and FiveThirtyEight averages showing a 1-2% advantage for Youngkin), and the energy of his support on the ground compared to McAuliffe’s was obvious. But what about New Jersey? Phil Murphy was supposed to have a 5-10% gap, but at 9 a.m. this morning the race was separated by 69 votes. He won by a fraction of a percentage point.
Creio ser justo dizer que os norte-americanos estão, mais agora, mas já há algum tempo, traumatizados com sondagens. Isto porque o debate tem vindo a intensificar-se desde 2016, aquando da eleição de Donald Trump. A questão é se “se corrigiram” metodologias tendo em conta futuras eleições. Este excerto do artigo Nate Cohn Explains What the Polls Got Wrong, da revista New Yorker é elucidativo:
But, if the state polls are just as bad as they were in 2016, despite steps that we know improved the President’s standing in the surveys, we can say with total confidence—and I know this was true in our data—that the underlying survey data has to be worse than it was in 2016. Or, if you prefer, if all the pollsters were using the 2016 methodology, the polls would have been far worse this year than they were in 2016.
O artigo é bastante interessante, abordando várias teorias para as falhas das sondagens tanto em 2016 como em 2020. Para concluir, destaco mais esta passagem:
One of the things about the polls being wrong, Isaac, is that then you’re never sure when they ever were right or wrong. Now, to be clear, that uncertainty always exists. When people get the polls right in an election, they assume that they’re right all the time, until they’re next wrong. But you never know that’s true. And I would say the same thing this year. It’s possible that the polls reflected attitudes about the President at many points over the last four years. And it’s very hard to tell when they were right or when they were wrong.
Do que li sobre o tema em publicações norte-americanas, gostava de terminar com esta frase: “I think that this is a very challenging time to use polling to predict the outcome of elections, because we live in a very polarized era.”
Como os jornais interpretam as sondagens
Na era polarizada em que vivemos (também acontece em momentos de outra natureza, mas sobretudo nesta conjuntura de extremos) é fundamental o papel mediador dos… media. Está no nome, não há como enganar. Ou há?
A julgar por algum tratamento jornalístico, e de publicações que supostamente seriam de referência em Portugal, sim:
Este é o título que o jornalista David Dinis escolheu para a notícia sobre a sondagem ICS-ULisboa/ISCTE-IUL/GfK Metris, de final de outubro. Até ver, nada a apontar.
Contudo, ao lermos o texto, percebermos que as subidas e descidas, em pontos percentuais, registam, nos partidos que registam variação, os seguintes valores:
PS: +3.3
PSD: -3.3
BE: -4
PCP: -1.1
CH: +4.4
Ora, com a descida de 1,1 pp. é preciso muita imaginação para colocar o PCP no título de uma sondagem com estes resultados. O caso agrava-se pelo facto de qualquer pessoa que trabalhe no meio jornalístico saber que imensos leitores se ficam pelos títulos e que o botão de partilha é uma arma.
Curiosamente também no Expresso, escreveu Henrique Monteiro o seguinte:
O problema não está em fazer sondagens, mas no modo como são feitas e, sobretudo, como se apresentam. Se a Comunicação Social continuar a fazer títulos definitivos com base em estudos de opinião frouxos, passa a tratar um auxiliar da compreensão política como uma espécie de oráculo.
Antes de outras histórias, esta (sem comentários, mas com muita reflexão silenciosa):
A Former Facebook VP Thinks Investing in Humans Is the Future of VC
Earlier this year, his venture firm, Slow Ventures, set aside $20 million to invest in creators themselves. Now the firm has gone and done it, joining a few individual investors in spending $1.7 million in the future of Marina Mogilko, a 31-year-old YouTube personality with multiple popular channels that touch on topics like life in Silicon Valley and learning new languages.
The decision to invest directly in humans brings about a host of legal, ethical, and moral questions that Lessin will surely need to confront head-on. The idea that someone might sign a 30-year employment contract and that society should explicitly value a human brings up questions of indentured servitude and worse—claims which Lessin sees as entirely ill-founded. (“it's def not indentured servitude,” he recently wrote in response to someone who said the legal issues seemed “daunting.”) He believes that he is setting society on a path where we are free to invest in our favorite humans through multiple venture rounds, providing young, brilliant people with the money to fund a path to success that doesn’t exist today.
By applying the venture model to humanity, Lessin believes he might have stumbled into the solution to one of society’s longest-standing and most burdensome issues: debt. “It's so fucked that young people have all the equity value—because they’re the ones that are gonna do the work, provide for the future, they have all the future potential—but they don't have any of the cash,” Lessin said. “They’re totally disadvantaged versus the boomers who are old and increasingly doing nothing but have all the money. It's completely socially imbalanced.”
Outras histórias
Da série ‘E-mails em que ninguém acredita’:
A história por detrás desta comunicação é triste mas infelizmente não surpreende. Peng Shuai é uma tenista chinesa que teve a coragem de falar sobre o assédio sexual que sofreu por parte de Zhang Gaoli. Agora, está desaparecida.
Um desastre de investigação e duas vidas encurtadas. 2 Men Convicted of Killing Malcolm X Will Be Exonerated After Decades:
A 22-month investigation conducted jointly by the Manhattan district attorney’s office and lawyers for the two men found that prosecutors and two of the nation’s premier law enforcement agencies — the Federal Bureau of Investigation and the New York Police Department — had withheld key evidence that, had it been turned over, would likely have led to the men’s acquittal.
À boleia da crise dos migrantes na fronteira Polaca, e além de condenar as ações de Lukashenko, importa pensar mais estruturalmente:
While plenty of justified opprobrium can be directed towards Lukashenko’s regime and his Kremlin backers, the EU’s dysfunctional asylum policies are also to blame. The first member state responsible for assessing an asylum seeker’s claim is generally the one in which the seeker first sets foot.
This inevitably forces people to languish in countries they never intended to seek asylum in. This system is designed by the wealthier, northern countries to outsource the delicate process of asylum protection to the bloc’s weaker frontier states in the Mediterranean and Eastern Europe. It also allows those countries to maintain a liberal façade of insisting on human rights in public, while the dirty work of fortification and pushbacks is left to poorer governments on the fringes
By consistently failing to offer meaningful opportunities or resettlement routes and denying asylum seekers access to their territories, democracies give the monopoly of movement to sinister dictatorships and smuggling networks that they purport to oppose.
Why Dictators Always Pretend to Love the Law pergunta o site Foreign Policy, partindo do exemplo do autoritarismo de Al-Sisi, ditador do Egipto. Destaquei o seguinte excerto:
This is not only an Egyptian phenomenon. Turkish, Hungarian, Russian, Polish, and other officials use the same kind of talking points. If one were to inquire with a Turkish diplomat why there are so many journalists jailed in Turkey, even though Article 28 of the constitution guarantees freedom of the press, they are apt to say that these guarantees do not apply.
That is because, the diplomat would argue, the reporters and editors in question sought to undermine the irrevocable Articles 1, 2, and 3 of the constitution, which establish Turkey as a democratic and secular republic governed by the rule of law. Many of Turkey’s jailed journalists did no such thing, of course. But because there are provisions in the constitution that limit the very same freedoms the documents set forth, the persecution of journalists for political reasons is all perfectly legal, which is what Turkish officials often argue.
Does anyone believe what officials representing authoritarian regimes have to say? It is hard to tell. Staying on the Turkey example, supporters of the ruling Justice and Development Party (AKP) look at you as if you are crazy for suggesting that Turkey is not a democracy and that in democracies journalists should not be in jail
Para fechar
Toward the end of the video, the character with Gosar's face photoshopped onto it flies through the air toward an image of Joe Biden, with sword raised, before a freeze frame concludes the video.
Yesterday, the House voted to censure Gosar, a formal slap on the wrist but one that carries significant consequences. It was the first such vote in over a decade, and cost Gosar two committee assignments, including the Committee on Oversight and Reform, which he sits on with Ocasio-Cortez. Shortly after the vote on Wednesday, Gosar retweeted a tweet that included praise of him and the video clip again.
Já vale tudo dentro do GOP e tempos sombrios esperam-se. Eis o vídeo: