“The 360” shows diverse perspectives on the day’s main news and debates.
What is happening
With less than six weeks to go before the midterm elections, Democrats are beginning to have reason to hope that a landslide Republican victory, often predicted, may not materialize. In recent months, once dire poll results have begun to change direction thanks in part to a handful of important legislative victories, falling gasoline prices and the emergence of abortion as a key issue for voters.
Forecasts now suggest that Democrats are favored maintain the majority in the Senate and take a chance on a puncher holding the house. But party leaders could be forgiven for looking at this potential change in fortunes with skepticism. In two of the last three election cycles, polls predicting big Democratic victories turned out to be inaccurate when the votes were finally counted.
In 2016, polls gave Hillary Clinton somewhere between a 70 and 99 percent chance of winning the presidency. These predictions, for several reasons, grossly underestimated Donald Trump’s support in key swing states. This massive misfire led to an intense period of introspection within the election industry. Many leading researchers have made significant changes to their methods to try to explain what they had previously missed. O semesters of 2018 gave reason to believe these adjustments worked—polls suggested Democrats were likely to retake the House, and they did.
So, in 2020, research was farther away than they were four years earlier. Joe Biden won the presidency and Democrats took back the Senate, but those victories were by incredibly narrow margins when forecasts indicated that a “blue wave” was the most likely outcome.
Why is there debate?
Polls are not intended to be rigid prophecies about what will happen – there will always be an inherent amount of uncertainty. But a pair of high-profile misfires in just four years, both in the same direction, have led many to believe that the questions behind the 2016 and 2020 forecasts remain, and polls may once again be exaggerating the strength of the Democrats heading into the middle of the semester.
In recent weeks, some of the most prominent U.S. pollsters have raised concerns that the polls could be wrong again. Nate Cohn of the New York Times wrote that the warning signs of previous polls are “blinking again” – specifically noting that Democrats appear to be exceeding expectations in many of the same states that pollsters misread in previous elections. Experts say one of the biggest problems is that so-called MAGA Republicans — who often harbor a deep disdain for conventional political structures — are much less likely to participate in elections than typical Democratic voters. This imbalance in “non-response bias” can make it difficult, if not impossible, to gauge enthusiasm among the Republican Party base.
But others say there is reason to believe that 2016 and 2020 were truly outliers that do not actually represent a fundamental flaw in political polling. They argue that polls have largely gotten it right in a long list of elections across the country in recent years and have even gotten it wrong by sometimes overstating the strength of the Republican Party. Some also believe that Trump himself may be the confounding factor that led to recent poll failures. He won’t be on the ballot in November, which could mean polls are primed for a more accurate performance this time around, they say.
What is the next
One thing experts agree on is that the battle to control Congress and win key governorships will be incredibly tight, even if the polls have the same margin of error as in 2016 and 2020.
Perspectives
You can’t kick out the entire election industry because of two bad results
“People’s concerns about the polls stem primarily from a sample of exactly two elections, 2020 and 2016. … It’s true that in 2020 and 2016, the polls were wrong across a large number of races and states. But the whole notion of systematic error in research is that it is, well, systematic: it affects almost all breeds, or at least the vast majority of them. There simply isn’t a meaningful sample size to work with here, or anything close to it.” -Nate Silver, FiveThirtyEight
Today’s polls look a lot like the ones that got it wrong in previous elections
“Democratic Senate candidates are exceeding expectations in the same places where polls overestimated Mr. Biden in 2020 and Mrs. -Nate Cohn, New York Times
Trump’s absence could make this year’s polls more reliable
“No one knows if this ‘non-response bias,’ as researchers call it, is something that only happens when Trump is on the ballot… or if it is also affecting the polls this year. And although researchers have tried several ways to improve their samples, there is no way to know in advance whether they made the right adjustments.” -David Lauter, Los Angeles Times
Political Polls May Be Fundamentally Broken
A polling error of about 3 points on average is quite normal. All polling is an inexact science that attempts to model the opinion of a large population based on a sample of a small part of that population. …But if polls consistently miss, over several cycles, in the same partisan direction, and often in the same states or regions, that could indicate a fundamental problem.” -André Prokop, Vox
Even if the polls are wrong, Democrats have reason to believe their fortunes are improving
“The wave of good news for Democrats is not a mirage either. The party is enjoying a mix of real political success and some genuine lucky breaks; the gains are real. …Most importantly, Dobbs’ decision made abortion one of the most pressing issues in the midterm elections. There’s no telling whether all this will lead to high voter turnout among Democrats in November. But what is undeniable is that none of these conditions existed six months ago.” -Alex Shepard, The New Republic
Polls may be missing a rightward shift among Hispanic voters
“[Hispanic] voters, often difficult to survey, were a reliably Democratic bloc for a decade through 2018, but less so in 2020. If they are shifting in the Republican direction this year, pollsters may be underestimating the size of that effect on the electorate — which would be very bad news for Democrats in crucial Senate races in Arizona, Nevada and Colorado.” -Dan McLaughlin, National Review
Even though he’s not at the polls, Trump makes it impossible to know whether to trust the polls
“The doubts we have about the polls reflect broader doubts about the ‘Trump effect’ in 2022. He is not on the ballot, but he is very present in political discourse and, even as a malevolent ghost, he may have permanently changed the way his supporters think about voting, answering surveys or (dis)respecting the election results.” -Ed Kilgore, new York
The polls will probably be wrong, but we can’t know how far or in what direction
“We don’t know if the polls will be wrong this year. And if they do, we don’t know which party they might underestimate. Those who reliably predict errors in polls often end up notoriously wrong. But the potential for errors in the polls on Election Day is high — and it can make political junkies queasy, regardless of any other election facts they know. -David Byler, Washington Post
The public and media need to stop asking polls to be something they are not
“Trying to determine whether the polls will be ‘right’ or not is the wrong approach. Instead, recognize the uncertainty that accompanies politics and look for polls that attempt to contextualize this unique election. Simple metrics will not be enough.” -Stephen Clermont, Political
The real mistake is treating polls as predictions in the first place
“At its best, research can empower the general public to help influence crucial decisions. But it is important to recognize the limitations of the research in addition to the strengths. The recent national elections have reminded us how problematic it is when we think of polls as predictions of the future, rather than a glimpse of where people stand at a given moment.” -Laura Santhanam, PBS News Hour
Is there a topic you would like to see covered on “The 360”? Send your suggestions to the360@yahoonews.com.
Photographic illustration: Yahoo News; photos: Getty Images
