Log in

Point & Counterpoint

Do election polls hurt or help? Two views

Posted

Rob Richie: Even when on target, polls hurt our democracy

Americans are blessed to have the First Amendment, including prohibitions against governments abridging freedom of speech and of the press.

Yet, freedom comes with responsibility. Our rampant polling may fill Americans’ hunger to know what might happen before we vote, but it’s a significant contributor to the deterioration of our democracy. Although we are unable to outlaw it, we can take more responsibility for its effect.

Polling’s problem isn’t quality. On average, pollsters typically get races within a few percentage points. We can generally trust polls as long as realizing a slight lead isn’t a guarantee of victory.

The problem is its effect on the quality of our political discourse and our openness to treating more candidates seriously. Take this year. A deafening drumbeat of polling virtually silenced what we might have learned about candidates.

It started before the primaries. Incumbent Joe Biden dominated among Democrats and his predecessor, Donald Trump, among Republicans. They were historically unpopular, but their big leads over their competition depressed interest in the primaries.

Biden and Trump refused to debate their challengers. Still, primary voters forgave them — if polls showed the results were a foregone conclusion, why worry about debates even if they would have allowed voters to better assess their readiness and engage in more policy discussion within their parties.

Democrats had a rude awakening with Biden’s poor debate performance against Trump in June. Biden’s candidacy was in trouble, but the debate’s relatively modest effect on polls kept him in the race for weeks. When he finally did withdraw, Kamala Harris immediately validated his decision, with Democrats having a surge of hope.

Would Harris be better-positioned now if having to win a “blitz primary” where voters could have learned more about her and her priorities? We’ll never know — polls showing she was the frontrunner justified her appointment.

By September, the combination of polls and relentless attacks had reduced potential independent and minor party candidates to the margins. The most reliably brutal effect of polling is to sideline minor pirates except for their “spoiler impact.”

Now, it’s all about polls in the dully familiar seven presidential swing states. Polls justify why the candidates spend nearly all their time there and why billions will be spent on the couple million swing voters who will decide who wins them.

In winner-take-all elections, campaigning only matters when results are in doubt. In Senate elections, Republicans seek an upset in Maryland, Democrats in Texas and Florida, and independents in Nebraska. But polls make them underdogs, and that Senate control will likely hinge on whether Democratic incumbents can carry Ohio and Montana.

Of 435 House seats, only about 40 might change parties — and far fewer will. Polls again drive where resources go, depressing debate, competition and turnout everywhere.

Reliance on polling shortchanges meaningful discussion about issues and character. It ensures most of us know we’re only symbolic participants and nervous spectators in deciding control of the White House and Congress.

So what to do? Given the First Amendment, we can’t easily regulate polling more than mandating shorter campaigning or limiting how much billionaires can spend on influencing opinion. However, we can still take more personal responsibility.

First, we can tune out polls and reward journalists who focus on substance more than the horse race with our subscriptions and clicks.

Second, we can refuse to participate in polls, which creates greater uncertainty by making it harder to get a representative sample.

Third, we can back election reforms that put voters more in charge. Pollsters thrive on binary elections where voters pick only one candidate. Used in Alaska and Maine and on the ballot in four more states this November, ranked-choice voting enables minor candidates to get a fair shake and incentivizes creative campaigning. Combining it with all-candidate primaries and multi-member districts will open up elections.

I’m generally trusting the polls this year — but eager for our collective work to reduce their influence.

Editor’s note: Rob Richie is a co-founder and senior adviser of FairVote, a nonpartisan election reform organization. Reader reactions, pro or con, are welcomed at AzOpinions@iniusa.org.

Chris Talgo: Polls provide data voters need for informed decisions

With days before the 2024 election, pollsters are working overtime to determine the state of what appears to be a very tight race. Throughout modern history, polls have played an essential role in the election process, providing real-time information that helps voters and candidates. However, in recent years, polls have come under increased scrutiny mainly because they have failed to measure accurately the pulse of the nation.

Despite the vast majority of polls in 2016 and 2020 miscalculating voter sentiment, that does not mean that all polls are untrustworthy and, therefore, useless. A more nuanced view shows polls that track issues instead of candidates are still meaningful.

For example, according to most polls heading into the 2024 general election, the top three issues are inflation/cost of living, illegal immigration and crime.

Information like this is helpful because it provides context for voters and candidates alike. Moreover, it can bring issues to the forefront of the political discourse that otherwise would not be highlighted. And, perhaps most important, polls like these provide much-needed feedback to the candidates so that they can focus on the issues that matter most to voters.

Aside from polls that track specific issues, general polls that measure the popularity of candidates are still worthwhile. This is especially true in primary races, where several candidates, many unknown to the public, run in closely contested races.

Polling that tracks where these candidates stand can provide valuable insights for potential voters, most of whom know very little about these would-be officeholders.

In recent years, research has emerged that shows pre-election polling is relevant across several fronts. According to a study, “Are Polls and Probabilities Self-Fulfilling Prophecies?” by Stanford University professor Neil Malhotra, some voters view polls as an insight into the “wisdom of the crowd.”

By this, less-informed voters can at least have an idea of where the masses stand regarding the election, which could sway their vote one way or another.

As Malhotra notes, “Majorities can cascade, which is not good if we want to preserve minority rights or worry about herding. But we see that people are also looking for information and try to learn from the wisdom of crowds.” Moreover, “The main reason why people conform to majority opinion in the political domain is that they perceive there to be information about the quality of the policies in learning about mass support,” Malhotra concludes.

Professor Jouni Kuha of the London School of Economics and Political Science has also researched the effects of polls on peoples’ voting habits. Like Malhotra, he found that “polls act as a feedback mechanism which could affect parties’ policy choices, whereas nearer to an election, they are feedback mechanism on how the campaign is going.”

Kuha also cites the wisdom of crowds, or what he calls the “bandwagon effect.” He is also careful to point out that pollsters should not try to influence the electorate by jigging the polls in favor of their preferred candidate.

This is a problem that has become evident throughout the Donald Trump era. In 2016 and 2020, most polls were way off the mark, partly due to the phenomenon of the shy Trump voter but also because pollsters intentionally oversampled Democrats.

In those two elections, Trump was down significantly right up to Election Day, according to the overwhelming majority of polls routinely cited by the media. Interestingly, some pundits believe this was done deliberately to tamp down support for Trump while others claim it was nothing more than an accidental oversight.

This year, most of the polls have Trump and Harris in a close race at the national level and across the all-important swing states.

Given this set of circumstances, I can’t help but wonder if the pollsters have learned their lesson after being so wrong the past two presidential elections or if we are in an enormous political realignment.

Regardless, I will continue to pay close attention to the polls, particularly consolidated polling averages, which have tended to be the most accurate of late.

Editor’s note: Chris Talgo is the editorial director at The Heartland Institute. Reader reactions, pro or con, are welcomed at AzOpinions@iniusa.org.

2024 election, polls, polling data, First Amendment, democracy, pollsters