Our Insane Addiction to Polls


 Frank Bruni 

REMEMBER the poll last week that had Bernie Sanders ahead of Hillary Clinton in New Hampshire by three points?

No, you’re thinking. I’ve got it wrong. Sanders was up by 27.

That’s true, if you’re talking about the figures that CNN and WMUR released on Tuesday. I’m talking about the ones that Gravis Marketing and One America News Network released on Wednesday.

There were three polls of New Hampshire voters over just two days last week, according to the archive maintained by Real Clear Politics. There were three polls of Iowa voters on Thursday alone. One had Clinton up by eight, while another had Sanders up by that same margin. One had Donald Trump up by 11. Another had Ted Cruz up by two.

Over a monthlong period ending Thursday night — a monthlong period, mind you, that included the Christmas and New Year’s break — there were 11 polls in Iowa, 10 in New Hampshire and nine nationally. There were polls focused on 10 different states.

And their findings were often treated as breathless news. On Wednesday evening, I visited the home page of the Politico website — I’m using Politico as a random example — and spotted four stories that were essentially about poll results.

My television was on: The CNN anchor Erin Burnett began her show with a report on the latest poll.

I’d say that we’re in a period of polling bloat, but bloat is too wan a word. Where polling and the media’s attention to it are concerned, we’re gorging ourselves into a state of morbid obesity.

“I’ve never seen anything like it,” said Ralph Reed, a longtime Republican strategist who thought that things were bad in 2008 and 2012 and realizes now that those were days of temperance and innocence, to be pined for and perhaps never savored again.

But it’s not the crazy bounty of polls that fascinates him (and me) most. It’s something else.

“There seems to be an inverse relationship between the preponderance of polling and the reliability of polling,” Reed said, nailing one of the most illogical, paradoxical dynamics of the 2016 election so far.

We’re leaning harder than ever on polling precisely when that makes the least sense. We’re wallowing in polls even as they come to wildly different conclusions that should give us serious pause.

Good polls are arguably more difficult than ever to do, for reasons I’ll go into later. And from 2012 forward, there have been prominent examples of how poorly they sometimes predict outcomes. They botched the most recent parliamentary elections in Britain and Israel. They botched the 2014 midterms in the United States, grossly miscalculating the margins in various congressional and gubernatorial races.

Last year, Reed noted, the Real Clear Politics average of polls in the Kentucky governor’s race had the Democrat, Jack Conway, ahead of the Republican, Matt Bevin, by five points.

Bevin won by nearly nine.

“It was 14 points off,” Reed marveled. “But everybody shrugs and moves on down the highway.”

There are explanations for those shrugs, and they speak to the quirks and flaws of political journalism in a wired, revved-up world.

There are consequences, too. An obsession with polls and a quickness to weave narratives around them bolster certain candidates and retard others, and could well affect the outcome of this presidential election.

If Donald Trump wins the Republican nomination — or, heaven forbid, the White House — it will be partly because we in the media justified saturation coverage of him by pointing to polls, which in turn legitimized his fixation on them as proof that he’s up to the job: He must be, because plenty of people apparently picture him in it.

“He caresses his polls numbers,” said David Axelrod, one of the chief architects of Obama’s 2008 and 2012 presidential campaigns, adding that Trump’s first order of business when he steps up to a microphone is “a recitation of poll numbers. He’s like a Lothario recounting his exploits every time he starts a speech.”

If Jeb Bush’s candidacy comes to naught, his underwhelming poll numbers — and how they were used to cast him instantly as an underachiever — will have been a factor. And if Clinton fails to win the nomination, the media’s embrace of certain polls among an ever-changing riot of them will have played at least some small role.

“If she’s three points behind in New Hampshire, it’s a close race,” Axelrod said. “But if she’s 27 points behind, her campaign’s in free fall. That’s a sexier story and the one that’s chosen. It becomes the meme. It becomes the prism through which everyone filters their coverage. It skews how people view everything that a candidate does: Is it conviction or desperation?”

Desperation makes a better story. So the media dwells on the most pessimistic projections, ensuring that polling, no matter how divorced from reality, shapes it.

Polls determined which Republican candidates participated in which debates, although just a couple of percentage points — the margin of error, really — separated a few of the prime-time debaters from the early birds.

Some of these polls were national ones, which have minimal relevance to the decisive, trajectory-setting contests in Iowa, New Hampshire and South Carolina.

“A national poll is absolutely meaningless,” said Stuart Stevens, the chief strategist for Romney’s 2012 campaign. “One of every nine Americans lives in California. So one of every nine voters in that poll is going to be in California. When’s the last time anybody read a story about the Republican primary in California?”

And if you dig below the surface of these national polls — or of polls in the states with the first contests — you find a crucial detail that we in the media blithely gloss over: Many, if not most, voters haven’t made up their minds. In last week’s CNN/WMUR poll of New Hampshire voters, for example, about one in three Republicans said that they had definitely decided on a candidate.

Part of what I find so jarring about the media’s insatiable appetite for polls right now is that it defies our past resolutions to go on a desperately needed diet. For all of my 30 years as a journalist, I’ve listened to reporters, editors and producers bemoan the “horse-race coverage” of campaigns and exhort one another to be better come the next election and concentrate instead on issues, records, biographies, substance.

Suddenly the apologies and exhortations are gone. We’re worse. Every candidate’s a thoroughbred, every day the Kentucky Derby and just about every other story on many newscasts and news sites an assessment of his or her odds. We’re resigned to treating campaigns as sport. It represents the surest path to a large enough audience to keep a beleaguered industry economically viable.

Some dispatches are overt in their reliance on polls, others less so, using polls as the prompts or context for discussions of stalled campaign momentum or new campaign strategies. The Times is guilty. I’m guilty. Polls are the seemingly irresistible argot of political coverage; to purge them from your vocabulary is to speak in an unrecognizable tongue.

BESIDES which, we need the content. In a fast-metabolism world of constantly monitored smartphones and routinely refreshed browsers, news organizations are under greater pressure to produce fresh nuggets of information even as their budgets and resources for reporting decline. So we readily trumpet and analyze polls, a practice that has even given birth and currency to a whole strain of news that’s survey-based, with practitioners who burrow ever deeper into numbers.

“You have a lot of people who don’t ever go out on the campaign trail and actually talk to voters,” Stevens said. “We now call those people data journalists. They don’t have to report. It’s a wonderful category.”

Our demand for polls guarantees a robust supply of them: Churning these surveys out is great guaranteed publicity for news organizations, research companies and academic institutions. How many Americans are aware of Quinnipiac University in Hamden, Conn., or Monmouth University in West Long Branch, N.J., only because they commission and put their stamps on polls?

But not all of the surveys that the media cites are created equal. Gary Langer, the founder of Langer Research Associates and a former director of polling for ABC News, compared the variable quality of polls to the variable quality of what you eat. “There are well-crafted, delicious meals,” he said. “Then there’s fast food. And then there’s listeria.”

Listeria is common. People’s reliance on cellphones has complicated polling: Federal law forbids automated calls to such phones, so pollsters must bear the considerable expense of dialing them by hand or, alternately, make do without them, which can lead to imperfect results.

Beyond that, there’s the question of whether the kind of people who consent to polls are true weather vanes.

“The entire industry rests on the idea that the people you get are representative of the people you don’t get,” said Jon Cohen, who supervised polling for The Washington Post from 2006 to 2013 and is now the vice president of survey research at SurveyMonkey. “I think that’s an increasingly questionable premise and one that I keep in mind every time I design a survey.”

“People don’t enjoy or even tolerate those conversations the way they used to,” he added. “Taking a survey is a participatory act, and different kinds of people participate in different things.”


Can we extrapolate from the kinds of people who have been professing adoration of Trump — or, for that matter, of Sanders — into the general population? Apart from that, can we trust the constancy of their affections? Will they actually cast votes?

Over the next few weeks, we’ll find out, and I promise you this: There will be some surprises. If they’re significant enough, polling could be as big a loser in the 2016 primaries as any candidate is. And we in the media will be forced to apologize anew for our poll-mad behavior. Sadly, that doesn’t mean we’ll change it.

http://www.nytimes.com/2016/01/24/opinion/campaign-stops/our-insane-addiction-to-polls.html?action=click&pgtype=Homepage&clickSource=story-heading&module=opinion-c-col-left-region&region=opinion-c-col-left-region&WT.nav=opinion-c-col-left-region&_r=0