THE BLOG

Featuring fresh takes and real-time analysis from HuffPost's signature lineup of contributors

Darcy Oberding Headshot

U.S. Polling Results Could Represent The New Normal

Posted: Updated:
REPUBLICAN DEMOCRAT
Hill Street Studios via Getty Images
Print

It would be unfair to say Donald Trump cannot win the election. At the same time, it would be inaccurate to say he doesn't face an incredibly steep uphill battle.

Clinton was the recipient of a very beneficial 4-point post-convention bounce, which gave her a 5.1 point lead following the Democratic National Convention. It has been nearly a month since the Convention and Clinton has expanded on her bounce to lead Trump nationally by 6 points.

What's even more troubling for Trump is that Clinton is leading in key swing states, such as Virginia, Iowa, and Colorado by 12, 3, and 2 points, respectively.

This comes from a new swing state poll released by Quinnipiac University. It should be noted that Quinnipiac polls are generally considered to be more favourable to Trump, because they typically show white voters as a larger proportion of the electorate than in 2012-- even if that assumption is at odds with reality.

It's not just in the aforementioned swing states where Clinton is leading either. Real Clear Politics has her beating Trump in Florida by 4.5, in Ohio by 2.6, in Michigan by 7.3, and in Pennsylvania by over 9 points!

Trump is one of the most poll-obsessed candidates in recent memory, and now, some of his biggest supporters are claiming polls are meaningless, and because he has large crowds at rallies, he is likely to win the election.

I don't think there is anything wrong with approaching polling figures with healthy skepticism, and while the science of polling can be imprecise, polls are typically quite accurate at predicting how close a race will be. In fact, political scientists, Christopher Wlezien from the University of Texas, and Robert Erikson from Columbia, assessed 60 years of election polling numbers and found that polls "gradually converge to a point that is close to the actual November outcome."

Trump supporters are often quick to point out that polls can be misleading based on the Bradley effect. The Bradley effect takes its name from the former Mayor of Los Angeles, Tom Bradley, an African-American who lost the election for Governor of California in 1982.

Essentially, what this describes is a social desirability bias. In the case of Bradley, some white voters intentionally gave inaccurate polling responses that they considered to be publicly acceptable or politically correct, fearing that if they had stated their true preference, they could receive criticism of racial motivation.

Savvy pollsters, such as those at the Huffington Post's Pollster trend, have put the Bradley effect to the test regarding Trump. This was done by comparing live telephone polls with online polls to combat any social desirability bias a survey participant may have. They found that both polls provide nearly the same results, concluding that "a lack of any difference is strong suggestion that social desirability bias is not a meaningful factor in any possible polling bias for the 2016 presidential election."

Political pundits often reference Trump's unpredictable rise as something that everyone got wrong. The Republican primaries are now a classic example of political predictions gone awry and how conventional political wisdom may be put into question. Although that may be mostly true, as Joe Wiener of The Nation points out, the polls' predictions largely matched the results, so the problem with the predictions had nothing to do with the polls.

Wiener goes on to reference Harry Enten of FiveThirtyEight who points out that of the 549 polls in their national primary polling database, Trump led in 91 per cent of them. In other words, even in the primaries, the polls were mostly correct about Trump.

It would be foolish for the Trump camp to ignore the polls and succumb to thinking that large rallies will equate to a large turnout on Election Day. As mentioned, Trump has shown that he is obsessed with polls and with the promotion of pollster Kellyanne Conway from advisor to campaign manager, this indicates they are paying attention to them. However, time will tell if Conway, and the new chief executive officer of Trump's campaign, Stephen Bannon, can do anything to positively impact the polls in Trump's favour.

The current polls may reflect the new normal, and if this is the case, this is bad news for Trump. It is hard to think of a modern candidate more qualified, in the traditional sense, to be president than Hillary Clinton. However, it is equally as difficult to think of a candidate that is more flawed. The Republican Party, and even Donald Trump, had, and maybe still have, an opportunity to win in 2016, however, it is becoming clearer and clearer, Trump being Trump will not get them there.

Follow HuffPost Canada Blogs on Facebook