A New York Times poll released yesterday showed President Obama ahead of Republican rival Mitt Romney by 10 percentage points in Ohio, 12 in Pennsylvania, and 9 in Florida. This and other polls have led Democrats to declare victory and Republicans to despair. But a closer look at the data suggests the results aren’t as accurate as they appear.
Most people assume that polls are accurate. It seems like any news report that includes the phrase “according to a recent poll” gets treated as the gospel truth.
Yet flawed polling is a problem, especially in an election season. Polls are often unreliable, misleading, and easily manipulated for partisan gain.
So what makes polls methods of spin rather than accurate foreshadowings? Humans conduct the polls. Human answer the polls. They are subject to human error and misjudgment.
Samples affect the legitimacy of a poll. In this cycle, Democrats are often oversampled. In Ohio, the New York Times poll skews toward Democrats: the sample was 35 percent Democrat, 26 percent Republican, and 35 percent independent. But this discrepancy went under-reported.
Wording is also crucial. Respondents will say one thing in response to a question about “entitlements” and another in response to a question about “Social Security,” even if they’re asking essentially the same thing.
But polling data is essential in presidential elections. Campaigns need data beyond their own internal polls. So the question emerges as to how to gain the most knowledge from a presidential poll in the middle of such a swamp. The only answer is: look closely. Every aspect of the poll matters. Perhaps the public should worry less about predicting the results of elections and more about the substance of policy debates.
As for 2012, the race is far from over. The misleading polls may continue to shape the media narrative in the coming weeks, but voters may yet surprise. It wouldn’t be the first time people have misunderestimated a Republican presidential candidate. Forget the polls.