Conservative Pollsters Grasping At Straws
In the classic Twilight Zone episode “The Midnight Sun,” a cataclysm has sent the Earth hurtling toward the Sun. As the planet grows nearer to the star, society is slowly breaking down, and yet, oddly, people try to hold onto hope. “I hear it’s a bit cooler in Toronto,” they tell each other, despite the ridiculousness of it. The alternative to hope is to give up, and people are bad at that.
I thought of that episode when I first heard about UnSkewedPolls.com, a website run by hardcore GOP partisan Dean Chambers. The site attempts to fix bias in polling by re-weighting the polls of the Gallups, Reuters and Rasmussens of the world, only with a more conservative partisan breakdown.
Chambers lays out his argument in a refiguring of a Washington Post poll:
The survey is skewed and an “un-skewed” analysis of their numbers showed Romney likely has a 50 to 43 lead right now. This difference of a skewed-result tie as opposed to an “un-skewed” 7 point lead for Romney has serious implications for all the conclusions drawn from the survey results.
In this survey, the sampling is based on only 24 percent responding as Republicans, while Democrats were 33 percent, Independents were 36 and the remaining 7 percent responded otherwise. No serious observer of American politics or pollster believes that Republicans make up only 24 percent of the population. Is it precisely this under-sampling of Republicans, and proportional over-sampling of Democrats, that skews this survey.
Unsurprisingly, if you assume more Republicans will vote, Mitt Romney does better. While the RealClearPolitics poll of polls has President Barack Obama leading Mitt Romney by 3.7 percent, UnSkewed Polls gives Romney a whopping 7.8 percent lead.
The methodology of Chambers is, to put it nicely, spurious. Chambers’ weighting is based on the partisan breakdown Rasmussen Reports showed for the 2010 election, a wave election that strongly favored the GOP. Additionally, Rasmussen’s party ID skews conservative. Scott Rasmussen himself told Buzzfeed that “you cannot compare partisan weighting from one polling firm to another.”
That hasn’t stopped Republicans from embracing UnSkewedPolls. Texas Gov. Rick Perry approvingly tweeted a link to the site. Drudge and Pajamas Media have also linked to Chambers.
It’s easy to deride this as yet another attempt by the right to ignore facts. While there’s a long time until election day, Barack Obama appears to hold a small but stable lead in the polling. The Romney camp certainly doesn’t appear to be acting as if they hold a 7-point lead. Just another day inside the conservative bubble.
That is unfair to Chambers, however, because Democrats have shown just as much willingness to will away bad polling. When calamity is staring you in the face, it’s much easier to hold on to a glimmer of hope.
In 2010, as Republicans were rolling up big leads in the generic congressional ballot polls, Democrats told themselves a simple story. Polls that did not call cell phones tended to tilt toward the GOP. Therefore, the polls were wrong. Democrats and Republicans were tied. The Democrats could hold on to the House after all!
Needless to say, the polls were not wrong. The GOP took back the House. Yes, polls that call cell phones may be slightly more Democrat-friendly than polls that don’t, but the effect was not so large as to make the polls invalid.
Were the Democrats in 2010 engaged in wishful thinking, just as Republicans who cite UnSkewedPolls are today? Of course they were. Democrats and Republicans may have significant policy differences, but they are both human. Humans, alas, are not strictly logical. We are hard-wired to be optimistic, to try to find hope in hopelessness.
This makes sense. If our forebearers were pessimistic, we as a species wouldn’t have gotten very far. When climate change was whipsawing humans in Africa back and forth between drought and plenty, it took an optimistic outlook, a hope that things would get better, in order for us to make it through the cataclysms.
As we are optimistic by nature, it makes it very hard to view a potential loss with clear-eyed acceptance. We seek a way around it, a way that victory can still be snatched from the jaws of defeat.
Ignoring the polls is an easy way to do this. Polls are, after all, as much art as science. Because we live in an age when it’s tough to get a truly random sample of people just by calling, pollsters are required to weight their responses based on a variety of factors, from party identification to gender to race to income. How they do that can affect the outcome of a poll. This means that anyone who wants to ignore a bad poll has a ready-made excuse — the polls are weighted wrong. If they were right, my candidate would lead.
Of course, polling firms have nothing but accurate polling to recommend them, and so they put a great deal of time and effort into getting their weighting right; the numbers are not just picked out of the air, or adjusted based on political bias, with the possible exception of Rasmussen Polls. By and large, polling is reasonably accurate, especially when looked at as an aggregate.
That won’t stop partisans from ignoring bad ones, though. Humans look for that glimmer of hope, and given an avalanche of bad news, it’s understandable that Republicans would be trying to spin themselves into a lead. It doesn’t mean they’re right of course — the UnSkewedPolls site is completely wrong. But they’ll know that come November; until then, they’ll be holding on to hope however they can.
Image Credit: Stock.Xchng