Prima Facie Politics

Predicting the outcomes to elections, as we have vividly seen these last three weeks, can be a messy and imprecise affair. Pick up a newspaper or turn on the TV after New Hampshire and you saw every sort of analysis explaining the failure of the pollsters and the national media at large. Pundits and pollsters showed a good deal of integrity and accountability by admitting that they were dead wrong in their projections, and the reasons for this failure were many. Despite all the elaborate mechanisms we have for analyzing the ups and downs of political candidates, in the end our trusted statisticians did their math wrong.

And a good thing this serendipitous error was. Far too often in politics, the game is over before it really begins: the media make proclamations that, true or false, become self-fulfilling prophesies. If Hillary is projected on the national stage as the "inevitable" candidate, the voters might begin to feel that she is in fact the most qualified for the job. After all, Brian Williams said so. Regardless of the mistrust many feel towards mainstream media (see "Lowering the Bar" post), they are the only way the vast majority of Americans experience the democratic process. Love it or hate it, we all rely of the media to present us with the realities of our political climate. If CNN is way off, we can't be blamed for believing it, nor for steering our electoral choices in a direction that validates CNN's claims. When the elaborate system of polls and projections falls apart, voters are reminded of the fact that they are in the driver's seat of this country. The failures of New Hampshire, therefore, breathed new life back into people's sense of purpose in our democracy; they demonstrated the absolute fallibility of the pundits and showed that, in reality, anything can and will happen.

But on to the major issue of this post. Polling failures raise a very intriguing question about how individuals vote: are voters' preferences so mercurial as to render any poll obsolete the day after it's taken? Campaigns readily acknowledge that any poll is temporary, but just how long of a time-horizon between the polling and the voting does it take to make the poll actually mean something?

In a perfect world, many argue, people would vote on the issues and the issues alone. The candidate's personality wouldn't be a factor; neither would his or her appearance, race, or gender. In such an ideal world, polls would be dead on - people would take sides according to their beliefs on the issues, and the candidate who supported those ideas would get the vote. Of course, we all know that people don't vote on the issues alone, and while this statement can be qualified and discussed ad nauseum, it raises a crucial question: just what do people vote on?

There are certain elements of an electorate's collection decision-making process that are completely and totally random. Even the most precise polling techniques ever concocted will fail to address the chaos of, say, the weather on the day of the vote. Studies have shown that cloudier, colder, nastier days encourage a sense of pessimism and hardness that will lead voters to choose the safest bet; beautiful, sunny days on the other hand tend to put people in bright, optimistic moods. Bad weather, therefore, encourages more conservative choices, while good weather tempts voters into taking chances. In a Democrat vs. Republican race, this weather effect often plays out in predicable fashion: rain helps the GOP and hurts the Democrats. In the primary coming up tomorrow, I think it's relatively obvious what sorts of effects weather will have when deciding between Obama and Clinton. (Weather.com is predicting severe storms in the south and the midwest for Super Tuesday. Go Hillary!)

Other factors that determine voter preferences are tied not to the randomness of the skies, but to the randomness of genes. An important study conducted at Princeton University demonstrates the profound effect of the appearance of a candidate's face on voters' perceptions. Researchers have discovered that people begin to make judgments about the leadership qualities of a candidate within split seconds of seeing his or her face; in fact, our decision-making is almost subliminal, as scales begin to tip in favor of one over another within the first 100 milliseconds of seeing the face! Showing individuals the faces of front-runners in national elections (unfamiliar to the participants), test subjects were able to predict the winner in 72.4% of senatorial elections. Issues schmissues: at the animal core of it, people vote with their eyes.

This fact is not limited to political choices. A fascinating recent study in Psychological Science (click here for a synopsis) indicates that people can predict the business acumen of CEOs simply by looking at their faces. Shown the pictures of executives at the top 25 and bottom 25 of the Fortune 1,000 list, respondents were asked to rate candidates based on their perceived competence, dominance, and facial maturity. Amazingly, the judgments made on the CEO's faces reflected the profits of their respective companies. The more powerful they looked, the more successful they were in business.

But looks don't end with power. Recently published, independent research carried out by different investigators in nations around the globe seem to indicate that those considered by their peers to be "good looking" also tend to have higher IQs. As the Economist magazine's tag line puts it, "To those that have, shall be given." The same studies indicate that "good looking" people actually earn more money for their companies than their less fortunate colleagues.

The previously cited studies carry profound ramifications, one of which I will illiberally suggest: if faces really do carry intangible but actionable information about the candidate's intelligence, skills, and ability to lead, is it really such a bad thing that people don't seem to vote on the issues? Millions of years of evolution has equipped us with mental hardware that is capable of penetrating through all the semantic elements of a political race into the core of what makes a person who they are. Of course, studies have yet to be done correlating those who look like leaders and are subsequently elected to actual success in office. Nonetheless, recent studies pose interesting questions: how much of electing a candidate is determined by prima facie facts and should we actually give those irrational perceptions credence? How different would our election results be if voters were simply given a facebook of candidates and asked to choose based on that alone? (We can bet that Mitt Romney would be doing a lot better.)

As Super Tuesday looms, it is important to remember that as much as we educated and civic-minded people like to put weight on the issues dominating the election, there is a lot more at play when we're in the voting booths than whose health care plan makes the most sense. Perhaps Obama, Clinton and McCain would do well to get a haircut before leaving their hotels tomorrow morning. And don't forget the umbrella.