During an election season, reading the exit polls becomes a sport in the media. Voters are stopped after casting a ballot and asked what influenced their thinking. The goal of interpreting exit polls is to determine why voters made the decisions they did.
Perhaps they only thing we've grown more used to hearing in election years than poll results is how unreliable polls are. But they're especially problematic during primary races, partly because the campaigns themselves tend to place too much faith in them. When candidates begin tailoring their messages to what voters in the early primaries say they care about, the campaigns can later be surprised to find those issues aren't as decisive as they'd first thought.
There's a psychological reason for this: the process of making complex choices, like the selection of a presidential candidate, begins with having a weak attitude toward a little information and ends up with holding a consistently strong attitude toward a lot more of it.
Voters are faced with lots of information about candidates right from the beginning. Many have just a superficial knowledge of the contenders—what they look like and how they speak—while some have a deeper knowledge of their policy positions. To some degree, people’s prior beliefs about many of those issues affect whether they like a particular candidate or dislike another. But it's also well known that the image a candidate projects—how much they look and sound "presidential" or the way they handle themselves in debates—shapes those initial preferences as well.
As soon as someone begins to lean in a certain direction, though, mechanisms start kicking in that make that candidate seem more and more favorable. The qualities a voter sees positively begin to hold more weight. Less favorable aspects become less important. Gradually, your initial preference is enhanced, turning a candidate you merely liked something about into one whom you thoroughly support. It's a process I've called "spreading coherence."
Early in the election, of course, these overall preferences are pretty malleable. A particularly bad debate performance by a candidate, a drubbing in an earlier primary, or a campaign gaffe might tilt a voter toward another contender. But when that happens, the issues a voter believes to be most important tend to change as well. Factors that favor the new preference become salient, while those that drove a love for the previous favorite begin to fade.
In fact, exit polls themselves reflect this psychological dynamic. In many ways, the attitudes voters express to pollsters aren't so much driven by their opinions on issues as they are by their preference for one candidate over another. It's often the candidate that drives someone's position on an issue, not the other way around.
That helps explain why the optics around a politician are so important on the campaign trail: that feeling of connection voters get when they see a candidate speak, the energy they impart to audiences. Their policy ideas notwithstanding, people like to support leaders they see as winners, so evidence that a candidate is actually going to win the election also drives more people to jump on the bandwagon.
The factors behind these interpersonal preferences can even shape how an election is remembered. After Ronald Reagan won the presidency in 1980, people talked about the way that his grandfatherly image had resonated with voters who appreciated his message of hope for the future at a time when the economy was bad and American hostages were being held in Iran.
In many ways, popular dissatisfaction with Jimmy Carter combined with the charm of Reagan to create an initial preference in many voters that led them to appreciate other elements of Reagan's message as the campaign progressed. Ultimately, many voters accepted his low-tax, pro-business plans at least as much because of his personal appeal as because they believed passionately in his economic policies.
But this decision-making process isn't just a political thing. The way we develop our attitudes from small, initial leanings to strong preferences happens for products and services, too. Advertisers make products feel familiar, and that familiarity makes those products feel comfortable. That effect is compounded when we factor a brand's popularity into our buying choices. If lots of people seem to like a product, we're more likely to give it a try ourselves. And when we perceive a product to be well-liked, we're prone to focusing on its characteristics and discounting its drawbacks.
For politicians and brands alike, it's tempting to try and win people over by creating messages that convince people to try something new and exciting. Instead, it may be more effective to make the new offering seem familiar and popularly accepted. The more comfortable people are with whatever it is they're being asked to go in for, the easier it is for the process of spreading coherence to take hold. Before long people will bring the messages they're hearing in line with their own emerging preference.