Pollster.com

Articles and Analysis

August 13, 2008

Taylor: What Motivates Voters?

By Guest Pollster

Today's guest pollster contribution comes from Humphrey Taylor, who has served as chairman of The Harris Poll, a service of Harris Interactive, since 1994.

When I started work in market research, I spent the first month learning to interview people face to face, in central Scotland. It was a great experience. At the first house, a woman answered the door and, as I nervously explained that I wanted to interview her, she shut the door in my face. My supervisor, a cheerful, middle-aged woman, rang the bell again and, all smiles and self-confidence, easily completed the interview. The same thing happened at the next house, and my self-confidence hit rock bottom.

At the third house, to my enormous relief, I actually managed to complete the interview with an elderly Scottish lady. But as I thanked her, she said, with a twinkle in her eye, "Och, surely you don't believe all the things the folks tell you, do you?" This may have been one of the most valuable comments anyone has ever made to me about survey research, and I have remembered it many times over my working life.

The sad truth is that all too often we researchers naively accept what respondents tell us without questioning if it is really "true." The following exercise explains part of the problem. Ask someone which candidate or party they prefer and they will usually give you an answer (and, by the way, it will probably be true). But then ask "Why?" and the conversation will probably go something like this:

Push harder and you will probably find that this voter really doesn't know what the candidate's economic (or most other) policy proposals really are. But this does not stop people from having a strong preference for one candidate over another, or believing that they would handle the issue mentioned better than their opponents.

One model of how voters choose candidates is that they are like juries. They listen to the candidates and carefully consider their policy proposals before deciding which way to vote. Unfortunately this theory is almost never true.

There are many reasons why it is so difficult to understand people's motives. One is that most people don't understand themselves and often rationalize their attitudes and behavior. Sometimes they surely deceive themselves and sometimes they knowingly bend the truth or tell outright lies. There is a growing body of literature that documents the unreliability of replies given to interviewers where there is a "socially desirable" answer. Large numbers of people lie in telephone and in-person surveys about whether they believe in God, go to church, give money to charity, clean their teeth regularly, drive over the speed limit or drink alcohol. Many people who do not vote claim that they do. And the number of people who say they voted for a sitting president tends to go up when he is very popular and down when his ratings fall.

Another problem is that people give inaccurate answers not because they are lying but because their memory is imperfect. And many people's honest predictions of their own future behavior are notoriously inaccurate.

When it comes to voting, there are many factors which influence voters' preferences, most of which they are often unaware of. While the candidates' positions on the issues (or voters' perceptions, which may be inaccurate, of their positions) are an important factor, there are several more powerful ones.

Voters often explain their votes based on the candidates' track record. Obviously these are important but voters' perceptions of politicians' track records vary greatly depending on their political and ideological views. Some voters think President Bush should have been impeached. Others think history will show him to have been a great president. While perceptions of politicians' record often influence voter preferences, the reverse is also true -- that voter preferences have a huge impact on perceptions of their track records.

So what other factors have a big impact on voting behavior? One is family, friends and people they work with. Most people vote the same way as most of the people they like and socialize with.

A candidate's voice, looks, style and rhetoric are all enormously important. Franklin Roosevelt, the only president to be elected four times, was the perfect candidate -- good looking, with a beautiful voice, a commanding presence and a wonderful way with words. But if you had asked people why they voted for him they would probably have referred to his policies or his track record and what they thought he would do.

One of the reasons Ronald Reagan was so successful as a politician was that he, and his pollster Dick Wirthlin, understood that "values" were more important than "issues." Reagan mastered the art of persuading people he shared their values, so that many people who did not support his positions on some key issues voted for him anyway. In addition he, and his aide Michael Deaver were masters of the photo-op, casting Reagan in great settings that made him look very presidential. But voters would not tell you this influenced their votes

There are many other favors that influence candidate preferences and voting behavior that are rarely mentioned by voters. Political advertising has a huge impact but few people believe, or tell surveyors, that they are influenced by advertising. The media voters are exposed to matter a lot. Those who read the editorials in the Wall Street Journal or The New York Times have their opinions shaped, or reinforced, by what they read. And those who watch Fox News get a very different world view than those who watch other television stations. As dictators and media moguls know well, the media's ability to shape public attitudes is very powerful.

So next time a voter tells you his vote is determined by the candidates' positions on the issues, treat this with a large dose of salt.