Riding the Bandwagon Effect

Do you ever get the feeling that the only reason we have elections is to find out if the polls were right? Robert Orben

During election season and even in the months before an election is even called, we are besieged by poll after poll weighing in on the relative popularity of various candidates, issues, or political parties. Though opinion polls date back to the 19th century, the rise of professional polling organizations, including the venerable Gallup poll, has made polling a crucial part of the election process in countries around the world.

Despite the reputation for objectivity surrounding opinion polls though, there have been some legendary failures due to problems with biased questioning and how "random" some polls really are. Though most polls, including the Gallup poll, predicted Harry S Truman's defeat in the 1948 presidential election, he won instead. To their embarrassment, some newspapers relied on poll results to print "Dewey Defeats Truman" headlines in their papers for the morning after the election (these papers have since become collector's items).

Perhaps more disturbingly, there is growing evidence that polls can also be used to shape public opinion. If opinion polls already state that a certain candidate is going to win, how likely are you to cast an opposing vote? Or for that matter, to vote at all? Public apathy is becoming a serious issue in many elections in which less than 50 percent of registered voters even bother to go to the polls. This means that polls can have a far greater impact than most people realize. Potential voters seeing their chosen candidate about to lose might well be dissuaded from voting due to the feeling that their vote won't matter anyway.

Then there is what has become known as the bandwagon effect. Researchers have long identified the impact of social conformity in shaping how people think and act. Along with explaining new trends in fashion or popular fads, this bandwagon effect can also influence how people would be likely to vote on important issues. Many voters often prefer not to make an informed choice before voting and simply choose to mimic the behaviour of other voters instead. If a poll predicts that a certain candidate will win by a landslide, could voters actually be persuaded to vote for this candidate themselves?

According to the theory of social impact proposed by psychologist Bibb Latane, attitudes and behaviour are influenced by how we view other people. If a candidate is seen as having majority support, he or she is viewed much more positively and is more likely to get the vote of uncommitted voters. Nobody likes to vote for the losing side, after all. Research studies already show that candidates leading in the polls are more likely to benefit from a bandwagon effect though how strong that effect is remains controversial.

A new research study published in the Journal of Media Psychology takes a comprehensive look at the bandwagon effect and how influential polls really are. A team of researchers led by Magdalena Obermaier of the University of Munich conducted an online experiment with 765 participants (56 percent were female, average age of 35).

In the experiment, the participants were told that they were taking part in a study of news coverage before political elections. They all read a news article about a fictitious mayoral election in a small German town. The participants were then presented information on the two candidates (both invented) including their history as politicians and whether they had won or lost previous elections.

The next part of the experiment involved assigning the participants to one of three experimental conditions: the first condition being that polls showed that one candidate was lagging behind the competition by a wide margin, the second being that the candidate was leading by a wide margin, or else a no-poll condition. Finally, each participant was asked to rate whether how they would personally vote in the election, which candidate they felt would win in the election, and their own estimate of that candidate's competence.

What the study results showed was that polling information has a powerful influence on whether or not participants expected a candidate to win. If no poll information was available, then they formed their opinion based on whether the candidate had won a similar election in the past. Still, while polls seem to shape how people believed an election would turn out, it didn't seem to have any impact on how competent they felt a candidate was.

Conformity Essential Reads

Dress Down for Success: When Nonconformity Signals Status

You Are a Conformist (That Is, You Are Human)

Using statistical modeling, Obermaier and her colleagues found that voters draw on different sources of information to decide whether to support a candidate. Along with opinion polls that indicate whether a candidate has majority support, voters also look at how the candidate did in previous elections.

While most voters prefer to weigh the election issues, including which political party a candidate endorses, opinion polls can have a powerful influence on undecided voters who aren't following the election issues that closely. Considering that election outcomes can often turn on a relatively small number of votes, this raises disturbing questions about the democratic process as it is currently practiced in many countries.

Still, while this research provides some evidence for a bandwagon effect, the researchers still had difficulty comparing these results to what happens in real-life elections. Voters often form real-world opinions about candidates based solely on what little information they can gleam from a morning newspaper, especially if they are feeling apathy over which candidate will actually win. For those who feel that "all candidates are the same", relying on polls or previous election results to decide who to vote for may relieve them of the responsibility to examine the election issues carefully.

In the months leading up to the 2016 U.S. Presidential election, opinion polls will likely be more important than ever. We are already seeing an overabundance of poll results comparing various Democratic and Republican candidates, many of whom are incorporating these numbers into their own campaigns. Will polls transform the election process? Only time can tell.