Even the country’s most politically savvy pollsters shrug their shoulders when pressed to predict the outcome of the presidential election. Absolutely no one really knows for sure what is going to happen next month, said Nicholas Clark, Ph.D., Department Head of Political Science and Director of the Innovation Center and Director of Public Policy at Susquehanna University. Clark spoke to the community Oct. 10 about voting behavior, polls and election forecasts.
In the lead up to election day, both political parties have been scrambling to win supporters, undecided voters, independents and new voters in crucial swing states. With the election this close and the clock ticking, Clark posed a pertinent question, “Why do people vote?”
Research dating as far back as the 1950s found that a lot of Americans, in fact, choose not to vote for various reasons: They don’t have an interest in politics, don’t have a strong incentive to vote, or don’t feel a sense of civic duty.
So, who does vote? Individuals over age 65 are most likely to participate while 18 to 25-year-olds are the least likely. Older, more affluent and educated adults who have a sense of civic duty and self-efficacy — the belief that their vote matters and that participating in the process matters — are more motivated.
In 2020, Clark worked with another associate professor at Susquehanna to identify an elusive subset of “voter” — the nonvoter. Using data from 7,365 respondents to the Cooperative Congressional Election Study and the Electoral Integrity Project, they identified four potential types of nonvoters. The data shows that about 40% of nonvoters may be interested in voting but are incapable of getting to the polls for various reasons such as jobs or travel; 25% are conditional, they express an interest in the political process and voted in past elections but actively choose not to participate in the most recent elections due to dissatisfaction with or ignorance about the candidates; 15% are apathetic, lacking any interest in the politics; and, 6% are obstructed from voting. Accessibility of the electoral process (such as overly long lines or intimidation at the polls) effectively prevents them from casting a ballot.
Of those Americans who vote consistently, the majority cast their ballots according to their party affiliation. And, in this close presidential election, partisanship is the greatest single predictor, said Clark. Democratic and Republican strategists know their candidate’s success depends on convincing as many of their supporters as possible to vote for their candidate. But hot-button issues, including the economy, abortion rights, gun control and the environment can weigh heavily on how people end up voting.
And public opinion polling is more complicated than simply asking people who they’ll vote for. Polls only tell us how the public would vote on any given day. The single strongest predictor is whether the people participating in polls have voted before and are likely to vote again, Clark said. For that reason, some polls might not include a pool of new voters, or those voters who plan on voting but whose decision might be affected by an incumbent’s popularity or disapproval rating. Additionally, people are becoming less interested in participating in polls; the response rate is between 1 and 3% currently, while it used to be about 30%.
“In this election where the candidates are this competitive with each other, polls can get us close but not close enough to know for sure what is going to happen,” Clark said.
Political forecasts predict future outcomes, a month or sometimes a year out and can be more reliable, falling within a slim margin of error. Yet even some of the top forecasters, whose previous election predictions have been spot on, say the race is just too close to call.
“Nobody really knows what is going to happen in this election and anybody that tries to tell you is trying to sell you something,” said Clark. “The number of variables in this election is just staggering and no one is willing to say for sure what they think is going to happen. I usually bet on every presidential election, and I win. But I am not betting on this one!”