Editor’s note: The polls prior to Election Day, as well as the circumstances in the country during weeks of voting, suggested to many that the presidential results would be different than they were. But predicting the turnout and vote in closely fought states is now notoriously difficult, says Scott Althaus, a professor of both political science and communication at the University of Illinois at Urbana-Champaign. And the role of the pandemic may have been different than it seemed. Althaus spoke with News Bureau social sciences editor Craig Chamberlain.
Pollsters have faced a barrage of criticism since the election. But setting aside the technical challenges in polling, what makes predicting the outcome at the state level so difficult?
Part of it starts with the changing nature of campaign tactics for influencing voters. Because we’re so polarized politically right now, there is little opportunity for one party to persuade those on the other side when it comes to high-profile races like the presidency. The better opportunity to really move the numbers on Election Day is to mobilize likely supporters who are not certain to vote.
So when the two parties go at those mobilization strategies differently from one another, it means that there’s uncertainty about how much mobilization effort is being mustered by campaigns in a particular state. That uncertainty may influence the accuracy of polls, which are based on assumptions about who is going to show up to vote, typically based on past voting records.
But the whole strategy for mobilizing likely supporters is to reshape past voting to change those probabilities. When those tactics succeed at bringing voters to the polls, it becomes more difficult at the state level to assess what the partisan mix is going to be on Election Day because these differential strategies are operating under the radar screen. We’re not going to see how effective they are until after the vote.
Are there other factors that might contribute to the gap between polling results and the Election Day vote?
The highest-quality polls for predicting election outcomes are also the most expensive. Since much state-level polling has been done typically by newspapers, and the news business is suffering economically, it is harder to come by high-quality polls that have the kind of information we might have expected in previous presidential election cycles. Those polls are also less frequent, making it harder to spot changes late in the campaign. This election was also complicated by the expansion of early voting options in many states, creating some uncertainty about whether people who ordinarily would stay home on Election Day might end up voting this time.
The Biden campaign clearly assumed that the pandemic, and President Trump’s handling of it, was an issue in their favor. But you point out that Trump and his campaign may have benefitted by the circumstances around it. How so?
Presidential campaigns generally try to define a policy contrast between two candidates, and we clearly had that to some extent. But what we didn’t have in play, because the pandemic was the dominant issue that kept changing and refocusing the attention of both campaigns, was an opportunity for candidates to articulate a broad governing vision of what the next four years would look like if they won.
This unusual aspect of the campaign played into the Trump administration’s preferred governing style, which is not based on setting out a broad governing vision but to respond in the moment, as crises erupt, and being adaptive as new situations arise. The Trump campaign has been very adept at responding to sudden curveballs thrown at it, and the pandemic provided those in abundance.
The pandemic also severely limited the most effective tool in the campaign arsenal for mobilizing supporters, which is door-to-door campaigning. It’s very clear from lots of social science evidence that in-person contacts make the biggest difference in turning people out to vote. It’s not who you register, it’s not the robo calls, it’s not direct mail, it’s somebody showing up on your doorstep and motivating you to commit to voting. The Trump campaign continued to do a fair amount of that; the Biden campaign did not.
You noted the part that polarization plays in campaigns and polling, and a lot has been said, post-election, about a stark divide in American society. But how unusual is this?
It’s the nature of politics to divide people into in-groups and out-groups. What’s different now is that the emotional attachments to those groups have become more intense, so there’s animosity based not just on disagreement about ideas, but on conclusions about the identity and character of those on the other side.
But this is not an unusual situation in the history of American politics. If we go back to the Founding Fathers, the ideological camps were quite starkly drawn. And all the way through the late 1800s, the personal animosity that separated Republicans and Democrats could be quite extreme. The degree of polarization decreased for a number of reasons in the mid-20th century, but what we’re now seeing is a swing of the pendulum back to the way things used to be.
But why now?
Clearly the media system that we have in place today is a factor. It’s easier to reach people with a diverse array of messages more rapidly than ever before. The rise of social media means that information flows about politics are no longer as tightly managed by either news organizations or the political parties. Another important factor is the tendency for political leaders to leverage anger, animosity and outrage to motivate short-term gains in elections and fundraising. When this happens cumulatively in election after election, this can create a deeper entrenchment.
What can we do about it?
It’s important for us is to listen for the ideas that we hear articulated by the other side, and especially the “why” behind the ideas, instead of assuming we already understand why somebody is thinking differently about politics than we do. We have a lot to learn from one another in a season where we’re so politically divided, and we may have more in common across lines of political difference than we realize. Coming together and better understanding what separates us is going to be crucial for repairing the damage in our political system, and hopefully for bringing us to a better season of political life in the United States.