The midterm elections are just weeks away, and much of Washington is abuzz about the possible outcomes. Will Democrats win control of the House, or will Republicans maintain their grip on the legislative branch? Prognostications aside, we can’t know the results until all the votes are tallied. But American University professor Jan Leighley can provide perspective about recent midterms, and she’s closely studied turnout and voting patterns in elections from 1972-2012.

In an interview with University Communications and Marketing—which has been condensed and edited for clarity—Leighley discussed the turnout age gap, a potential “Kavanaugh effect,” and who typically votes in congressional midterms.

Leighley, a professor in the School of Public Affairs, is coauthor of the book Who Votes Now? Demographics, Issues, Inequality, and Turnout in the United States.

UCM: There seems to be a belief that Republican voters are more likely to turn out during midterm elections. Is that true?

Leighley: Well, it’s not exactly correct. I think it’s an inference that’s drawn after looking at who shows up at the polls, and some of that is based on exit polls. But a lot of it is also based on the demographics of who votes. And in the past, we’ve had higher-education, higher-income individuals turn out at higher rates in midterms than lower-education, low-income individuals. And that has tended to favor Republicans. But, that said, the strongest evidence we have for a pattern is that the president’s party is the one that tends to not show up in midterms. If there is a Republican president in office, then it’s the Republicans who tend to stay home. Both of those things can be going on simultaneously, so it’s hard to know if it’s just demographics or a partisan effect.

UCM: You’ve noted that wealthier people are more likely to vote than working-class and poor people. Do we know why that’s the case?

Leighley: There are various explanations. One is that high-income people also have more education, and that means high-income people spend more time learning when to register, where to vote, and just have much more information about the whole process. Some would argue they feel more of a stake in the system. Others would argue that the higher income gives individuals a greater likelihood of being contacted by political organizations. We know there’s a relationship between income and donating to political groups. If you have the income to write out a big check, you might as well go cast your ballot. There’s a general theory that this is a rational, cost-benefit decision.

UCM: There’s also an age gap in turnout, which is relevant for AU students. Why do you think young people have been less likely to vote, especially in midterms?

Leighley: It’s all the reasons that we have explaining who votes. Youth are in some senses at a disadvantage there. They typically have less information than older adults. They are more mobile. One of the things we know is people who move around a lot have more pressing things to do on day one—like unpacking boxes—than registering to vote. In terms of the skills to know how to vote, it’s incredibly complicated for youth to figure out. Where is their legal residence? Each state is different in terms of what students have to do, and that means it takes extra time and extra information. And they might wake up a week before the election when all the campaign activities are in high gear and realize, ‘Hey, I want to vote.’ Not an unrealistic reaction or thought. But if the voter registration deadline was 30 days before the election, it’s simply too late.

UCM: In terms of midterm voters, ideologically and demographically, how are they different from the general population?

Leighley: The same biases we see in presidential elections exist. Compared to the population as a whole, midterm voters are wealthier, they have more education, and they’re older. So they’re not representative of the population. They never have been. The biggest change in representation has actually been with age. Older people are increasingly voting at older ages, and youth turnout continues to be really spotty. Interestingly, the argument is always made that midterms are less representative than presidential elections. That’s not quite true. Are the turnout levels different? Yes. But if you look at voters with higher income and education in midterms, and you compare them to those casting ballots in presidential elections, their share of the vote is actually quite similar. But who really loses are youth in midterm elections.

UCM: So, with young people’s share of the vote, there’s a definite decrease in participation between presidential and midterm elections?

Leighley: Yes. And in part because they don’t have all the other experience, or social connections, or community commitments, that older people have. So if anyone’s going to drop out at a higher rate than others, it’s going to be youth.

UCM: In your book, you show that older people make up a larger share of the vote than they did in 1972. Is that related to the graying of America?

Leighley: I don’t know. People are living longer, they’re healthier, working longer, and maybe more engaged. There’s also this organization called AARP that is inherently political, and it seeks to inform and mobilize its members. And older people certainly report a greater vulnerability or reliance on government benefits and programs with Medicare and Social Security. So there are all sorts of reasons why they would participate more.

UCM: What kinds of issues animate midterm voters? If you had to paint a broad brush, are they generally cultural issues or economic issues?

Leighley: In the past, social issues or cultural issues have not had a huge effect on turnout in midterm elections. Now that’s partly because parties go back to people they know are going to vote, and they try to mobilize them to vote. A lot of those decisions reflect a dominant importance of economics in vote choice. Another point is that we have had an increased nationalization in politics. This isn’t new with Trump. Tip O’Neill, speaker of the House of Representatives, had the famous phrase ‘all politics is local,’ which may have been the case 20, 30, 40 years ago. But one of the big changes that’s happened is the decline of local newspapers and local news sources. And I know the internet affords us a lot of ways we can search for information, but that’s very different than having news or community events in a central news source. And we know that decline simply leads to less information about local politics today, and we’re less likely to participate. Now, there is a footnote to that: We do have some evidence at the state level that when there are initiatives—like gay marriage—on statewide ballots, often times you’ll see a spike in turnout in that election. Why? Because parties or interest groups are spending money. Telling people, ‘If you care about this issue, go and vote.’ And then you have Donald Trump, who has handled his media relations and public information in a new and different way. And, of course, he’s crossed some people’s lines in terms of what’s appropriate to say or believe in politics. So it’s led to this high level of antagonism and conflict. People might not like hearing that stuff, but it’s sort of like a football game. You might be more likely to sit through cold football games if you love those Patriots, right? Or if you hate those Patriots, you’re going to show up and root against them. This is the general take on why the Democrats expect high turnout, because the reaction against him has been so strong. And they’re investing in organizations and communications systems to reach out to a broader group of individuals.

UCM: There’s been a lot of talk about the ‘Brett Kavanaugh factor.’ Pundits are speculating that the Supreme Court fight will energize evangelical voters to come to the polls for Republicans. Conversely, it could possibly bring out more women voters to the polls for Democrats. Do you think the confirmation battle will have any measurable impact on turnout?

Leighley: I think the Kavanaugh hearing, along with other discussions or actions taken by Congress, will likely mobilize more women into politics. One of the principles we know from political psychology is that threat tends to mobilize more than accomplishment. So I think some women feel threatened by various things, and may feel threatened or object to how the hearing was conducted. If you are an extreme partisan, on the Republican side, you probably always vote anyway. The interesting action will be for independents and weak Republicans. So, will weak Republicans, whether male or female, think about voting for another candidate? Will they think about voting for a Democrat instead of a Republican? Or maybe they’ll just stay home, because it’s just confusing or unpleasant or not a game they want to play.

UCM: Could the GOP benefit from a good economy? Doesn’t the party in power usually get credit when there’s low unemployment?

Leighley: Yes, absolutely. Yet this goes back to the nationalization of midterms. We have some pretty good evidence that in the midterms in the 2000s, people who disapproved of George W. Bush’s handling of the Iraq war, even if they were Republican, voted Democratic. And that’s part of nationalization. Bush was a very polarizing president and had a very high disapproval level. So that’s some of the information people are taking in—whether I like or dislike the current president, whether I want to elect another person in the same party.

UCM: Do you think either party has an advantage with organizing and getting out the vote? Or has that varied from election to election?

Leighley: There’s a huge political consulting industry now that takes up a lot of campaign contributions. And they pretty much know how to do this stuff. Along with the consultants, parties are now finally investing more and more money to collect data. Technological advances allow parties to track people. I think this is where the Democrats have done a little better at mobilizing new voters. The belief is that those low-income, lower-education people need more support, more information, more contacts, more touches. Maybe it’s the difference between one phone call or 10 texts, or maybe it’s the difference of people going door-to-door and meeting. That’s what we know is the most powerful thing. That’s a person-power issue. And so both parties have to allocate bodies where they think they can make a difference. That’s a strategic decision, and those are hard calls to make.