What early voting data can and cannot tell us

A version of this story appeared in CNN’s What Matters newsletter.To get it in your inbox, sign up for free Gentlemen.


When Democrats won the House of Representatives in 2018, they did so with the help of a surge in turnout, the highest turnout for a midterm election in more than 100 years.

Still, half of the voting-eligible population did not participate.

This year, early voting in some key states, but when I spoke with Michael McDonald, a University of Florida political scientist known for tracking early voting data, he predicted turnout would be lower than 2018 levels.

MacDonald has a new book dissecting the huge achievement of the 2020 presidential election, when nearly 67 percent of eligible voters cast their ballots. There is more information about the book and his early voting tracker on his American Elections Project website.

We discussed what people should take away from the last election and what he saw when tracking early voting data for the current election.

Below is an abridged version of our longer phone conversation.

Wolf: You wrote a book about this amazing democratic achievement of voting during a pandemic. What do you hope people take away from this research?

McDonald’s: We must give a lot of credit to election officials, the volunteers working at the polls, and the voters themselves, who participated in the presidential election with the highest turnout since 1900.

In the 2020 election, no one voted in our last high turnout election. This is indeed a remarkable achievement. In exceptional circumstances, we managed to do something historic. This is very positive news.

Another takeaway from the book, unfortunately, is the relentless attack on voting that took place during the election, from rhetoric with (former President Donald) Trump, then just filter through this party.This destroys democracy, we can see happens in real time 2022 election.

Wolf: You mentioned the highest turnout in 2020 in 100 years. I read in the book that the 2018 midterms had the highest turnout since 1914. We are seeing more and more people questioning the integrity of elections, but they are also voting more. What do you think?

McDonald’s: The last time we had unusually high turnout was in the late 1800s, and that period also saw strong polarization. We don’t have any survey data, so we can’t go back and ask if voters are polarized, but we can guess that what happens among our elected federal officials mirrors what happens among voters.

So we’re entering a period of high polarization, and you can pinpoint the culprit behind that. But whatever the reason, we’ve definitely gotten to a point where people really believe it matters who is running the government, and it really matters that their side is the side that is running the government.

People are more likely to vote when they realize the differences between parties and the importance of political differences to their lives.

It’s that old curse: may you live in interesting times. We live in interesting times. People are very interested in politics, so they are highly involved in elections.

Wolf: exist 1880s, America Have Close to 80% turnout. You could argue that higher turnout is in some ways a wake-up call for democracy.

McDonald’s: You would want people to get involved for altruistic reasons, they want to be good citizens, they are carefully weighing their options, and rationally deciding who they are going to vote for.

Some people are looking back at political science reports done in the 1950s and lamenting that there is no difference between parties, we are falling apart, and unless we fix our problems, America is going to democratize those parties, those parties.

Look, you have to be careful with your wishes, because the political parties in the electorate are stronger than ever in modern times, and now people are thinking, well, maybe this is going too far.

What is a happy medium for engaging voters, but not so partisan that in some cases they want to act violently because they think politics are so important?

Wolf: You are known for tracking early voting data. What can it actually tell us before Election Day?

McDonald’s: I first started tracking early voting in the 2008 exit polls. They want to know the size of early voting so they can properly weight their polls.

Like a lark, I posted it online. Then a website I created got a million hits and I knew I was doing something different and special in some way. If you look at a lot of data journalism happening today, it’s more like what I do, which is take some administrative data and tell a story in some way.

To answer the question about where we are in the early voting… what you want to do is you want to gather all the information that can be woven together and try to understand where we are. So I don’t think early voting alone will tell you anything, just like I don’t think polls alone will tell you exactly what’s going on in the election.

The polls are wrong. Early voting has its nuances and measurement issues.

Wolf: What did you see in early voting?

McDonald’s: It’s not just because they get votes, or they get a chance to vote. They actually have to vote, and we’re definitely seeing a lot of interest in voting, especially in those very high-profile high-level elections that are going on in the U.S. Senate or certain gubernatorial races. These appear to be attracting voters.

What we’re seeing in these states is a high level of early voting. We are seeing a lot of democratic participation.

What we usually see in midterm elections is that there is some kind of punishment for the party that holds the presidency. Whatever the reason, people will find reasons to be angry and engaged because of what the government does.

But in these games, we are not See a kind of referendum on the Biden presidency.In fact, look at the polls: People who highly disapprove of (President Joe) Biden still says they will vote for the Democratic nominee. What’s happening here is that the election has moved into a choice between candidates, not a referendum on Biden.

If you look elsewhere in the country, we’re not seeing the same level of engagement. In the absence of this participation, the election is more of a referendum on Biden, and this is where we can see divided results, as many polls have shown.

If Democrats do lose the House, it may be at least in part because their voters haven’t found a reason to vote in a state like California.

That’s the challenge for Democrats if we vote early last week. How do you motivate your voter turnout to the same level as the Republicans, where you don’t have this high-profile marquee contest to drive people to the polls?

Wolf: can us Now suppose some of the concerns about restrictive new voting laws are unfounded due to high turnout in some states and the fact that many people use early voting?

McDonald’s: I’ll give you a silly and totally ridiculous response. But it makes sense. Do you know what this election is about? I saw massive voter suppression in this election.

As I look back at the 2020 presidential election, turnout is down across the board in every state. There was massive voter suppression in this election.

Of course, you think it’s ridiculous. It’s ridiculous because people turn out more in presidential elections than in midterm elections.

Just because you have a fun race to get people to vote in a state like Georgia doesn’t mean SB 202, which is a law passed in Georgia, somehow makes it easier for everyone to vote in the state. That doesn’t mean some communities aren’t being left behind.

A good example of this, if you look at Georgia, while we’ve seen record numbers of in-person early voting, we’ve seen mail-in ballots drop by about half. You might say, well, that’s okay.People who would have voted by mail, they just had to vote in person or they would Vote on Election Day or earlier.

There may be people who, for whatever reason, are quarantined at home, unable to go to polling places, and they have to vote by mail. And for those people, they may not be able to participate as well as others in Georgia.

I wouldn’t say that just because Georgia has a high early voting rate means that the law isn’t disincentive for any particular community in Georgia.

Wolf: Another new storyline for this election is Florida’s turn to Republicans, which was driven by Hispanics and Latinos vote Republican. Was there anything for or against in the early voting? Do you agree with this larger narrative?

McDonald’s: We really can’t answer this question with the data we have, because we don’t know how people vote.

By and large, early voting in a typical election is usually won by Democrats, or at least registered Democrats. In this election cycle, it is the Republicans who have won the early vote.

So far, as of Nov. 2, registered Republicans have a nearly 180,000-vote advantage in both mail-in ballots and in-person early voting, with most of that advantage actually coming from in-person voting.

However, all of these Democrats have mail-in ballots. Strange thing: they don’t return them. Not to the same extent or rate as Republicans.

So if you look at the return rate, as of Nov. 2, 48% of Democrats have returned their mail-in ballots, compared to 55% of Republicans. So these people have a mail-in ballot in their hands, and you’re going to see a big difference in these return rates.

Part of what’s happening in Florida is a self-fulfilling prophecy that those who don’t believe Democrats can win won’t vote. And because they didn’t vote, Democrats couldn’t win.

Wolf: Will the 2022 turnout surpass the 2018 midterms?

McDonald’s: We will see high turnout. Georgia could surpass their 2018 turnout. Some other states like Pennsylvania may.

But there’s a good chance we won’t see the same level of engagement in some of the larger states like California, New York, and Texas. And since the majority of the population lives in these large states, and they don’t have competitive elections that pull turnout, we’re likely to see some imbalance.

I don’t think it will go all the way back to 2014. That was the lowest turnout since 1942. But I wouldn’t be surprised if we were lower than 2018.

Source link