A Democracy, If You Can Keep It

In my mid-20s, I moved from Mesa, Arizona to Tempe, Arizona. On election night, I walked to the polling place I’d seen closest to my new house, since I hadn’t received anything in the mail about the election. When I arrived, I learned they didn’t have my name on the voter rolls. Odd, but not that odd when I’d just moved 2 months before. Everyone agreed that it was likely my updated info just hadn’t come through.

The poll workers let me vote with a provisional ballot, and I left. This was in the days before smart phones, so I cracked open my laptop when I got home to double-check the location. Turns out, that wasn’t my polling place at all. A different location, a mile further away, was where I was supposed to vote. I hopped in my car and made it 5 minutes before the polls closed…where I had to vote with yet another provisional ballot.

While no one implied that I was trying to vote fraudulently then, 18 years later we live in a wildly transformed world. With so much riding on the upcoming presidential election, it's more important now than ever that we know we can trust our voting process.

I’m your host Jess Phoenix, and this is science.

Jess: The Union of Concerned Scientists has just released a new report as part of our Precinct Analysis Project. With me today is Dr. Liza Gordon-Rogers, research associate here at UCS and one of the authors for this critical study. Liza, nothing could be more timely right now than for us to talk about some of the lesser-known aspects of our democratic process. Who gets to exercise their right to vote? What are the factors that either hinder or help voters in getting their ballots in and counted? So to start, why don't you give us some background on why this type of detailed examination of one of our most fundamental rights is so important right now?

Dr. Gordon-Rogers: Yeah, you're exactly right. Now it's more important than ever. So one of the really unique things about this study is the level of specificity, right? So we're looking at precinct-level data, not just county level and not just state or national level data. And I think that's really important because elections are local. We really want to know what's happening in elections. We need to get to that granular level to understand people's real experiences when they go to vote.

Another really important facet of this study was that we're not just looking at turnout, but we do look at that. We're looking at whose votes are actually counted when all is said and done, right? So you can go through the whole process of either voting by mail or absentee or going in in person. And what matters is after you cast that ballot, is it actually counted in the election? Do your preferences actually matter at all? Does your voice actually get heard in elections?

Jess: I know that's a question that I think that a lot of young people, particularly first-time voters or potential first-time voters have. It's a question that I would like to be able to answer as a parent as well as a scientist. So how fundamental to our democracy for people both within and outside of the U.S. is it for us to have confidence in the integrity of our electoral process?

Dr. Gordon-Rogers: So trust in the process is essential, right? People are less likely to participate if they don't think that the process is fair, if they don't think it has integrity. That's kind of one of the major reasons one of our big buckets of work is greater election data transparency. The idea is that if you're more transparent with the election process, right, how people cast ballots, the process of counting them, everything that goes along with elections that people will trust the system more because they see that it's well-meaning election administrators that work really hard at their jobs, that if you're transparent about the numbers, there's no room for people to be like, "Well, we don't know what's going on. There's a whole bunch of fraud," right? That's one of the things we don't find in this report, is evidence of mass voter fraud.

And many studies on voter fraud have failed to find any evidence of widespread voter fraud. But if we just become more transparent and release more election data, I think it just makes people trust elections more and more. It's interesting that you mentioned young, inexperienced voters because young and first-time voters and inexperienced voters are more likely than other voters to have their ballots rejected because of mistakes, right? So it's really critical because they don't know a lot about the process, right? So they get their absentee ballot in the mail. They try their best to follow the instructions and do everything the right way. But if they forget to put the date on, if they don't use a secrecy envelope in some places in Pennsylvania, right, if they don't do that, their ballot can sometimes not be counted.

There's also a lot of studies have found that historically marginalized communities and people of color are also more likely to get their ballots rejected. And we see that in this analysis that precincts with higher numbers of minority voters do have higher ballot rejections, right? So we see that here in our study as well.

Jess: Yeah. I'm definitely going to want to dig into that in a little bit. I'm glad you mentioned it because it's, I think, probably the most striking thing that I took from the report. But I wanted to highlight what you said, which is that there's no evidence of widespread voter fraud. Louder for the people in the back.

So no widespread voter fraud, but are there people and entities working to limit voting access in a targeted fashion? And if so, what evidence do we have of those efforts?

Dr. Gordon-Rogers: With elections, are always going to be bad actors, for political reasons, trying to limit people's voting rights. We do see evidence of that in a lot of different states right now. There are kind of certain tactics that they tend to rely on. One is voter challenges and mass voter purges, right? So we even see some advocacy groups using training their volunteers on how to use AI.

It's not really AI, but they call it EagleAI, they call it. It's basically a program that they use to troll state voter files for people they think are suspect and then challenge those registrations. And those people can get purged or removed from voter lists, right? And sometimes that can happen without somebody's knowledge for a lot of different reasons. And if that happens, then someone can go to their polling place on election day, only thinking that they're registered to vote and finding out that they're not on the list and not being able to vote.

Sometimes they can fill out a provisional ballot, but those aren't always counted, right? If there's a problem, they're not on the voter file on the registration list and they fill out a provisional, that won't get counted, right, if they're not registered to vote. You hear a lot about purging, but that is just essentially removing people's names from voter files. You have to be on a voter file to vote.

Jess: So how do we know about the EagleAI use and what do we have that we can point at and say, "Ah, there's the smoking gun that this is happening?"

Dr. Gordon-Rogers: It's actually these organizations are pretty vocal about the fact that they're using these kind of tactics. They're very proud of the fact that they're harnessing the power of technology to help them purge people. From their point of view, they think that they're helping elections be more secure, when in actuality they're not. And some of that might be a political front. They know they're not really doing anything to make elections more secure, but they're targeting certain people, right? So there is evidence that these organizations do target voters in communities where you have higher populations of people of color. And so there's most likely a political stratagem buried in there, right? You have that they assume that those people are more likely to vote for certain candidates and therefore don't want them to be on voter rolls so they won't be able to vote, which will influence the results of elections.

Jess: What do you see as the biggest vulnerabilities in our current voting system?

Dr. Gordon-Rogers: Oh, my goodness, that's a big question.

Jess: Yeah, we like the big questions here.

Dr. Gordon-Rogers: So there's a lot. My mind is going towards our work on election data transparency because it kind of...it's more than that. It's more than just releasing the data. But part of that is that we think election administrators should, and a lot of them do want to, just a lack of resources or staff or capacity or the infrastructure to do these things, but to frequently release the names of voters who are going to be purged from voter lists and actively work with community organizations to try to do outreach to those people to warn them that they're on the list of having their name removed. So either they fix whatever error, right, if it's administrative error, their name is wrong or anything like that, or re-register to get their name re-on voter rolls so they can vote to try to limit the amount of people who go on election day and are kind of out of luck, right, because they're not going to have their vote counted.

There's also ballot curing, which is a fancy way of saying a lot of states allow voters to fix mistakes on their absentee and mail-in ballots, which I think is so critical, especially as that's becoming more popular. Especially it was super popular in 2020, right, with the pandemic, and remains pretty popular to this day, especially for certain communities. So some states don't have a statewide curing policy. I could go down the rabbit hole of Pennsylvania, which is where I'm from, which does not have curing in its state election code. So every year, it's just chaos, every election year of court cases trying to decide should election officials allow people to fix errors, and if they do, what errors count as being allowed to be fixed and what aren't.

But I think it's so important that states allow voters to cure their ballots, fix their mistakes, have really expansive lists of what those mistakes entail, and allow them plenty of time and a lot of ways to cure their ballot and reach out to them a lot of different ways. So North Carolina is actually a pretty good model for ballot curing, interestingly enough. I will say that a lot of the improvements that they've made since 2020 have come out of court cases dealing with their different election policies, but ultimately what they have right now and what they did in 2020 was great in terms of they really did a lot of outreach to people who needed to cure their ballots. They texted them if they had their phone number and their voter file. They emailed them. They sent mailers. They allowed various ways for people to correct their ballots. They work with community organizations to get the word out. So I think that's incredible. So I hope that more and more states do those kind of things.

Jess: That's actually fascinating that the courts are propelling some of this, and North Carolina is setting an example here. I think that's great that you can text people and say, "Hey, you've got an issue. You need to fix it." But I want to dig into the report a little bit now. So from this report's data, who do the data indicate are the people who are the victims of disenfranchisement?

Dr. Gordon-Rogers: Yeah. So, unfortunately, it was what we expected. We being the research team who worked on this, right? Part of what we wanted to make sure was we have national studies that find that people of color are disenfranchised at our election system, we have state-level studies that find that, we have county level. So we assume that we were going to see the same thing at the precinct level, and we do. And the states and counties we look at, we find that precincts that were majority Black, majority Hispanic had lower rates of turnout than majority-white precincts, which is unfortunate but it is what we find. And it is significant even for controlling for county and state-level fixed effects. So that's just a fancy way of saying it's not just characteristics that are unique to the state or county, right? It's not just election law. It's not just income. We try to control for all those things, and even after we do, we still see these stark differences.

Jess: Wow. So bottom line, these are the trends that you thought would be true going in. And unfortunately, the data proved you right.

Dr. Gordon-Rogers: Yes, unfortunately.

Jess: Okay. Can you explain if there are...? I mean, obviously, we just spoke a little bit about efforts to remove people from the voter rolls, which is more of like an active way of discouraging voting or preventing voting. But are there passive ways that people lose their ability or their desire to vote that you see reflected in the data? I mean, are there community problems or conditions that make it so people don't vote?

Dr. Gordon-Rogers: I can't say with any level of certainty because we didn't look at this specifically, right? But we know from other science that in places with lower turnout, you're going to have less mobilization efforts by campaigns. It's the idea that they don't want to like "waste" resources in areas where they don't think people are going to vote. So you have those communities and those people kind of left out of that campaign process, right? You don't have political parties going in there trying to educate voters. You don't have them going in there trying to mobilize voters or really listen to their needs and concerns and try to advocate for those needs and concerns so people vote for them. So that's one of the aspects.

Jess: How about voting desire? Some people, I think, and some segments of the population are very motivated to vote. And sometimes it's passed down from your parents or, you know, your community. And sometimes it's, I guess, maybe from the education system. But do you see that people lose the desire to vote? Like, say, if they get purged from voter rolls, has there been anything, any inquiry into that in a scientific way, whether it was in this report or in other work that you're familiar with?

Dr. Gordon-Rogers: Right. So we don't really get into anything like that in the report. But I will say very like, broadly, a person's socioeconomic status is the most predictive variable in determining if they're going to vote, right? So people with higher socioeconomic statuses are more likely to vote and people with lower socioeconomic statuses are less likely to vote. Across the board, that has been like a definitive finding in kind of science research.

Jess: That is actually scary because, you know, it's supposed to be the great equalizer. And if money is what is determining who is voting, then we have problems. I think we know we have problems, but…

Dr. Gordon-Rogers: Right. Because voting isn't a free activity, right? If you think about it, there's this whole resource model in political science where we think about each way that we participate in government takes resources, right? And people think voting is easy. And comparatively, to other ways of political participation, it is maybe easier, but it's not easy in any way, right? You have to be registered to vote. You have to make sure your name is still on there. You have to find your polling place. You have to go to that polling place during the hours that they say. You might have to take off work. If you're paid by the hour, that might be really difficult. If you have kids, you have to find childcare. If you don't drive, you have to find public transportation and pay for that public transportation, right?

So it's all these kind of cumulative, additive resources that it takes to vote. And if people don't have those resources, and there's even like resources outside the traditional kind of thought of resources. Do you have the...like you said, do you have a social network that discusses these things, promotes voting, thinks of voting as a responsibility? Right? Do you have those resources in your life to even want to vote in the first place? Right? So if you lack those, all those different kinds of resources, it's difficult.

Jess: It really is. And I've just contrasted. I lived in Australia for a while. I couldn't vote there, but I could see everybody else voting and it was mandatory. Like you had to vote.

Dr. Gordon-Rogers: Yeah, they had mandatory voting.

Jess: And that, I guess, would cut across the socioeconomic barriers for sure. And I mean, I'm sure it's still a burden, but at least it's everybody does it. And you get everybody's voice in there. So can you explain for our listeners who maybe just aren't as familiar with voting issues in general, why seemingly innocuous voting requirements? Like, I mean, you just did why going to a polling place in person can be problematic. But things like showing a picture ID, because I know that's been in the news a lot. So why does showing a picture ID place an unfair burden on eligible voters?

Dr. Gordon-Rogers: The thing with voter ID is that a lot of people look at it and say, "Well, I have a government-issued photo ID," or, "I have a photo ID that would allow me to vote." So they assume that most everyone does. And that's just not the case, right? You have specific segments of society less likely to have an approved ID, right? And it really depends on what kind of...we have strict voter ID law states and non-strict voter ID law states, right? So some states say it has to be a government-issued ID.

If I was in college and I had my student ID with my picture and my name on it, that would not count, right? I would not be able to vote with that ID. I would be turned away, right? So I need a government-issued. And some people just don't have that, right? They may not have gotten it at all in the first place. If you live in a city, you might not need to drive. You never get your driver's license, right? So you might not have ever had the need. You don't have the resources maybe to travel. You don't have a passport. You never left the country. You don't have anything, right? So that would actually add more cost if we're talking about the resource model, right? You have to go out, apply for an ID, again, find the office.

Maybe they're not open every day of the week. Maybe they're not...they're only open when you work. You have to take off of work, right? You have to pay for the application in some states to get it. You have to have all your paperwork in order, your birth certificate, or whatever required paperwork there is, and go through that whole process. And people shouldn't have to do that just to be able to vote.

Jess: Yes.

Dr. Gordon-Rogers: Especially when we see no evidence that people are showing up in large numbers trying to vote as somebody that they're not, right? So it's kind of...and people have said this. It's a fix to a problem that doesn't exist that creates more problems.

Jess: So this report in particular, and you mentioned it at the beginning of our talk, that this looked at precinct-level data. How were the precincts you examined for the study? How were those selected?

Dr. Gordon-Rogers: So that's a really great question for science nerds like me who are all into, "What's the sample like?" So we couldn't, right, with time and getting all the data and examine all 50 states, right? So what we decided to do was focus on the most pivotal states in elections or the last few election cycles, right? So swing states that are really critical to determining who the next president is going to be. And we didn't do every county in those states. So we decided to focus on large populist counties, counties that have been targets to election fraud allegations who have had election problems before, and really focus on those counties.

So, for example, in my home state of Pennsylvania, we had Allegheny County, which is home to Pittsburgh, where I live. And then we also did Philadelphia County. The reason we did counties instead of just cities, right, we could have just looked at the precincts in Pittsburgh, right, doing Allegheny County gets you the suburbs and some rural parts of Pennsylvania, right? So it's a better sample of people and communities than just looking at a city would give you.

Jess: That is actually really strategic. I love it because cities are really diverse places, too. And I live in L.A. and I'm thinking about L.A. County versus L.A. City. And it is the county is vast. It includes these areas of Acton and Palmdale and Lancaster. These are places where some of the lots are minimum five acres. And then you think about L.A., you're thinking about Downtown L.A. or Compton, or... You know, it's so diverse. And I think that's a great idea to be able to capture the true diversity within each county by looking at the precincts. Oh, brilliant. I love it.

Dr. Gordon-Rogers: Yeah. That's one of the really cool parts, if I can geek out for a second. The cool parts of us making the maps that we did, we could have just released the report, which is interesting by itself, right? But the reason we did the maps was one, so it was more accessible for everybody, but also to visualize that, right? You can see what the different colors on the map and turnout rejections, right? You can see patterns. You can see the differences of just what five miles in a city will give you, right?

So, for example, in Wisconsin, I wrote about this in the blog that I did for the release, you have one community that has amongst the lowest turnout in that county, right, for 2020. Just like four or five miles west, you have the precinct with the highest turnout in the county. Like 84%, I think, it was around, compared to the lowest. So you see those things in communities and people who live in these counties or know people in these counties are just...are interested. You can zoom in and see the streets, right? You can locate the communities. So I located my street in the map of Allegheny County in 2020 and I can see that information, right? So I think it's a great way to bring people into our election science and say like, "This is impacting people. Like these are communities with your friends, your family, maybe even yourself." And this is what happened in 2020.

Jess: So our listeners who live in battleground states or states that have had issues with voting processes before definitely need to check out both your blog and the maps. And so I'll link them on the page for the show, which is sciencewithjess.org. And we'll have that with the episode information. So along with our transcript, we will have those links because I think that's great for people to access. And data visualization is half the battle. Data tells a story. But if we don't tell that story effectively, then why do we have the data?

Dr. Gordon-Rogers: It's such a great point. We talk a little bit about that in our Election Data Transparency Report because you have election officials come to us even and say, "We have all this data," but they're just Excel spreadsheets, essentially, right? That wouldn't mean really anything to anyone. How can we translate this data in a public-facing way that's really accessible and makes sense? So you have states, California is a good example. They have a statewide dashboard with maps on it that lets you filter pretty specific election data for people, anyone to see.

So it is kind of a problem that we need to workshop with election officials, which is why part of our work on this is working with the Election Science Task Force, a task force full of community organizers, but also election officials and academics and kind of getting everyone in the room and say, election officials, say, "We have this problem. We don't know how to visualize this data in a way that's good for our voters in our community." And you can have academics and nerds like us say, "Well this is what other states are doing," or, "This is really interesting. You can use this, a system like ArcGIS like we did, look at what we can do with maps," and you can show this in a way that people can understand.

Jess: That is genius because that goes to the end goal of using the data for action. So then what's the clearest message you took from the data you examined? Can you distill it to a sentence or two?

Dr. Gordon-Rogers: I think I keep harping on it but it's true. It really shows the need for a better election data transparency because there were some...we examine 11 counties in seven states for turnout, right? We only examine eight counties across five states without rejections because we just couldn't get that data, right? Even working with election officials, we couldn't get it. That's not to say it doesn't exist. But there's so many hurdles for sharing and generating data when it comes to elections. And as someone who works with data, we know how important data are for everything, right? That's how we know who's voting, whose votes are counted, right? That's how we know how elections are being conducted. If we don't have the data, we can't approve elections. And that's really our goal here at the Center for Science and Democracy, is to improve elections and make them equitable and fair. And we can't do that if we don't know what's going on. So election data transparency is so key to moving forward in elections before we can really do anything. We need to know what the problem is.

Jess: I did see that in some of your past work, you've really looked at the intersecting factors like race, class and gender and how political rhetoric shapes public opinion towards traditionally disadvantaged and under and misrepresented groups of people. So how does looking at data through an intersectional multifactor lens help us identify the heart of issues like voting access?

Dr. Gordon-Rogers: So critical. We need better data, period. But in general, the data also needs to be more representative, right? Because people aren't singular, right? We're multifaceted. We have all these different identities that shape the way we view the world and the way we move through the world. And if we don't have intersectional data, then we can't really see those different components, right?

A big piece of our recent report on ballot design focused on disability access and language access. So the only way that we can really make improvements in that area is to have good data about how people with disabilities experience voting and how that experience is very different than someone like me. And then moving forward, if we have that data, we can try to say, "Okay, this is that experience, what's going right, what's going wrong. What can we do to improve voting for everyone, so every eligible voter, regardless of what their main language is or what disability they may have or may not have, can vote?"

Jess: we're in an interesting time and I'm glad that folks like you are out there digging into the data. But I do have a final question that I ask our guests on, "This Is Science." It is a two-part question. The first part is you are a scientist. We work at the Union of Concerned Scientists. So, Liza, why are you concerned?

Dr. Gordon-Rogers: Why am I concerned? Because I know that our election system isn't equal. That's a really simple way of putting it from my own personal experiences and from what I see in research. I know it's not equal. And I think it should be. That's really simple. But there it is.

Jess: It's perfect. It's completely valid. And I love the short and sweet. So I do love a positive ending as well. So what are you doing about that concern?

Dr. Gordon-Rogers: I'm doing my work every day. I come from an academic background. I was a visiting assistant professor before I came to UCS. And while I did love that job because I got to talk about what I love in research, what I'm really interested in, what I love, that's kind of where the journey stopped. You might present a paper at a conference, get it published, and that was kind of it. And you moved on to the next thing.

But at an organization like UCS, next year we're moving into our advocacy phase, right? This is a research every year. But then the coming fiscal year, we're going to put that research into action and say, "This is what science shows us that we think election officials should be doing for elections to make them more fair and equal," and working in states with election officials to try to get those recommendations implemented. So we do see some real improvement in elections.

Thanks to Liza Gordon-Rogers for sharing her time and expertise. Thanks to Matt Heid for suggesting this conversation, and to Omari Spears and Abby Figueroa for production help.

See you soon, science wonks!