Trust in American elections has declined since President Trump failed to concede the 2020 election. But voters do have real concerns about electoral institutions born of not knowing the process well. Jennifer Gaudette finds that election officials can increase trust in elections by informing citizens of basic facts, such as that election machines are not connected to the Internet. There are partisan patterns of mistrust but that doesn’t mean people can’t learn to trust election procedures, even in jurisdictions that typically vote for the other side.
Guests: Jennifer Gaudette, UC Riverside
Study: Can Public Information Campaigns Restore Trust in American Elections Across Polarized State and Party Lines?
Transcript
Matt Grossmann: Can we improve trust in elections if the losers don’t concede? This week on the Science of Politics. For the Niskanen Center, I’m Matt Grossmann. Trust in American elections has declined since President Trump failed to concede the 2020 election, but voters do have real concerns about the electoral institutions born of not knowing the process well. Perhaps information can be part of the solution. This week, I talked to Jennifer Gaudette of UC Riverside about her co-authored election law journal article, “Can public information campaigns restore trust in American elections across polarized state and party lines?” In that and related work, she finds that election officials can increase trust by informing citizens of basic facts, like election machines aren’t connected to the internet. There are definitely partisan patterns of mistrust, but that doesn’t mean people can’t learn to trust election procedures, even in places that typically choose the other side.
As we gear up for the midterms, it will be important to ensure that the winners and losers feel like the votes were counted fairly. Listen in for a bit more hope on that front. Tell us about the findings and takeaways from your election law journal article on information campaigns to restore trust in elections.
Jennifer Gaudette: Yeah. This article was part of a broader agenda where we’ve been interested in seeing how to restore trust in elections. Obviously, in the wake of 2020 and 2021, this is a big question that a lot of practitioners and academics are interested in. What we wanted to establish was both, do people have less trust? We found that people have less trust in other states, but do they have less trust in other states based on the partisan leaning of that other state? And then, can the informational treatments that we’ve previously shown can work? Do those work on partisans for that polarization across state lines? We find that they do. We use California and Texas here, because they’re a big blue state and a big red state. We have a national sample, then we have these California and Texas samples, and we break out bipartisanship, and we find that when California respondents are shown a video that explains how elections are kept secure in Texas, they become significantly more trusting in elections in Texas.
The same thing applies in Texas. And you might think, “Well, this is probably just Republicans in California being more trusting of Texas or Democrats in Texas becoming more trusting of California.” But no, all partisan preferences, so Democrats, Independents and Republicans, all become more trusting in California elections when they learn about how California elections are kept secure and vice versa. We find that both Democrats and Republicans, that they have less trust in those outpartisan states. But then, you can increase that trust simply by explaining how those states keep their elections secure.
Matt Grossmann: The context here is that we have very decentralized election administration in the US compared to other places and especially that the states are very involved. You also tackle this as a theoretical in addition to practical issues. What are the big challenges for public trust from that federalism of election procedures and what are the benefits?
Jennifer Gaudette: I think the big challenge is simply the fact that we are a huge country, there are 50 states, and most people don’t even know how their own elections are kept secure. They tend to think that where they live, their vote will probably be counted, but they have growing skepticism the less they know. And again, people know very little even about where they live. What we think is happening here is essentially that because people don’t know very much about what’s happening in other states, they’re substituting what they do know. If you’re a Democrat in California and you know that Texas has policies you don’t like because it’s a big red state, then you substitute that and you think, “Well, I assume they’re also doing maybe something sketchy with their elections or they’re not very secure.” And again, exact same for Republicans in Texas.
There’s just a huge information vacuum here where voters really don’t know that much about how elections are kept secure. What we think is interesting is that if you provide that information, it seems to help a lot. One thing that I think like a big benefit of having federal elections is that we saw it in 2020 where there was obviously a push to overturn election results and there were so many different officials who would’ve had to toss out ballots or engage in fraud in order to allow for election results to be tossed out.
From one perspective, they’re much more secure, but because people already don’t know that much about how elections are kept secure and because there’s 50 states and over a thousand counties that are all administering elections, people just don’t know about what’s happening. It presents an information and knowledge problem, even as it does create a situation where it would be a lot harder to actually steal an election or change election results on a large scale.
Matt Grossmann: This isn’t an insane purely partisan idea that voters have. California and Texas did just recently re-gerrymander to try to gain or react to partisan advantage. They often enact voting rules that they think are going to be more favorable even if they’re not for their own political side. Why is election administration and vote counting and all that different from all of that? What are the key protections that states have that would matter to voters?
Jennifer Gaudette: Two of the big pieces of information that we typically try to test when we’re using videos that convey information are that, first of all, the voting machines that people use are not connected to the internet, and the ballot counting machines are not connected to the internet, so you can’t just hack in. That is a really persuasive piece of information in general that we found people seem to feel a lot more confident once they know, “Okay, these machines are not attached to the internet.” And then, the other thing is that both parties oversee vote counting. Either side could raise their hand and call out if some kind of fraud was happening. Those two pieces of information are, I think, really persuasive to people and they suggest the security of the process. That said, I think that a lot of times people hear stories about the lead up to the election how easy or hard it is to vote the gerrymandering, and then they kind of attribute that to the election security.
And that’s again, why I think providing this information can really increase people’s trust in the elections because it doesn’t mean that they trust that the elections are fair in terms of who can turn out, but it means that they’re more trusting in the results.
Matt Grossmann: In this experiment, you actually used public official election videos. I know you have some varied projects that try to convey the same information, but you express a willingness to think that these are effective messengers and that we have identified a couple of the important factors. What matters in terms of the messenger and the message so far? Is it just a couple key facts? Is it actually showing off the security or humanizing the officials? Is it expertise? What matters in terms of convincing the public?
Jennifer Gaudette: We have tested over a variety, five, six, seven papers, we’ve tested a lot of different videos. One of the least effective videos was a video from Colorado, where it was a former Secretary of State and a current Secretary of State identifying themselves as, “I’m a Republican, I’m a Democrat, but one thing we agree on is our elections are secure, but there’s a lot of misinformation out there.” I think, again, this goes to the information provision aspect, videos where you’re telling people things you’re secure without giving them an explanation do not appear to be persuasive. The kinds of videos, these are really … the kinds of videos we use, they’re very boring. They’re very low-fi. We don’t have a concern that there’s really snazzy production value that is persuading people because these tend to be very low budget.
Typically, you’ll have maybe a pan shot of some voting machines and a voiceover or somebody that nobody’s ever seen before, some county election official is talking about how they keep elections secure and providing information. It is actually, it seems that boring, low-production value is part of the reason probably that people don’t think that this is somebody with ulterior motive or something trying to convince them that elections are secure when they’re not. I do think the fact that typically we have election officials, and we can dive into the demographics. We’ve gotten into that in a newer working paper. Typically, a basic county election office seal and then some basic information about how these voting machines are kept secure and that both parties get to oversee, it seems to be pretty persuasive to people across such a range of videos. And then, I’m happy to get into some of our new working paper if you’re interested in diving into that, where it suggests that the authority is not always, the authority of the person speaking is not always the thing that is most persuasive, but rather it is more the information people are receiving.
Matt Grossmann: Well, why don’t we do that now? You starred in one of these videos in the latest working paper. Tell us about that experience and what it tells you.
Jennifer Gaudette: Yes. Yeah. One thing that some of my colleagues and I had been really interested in was this sense that there has been a decrease in trust in authority, particularly among maybe younger voters. One thing that we really wanted to test was the medium and the messenger. We have done actually previous, we have something that was just published at the Politics and Governance, where we looked at, for example, a former election official, Stephen Richer from Maricopa County. He was in San Diego and he recorded a version of a exact same message. This is the benefit of working with an election official. We got an election official to record a message that was the exact same as an interview he had done, but this was then we put it on a, we turned it into an ad. It was paid media versus earned media, exact same effects.
One thing we really wanted to think about was the social media effects and social media videos. Because of the nature of a grant, we had one week to do it, otherwise, I would not have been excited to be recording and making myself part of the experiment, especially because we’re asking, how much do you trust the speaker?
We compared three different videos. One was just, again, very boring, standard Maricopa County video, a pan of these election machines, and all three videos we used had the same information. Voting machines not connected to the internet, ballot counting not connected to the internet, and both parties get to oversee ballot county. But one of them was that boring video. One was this humorous video that they’d done in Dane County in Wisconsin, where there was a comedy series character, Chad Vader, where instead of Darth Vader, it’s somebody who’s … It’s supposed to be humorous. I do actually think it’s pretty funny, but again, conveying the exact same information. It’s just Vader is trying to take over a polling station. They’re like, “Actually, that’s really hard because of all the ways we keep elections secure.”
And then, the third video was me talking to my phone in my car. It’s supposed to be like a TikTok video, so we put the subtitles on it, and so it’s a vertical orientation. Again, I am talking in my car. I do not identify myself as a professor or a researcher. I literally start the video by saying, “I was talking to a friend and I realized a lot of people don’t know,” and then, I go into the information. It’s about a little over a minute. We find that that video of me just talking to my phone in my car is equally persuasive at increasing trust in elections as the Vader video.
… at increasing trust in elections as the Vader video. And the sample size is small, but for under 25s, it appears to be, it’s a much more positive direction. It’s just, again, large measurement error potentially. But we find that it does appear to be equally persuasive to the Wisconsin video, statistically significant. And then both of those videos are slightly less than that boring Arizona, Maricopa County video. And so part of the takeaway here is that Wisconsin video probably has the highest production quality of almost any of the videos we’ve done, so that probably costs a lot of money versus me literally recording a video in my car. And so that is kind of, for me at least, it’s really made me think about where especially young people get their information and how trusted the authority source needs to be. And that’s something we want to dig into a lot more in the future versus just the literal provision of information to people.
And so again, that’s something we want to understand better. How much is, oh, it’s an election official versus, oh, that information, I didn’t know it before, sounds right. Interestingly, when one of our behavioral measurements was, if you want to learn more from the source of the video, click here, people were not interested in learning more from who saw my video. And it does, I think, potentially speak to where people receive information through social media maybe… Oh, and they remembered the details the best from the social media style videos. So they remember the details the best, but were the least interested in following up on the information. So I think especially because so much communication now, especially political communication is through these mediums, we really need to better understand how persuasive they are and how they’re working. So that’s something we’re looking at digging into a lot more.
Matt Grossmann: So in the Election Law Journal article, we’re comparing California and Texas and using kind of cross-state information. And the baseline there is that partisans don’t trust the sort of stereotypical other party state. And that fits with the sort of longer running context for this, which is that the loser of any election tends to be more concerned about whether the election was carried out fairly and people may be less likely to trust a place where they think the results are going to go against them. So put the kind of current focus in the context of that longer running story. What’s happening right now in the United States with regard to kind of trust in elections and how out of the norm is it compared to this kind of normal reaction that people have to elections that they lose or think they might lose?
Jennifer Gaudette: Yeah. I think that the winner effect is pretty clear in elections where people in general are just more skeptical if their candidate didn’t win. And that goes back quite a ways. But I think because we have prominent members of our main political parties now talking about fraud, we have seen a substantial reduction in trust. And part of the reason we wanted to sort of investigate social media is because in 2024, in the wake of the Democrats’ loss at the presidential level, there was a pretty substantial drop in trust in elections by Democrats as well. And there were a lot of viral claims of voter fraud on social media. And so part of the reason that I think that this paper is so important, and we ran the study a few years ago, it was prior to the 2024 election. And part of the reason we think these cross state findings are so important is because most of the time when people hear claims of election fraud, they’re in another state.
And in fact, California, which we’re studying here, has been a target of claims of voter fraud by then candidate, now President Trump. So using these states that are, again, very emblematic of the partisan tilt, allows us to test, it’s kind of a pretty tough test of whether information can change people’s partisan priors because this is the big blue state and the big red state. And so in the context of how people have historically thought about elections and had in general been more skeptical when their candidate didn’t win, we are seeing much larger shifts than we’ve seen previously. And again, there’s the party elites part of the picture and there’s the social media part of the picture. Disentangling all that is a little difficult, but these are the kinds of interventions that hopefully don’t even have to cost a lot for an election official that can potentially reduce the amount of distrust at least.
Matt Grossmann: So you also look at other outcomes like voting turnout and willingness to serve as a poll worker, but these are just sort of survey expressed willingness. How connected are these outcomes? Should we be concerned that this loss in trust is leading people to not want to participate in elections at all or not be part of the process? And to what extent is information a solution to those aspects?
Jennifer Gaudette: Yeah. So I mean, the behavioral questions, I think they’re most suggestive of an attitudinal shift. It’s very difficult, obviously, with a survey to actually know if behavior is going to change. But I think as a basic question of democratic norms, if people don’t think elections are fair, on the margins, you’re going to have people choosing not to turn out to vote. And just generally for democratic functioning, people need to think that elections are secure or they lose faith in the process. Now that doesn’t mean that everyone who has questions is not going to vote, but it is sort of like a long-term rot in the foundations that you want to try to combat because the lower the faith in democracy in general, it’s from a normative perspective, that’s just not a good thing. We want people to have faith that their votes count in some way. People already think their votes don’t really count because there’s so many people who vote. If they think that quite literally their ballot will be tossed out, it has a very negative impact on how people perceive the political process and their voice being heard.
Matt Grossmann: So you also have another article that looks at one of the reasons that California has been distrusted, which is that it takes the results so long to come in because they allow mail-in votes to be received after election day. So you find that you can sort of change people’s reactions to that if you sort of tell them in advance about the delayed results. So what should we conclude from that and how widely is it applicable to situations like the Midwest where we have early votes counted differently than election day votes, and so results seem to change during the night, since that does seem to be a source of a lot of these conspiracies?
Jennifer Gaudette: Yeah, absolutely. I mean, the way we characterize it is as inoculation. So people have a lot less trust in elections when results change after the fact or there are significant delays. And this is actually a case that is… There’s a case at the Supreme Court right now that is talking about this kind of research, and this paper was actually cited in a couple of amicus briefs because it’s some of the evidence out there that shows that yeah, these delayed results, they do have real impacts on people’s trust in election results. But what we find is that to the extent you explain in advance the cause of delays and the cause of changes and people understand what’s happening, it actually reduces much of that distrust.
And the way I kind of characterize it actually, I recently had an experience that was very illustrative, I think. So my husband got sick five days before my daughter’s birthday party with the flu, tested positive. And on that day, sat my daughter down and was like, “We are probably going to get the flu and here is how the flu works. Here is how it’s contagious.” She’s six or was turning six. “And if you’re contagious, here’s why it’s probably not a good idea for us to still have your birthday party.” And then a couple days later I get sick. I’m like, “Hey, remember what we talked about. There’s a good chance.” Explained it again. Day later she’s sick. And then the day before her party, she’s testing positive for influenza A. And we discussed it. I was like, “Okay, look, you tested positive. I think we have to postpone your party. What do you think?” And a six-year-old is like, “Yes, we have to postpone.” And the reason why is because she had gotten the information explaining the situation so that when the situation happens, she understood the context and could understand what was going on, that we weren’t just randomly canceling her birthday party.
And similarly, what we think is happening here is, again, when people understand the context of why there’s a delay… Because I think as political practitioners or political scientists, it can be kind of hard to understand. Sometimes we’re so in the weeds, right? But for the average person, they’re like, “Wait a second, the vote totals have changed six days after the election?” It just doesn’t ring true, and so especially again, when we live in a modern media environment when there are so many claims and you can see so many different sources of misinformation here. And so the basic idea is simply that if you explain what’s happening, if you give people that information, it is essentially like an inoculation or like a flu shot or something where you’re saying, “This is what’s happening. Here’s how you can understand and make sense of these headlines and news stories that you’re going to see and what’s going to happen.”
Now, I do want to note, just as a point of personal privilege, I think California should probably change some of its laws to count ballots faster because we do see these negative effects from the delays. And there are ways California could count faster and do certain things better. But where there’s just no getting around it, and obviously 2020 was just this incredibly unique circumstance because of COVID, to the extent that there had been a lot more effort to educate the voters about why there could be delays, I think that would have been really helpful, at least that’s what our research shows.
Matt Grossmann: So you have done several of these studies, and as you mentioned, they’re often in consultation or directly working with election officials. So you’re kind of tied into this community. As we go into another election this year, what is kind of the state of the work? Are they getting the information that they need? Is it likely to make any difference in kind of what they do and how much voters trust elections this year?
Jennifer Gaudette: I think absolutely. I think there are some offices… Again, the fact that our elections are so decentralized. So some offices really care about this and they’re really consulting, looking at different sources to figure out how they can maximize voters’ trust in what they’re doing. It’s worth noting that election officials have received an incredible number of threats over the past few years. I think that kind of goes to your earlier question about how much things have changed or historically, the amount of threats people working in the election space are getting now has just really increased. And so I think partially because of that and just because of the overall narrative about elections, there is a lot of interest in this. And so I think a lot of election officials are working on these kinds of information campaigns. But honestly, I think, again, our media environment is just changing so rapidly and has become so diffuse that there is a certain amount of questioning of how to best address voters and how to best get this information to them.
And that was part of the motivation of that recent work where we were like, “Well, how effective is it if somebody literally just records a video in their car?” And so we, in our conversations with other practitioners who are working with, or excuse me, other researchers who are working with practitioners and then other practitioners, one thing we really stress is everything you do to secure elections can be content that you can try putting out there and you never know how many eyeballs something will get, but the more content you can put out showing how elections are kept secure, the better, because you just never know what will hit with a different demographic or audience. And so we think sustained information campaigns can have a really-
We think sustained information campaigns can have a really positive effect on the information environment, but it doesn’t have to be a paid advertisement campaign that costs hundreds of thousands of dollars. It can be a combination of communication methods.
Matt Grossmann: So there is declining trust in institutions almost across the board, but this election trust, it takes a different form. It really changes a lot after the 2020 election, and we’ve been talking about it as elites are changing their messaging, but it’s really President Trump refused to concede, tried to challenge thing in all six close states in its pretty unprecedented fashion, and endorsed every conspiracy under the sun about election results.
So, how much is this, really, this Trump-specific story? Now, you’ve mentioned that Democrats also had a lot of concerns about 2024, so maybe it has moved back in the other direction as well, but how much can we expect this to sort of be a Trump phenomenon that when he leaves the scene, trust will come back? And how much is this sort of a new permanent state of American politics?
Jennifer Gaudette: So I think there is something unique to Trump in that he reshuffled somewhat our coalitions. So people who had typically been maybe more Republican, like suburban college-educated voters, who had leaned more Republican previously, became more Democratic. Those are people who have a lot of faith in institutions. Institutions have tended to work out well for them. And people with low trust who maybe historically had voted Democrat have shifted significantly toward the Republican Party, but specifically to presidential elections where Trump is on the ballot. And so you have a group of voters who are very primed to be distrustful of institutions and authority, and that is a group of voters who president Trump is specifically speaking to.
So I think it makes a lot of sense that we have seen with these reshuffled coalitions that Trump’s coalition is particularly mistrustful of elections. I think given that actually our results do … I don’t know, they make me feel better about the future, the fact that to the extent we don’t have as much of this messaging coming out of President Trump who is no longer … I think presumably he will remain active on the political scene after he’s no longer in office, but to the extent that he’s no longer the leader of the party potentially or constantly bringing these up, I think that we will hopefully return maybe to a historical baseline, although I think there will probably be a continued lower level of trust within that coalition, simply because it’s like a lower trust in institutions group.
But that said, I do think it’s imperative that our elites on both sides really don’t enhance this kind of messaging. On the Democratic side, I think a lot of people, California voters, including Democrats, were very disappointed when then Congresswoman Katie Porter suggested she had lost the primary in 2024 to Adam Schiff because the election was rigged. And then she later clarified, oh, she’s talking about money.
But I do think it’s very important that our elites, if we want to move past this, that our elites talk about elections without using that kind of language. But then again, I think it’s very important that we also ensure that voters just have the information of what is happening. I think the overwhelming result across, again, multiple papers, at this point, probably 30,000 respondents across a variety of platforms is that when we give voters information that they didn’t previously have, and it comes from a unbiased source, they tend to increase their trust in elections because there is data from Charles Stewart that suggests, I think it’s like 15% of voters know about signature verification.
So our elections are by and large extremely secure, and people don’t understand the ways we keep them secure, and literally just explaining it to them can increase a lot of trust. So I certainly think that this is a Trump-specific phenomenon in the coalition he’s amassed, but I also think that it can change even with him on the scene if there is more of an effort to give voters information as opposed to just lecture them about how elections are secure. To me, the information part is just really key here.
Matt Grossmann: So the good news is that people do respond to information, but this is, it’s a particular … You’re giving them only that information, you’re looking at the effect of that information. You could see your social media video is a good sign that we can get across the information anyway, but you also might be able to see it as a bad sign. Well, if any self-recorded video saying anything without a source is potentially influential, then in the course of an election where it’s the day after, we have some real uncertainty about results and we have lots of different pictures and messaging being shared that it’s going to be hard for these basic messages to break through compared to all the others, which are kind of aligned with partisan identity coming from partisan elites that people trust. So how much can information matter in that kind of context?
Jennifer Gaudette: Yeah. I mean, I think it’s a tough question. We have a paper that’s under review where we tested how well one of these videos held up when respondents watched the video, and then they watched a clip of President Trump alleging voter fraud. And we found that their increase in trust remained.
And again, I think part of it is the voter fraud allegations, they’re kind of baked in now. So they were to a certain extent new information in 2020, 2021, but they’ve become kind of baked in, like every election cycle, there’s going to be allegations of fraud versus a lot of people, again, don’t know about how elections are kept secure. So it can be new information potentially.
But I do just think that we are operating in an information environment, and I think algorithms in particular make it very difficult to break through. It is actually a reason I encourage everyone I talk to about election security when they have a question that they get answered, like, “Hey, share this with your network.” Are you trying to get breakthrough social media bubbles is a really important part of this process, but it is kind of incumbent on individuals to be active on social media and then sharing and hopefully crossing some of those boundaries because again, people tend to be in these kinds of polarized bubbles on social media and not be exposed to that information that is contrary to what they think.
So I just think that’s a really difficult problem to deal with in general. And so we just kind of have to try our best to break through those bubbles.
Matt Grossmann: So, how has working with Secretaries of State and other election officials kind of shaped what kinds of questions you ask, what kind of interventions you can do, and what methods you can use? I know that there’s some good side, but we also might worry that had you been entrusted with trying to decrease voter trust through a partisan lens, that you would’ve tested different things, and you may have found that that’s easy as well.
So yeah, how does the perspective that you’re starting from change what you’re able to test and show?
Jennifer Gaudette: Yeah, absolutely. I don’t know if we could have gotten that one through IRB, but I actually, honestly, I would love to test whether it’s easier to decrease or increase trust in elections, but yeah, I mean, I think that there’s huge pros in that it’s just personally very motivating to be like, “Oh wow, practitioners in the real world are going to use these findings.” But there are obviously… Is a certain kind of person less persuasive? That’s kind of hard to test, right? You’d have to really clear that with a person before doing that, and you can only work with the videos that they produce. That was actually one reason we were like, we have one week, rather than trying to get some kind of influencer or something, let’s just record the video, and then it’ll be exactly what we’re thinking of. And that is an element.
These are very bundled treatments, obviously. What was so cool was that some of these practitioners have been incredibly willing to work with us. And again, Steven Richer was a great example of that. Unfortunately, he lost his reelection again because of all these claims about voter fraud in Arizona, unfortunately, but his willingness to do something like record the exact same message, but then do it in an advertisement format, so that we could then compare earned versus paid media and find that there are no differences in persuasiveness.
That’s pretty exciting, but in general, they really want to get the information, but most of the time they’re not going to be quite that willing to do a bunch of treatments where you’re changing just a small, subtle aspect of it.
And so we have great external validity. Obviously, there are downsides from an experimental standpoint, but I think they really feel like they get information that they can use. And so I think we find, especially given we’ve done at this point, so many studies with so many different kinds of videos, that we are providing information that’s valuable in the real world. So it outweighs the negatives from the perspective of a perfect experimental design.
Matt Grossmann: So, to try to summarize those takeaways for the practitioners, let’s pretend that you have a multimillion-dollar budget for a state election official. How would you use it? What do we actually know that would work and last?
Jennifer Gaudette: Yeah. I mean, the thing is that to the extent you can get earned media where you can get interviewed about that, they appear to be pretty effective. So we have samples of local news, Fox News interviewing the Texas Secretary of State. We tested that. That was quite persuasive to people when he was just explaining how elections in Texas were kept secure.
So, trying to get your communications team to get you on TV, local news, et cetera, really valuable, especially with older voters who maybe watch TV more. If you have that money, if you have a million dollars and you can run a couple ads, they appear to be equally effective despite the fact that they’re paid. Again, though, the kind of ad you would want to run, you don’t want it to be slick. You don’t want to be out there talking about how you’re a Democrat or a Republican, and people need to trust you.
You want to make it. Here’s how our elections are kept secure. And then I really encourage election officials to just record themselves doing things like, “Oh, here’s where we do our random audits of machines. I’m heading in to double check this,” and, “Oh, here’s a Democrat and a Republican who are overseeing this process.” So producing content that can go out on social media for people who consume mostly social media, producing content that can potentially be a paid video campaign, video advertisement campaign. The one thing I would advise not to do that we have found is not effective are static images. So, like poster ads, those do not appear to increase people’s trust in elections.
Now, one thing again we would like to test is some of those Instagram people, slide through infographics. We’ve done some get-out-the-vote things previously that didn’t seem to have an effect.
The informational posters don’t seem to have an effect, but does a social media … Does an Instagram slideshow of infographics, is that persuasive? Based on the previous results, my guess would be not particularly, and there is something about a video that is more persuasive to people, but in general, the more information you-
But in general, the more information you can give people, and in a variety of mediums, the better, because the information really does seem to be driving how people feel about elections. And then again, authority is still persuasive. Being from an election official does still appear to be more persuasive to older voters, but younger voters seem to be very much more persuaded by some of these social media accounts.
Matt Grossmann: So as you mentioned with the Katie Porter example, there’s often a fallback after pure fraud doesn’t hold up that there was still something unfair about the process. And these takes on partisan complexion, Democrats tend to complain about what they call voter suppression, which could be just making it slightly harder in particular areas to vote, all the way to more direct efforts. The Republicans tend to complain about the opposite, making it too easy for unqualified or people who got them in late to succeed. But people also complain about everything from media coverage being unfair to campaign finance being unfair to gerrymandering.
In other words, there’s enough out there that we always kind of have a fallback position to complain about unfairness. So how important is it that we change this specific view that people have about election procedures versus that we address this broader issue that people think that elections are unfair in some broader sense?
Jennifer Gaudette: I mean, I think that’s a great question. I think that when people’s preferred outcome does not happen, they always have a reason it was unfair. Everyone, that’s kind of a basic human response to your side not winning. I think going back to these democratic norms, I think for the healthy functioning of our democracy, people, while they might think things are unfair and find some other reason to be unhappy with the outcome, especially because every election feels so incredibly important at this point to people, and that’s how it’s communicated. I think that people are always going to have a sense that it was unfair in some way. Obviously, if you’re a Republican voter in California after Prop 50, you’re going to say it was unfair. And if you were a Democratic voter in Texas, after they re-gerrymandered, you would say it was unfair.
But I think the basic questions of, are ballots literally being properly counted? People feeling like their vote, whatever the process is, their vote is going to be counted and that the votes in other states are going to be counted accurately. I think that goes to just a bedrock principle of democracy. We talk about Hungary is a great example here where they had changed a lot of their processes of their elections to make it easier to have a super majority. And ultimately though, because they had not changed the actual systems of elections such that they were fraudulent, you could change the party who was in power. And so if you thought that your ballot literally wouldn’t be counted, even though it was an unfair system, if in Hungary they had thought, “My ballot just literally won’t be counted as fraudulent,” they wouldn’t turn out.
And so I think there is that element of people need to understand that their ballots and ballots in other states are being accurately counted. And the lead up might be unfair, but if you can get your ballot in the door, it will be counted accurately. I think that’s important. And I think that people will always find something to think was unfair in any election if their side doesn’t win, but it’s just such a bedrock principle that the more people understand that part is safe, I think it’s just the better for our democracy.
Matt Grossmann: And what about real threats to elections? Is there a concern that we spend a lot of time telling people, “Here’s why the elections are secure.” And then we do have examples of attempted interference, attempted false voting. We have concerns about AI, all kinds of real potentials that election officials are worried about or are working on. So what’s the relationship between addressing public information and misinformation versus actually making elections secure? And is there a danger that we have two or three incidents that are real threats to election security that kind of undermine all the work that we’ve been doing?
Jennifer Gaudette: I actually think it’s really important for election officials to acknowledge that, yeah, there are a couple cases that have happened. But also say, “And here’s how we found it and here’s how we can be confident this isn’t widespread.” I think acting like there has never been a ballot cast inappropriately is just… It’s not the truth and then people won’t trust you because people can do a simple Google search and find these stories. They do exist. There was a story recently about somebody in Southern California where she had cast a vote for, I think it was city council, in her dog’s name, which is a really strange thing to do and a terrible reason to end up in prison. So these stories, they exist. There’s never been any evidence that happens at a broad scale.
So I think, as a first step, it’s important to acknowledge that there are cases like this. It’s just that there is no… Explain how they were caught and how we know that they’re not happening at a wide scale. But I think it’s also really important with technological advances that money is put into thinking through how can we ensure that, yes, this is how we’ve done things and how we’ve kept them secure, and how can we continue to do that? And states are putting money into this. It’s something that’s being considered. But talking about that I think also helps, “Here’s all the ways we’re forward looking in ensuring that we’re staying on top of this.”
But again, that’s where internet not being… The voting machines not being connected to the internet is just incredibly important. A lot of these things… If there’s no way other than the chip that’s inserted to pull the numbers, if there’s no way else to get into the machine, then it’s going to be pretty secure. And these vote-counting machines are in these locked rooms, but they’re surrounded by glass. It’s very mission impossible or X-Men style where everyone can see in to see what’s happening, but there’s no way unless you can get in, and these are not, again, something where you could just on a wide scale try to hack in with the internet, it just wouldn’t work.
So I think the way the systems are set up, they’re quite secure, but it’s absolutely an important thing that money continues to be put into thinking about future ways that ballots could be messed with in some way. And I think people have valid concerns about voter intimidation at the polls, right? So not the actual ballot, but are there… And there’s been a lot of talk about this, are there ICE officers outside the ballot place? Those are a different concern from how our actual ballots are counted, but obviously those would impact people’s impressions of the election and could affect whether people show up to vote. But again, I think that’s the lead up question where we should still be paying attention versus the pure security of people’s votes.
Matt Grossmann: And anything you want to tell about what else you’re working on or that we didn’t get to that you want to leave us with?
Jennifer Gaudette: It’s kind of funny because I have two different strands of research that can seem kind of disparate. So on the one hand, I study the influence of interest groups, particularly at the local level. And then on the other hand, I’m looking at elections, secure and trust in elections. But I actually think that a lot of my work boils down to polarization and how democracy is functioning and how it works.
And so I think that in general… I was a lawyer prior to becoming a political science of public policy professor. And what really inspired me to shift, besides paying off my law school loans, was I was just really fascinated by the changes I was seeing in our politics and how our democracy was functioning. And I think that interest just really drives both of the strands of my research, just better understanding what is driving the polarization in our country and the different elements that impact how our democracy works.
Matt Grossmann: There’s a lot more to learn. The Science of Politics is available biweekly from Niskanen Center, and I’m your host, Matt Grossmann. If you like this discussion, hear the episodes you should check out next, all linked on our website. Partisan election administrators don’t tip the scales. When information about candidates persuades voters. How early voting is changing American elections. Do Americans implicitly trust government despite our public anger? And how the left and right undermine trust in government. Thanks to Jennifer Gossett for joining me. Please check out, can public information campaigns restore trust in American elections across polarized state and party lines? And then listen in next time.
