Donald Trump is trying to link the COVID-19 pandemic to fear of immigration. There is precedent for linking infectious disease to that issue and winning elections as a result. When an Ebola outbreak came to U.S. public attention just before the 2014 midterm elections, Republicans were able to use it to their political advantage. Claire Adida finds that Republican politicians successfully raised fear of immigration by politicizing Ebola. Filipe Campante finds that local Ebola outbreaks gave Republicans an electoral advantage by changing immigration attitudes. They both see big differences this time, even though Republicans are trying to repeat the success.
Matt Grossmann: This week on the Science of Politics. Republicans successfully politicized Ebola. What about COVID-19? For the Niskanen Center, I’m Matt Grossmann. Political conflict over the COVID-19 pandemic has us thinking back to 2014 when an Ebola outbreak came to U.S. public attention just before the midterm elections. How is that infectious disease interpreted by citizens, and how were Republicans able to use it to their political advantage? Will this time be different? Today I talk to Claire Adida of the UC San Diego about her Politics, Groups and Identities paper with Kim Dionne and Melina Platas, Ebola, Elections and Immigration. They find that just hearing about Ebola was not enough to change opinions, but Republican politicians could successfully raise fear of immigration by politicizing Ebola.
I also talk to Filipe Campante of Johns Hopkins about his NBER paper with Emilio Depetris-Chauvin and Ruben Durante, The Virus of Fear. They find that local Ebola outbreaks gave Republicans an electoral advantage in the 2014 elections by changing immigration attitudes. Changing those views was enough to change votes. They agree that Ebola was a big U.S. news story just before the election. Despite a huge outbreak, U.S. attention was tied to those U.S. cases.
Claire Adida: The Ebola outbreak of December 2013 was a huge news story for those of us interested in Africa, and specifically West Africa, which is where it started, and really where it was prevalent. It was a horrible outbreak. It killed more than 11,000 people across three countries, Guinea, Sierra Leone and Liberia, and it didn’t actually officially end until mid 2016. It was the deadliest and most widespread Ebola outbreak on record at the time, but it didn’t catch anyone’s attention in the U.S. If you look at trends in Google searches for the term Ebola in the U.S., you see basically just a horizontal line at zero for the first half of 2014. Six, seven months into the outbreak, there’s zero interest in this as measured by Google searches for the term Ebola in the U.S.
Then in August 2014, you start to see a little spike. That’s just after the first American was infected in Liberia on July 26th of 2014 and brought back to the U.S. for treatment. Then the next big spike, the biggest spike actually, comes in October and November of 2014, when the virus appears in the U.S. for the first time because of a Liberian national traveling to Dallas to visit family and then developing symptoms once in Dallas. Then one of his nurses eventually also catching Ebola. That’s right around the time of the midterm elections. It’s October, so fall of 2014, and that’s when you see this huge spike in searches in the U.S. for Google searches for the term Ebola.
Filipe Campante: The first thing to remember is that it was a big public health crisis in West Africa. Over the first few months of 2014, there were a substantial number of cases, thousands of cases of Ebola, particularly affecting Sierra Leone, Guinea and Liberia, three countries in West Africa. The World Health Organization, at some point, declared this a global health emergency. That was around August. That became an international piece of news, but it never really, if you look at the data, it never really became a public health crisis anywhere else. But it did reach the U.S. and it reached the U.S. in a particularly sensitive political timing. In late September 2014, one Liberian national was visiting the U.S. and fell ill and was diagnosed with Ebola, and that was in Dallas, Texas, and that was big news in the U.S. If you look at the level of coverage in cable news, in broadcast news, that was huge. It’s kind of understandable given it’s a very gruesome disease, very little in terms of treatment, very high mortality rate, so it was a big deal.
Then after that first case, there were two nurses who were involved in treating that patient, and they were infected as well. It so happened that one of them, between being infected and falling ill and being diagnosed, had traveled from Dallas to the Cleveland, Akron area in Ohio, so that was kind of a second area affected in the U.S. Then there was a third or a fourth, rather, patient who was a doctor that had been in West Africa helping treat the Ebola crisis, and he came back to the U.S. and he was diagnosed with the illness as well. Around each of these individual cases, there was a huge media mobilization that really seeped into public discourse. If you just look at the media impact of that, it was quite substantial and very much out of proportion with the actual public health risk that it represented at any given moment in time.
Matt Grossmann: Adida says that set the stage for Republicans to connect Ebola and immigration, which they found was successful.
Claire Adida: The paper investigates whether the threat of Ebola, which became a real threat in the U.S. in the fall of 2014, has any effect on American attitudes towards immigration. The outbreak actually started in West Africa in December 2013, and really devastated West Africa, but it didn’t touch the U.S. until the summer of 2014. That coincided with the coming of the midterm elections of 2014. You started seeing U.S. politicians, overwhelmingly Republican politicians, calling for more restrictive measures, such as ending all flights from affected countries and closing the southern border. You also started seeing increasing anti-African sentiments, particularly in Dallas where the U.S. Ebola outbreak began. In the media, the term that came up over and over again was the Ebolification of immigration reform in the U.S.
Our interest in this study was to really investigate whether the fear of a public health threat coming from outside the U.S. could be exploited by political actors to shape American attitudes towards immigration. It is certainly what some politicians were trying to do in their communications with their constituents. They very much were talking about the Ebola response, and we wanted to basically know if that worked, if they were able to shape American attitudes towards immigration. That was the objective of the study.
The findings, in a nutshell, there’s two broad findings. The first is one that surprised us. We found that just mentioning Ebola did not change people’s attitudes towards immigration. A big null effect for this first question. However, perhaps more interestingly, we found that the politicization of the Ebola response, in other words, a political actor exploiting, criticizing the Ebola response, that did have an effect on respondents by making them less open to immigration, and the effect was particularly strong for Republican respondents.
Matt Grossmann: Campante says that even led to big Republican electoral gains in 2014.
Filipe Campante: The big question we were interested in is what’s the impact of fear on elections in general? That’s kind of the big question. Actually it’s very easy to motivate in the current context, of course. When you think about the impact that COVID-19 might have in November, on the one hand you have that in all likelihood it’s going to lead to a massive recession and we know that that affects elections. It is requiring a massive policy response, which presumably people judge incumbents on. We can also imagine that there’s a pure fear anxiety element to it. People are anxious about this risk, this virus that is out there, and that might affect how they behave when it comes to voting, and politicians may respond to that. If we’re interested in it in this third aspect, the Ebola crisis of 2014, or the Ebola episode, perhaps, we should say, of 2014, it provides us an almost ideal context because we have a relatively small shock to actual or public health risk environment, but we had a very out-sized, very large impact on public discourse and generated quite a lot of public anxiety, which happened over the month before the midterm elections in 2014.
What we ask is what was the impact of Ebola, concerns surrounding Ebola, on the election that year? The first key finding is that Ebola concerns caused a substantial drop in Democratic vote shares. Just to give you an idea of the magnitude, if we take a one standard deviation increase in Ebola concerns, and we can talk a bit about how we measured those, that would be associated with a 4.5 roughly percentage point drop in the Democratic vote share, which, if you look at the numbers from that election, would have swung 15 House races that were won by Republicans, which is about the size of the gains that Republicans had between 2010 and 2014, so that’s the first key finding. This is associated with a drop in turnout as well. By the way, the 2014 midterms had the lowest turnout in many, many decades.
Then the question is what was going on that might be leading to this effect? We also find that politicians, in particular Republican politicians, responded to these Ebola concerns by exploiting them strategically. As Ebola cases emerged, they start talking about it in a number of different ways, and doing more of that in particularly competitive races, which indicates the strategic use of those concerns. We also find that they’re not just talking about it, but they’re talking about it in association with specific themes. They talk about it in association with immigration, terrorism and President Obama. Then the question is did that work? One thing that we find is Ebola concerns, they didn’t make people more conservative in general, or they didn’t really affect President Obama’s approval ratings, but they did make people more conservative in terms of their attitudes towards immigration. This is the broad picture that the paper points out. There was this negative impact on the Democratic co-chair, and this seems consistent with the strategic exploitation that we find Republicans pursued, and it seemed to have an impact not across the board, but in this specific topic of immigration.
Matt Grossmann: Adida says politicization was a common Republican tactic that year.
Claire Adida: What we were reading was these anecdotes of mostly Republican politicians complaining, criticizing the Obama response. They were saying that the Obama administration needed to end all flights from the affected countries in West Africa. I think Trump at the time tweeted that even U.S. citizens coming from those places needed to be banned, to not be let in. There were a number of claims that were made about the migrants coming in from the southern border, and a number of people, a number of Republican legislators linking Ebola to migrants from Mexico, to ISIS all in one breath, and talking about the dangers coming from the southern border. Again, these are just kind of anecdotal examples.
There is a more systematic study that was done, not by us, but that looked at the email communications, the e-newsletters that representatives sent their constituents, and the number of times that different issues were mentioned, and it found just descriptively that GOP legislators, GOP representatives mentioned Ebola significantly more than their Democratic counterparts, and that they mentioned it significantly more than they mentioned other issues at the time that were salient, the legalization of marijuana, for example. There is a sense that in terms of communicating with their constituents that these politicians were trying to make Ebola really salient in people’s minds.
Matt Grossmann: They used a survey experiment to compare the effects of information about Ebola and its direct politicization.
Claire Adida: With support from the National Science Foundation, we fielded a survey experiment among a national sample of about 3,800 Americans between November and December 2014. Our goal was to test whether mentioning Ebola and emphasizing various aspects of the crisis had any effect on people’s preferences toward immigration. To do this, we randomly assigned our respondents to various versions of the survey, and each version of the survey emphasized something different. We would be able to identify the causal effect of what we were emphasizing on people’s immigration preferences. What were these different versions? Two thirds of our sample was randomly selected to read a fact sheet about Ebola before answering questions about their attitudes towards immigration. That’s two thirds, and then the remaining third did not read a fact sheet before answering questions about their immigration preferences. By doing that, we can measure what mentioning Ebola does to someone’s preferences for immigration.
We did more than that. Among those reading the fact sheet, we had four groups reading four different versions of the fact sheet. The fact sheet had five bullet points about the Ebola crisis. The first four bullet points were all the same for all respondents. They covered where Ebola originated, what Ebola is, how it’s transmitted and what it’s fatality rate is. The fifth bullet point was different for each of the four groups of respondents. One of the bullet points, one of the versions emphasized the American origin of one of the first cases of Ebola in the U.S. Another one emphasized the African origin of one of the first treated cases in the country. By the way, both actually are correct. We didn’t use any deception. We didn’t lie at all in our survey. It was the case that one of the first cases was an American doctor, and another one of the first cases was a Liberian national, so both of these are technically true statements. What those two versions of the fact sheet allow us to do is to test whether people respond differently when we emphasize the African origin of the disease.
Then we had two other versions. One quoted in the last bullet, it quoted an American politician criticizing the Obama administration’s response to the crisis. The other did the exact same thing, but emphasized that the politician was Republican. Again, here we didn’t use any deception. We used an actual quote by Republican governor of Louisiana at the time, Bobby Jindal, who criticized the Obama administration. With those two versions we can test whether the politicization of the public health crisis, and by that I mean that a politician publicly criticized the Obama administration’s response to the crisis, and then we can also test whether this happens via partisanship and whether that has any effect on attitudes towards immigration.
What you have is you have a group of respondents who don’t read anything about Ebola before telling us how they feel about immigration, and then we have these four groups of respondents who each read different versions of a fact sheet about Ebola. That’s how we’re able to come up with the findings that I mentioned before. We measured attitudes towards immigration by using these traditional measures that are used by the Pew Research Center, for example, where we measured how people feel about whether immigrants strengthen or weaken the country, whether legal immigration levels should be increased or decreased or kept the same, and whether illegal immigrants already in the U.S. should be allowed to stay, granted temporary work status or required to return home. We have these three questions. We can create an index or we can look at these questions separately.
Our first finding, which is this non-finding, looks at the effect of receiving any version of the fact sheet on immigration attitudes. Here we compare the two thirds of the sample that got any version of the fact sheet to the one third of the sample that didn’t get anything, that didn’t read anything about Ebola before saying anything about their immigration preferences. Again, we didn’t find any effect here. We find no differences in immigration preferences between these two groups. This was actually surprising to us. We were expecting that mentioning Ebola would make people more conservative and more closed to immigration.
When you start to look at the different versions of the fact sheet, it gets interesting. This is where we find that respondents who received the fact sheet that politicizes the crisis show less openness to immigration across the board. This effect is pretty unsurprisingly strongest for Republicans, but we also do find an effect for independents, and although it doesn’t reach statistical significance, we see some negative trend also among our Democratic sample.
Matt Grossmann: Campante’s analysis used online activity to show that Ebola concern had a real local impact.
Filipe Campante: When we’re interested in a question like did anxiety around Ebola cause whatever impact we’re interested in, in terms of the electoral outcomes, there’s a question of measuring those Ebola concerns. How do we know that in certain areas people were more concerned with Ebola? Our approach is to look at people’s online activities. We use these two kinds of online activities. On the one hand, Google searches, which we can measure and we can locate where these searches were coming from. The second one is tweets, tweets that mention the word Ebola or hashtag Ebola. Out of the total tweets, we also focus on the ones that we can geolocate and we can pinpoint where they were being sent from.
That’s very useful because it gives us this geolocated measure of Ebola concerns. We know that there were people in these areas that were thinking about it, they were looking it up online, they were talking about it in social media. That’s the crucial advantage. Now the limitation clearly is that the people who are active online aren’t necessarily your regular folks, your average voters, so we’re taking this slice of the people in each particular location, and we have to make inference based on that, but that’s another reason why we may want to use empirical methods that can allow us to deal with concerns around measurement error and imprecision in the measurements.
Matt Grossmann: They tried to show a causal electoral impact by linking geographic differences and checking for the same mechanisms that Adida found.
Filipe Campante: The first thing to think about is how do we know that these Ebola concerns were causing these effects? We can’t just look at the measure of concerns and correlate them with electoral outcomes, because it may well be the case that people who are likely to be concerned with Ebola in that situation, they were already likely to vote one way or another, or to think about President Obama in whichever way this might have gone, so we need to dig a little deeper and try to come up with a strategy that allows us to disentangle the causal impact of Ebola. For that we use the geographical variation that Ebola happened to feature in terms of how it affected the U.S. We have these three locations, Dallas, Texas, the Cleveland/Akron area in Ohio and New York City, that were, quote, unquote, affected directly by Ebola. That’s where the cases emerged or there was risk of exposure. We use proximity to these Ebola cases as the source of variation in Ebola concerns, and we show that the closer you are to one of these areas, the more concern you observe in terms of those measures that we use, the searches and the online activity.
What we find is in these areas that were shocked by the Ebola threat in this way, we find that Democratic vote shares went down. That’s the first key result. Then it raises the question of what’s driving this drop in Democratic vote share? One hypothesis would be maybe people are feeling anxious and upset and they’re kind of projecting that anxiety and that fear on the performance of incumbent politicians. They are looking at the president or governors or what have you, perhaps being frustrated by their response to this crisis. That’s one hypothesis. We tried to look at what is the impact that Ebola concerns have on the judgments that people were making about incumbent politicians, and the first thing is that we find that there is no impact on incumbent politicians if they are Republican. It’s just the Democratic side that is being affected.
Maybe the Democratic candidates are suffering by association with President Obama, and people are responding to Ebola concerns by having more negative feelings towards President Obama. Well, we can look at that because there is quite a lot of data on presidential approval ratings. Gallup runs daily polls about presidential approval ratings, so we can look at that and see how it responded to the emergence of these Ebola cases. We find zero effect. There is no effect on the approval ratings of President Obama as measured by Gallup. We have another survey as well, not in the same frequency, which we can use, and we find the same effects, so we really find no impact on the evaluation of President Obama. The question is what’s going on in the minds of voters and what’s going on in the political environment that might explain this ultimate impact in the vote share of Democratic politicians?
The other possibility is politicians are exploiting those things and raising those concerns and doing so in ways that are strategically in their interest. We look at the behavior of politicians and how do we measure that? Well, for members of Congress, they send these newsletters to their constituents, and we have the content of these newsletters, so we can actually look at the text and see whether they are mentioning Ebola. We can search for the word Ebola in those newsletters. We can also see if they are mentioning other things as they are mentioning Ebola as well. What we find there is that politicians, and Republican politicians, they immediately respond to the emergence of the cases in the U.S. by talking about Ebola and talking about Ebola not just in general, but in association with traditional Republican themes such as immigration or terrorism, and it’s very easy to come up with examples.
In the paper we have an example from then candidate Thom Tillis, he won the election in 2014. He’s now the senator for North Carolina, and he’s talking about how we have this Ebola crisis and the borders are a mess and we need to control the borders because this is coming from abroad. We have examples from Rand Paul as well. You had this anecdotal perception at the time, and what we see in the data using these newsletters is that they were doing that systematically. This is not just anecdotal examples. Newsletters are one source of information, but we also look at campaign ads. There is data available on the ads that candidates were placing at the time, and what’s very interesting for that is that these data, they code the appeal to fear that these ads made. They do that based on things like the use of ominous music and the language and the like. We can actually see whether they are responding by appealing more to fear. We find that they are doing that, and they are doing that, again, in connection with immigration and terrorism and President Obama as well. The Republican politicians are the ones that are doing that as well.
We’re actually adding now the analysis, so this is not current working paper version, but we’re working on that, shows similar results for Twitter activity of candidates. We have a number of different measures that all point sort of in the same direction, which is that Republican politicians were talking up the issue of Ebola and doing that in conjunction with themes that they tend to think are favorable to them. Then the question becomes did it work or not. Well, on the one hand, we saw that it didn’t seem to have an impact when it comes to President Obama’s approval ratings, even though Republican politicians were trying to tie Ebola to Obama in their tweets and ads and communications. We do find evidence from survey data that voters became more likely to report anti-immigration sentiments. This is not true for other attitudes, by the way, general conservative ideology or attitudes towards guns or gay marriage or religiosity, which are things that you might think of as having this conservative balance as well. We did not see action in any of those things. We do see action specifically related to immigration, which we also think is very intriguing and interesting from a policy perspective.
Matt Grossmann: Adida says that the electoral and experimental effects match up, though the survey experiment didn’t find localized effects.
Claire Adida: I think that the two findings nicely complement each other. The Republican position since the early 2000s on immigration has been increasingly restrictive, and we show that politicizing the Ebola crisis to influence attitudes towards immigration was effective, especially for some groups of people, but even in general, even among independents, it drew people closer to the Republican position. Now with our study, we can’t make the jump and say that’s going to have effects for actual vote share, but it is consistent with what this other study found, which is that the Democratic vote share and turnout went down.
One of the things that we did in the paper that I don’t know that it made it, I think it might have made it in a footnote in the published version, is we looked to see if there were stronger effects in Texas, since that is where the Ebola outbreak occurred, that’s where it started in the U.S., so we would expect maybe that we would find stronger effects among our respondents living in Texas. We didn’t get a special effect, a stronger effect for those respondents, so we couldn’t find any evidence of a more localized response.
Matt Grossmann: Campante agrees that the mechanisms and the findings line up, with the research confirming that the experimental effect shows up in the wild.
Filipe Campante: Generally speaking, the findings they line up, they are consistent. We’re finding that there is this impact on attitudes towards immigration, and we’re also finding that there was this strategic exploitation by Republicans which I think lines up well with the results from the survey experiment. Now what we had here is first thing, and I think that’s what we had with respect to the previous literature in general when it comes to thinking about the impact of fear and anxiety is that a lot of that literature is experimental, particularly you’re in a lab context or in a survey experiment in the case of that particular paper, but you’re priming people to think about certain things and then you’re measuring their response. What we’re adding here is that we’re looking at the impact of that not in the lab, so to speak, but in the wild, to put it that way. You might say, well, people might respond in a certain way when they’re being surveyed or when they are in a lab experiment, but does it really matter in an actual election? What we find is that it does, and I think that’s very much what we think we’re bringing to the table here.
Matt Grossmann: How about potential political effects of COVID? Both see a lot of differences. Adida says there are attempts to politicize COVID, but they might not succeed this time.
Claire Adida: I think we’re seeing attempts of politicizing this. President Trump has called the virus the Chinese virus. Many other Republican politicians have jumped into that, certain Fox News has, and we’re also seeing a jump in xenophobic and racist incidents reported against the Asian American community in the United States. There’s definitely the ugly racism is rearing its head here, just as racism has always been tied to the threat of public health and to disease. We are seeing that.
I do think that the context, though, is different in three ways. The first is that the threat is much more salient, it’s a much bigger scale. Only 11 people were treated for Ebola in the U.S. during that outbreak, and only, I think, one died. Compare this to the current crisis in the U.S. More than 550,000 cases, close to 22,000 deaths. The enormity of the crisis is much greater. It’s also global. It was not global for Ebola. Ebola was an epidemic, it was not a pandemic. Part of this may be due to the political response itself, but it’s also probably due to the nature of the virus itself, which my understanding is that it’s very contagious before symptoms emerge. The virus is very different that we are much farther into the crisis than we ever got with Ebola, so I think there’s a lot less room for shifting blame, even though you see these attempts. I don’t know that those attempts are going to be effective because we’re dealing with a very important crisis with people dying, and people are expecting a response. I’m not sure that the nature of the virus is really going to give politicians much room for shifting the blame.
The second question is just the polarization in the country right now, the political polarization. It was already pretty bad in 2014. It’s even worse today. There’s a new working paper out that has surveyed 3,000 American citizens just in March of this year. This is by Shana Gadarian, Sara Wallace Goodman and Tom Pepinsky. They found that the single most consistent factor differentiating Americans’ health behaviors and preferences in the midst of this COVID crisis is political difference, partisanship. My sense is that the polarization that we’re seeing in the political landscape that’s just been getting worse and worse is reflected again in how people are responding to the crisis. You find Republicans being quite satisfied with the Trump response, everybody else being dissatisfied. It just seems like there’s a lot less movement, there’s a lot less possibility to move people around on issues these days, so I don’t know that you would be able to see the kind of effects that we’ve identified.
The third difference is just the response. The Obama administration did not reject science at the time, whereas the Trump administration rejects science and vilifies the experts. I think what’s going to be, I think this is a mistake for the Trump administration. If we look at the political science research on this, and here I think Shana Gadarian and Bethany Albertson’s work on anxiety and politics is really informative. They show us that in a time of a public health crisis, the public wants to see the experts, the medical experts. They want to see protective policies. They want to see Dr. Fauci, they don’t want to see President Trump. President Trump is doing a lot of the President Trump show, and is even hinting at firing Dr. Fauci. I think this would be a terrible mistake for him. The response is different between then and now, and if we follow, if we believe the political science literature on this, I think that this is a mistake for the Trump administration.
Matt Grossmann: Campante agrees there are a lot of key differences between COVID and Ebola that changed the likely impact.
Filipe Campante: There are important distinctions here that I think make the outcome a bit hard to predict, the ultimate impact of coronavirus a bit hard to predict. I think the forces at play with Ebola when it comes to the fear and anxiety are going to be in play here, and even more so because this is a much, much bigger shock to the actual health risks that people are facing. This is a real large, large scale public health emergency in this country in a way that Ebola never came close to being. In that sense, this aspect that we isolate I think is going to be very important because the shock is going to be very big.
Now what’s different right now is that the shock is not just this psychological shock, but there’s also the economic shock. There’s also the policy response and the evaluation of that policy response, which can push in very different directions, so that also makes the ultimate effect hard to predict. If you focus on the fear aspect and the anxiety aspect, I think there are similarities and there are distinctions. The similarities, that again we have this virus that is coming from abroad, and I think here the association with foreigners and with immigration and so on and so forth is something that I think is very reasonable to believe that these associations will be triggered again, and I think we can see politicians already trying to exploit that. What’s different here is that whereas the incumbent in 2014, the president, was a Democrat, right now the president is a Republican, so that changes the way the political incentives are playing out here.
Now I think what we find is that the pure anxiety shock doesn’t seem to connect directly to the way you think about the incumbent necessarily, so it’s not as if it’s leading to a more negative evaluation of the incumbent, even if at the time the opposition party was trying to build that association. On the other hand, obviously right now, again the impact is much bigger than it was for the case of Ebola in 2014, so the importance of the policy response is much greater now, which I think, again, makes it harder to predict which political forces might be harmed or might benefit from this shock now.
Matt Grossmann: He says Republicans are trying to make the connection to China, which gives them a clearer strategy than Democrats, but it’s unclear if it will work for either side.
Filipe Campante: On the Republican side, it’s pretty clear from the actions of many Republican politicians and President Trump as well what the playbook is, which is a very similar playbook from the Ebola, which is trying to tie this to foreigners and to immigration, so talking about the quote, unquote Chinese virus or the Wuhan virus and making that type of association and pointing to the response of, oh, we closed down the border to China, whatever the actual nature of the policy response was. I think that part of the playbook is easy to draw, and I think they’re already doing it.
On the Democratic side, I think it’s more complicated because I think the message that comes from our paper is that you have to find an association that resonates with voters. Obviously the immigration and the foreigner association is not one that I think plays to the benefit of Democrats, so if I were to put my political strategist hat on the Democratic side, I would be like, okay, we need to think about something that can resonate with topics that are favorable to us, so maybe something about anxieties associated with health care or the economic fallout of this. Those were the things that I would speculate, wearing that political strategist hat.
Matt Grossmann: Where do we go from here? Adida plans more research on immigrant scapegoating, but she sees possible solutions in some framings of the COVID crisis.
Claire Adida: My research topic of interest, what really gets me up in the morning and excited to do research is the study of migration and how people respond to migrants and understanding under what conditions migrant communities face exclusion or assimilate. I think there’s been a lot of very rich literature and rich research in the social sciences to understand what leads to exclusion. When are migrants scapegoated, when are they excluded? Just how powerful threat narratives can be to exclude people who aren’t like us. We know a lot about these questions. My recent work looks at how we can foster inclusion instead. I look at how things like perspective taking, putting yourself in someone else’s shoes can mobilize people to act on behalf of vulnerable minorities or can move people to hold more inclusionary attitudes towards immigration. Right now, in light of the COVID crisis, I’m looking at how narratives of refugees contributing to the fight against COVID could influence public opinion towards refugees and towards refugee policy.
More generally, can you move these boundaries, these in group, out group boundaries for people? Can you shift how people think about who is a member of my group or not? Who do I include when I think about solidarity, and especially in a time of crisis like COVID, which is affecting everybody at the same time, is this an opportunity to increase inclusion, and if so, how can we do that, since there’s so much work already showing that you can do the negative stuff. You can scapegoat migrants, you can scapegoat refugees, you can blame racial minorities. Are there messages, is there political communication or rhetoric that can do the opposite, which is to increase inclusion of these minorities?
Matt Grossmann: Campante says coronavirus research may be harder, but there could be a chance to compare the effects of media coverage of disease on public opinion.
Filipe Campante: One challenge that will obviously face the study of the impact of coronavirus is that it’s just this massive shock that is taking place all over the country. I mean there’s variation in that, but I think it’s a little bit hard to think about sources of what we would call exogenous variation, things that induce more or less exposure to the coronavirus, but that are not correlated with other features that also affect political behavior. I think there are ways to think about that, but that will be one major challenge that is I think harder to solve in the current context.
In terms of our own research, the one aspect that we’re interested in and that we haven’t dug into as much so far, and we plan to, is to think about the role of the media in all of this. I think there is a lot of anecdotal evidence, and if you just look at the data on the mentions to Ebola in TV news coverage, it reaches this peak right in the middle of the last month of campaign, and then it just collapses and goes to basically zero right after the election. Then there’s a question of to what extent or what was the role that the media play in amplifying the concerns with Ebola, and did that help explain the patterns of political impact that we find in the data? That’s something that we’re very much interested in digging into going forward.
Matt Grossmann: There’s a lot more to learn. The Science of Politics is available biweekly from the Niskanen Center. I’m your host, Matt Grossman. Thanks to Claire Adida and Felipe Campante for joining me. Please check out Ebola, Elections and Immigrations, and The Virus of Fear, and then listen in next time.