TODD LANDMAN: I'm Todd Landman.I'm a professor of political science.And for the last 20 or so years, I'vebeen looking at the question of human rights more generally,but I've also really taken an interestin the systematic study of human rights problemsas they relate to development, democracy,and other global issues.And in order to do that, I've looked at comparative methods.
TODD LANDMAN [continued]: And one part of the comparative methodological storyis the importance of case studies.So in this tutorial, I'm going to talk about the value of casestudies, the selection of case studies,what case studies can do and what they cannot do,and of course the limitations of case studies.So I hope you find that useful, because case study'sa very popular form of political analysisand comparative analysis more generally.
TODD LANDMAN [continued]: It's very popular these days in political scienceto be rather skeptical of the value of case studies.And by that I mean the sort of single-country studythat is, some would say, quite descriptive.And there was one study that was criticizedas being lots of evidence but not much inference.And that study is John Womack's Zapataand the Mexican Revolution.
TODD LANDMAN [continued]: This could be described as a descriptive or narrativehistorical account of the state of Morelos in southern Mexico,and this character of Zapata and how he contributedto the Mexican Revolution.That could be one simple reading of this book.However, I would make the claim that there'squite a lot of causal inference in this book,and it does relate to theories of revolution, et cetera.
TODD LANDMAN [continued]: And if I read just the opening sentence from the prefaceto start, there's actually a really interesting statementhere.Womack says, "this is a book about country peoplewho did not want to move and therefore gotinto a revolution."That single sentence contains a hypothesis, or a propositionif you will.It's about country people, so we have a definitionof a type of people.
TODD LANDMAN [continued]: And that could be further refined.They didn't want to move, so theywere resisting some outside force wanting them to move,and therefore got into revolution, which suggeststhat because they were being asked to move,they got involved in revolution.And there is a passage at the end of the bookthat I'm fascinated with and it's gripped me for years.
TODD LANDMAN [continued]: Womack says, "new attitudes, new policies, new laws,new agencies, new authorities, and of the plain country peopleof 1910, about 3/5 remained.They had won a victory too, simply in holding onas villagers, not in refuge in the state's cities or huddledinto the haciendas, but out wherethey felt they belonged, in the little towns and pueblosand ranchos, still reeking at least a "pacific Zapatismo.
TODD LANDMAN [continued]: In 1910 the bases of the only life they wanted to livehad been breaking down.Although they wore themselves out dutifullytilling their scattered patches of corn and beans,now and then trading a horse or a cow for a few pesos,marketing eggs, tomatoes, onions, chilies, or charcoal,tending their scrubby orchards, desperatelysharecropping on the planters' worse land,they had nevertheless lost the struggleto keep their communities going."This rich description, if you will, of the life of a peasantor a country person in Mexico tells usa lot about the experience of those lives,and the experience of change in those lives,and how that change led to peoplebeing mobilized for revolution.
TODD LANDMAN [continued]: And it's that sort of relationshipbetween people's lives, external forces that change those lives,and the motivation to be involvedin violent political activities such as a revolution asexperienced in Mexico is something thatshould be of interest to scholarsof comparative politics and international relations moregenerally.So what I try to do is explain how a case study canmake a contribution to these types of statements thathave larger implications for political processesin countries outside the case that was originally studied.
TODD LANDMAN [continued]: The Mexican case is a simple illustrationof the plethora of case studies available to us as researchers.There are 194 independent nation-states in the world.And of course, the history of the worldis very rich and very deep.My students often come to me and say,I'd like to do a case study on Guatemala.I'd like to do a case study on South Africa.And then I say, what would you like to researchin Guatemala or South Africa?
TODD LANDMAN [continued]: Well, I want to work on human rights.Well, why have you chosen Guatemala or South Africa?Well, the United Nations says these are terrible places.And then I say, well, there are a lot of terrible placesin the world.So why have you chosen Guatemala and whyhave you chosen South Africa?Oftentimes students stumble at that very basic question.What is the function of the case study?What is it for, or what is it a case study of?
TODD LANDMAN [continued]: In the case of Mexico, we know it'sabout the Mexican Revolution.We know it's about a particular character, Zapata,and the state of Morelos.But we also know it's about land,people being force from that land,and the motivation for being involved in revolution.But when I probe my students' mindsand say, why is it that you reallywant to work on that country, I'mreally asking them, what's the puzzle that needs to be solved?
TODD LANDMAN [continued]: Have you found some sort of anomalyor some sort of question that youwant to answer by researching that single case?And if you do it through that case, what willyou learn from that case?And what might you learn from that case that couldbe applied to other cases?In a seminal essay in 1975, which in my viewis still the very best essay ever writtenabout case studies, by Harry Eckstein,he says that there's a set of functions around case studies,that they serve different functions.
TODD LANDMAN [continued]: There's sort of the clinical studyof a one-off, single descriptive account,a kind of in-depth clinical analysis of one case.But case studies can also do thingslike weaken theories, infirm theories, he says.So you have a dominant theory about social relationships,and yet in one of the cases you find it doesn't work there,so that kind of weakens a general proposition.
TODD LANDMAN [continued]: They can also be theory-confirming.So you might have a general theory about somethingand you find a case that upholds that theory,so that's a theory-confirming function of a case study.Case studies can also be used to generate hypotheses.So for example, if I had gone to Mexico in 1910and sort of walked around for a whileand looked at what was going on, I might think, hmm,I think there might be a revolution brewing here,because the country people are getting agitated,they don't want to move, et cetera.
TODD LANDMAN [continued]: So imagine if we cast ourselves back in historyand we looked at the circumstances within that case,we might say, there's a hypothesis hereabout the relationship between people,land, power, and violence that wemight be able to test in this particular case study.So in that sense, the case study is nota sui generis it's an interesting place to work on.
TODD LANDMAN [continued]: But it is related to social theory about change,about power, about conflict, about all those topicsthat interest us in political science.So I go back to my students and say,where is that case study located in a larger set of questionsthat you're really motivated by?And what will you learn by studying that case study thatmight apply to other cases?
TODD LANDMAN [continued]: Now two of the main functions that I reallylike about case studies include the notion of a most likelyor a least likely case study.So let me explain what that means.And I'll just give you a little illustration to do that.So if you could imagine two variables-- and I'mgoing to draw a little graph here in a minute.But if we had a measure of-- and I'm writing here "development,"and we had a measure of democracy.
TODD LANDMAN [continued]: So I'll just draw these for the momentand hold this up so you can see.So at the bottom we have development.At the side we have democracy.Imagine you had those two variables.And you had a measure of development,let's say it was the per-capita GDP in a country,and you had a measure of democracy.One of the popular measures we use in political scienceis the Polity IV measure of democracy.
TODD LANDMAN [continued]: It's kind of a negative 10 to positive 10measure of democracy.A negative 10 is authoritarian, positive 10is fully democratic.Now we know statistically ever since really the late 1950suntil today, that there is a positive and significantrelationship between the level of economic developmentand democracy.So positive and significant relationshipis captured by this straight line rising up in this manner.
TODD LANDMAN [continued]: We know that as levels of development increase,countries tend to be more democratic.Some say it's a causal relationship.Some people think it's just a sort of correlation,if you will, an elective affinity between these twovariables.What we do know is that rich countriestend to be democratic.Now what would a most likely and least likelycase study look like?
TODD LANDMAN [continued]: Yes, there is a positive and significant relationship.But not all the countries in the world are on that line.So we would have some countries, whichI will represent by dots in this graph, that wouldbe quite close to the line.So the dot would be a country location.This would be its measure of development.This would be its measure of democracy.So that would be its unique location within whatwe call a scatterplot.
TODD LANDMAN [continued]: The line itself describes the relationship more generally,but the countries make up the scatterplot itself.Now, they're quite close to the line.And we could say that they confirm the relationshipbetween development and democracy,because they're very close to that line.However, there might be some countriesthat are off the line.
TODD LANDMAN [continued]: So I call this my Saudi Arabia, Costa Rica problem.So I'm going to write Saudi Arabia here,and Costa Rica here, and then show those to you.Now what I'm saying here is that here wehave an example of a country thathas high levels of economic development,but isn't particularly democratic.
TODD LANDMAN [continued]: And here we have a country that haslow levels of economic development, but is democratic.These are what we call outlier cases, or deviant cases.They're very far away from the overall relationship,so there's a distance between the position of this countryand the line itself, a distance between this countryand the line itself.
TODD LANDMAN [continued]: The distance here is quite negative.We call this a negative residual.And this is positive.We would call this a positive residual.We call them outliers because they in partconfound our general relationship.It leads us to ask more interesting questions,like in the case of Costa Rica, why is it democratic and poor?
TODD LANDMAN [continued]: How did it become democratic even though it wasn'tparticularly well developed.The per-capita GDP in Costa Rica over the 20th centurywasn't very high.It made a kind of upper low-income countryrather than a high-income country, measured by World Bankstandards.So what was it in Costa Rica thatled to its democratization?So Latin American scholars would say, well,there was a big peace agreement in 1948.
TODD LANDMAN [continued]: They got rid of their military and police force.And they invested that money in education,and that led to the development of a democratic cultureand a kind of elite bargain that hasworked since 1948 through the rest of the 20th century.International relations scholars would say, yes,and then you add in the protection of the United Statesfrom the north.They didn't need an army.If they ever got invaded, the United States would help themout, et cetera, et cetera.
TODD LANDMAN [continued]: So there was a sort of set of explanationsthat sort of accounted for the exceptionalityof the Costa Rican case.So here, this deviant case is another anomaly.Why in the case of Saudi Arabia is there great wealth, but notparticularly democratic?And of course, we would then lookat things like the concentration of oil, the factthat the development of Saudi Arabiais very much based on the concentration of oneasset, the oil.
TODD LANDMAN [continued]: There might be cultural explanationsaround the way in which the Saudi kingdom runs its affairsand the sort of, if you will, skepticismabout democratic institutions and the full protectionof human rights, et cetera.So it becomes an outlier case as well.Now when we talk about most likely and least likely cases,we're using that kind of logic and we'resaying of all the places in the world we expect a country notto be democratic, and it is.
TODD LANDMAN [continued]: So in the Costa Rican case, we'd saywe don't expect Costa Rica to be democratic because it'srelatively lowly developed.We do expect Saudi Arabia to be democratic, but it's not.So that's the obverse of the argument.So here we have this mixture of most likely and least likelycases.The intense investigation of a single case like Costa Rica,or an intense investigation of Saudi Arabia,will tell us things about this overall relationship thatenriches our understanding.
TODD LANDMAN [continued]: And for me, that really representsthe value of a case study.It's not a sui generis case.It is sitting there in relation to a larger set of questionsand relationships in the world.By investigating Costa Rica or investigating Saudi Arabia,we would learn something about that overall relationship.And we would have a conversation between the in-depth countrystudy done in either of these cases,and a global study looking at relationships all together.
TODD LANDMAN [continued]: Now despite all the many benefitsand the value of case studies that I'vebeen discussing in this tutorial,there are naturally some limitations.We have to be careful about the types of inferencesthat we draw from single case studies.So there are all sorts of questionsaround which case study was selected and why.Was it a most likely case study, a least likely?Was it an infirming or confirming case?
TODD LANDMAN [continued]: And why was that case chosen, or was itjust an in-depth clinical analysisor a hypothesis-generating case?I think the more explicit a scholar canbe about what the case study is trying to achieve,then that actually locates the case study much better.Clearly there are limits to the inferences we canmake from a single case study.I might learn a lot about democratic transitionin Chile by an in-depth study of military documents and factionswithin the military, mobilization from below,from social movements, pressure from the international systemthat brought about change in Chileduring the 1980s, a plebiscite in 1988, the defeat of Pinochetin an electoral process.
TODD LANDMAN [continued]: All those sorts of things would be verydetailed in the Chilean case.I might learn that there are some inferencesfrom the Chilean case, so I mightlearn that the conflict between those factions led to a break,that the rule-oriented faction within the Chilean militarywas more interested in international legitimacy,et cetera.And there's a wonderful case studyby Darren Hawkins to this effect thatgives us that deep history.
TODD LANDMAN [continued]: But what he does in that case study, which is interesting,is he says, I think these inferences from Chilemight apply to the case of Cuba, no democratic transition,and the case of South Africa, democratic transition.So he takes the inferences from Chileand then applies them to these other two countries.However, that's where he stops.And I think it's really intellectuallyhonest for a scholar to say, I'm not sure if what I've said hereapplies to the whole world or to the universe of cases thatinvolve democratic transition.
TODD LANDMAN [continued]: And it's that self-limiting nature of the study that Ithink is really important.Again, if we go back to Zapata and the Mexican Revolution,the story that was told about the state of Morelos,the country people not wanting to moveand therefore getting involved in revolutionhas a set of universal aspirations in terms of itmight apply to other cases of revolutionin Vietnam, Angola, Peru, et cetera.
TODD LANDMAN [continued]: But it might not.And there's a really interesting sortof conversation between this case study and future casestudies.So we have to be very cautious in making inferences,be very honest about the degree to which those inferences applyto the rest of the world, and the degree to which theydo not.There may be features to the case studythat really are unique to that country.
TODD LANDMAN [continued]: And we don't want to over-egg our explanationand say that it might apply to the whole world.So there are limitations to case studies,but nonetheless I think the value that I'vedemonstrated in this tutorial is definitelyserious food for thought.This tutorial has been about the value of case studiesand comparative research.
TODD LANDMAN [continued]: I made reference to case studies reallyas a single-country study.They might also refer to a subnational partof a single country.But by and large, we're looking at the valueof looking at single countries and whatthat might tell us about larger political processes.I talked about the selection of cases,that there are functions, if you will,of case studies from the purely descriptive clinical analysisall the way to a most likely, least likely deviant caseanalysis, where the goal of the analysisis really to make that case study talk to largerquestions in political science and comparative politics moregenerally.
TODD LANDMAN [continued]: I've also talked about the limitations of case studies,that we don't want to over-egg inferencesand think that we've discovered the full proof of a theory thatwill apply to all countries in the world.Equally, we may not want to say that our case study rejectsforever a particular popular theory.So we need to be a little bit humble about our case studies
President Donald Trump is a Republican. But often, and more so than any president in memory, he lacks a consistent political ideology.
During the campaign, Trump took five different positions on abortion in three days. On other issues, his policy preferences have been clear as mud: “I don’t want to have guns in classrooms, although in some cases, teachers should have guns in classrooms, frankly,” he told Fox News in 2016. He’s quick to engage in public feuds with members of his own party. He’s willing to rebuke his own attorney general, and has shown willingness to work with Democratic leaders Nancy Pelosi and Chuck Schumer on legislation to protect undocumented immigrants who came to the country as children.
And that makes this period of history extremely interesting for political scientists and psychologists to study.
“We’ve never had a federal elected official, let alone the leader of a party or the president of the United States, who is so easily moved from one position to another without offering any sort of justification or apology or explanation,” Michael Barber, a political scientist at Brigham Young University, says.
Researchers like him have long tried to understand the power of leaders and the willingness of the public to hold them accountable. And rarely do they get a real-life experiment like Trump to help them answer some huge questions at the heart of democracy:
How much power do presidents have in swaying public opinion? Will the base always follow even if a president swings wildly from one position to another?
And there’s another key question at play here that matters a lot for the future of Trump’s presidency: If prominent Republican leaders start breaking away from Trump because they don’t like his ideas or temperament, will the base also break?
In essence, Trump is providing an unprecedented opportunity to test the “follow the leader” theory of political science — and to see if Trump will push his followers to a breaking point.
A recent experiment found Republicans are more likely adopt liberal policies when Trump supports liberal policies
Yes, Trump has been extremely consistent in some areas: He’s made policies that make it harder for people who live in Muslim-majority nations to enter the United States, he shares Republican desires to end Obamacare, and he’s reducing the role of the Environmental Protection Agency and other federal agencies. But for a president, Trump is historically scattershot.
In January, Barber and his BYU colleague Jeremy Pope designed an experiment to take advantage of that fact. They wondered: Are Trump’s supporters ideological, or will they follow him wherever his policy whims go? Right after Trump’s inauguration, they ran an online experiment with 1,300 Republicans.
The study was pretty simple. Participants were asked to rate whether they supported or opposed policies like a higher minimum wage, the nuclear agreement with Iran, restrictions on abortion access, background checks for gun owners, and so on. These are the types of issues conservatives and liberals tend to be sharply divided on.
Barber and Pope wondered: Would Republicans be more likely to endorse a liberal policy if they were told Donald Trump supported it?
One-third of the participants read statements where they were told Trump supported a liberal position, like this:
Please indicate whether or not you support or oppose the statement.
To increase the minimum wage to over $10 an hour. Donald Trump has said that he supports this policy. How about you? Do you support or oppose increasing the minimum wage to over $10 an hour?
The control group of the experiment saw these questions, but they didn’t mention Trump. And another arm of the experiments tested what happened when Trump was said to support conservative policies.
The answer: “On average, across all of the questions that we asked, when presented with a liberal policy, Republicans became about 15 percentage points more likely to support that liberal policy” when they were told Trump supported it, Pope says. They follow their leader. “The conclusion we should draw is that the public, the average Republican sitting out there in America, is not going to stop Trump from doing whatever he wants.” The effect even held true on questions about immigration. If Trump supported a lax immigration policy, his supporters said they did too.
(Past experiments with liberal participants have found a similar effect: Liberals are more likely to support conservative policies when told their leaders support conservative policies.)
It’s important to note that not everyone is so easily swayed by their leaders. In their study, Barber and Pope found the people most likely to follow Trump’s lead were those who didn’t know much about politics but still strongly identified as Republicans. The most knowledgeable in their sample were hardly swayed at all. To a certain extent, the effect may be the result of people not thinking too hard when filling out a questionnaire. But then again, that’s how most of us come to our political opinions: by not thinking too hard about them.
Another note: This paper has yet to be published in an academic journal. However, I asked several other political scientists about its merits, and they thought the experiment was well designed and conducted.
There are real-life examples of this leadership-down process of opinion that go back years. In 1971, Republican President Richard Nixon made a surprise decision to impose a 90-day freeze on wages and prices (to halt inflation, a move that runs counter to conservative economic policy), and even ardent conservative supporters followed suit.
“A Columbia survey ... found a 45-point increase from 32 percent to 82 percent ‘virtually overnight’ among Republican activists — precisely the people who ought to have been most resistant to the policy shift on ideological grounds,” explains Larry Bartels, a political scientist at Vanderbilt University and a leading expert on how public opinion forms.
Other research has found that many of us will change our opinions merely from learning a leader supports an idea.
In 2014, political scientists David Broockman and Daniel Butler conducted a field experiment where they had real state legislators’ offices send letters to constituents who the researchers knew, via another survey, disagreed with the legislators’ policies. The letters simply stated what the legislators believed. A control group got no letters. The letters were sent by Democratic lawmakers, but they didn’t state their affiliation in the letters.
The letters — sent to people regardless of the political affiliation — made a small but meaningful effect. They “caused the voters who disagreed with the legislator to become about 6.5 percentage points more likely to agree with the legislator,” Broockman and Butler wrote.
A follow-up experiment added another variable to the mix: Some of the constituents got letters that contained lengthy explanations of their leaders’ policy views. Others just got short summary statements. And “constituents who received lengthy arguments from legislators justifying their positions were just as likely to change their opinions as constituents to whom legislators provided little justification,” they wrote.
The experiment is a demonstration that it doesn’t take all that much to get voters to flip-flop their political opinions. They might not even realize they’re doing it at all.
Trump may have caused some massive swings in public opinion among Republicans
Consider this recent one.
In 2015, just 12 percent of Republicans held a favorable view of Russian President Vladimir Putin, according to Gallup. It’s not surprising this figure was so low. Just three years earlier, Republican presidential candidate Mitt Romney had declared Russia to be our “top geopolitical foe.”
Flash-forward to 2017. Now 32 percent of Republicans have a favorable view of Putin, according to Gallup. That’s a huge jump. And during that time, if anything, Putin lived up to Romney’s prediction.
What changed in just two years? Donald Trump, and his reluctance to criticize the Russian leader.
Or take the issue of free trade: Typically conservatives are in favor of it. It lowers costs for American businesses that want to outsource labor or buy materials from overseas.
“Since Donald Trump entered the 2016 presidential campaign by descending a golden escalator, Republican support for free trade has similarly declined,” Barber and Pope write in their paper. As Pew finds, from 2015 to 2017, Republican support of free trade plummeted from 56 percent in support in 2015 to just 36 percent in 2017. That’s a 50 percent change in opinion.
Trump has massive power to sway public opinion among his followers.
In March, a CBS poll found that 47 percent of Americans — and a full 74 percent of Republicans surveyed — believed it’s “likely” or “somewhat likely” that President Trump’s offices were wiretapped during the 2016 presidential campaign. Trump had tweeted that “Obama had my ‘wires tapped’ in Trump Tower” during the campaign. But there was no credible evidence this happened. Since then, it’s been revealed that Trump’s former campaign manager Paul Manafort had been under surveillance, which confuses the matter. Still, Trump said it was so, and many believed.
And there’s an opposite of the follower-the-leader effect at play too. When Trump takes a position on something, liberals are sure to oppose it. Vox’s Matthew Yglesias has documented these swings in public opinion here.
Very few people have stable policy views. They do have stable party affiliations.
We have a vision of democracy where a virtuous public steers the decisions of leaders, Barber says. But over and over again in experiments — and in the real world — that vision turns out to be a mirage.
The researchers I spoke to admit there’s a bit of a chicken-and-egg problem in this research. Can we really know that Trump caused the change of opinion on free trade, or did Republican voters change their minds and then endorse an anti-free trade candidate?
Both seem to be going on. But there’s a compelling case that Trump caused it.
Gabriel Lenz, a political scientist at UC Berkeley and author of Follow the Leader? How Voters Respond to Politicians' Performance and Policies, has analyzed decades’ worth of data from the American National Election Studies, a long-running study assessing voter preferences and knowledge about politics. The study data allows Lenz to track the very same voters over decades of political change.
One of the biggest findings in all this data: Voters aren’t all that knowledgeable about politics. Lenz finds only 50 percent of the public can accurately match policy preferences to the correct party or candidate. And about 20 to 40 percent of the public holds stable views on policy, Lenz finds in a forthcoming paper in the Journal of Politics.
“A lot of people don’t know how the parties describe themselves,” he says. “And when they learn my party’s conservative [or is against free trade, etc.], they start saying they’re conservative too. ... It’s very hard to find instances where prior policy views seem to drive later voting or decisions.”
A lot of us are like Trump in that regard.
In his research, Lenz finds some people are more likely to be consistent on certain topics, like marijuana legalization, abortion access, and anything related to racial identity. But overall, “we have ample evidence that people are amazingly good at ignoring contradictions.” People can say they support free trade one year and say they’re against it the next, and not really notice their opinion has changed.
When a party leader vocalizes a strong opinion on an issue, many people will just accept that view. In the mind of the voter, it’s replacing a hard question (“How do I feel about international trade taxes?”) with an easy question (“What team am I on, and how would they answer this question?”).
One reason Trump is particularly influential: he’s a “toxic meme” machine
Trump has a powerful ability to sway opinions among his base — and create polarized opinions among his detractors. We see this play out all the time.
Think about his recent fracas calling out NFL players protesting the national anthem.
The message to Trump’s base is clear: Only ungrateful un-Americans would dare protest the national anthem. His supporters — many of whom perhaps haven’t even been paying attention to the anthem controversy — will believe that too.
When Trump comes out with a strong opinion on a subject, it’s a signal to his base saying, “This is what we believe.” He draws a clear line of what’s “good” and what’s “disgraceful.” It draws a massive amount of attention to a subject, and drives a wedge.
A common Trump tactic is to take something nonpolitical, or not always obviously political (like the NFL) and use it to turn people against another. This isn’t to say there’s nothing worth debating or caring about in the NFL protests. Trump just amps up the temperature in a way that’s sure to pit people against one another.
As Yale professor Dan Kahan puts it, Donald Trump is a “toxic meme” generator.
Toxic memes are stories or ideas “that are predictively likely to trigger the sense that it is us against them,” Kahan explained earlier this year at a scientific conference. “The special danger of Donald Trump,” he said, “is that he can drag issues across this line. He’s the president of the United States. He’s going to get publicity for these kinds of statements, ones that he knows will end up dividing people.”
Look at what’s happened. After Trump directed a massive amount of attention toward the NFL protests, and became so outspoken about it, a Reuters poll found 82 percent of Democrats disagreed with the president, saying football owners should not fire players who kneel during the national anthem. Twenty-nine percent of Republicans said the same. Suddenly the whole country is screaming at one another over football. The issues at the heart of the NFL protests are important. But Trump has turned them into a provocative distraction.
All political leaders have the ability to polarize. But few have done what Trump has done: built a massive social media presence around drawing stark lines between the people who believe in “Make America Great Again” and the “losers” who don’t.
Trump makes it exceedingly clear to followers where their opinions on matters like the NFL anthem protests, Obamacare, and other weighty matters should lie. He’s a master at tying these issues to a proud American identity. If you identify as a proud patriot, and are told that patriots despise NFL owners who let players kneel during the anthem, you’re going to despise them too.
Prominent Republicans and conservatives have spoken out against Trump. It’s (probably) not enough to break Trump’s influence over Republicans.
Politicians are usually ideological. They usually conform nicely to what the party believes. There haven’t been too many tests (aside for the ones mentioned above) of the follow-the-leader theory in the real world. “It’s so rare for a leader of a party to come out and say, ‘We’re going to do this thing the base doesn’t like,’” Barber says.
It’s tempting to imagine a scenario in which Trump’s base leaves him — that if enough prominent Republican leaders and conservative thinkers make a clear, damning break with the president, they can lead Trump’s followers to a more ideologically pure pasture. In that scenario, Republicans hold on to their group identity while decrying Trump as a betrayer.
There are a few recent instances that suggest some cracks are forming. After Trump held a meeting with Pelosi and Schumer on DACA, some conservatives spoke out, loudly. Uberconservative Rep. Steve King of Iowa tweeted that Trump had “blown up” his base.
Trump is even losing Ann Coulter on the DACA issue.
And now Trump is publicly feuding with Republican Sen. Bob Corker. In an interview with the New York Times, Corker said he feared Trump’s temperament could set the world “on the path to World War III.”
But let’s not hold our breath and wait for the moment when the Trump base leaves him en masse, though it would be a fascinating moment for political science if that happens. And there are some signs it could. Recent polls are showing signs his base is becoming disenchanted: 67 percent of Republicans approve of the president’s job performance, an AP-NORC poll finds. That’s down from 80 percent in March.
For now, here’s what the political psychology suggests: A mass exodus ain’t going to happen. Recently, Broockman and his colleague Josh Kalla published a meta-analysis on 49 field experiments in which campaigns attempted to sway voters’ minds. They found evidence that voters’ opinions on issues can change, but not their opinions on candidates. That is to say: There’s scant scientific evidence to say that door-to-door political campaigns can dissuade voters from liking their chosen leader.
It’s also not certain that prominent Republicans could be the catalyst to lead a revolt. In their experiment, Barber and Pope found no evidence that “congressional leaders” could sway Republicans to adopt liberal views. The trick only worked with Trump’s name.
“Only the most vocal, loudest messages come through, and on that, it is hard to beat the president,” Lenz says.