One interesting thing about the recent scandal of the retracted study on voter persuasion (see my earlier post for details and Will Moore’s post for background and perspective) is that, as a rare example of political science in the news, it gives us some sense of outsiders’ perspectives on our field.
And some of these perspectives are waaay misinformed. Which, I suppose, is ultimately out fault for not communicating well.
Communication to the general public is part of the Monkey Cage’s mission, so I wanted to clear up one of these points of confusion here.
In a recent article in the Weekly Standard, Andrew Ferguson reviews the scandal in question and writes:
For years, social scientists have believed (scientifically) that it is extremely difficult to change another person’s political opinions, especially if the other person is an “everyday American” and not a social scientist.
So far, so good. Indeed, David Rothschild, Sharad Goel, Doug Rivers and I recently wrote a paper showing evidence that very few Americans changed their mind during the final month and a half of the 2012 campaign, and we argued that, more generally, public opinion swings are smaller than we have all been led to believe. So, yes, I do think it is difficult to change another person’s political opinions.
I don’t know where Ferguson got the quote about “everyday American” as this seems to be more of a journalistic phrase than something from social science, nor do I know of much research claiming that it’s easy to change social scientists’ views, but I’ll take that last phrase of his as just a bit of poetic licence.
Ferguson continues:
Multiple studies and mounds of research—by political scientists, sociologists, social psychologists, all kinds of scientists—have confirmed the bullheadedness of everyday Americans.
I’m not sure what is meant by “bullheadedness” here, but if it means “hard to persuade,” than fine. Political scientists talk a lot about “nonattitudes,” but when it comes to the decision of how to vote in a general election, persuasion is difficult.
Opinions do change, though. For example, something like three-fourths of Americans supported health care reform, but that went down to half during the specific battle over the Obama plan. Other well-known examples of public opinion swings include interracial marriage, same-sex marriage, and various foreign policy issues.
So I think “bullheadedness” is a bit strong. I’d prefer to say that, on certain issues, politically active people tend to have strong opinions which of course are not so easy to change.
But then, then comes the kicker. The place where I realized how poor our communication has been, that political science research has not even reached political journalists, which makes me scared to think what the general public might think.
Ferguson writes:
People’s attitudes and opinions are not the consequence of argument or experience, research revealed, but rather of unreasoning bias and emotion. The studies proving this, by the way, are not the consequence of bias and emotion. They are designed by scientific researchers studying other people.
The last two sentences here don’t have much content. I mean, of course we study other people. If we didn’t study other people, our work would just be introspection. If you’re a reporter, you study other people every time you go out on the street to observe or interview people, and every time you report a poll result. Bias and emotion are involved—they inevitably are in just about everything we do, we’re only human—but social science without studying other people would be pretty limited.
Anyway, that’s not my point, I just wanted to keep Ferguson’s remarks in context. The line of his that really bothered me was this:
People’s attitudes and opinions are not the consequence of argument or experience, research revealed, but rather of unreasoning bias and emotion.
No! Nononononononononono. No no no. No.
I have no reason to think Ferguson is being malicious here; I can only assume he is summarizing social science research, as he sees it.
But this is not what is found by mainstream research on opinion and voting behavior. For example, in our 1993 article, Gary King and I write:
Based on the enlightened preferences hypothesis, we conclude that the news media have an important effect on the outcome of presidential elections—not through misleading advertisements, sound bites, or spin doctors, but rather by conveying candidates’ positions on important issues.
The basic idea is that people have strongly-held beliefs and act on them in the voting booth.
“Unreasoning bias and emotion” (in Ferguson’s words): sure, these are part of life. But political scientists do not see them as the main cause of people’s attitudes and opinions. As emphasized by the authors of Freakonomics, people respond to incentives. And they make their decision in light of partisan identification and political ideology.
But I can see how Ferguson got his wrong impression.
Consider the following headlines:
“The Ancestral Logic of Politics: Upper-Body Strength Regulates Men’s Assertion of Self-Interest Over Economic Redistribution”
“The Fluctuating Female Vote: Politics, Religion, and the Ovulatory Cycle”
“Female hurricanes are deadlier than male hurricanes”
“Here’s how a cartoon smiley face punched a big hole in democratic theory”
“Liberals smell better to other liberals than to conservatives”
These headlines came from Psychological Science (the official journal of the Association for Psychological Science), the Proceedings of the National Academy of Sciences, . . . and. ummmmm, the Monkey Cage! All these claims are, in my judgment, neither plausible nor well supported by data (see discussions here, here, here, here, and here). But these sorts of dramatic claims get a lot of attention. Enough so that I have to keep reminding people: Just cos something’s counterintuitive, doesn’t mean it’s correct!
A common feature of much of these headline-social-science claims is indeed that ordinary people are “biased and emotional” as Ferguson puts it and probably shouldn’t be trusted with anything as important as the vote. I share Ferguson’s irritation at such an attitude; as he puts it, there’s something odd about assuming that ordinary people are full of biases without recognizing one’s own problems.
But this work is not the mainstream of social science! Far from it. Yes, it’s well known that people are misinformed about a lot of things; lots of survey research shows high levels of confusion on various important issues such as the ethnic composition of the country (people tend to overestimate the percentage of minorities) or the Federal budget. But when it comes to partisanship and core ideological issues, people tend to have coherent views and are not easily persuaded. Same-sex marriage has received lots of attention recently in part because it is unusual in that there has been a big change in attitudes in America and elsewhere on the topic during the past two decades.
In summary, I think we have to do a better job of conveying the mainstream of social science, our core views.
I have no animus toward Ferguson here. He was legitimately annoyed as what he saw as the silly and biased claims coming from credentialed social scientists. And it is a problem, that often it seems that the silliest, shakiest work is what gets the attention. Debunking silly claims one by one, that’s fine, but the big picture is that some of the messages we are sending are counter to the core of social science.
Let’s do better, and let’s thank journalists such as Andrew Ferguson, who, by honestly sharing their misconceptions, can remind us how garbled our message has been.