ABC News webcast interview on corrections research

I did an interview about my corrections research (co-authored with Jason Reifler) with ABC's Ron Claiborne for the World News Tonight webcast series The Conversation -- here's the video and the accompanying article:

Can you admit it when you're wrong? Do you believe new facts will help you change your opinion?

Turns out, probably not.

A growing body of research has found that we're even more stubborn and steadfast than we realize, and this can have troubling implications for a democracy based on an educated society.

"This isn't just about being educated or knowledgeable," said Brendan Nyhan, a political scientist who studies political misperceptions. "Those are often the people that are more resistant to information they don't want to hear."

You might believe you base your political views on hard facts, but in many cases, our voting behavior stems more from our beliefs.

Nyhan's paper, titled "When Corrections Fail: The persistence of political misperceptions," found that when we encounter facts that contradict those beliefs, the facts are either ignored or twisted to support our positions. And it's not limited to one political party -- both liberals and conservatives share the same problem.

"There's an idea out there that if we just give people the facts, they can make good decisions," said Nyhan. "I think people can make good decisions, but just giving them the facts is not going to do it."

What can be done about it? Nyhan has some ideas, which he shared with ABC's Ron Claiborne in today's Conversation.

So consider the facts, and watch today's Conversation with an open mind.

Comments

[David in Cal]

Nice interview, Brendan.

I don't disagree with your general thesis. In part, it repeats findings reported in the 1993 book How We Know What isn't So,.

Still, there's something about the research that bothers me. That is, the idea that the researchers are giving the subjects "the facts." Actually, the subjects are being given some facts -- facts selected to support the side the researcher deems "true."

This distinction is particularly important when the issue cannot be decided with absolute certainty. E.g., Brendan believes that it's most likely that there were not large stores of WMDs in Iraq prior to 2003 (and I agree). But, that's not suseptible of being proved. No inspections were done at that time. Inspections done later cannot absolutely rule out the possibility that WMDS had been moved out of the country.

People who disagree with the researchers aren't apt to change their opinion faced with the researchers' facts, if they already decided that those facts are outweighed by other facts.

Here's another way to look at this research, via an analogy:

Soap company A hires a salesman. His job is to call on customers who think soap B is better than soap A. He is to give them facts about soap A's superiority and convince them to switch to soap A.

He returns from his sales trip and reports that the customers he called on didn't change their minds. In fact, after his call, they were more committed to soap B than before.

The salesman blames the customers. He says they're irrational. However, the salesman's boss is likely to blame the salesman for doing a poor sales job.

I think Brendan's point about how subjects who feel good about themselves are more apt to change their mind is analagous to the salesman finding a more effective sales technique.

[Rob]

Brendan, your interviews about your research have been fine, but aren't you drawing general inferences that go well beyond your research findings?

You've stated here and in previous interviews that what Kent Brockman Ron Claiborne referred to here as being obstinate affects both liberals and conservatives. That's admirably even-handed, but didn't your research fail to find a backfire effect among liberals (at most corrections were neutralized, which I guess means they were ineffective) and in some models, among less committed conservatives?

In addition, these interviews are promoting the idea that your research is true of the general population, presumably the general American population. But your research paper is conscientious in pointing out that it was conducted among undergraduates at a Midwest Catholic university, and you point out that "it would be valuable to replicate these findings with non-college
students or a representative sample of the general population."

They say that mighty oaks from tiny acorns grow. In the world of social science research, it seems that mighty memes from tiny studies grow.

[JP]

@David: You bring up a great point! This is why I never trust anyone who starts a sentence with a compliment, especially waitresses and salesmen.

@Brendan: Your research reminds me of breaking up with my ex. No matter how many facts and rational arguments I slammed her with she could never see the truth that she was a selfish blood sucking vampire of a person.