Why did birther support drop so much?

When President Obama released his long form birth certificate, I was skeptical about its chances of killing the myth that he was not born in this country:

Will Wednesday's release of Obama's long-form birth certificate put an end to the birther myth?

The odds aren't good. The problem is that people can be extremely resistant to unwelcome factual information...

Given how much evidence is already available, it's hard to see why a long-form birth certificate would suddenly change the minds of people who are predisposed to believe in the myth. The hardcore are already shifting to new rationales for questioning Obama's right to hold office and deconstructing the PDF released by the White House for supposed evidence of forgery.

As expected, the fringe continues to search for rationales to discredit the birth certificate -- the birther movement isn't going away. However, the Washington Post released a poll conducted after the birth certificate's release (but before news of Osama bin Laden's killing was released) showing that that birther beliefs dropped by half compared to April 2010:

I've been waiting to see if any other polls were conducted, but it looks like the Post is the only media outlet that polled on this issue. Since bin Laden's killing created a bounce that will affect responses to questions about Obama for a while, it's probably the only clean pre-/post-comparison of birther beliefs that we're going to get.

So why was this correction so effective when others tend to fail? (PDF) The answers aren't entirely clear yet, but here are some initial thoughts. First, the birth certificate's release was an unusually definitive debunking that became a major news event, so there was saturation coverage of some very strong corrective information. Second, no prominent elites on the right contested the validity of the birth certificate, which meant that coverage of its release was almost entirely one-sided. Finally, it's possible that support for the myth was soft because poll respondents didn't really believe it but were using poll questions about Obama's religion and place of birth as a way to express disapproval (as some commentators and pollsters have argued).

Does this mean that people's minds can be changed? Yes (though I will be curious to see if these effects hold over time). In cases where the first and second conditions described above hold (very strong corrective information, saturation coverage, no elite controversy), it is possible to convince people who aren't hardcore believers and conspiracy theorists. However, these conditions rarely materialize for prominent political misperceptions, which tend to be controversial among elites and harder to definitively debunk.

To underscore why misperceptions are so difficult to correct, consider the case of the Duelfer Report, a comprehensive CIA-commissioned report released in September 2004 which found no evidence that Iraq had weapons of mass destruction or an active weapons of mass destruction program. Even though the release of the report was a major news story confirming that Iraq did not have WMD, no one could not prove they hadn't been hidden, moved to Syria, etc. and conservatives continued to dispute the claim. As a result, half of Americans were still telling pollsters that Iraq had WMDs as late as 2006.

In short, the release of the birth certificate is something of a best-case scenario for misperceptions. The Post poll results are a victory to be celebrated, but we should be cautious about extrapolating from this case to other, more stubborn myths.

Update 5/13 2:03 PM: Gallup just released a poll conducted May 5-8 (after bin Laden's killing) confirming that birther beliefs were cut roughly in half:

As in the Post's poll, Gallup found that the decline was most significant among Republicans (who were most likely to endorse the claim before the release of the birth certificate):

These results are an encouraging sign that the effects of the birth certificate's release are persisting, though we should be cautious given that the Gallup poll was conducted after bin Laden's death.

[Cross-posted to HuffPost Pollster]

Comments

[MartyB]

I suspect support for the "birther" myth, other than on the fringe, was because poeple wondered why the president didn't simply release his long form birth certificate a long time ago, which created the impression there was something to hide.

It was simple step he coud have taken any time in the past two years, and the longer he didn't made people who dislike or ditrust him or his policies in the first place distrust him even more.

This is the "Occam's razor" answer IMO.

[bnyhan]

Well, it's not like he could have released it whenever he wanted. He had to get special legal permission from the state of Hawaii to obtain and release it.

[Rob]

Please, Brendan. Getting permission from the State of Hawaii was about as much of a slam-dunk as you'd expect it to be. Anytime the President wanted to release his long-form certificate, he could have.

I do have a question about your research, which you cite again in this post. One of the fundamental principles of the scientific method is reproducibility or replicability. In your paper, you point out, "In addition, it would be valuable to replicate these findings with non-college students or a representative sample of the general population." But of course, even beyond the value of replicating your findings with non-college students, there would be enormous value in simply replicating your findings at all. That's what science is all about. I wonder, have there been studies that replicated your findings of a backfire effect?

Let me add as the comment of a casual observer that in the social sciences, there doesn't appear to be much interest in replicating results. Whether that's because little glory and honor attaches to replicating others' results or because there's a reluctance to appear to be challenging another's findings or because there's a fear that a lot of what passes for conventional wisdom in the field is based on studies that are likely not to be reproducible, I don't know. But it seems to me that failure to replicate results ought to make social scientists a lot more cautious about accepting findings as valid than they are, and that's worrying. This is a big subject, and it's one I'd love for you to address in your usual thoughtful manner whenever time and energy permit.

[David in Cal]

I agree with MartyB that Obama could have released his LFBC whenever he wanted, because he could asked for special legal permission and gotten it whenever he wanted.

Those who promoted the myth that Obama was unable to release his LFBC were perhaps acting out of loyalty to Obama. Common sense should have told them that regardless of normal procedure, if the President of the United States wanted his LFBC released it would be released.

[bnyhan]

We replicate two backfire effects in Political Behavior article, find one more (when white experimenters are present) in the Obama Muslim paper (non-college students), and have another study in draft form that finds a similar effect in another context (also non-college students). I agree there should be more replication, but I can't make other people do it. Also, the backfire effect may be contingent, as we discuss in the article; you shouldn't expect it in every situation.

[Mra]

Dumb shit, it's because no one ever saw it, he spent money to conceal it so there was legitimate questions. When he was open and honest about it and actually showed it that took all reasonable doubt off the table. No need to analyze it further, or make lame hypothesis, its just that simple. Don't try to be too smart, cause youll just look stupid.

[David in Cal]

I suspect the backfire effect is affected by who provides the corrective inforation.

Try this mental experiment:

Suppose subjects are asked about some myth that's primarily believed by liberals, e.g. the myth that Medicare is not unsustainable. Then, as a correction, the group is given a monologue by Rush Limbaugh pointing out that Social Security Adminstration projections demonstrate that SS is indeed unsustainable.

I think some of the liberals who weren't sure Medicare was unsustainable might then decide that anything Limbaugh believes is probably wrong. Thus, a correct explanation from Rush Limbaugh might actually increase belief in this myth.

In Brendan's experiments, the corrections were provided by Brendan and his colleagues. I suspect that some conservatives consider the typical college professor to be unreliable on political issues, just as liberals think Rush Limbaugh is. In other words, if Brendan's experiments had the same questions, but were run by conservatives and included corrections from conservative sources, perhaps the backfire effect would disappear.

[David in Cal]

In a comment to the prior post, I said that the methods used by the SS actuaries underestimate SS's future costs. As a result, each new year of actual data tends to be worse than what was projected. Thus, each new year's projection tends to be worse than the prior projection, as worse-than-expected data works its way into the model.

That indeed is happening, according to the new report of the SS Trustees.

According to the report, Social Security is now permanently cash negative and can no longer be funded solely from the payroll tax. It is projected to exhaust funds in 2036 — one year earlier than the report predicted last year.

A similar situation exists for Medicare:

Medicare isn’t faring any better. The trustees expect the fund to run out of money in 2024, and not 2029 as was previously expected.

These two funds are projected to run out of money fairly soon. Furthermore they will probably run out of money even sooner than the current worrisome projections.

[bnyhan]

Our Political Behavior article varied whether the articles were attributed to Fox News or the New York Times and found that it didn't matter. Research by another political scientist (in progress) finds that the corrective information itself coming from a conservative helps a bit, but I agree we need to study this more.

[David in Cal]

Brendan, perhaps the reason that attributing articles to Fox News didn't matter was that subjects knew that the facts were selected by you and your colleagues. Rush Limbaugh quotes liberal sources all the time, but that doesn't give him credibility among liberals. Conservative subjects may have believed that you and your colleagues would cherry-pick the facts -- presenting those that supported your POV and not those that would tend to contradict it.

Such a belief would have had some basis IMHO. E.g., you provided subjects with the Duelfur Report, which supports the likelihood that there were no WMDs in Iraq before the war. OTOH you did not provide the fact that chemical weapon stores were known to be in Iraq in 1998 when the UN inspectors left, and those WMDs were never accounted for.

[kestrel9000]

mra says:

he spent money to conceal it

which is not true.
That is a falsehood arising from World Net Daily who essentially asserted that the entirety of money paid to every attorney and court cost incurred by the Obama campaign was directed toward that single purpose.
Which is, of course, ridiculous on its face.
But "ridiculous on its face" never stopped Joseph Farah- or Jerome Corsi - from doing anything.
Think about that one. Why should the Obama campaign have to spend so much as a dime to conceal something the law said people couldn't have in the first place, even if there was something to hide, which there wasn't?

[Colin]

I would speculate that another reason the numbers dropped is not just because the long-form got so much publicity, but because the short-form did not. I think a lot of people heard that Obama had not released his birth certificate and did not realize that, in fact, he had released the same birth certificate that anyone ever born in HI would get, and could use, for any reason. The "short form" characterization implied that he had contrived some summary document that was not valid, when, in fact, it is valid for everything a birth certificate is valid for. I don't have any evidence that this is a reason, but it's my hypothesis.

Also, it's worth noting, as even the comments to this post suggest, that the birthers have not stopped believing any of their original conjectures, such as that the short form was insufficient, or the B.S. about the millions he spent to defend (which was actually to get them dismissed. What was he supposed to do, not defend those lawsuits?) And I think to say all along "release the BC" and then to say it doesn't count was more conspiracy theorizing than at least half of them could stomach.

Love your work. Thanks.

[MartyB]

Colin -

Good point on the "long-form publicity" angle, although I suspect the only reason it got so much attention was because it wasn't released to the public - the whole "he must have something to hide" feeling some may have had.

That said, I am not sure which comments you refer above that indicate anyone who commented on this site could be classified as a "birther". Speculating on why support has dropped for this myth doesn't mean a person agreed with it.

[Colin]

MartyB, what wasn't released to the public? That's exactly my point. The distinction between the long form and the short form is a distinction without a difference. There is NO legal distinction between them. The short form IS the legal BC. The idea that the long form mattered at all was entirely a fabrication of the birthers, and therefore to say "he could have released it at any time" is entirely irrelevant. He released his legal birth certificate in 2008.

My theory is that the long-form vs. short-form framing caught hold, and that people began to believe there WAS a legal distinction between them, and therefore that he hadn't released his BC. However, the birther framing after the release of the long-form (that it's forged or that he has other eligibility problems) is not as compelling as the long vs. short framing, and so to continue being a birther takes a new level of suspicion that I think is what dropped the numbers.

The comments of MRA, above, for example, are the perfect illustration of what I mean.

[MartyB]

Colin -

No argument from me on the legal distiction (or lack therof) between the two forms. The whole thing was never an issue in my book.

However, legal distictions hold no sway over PR problems which can arise from perceptions, and this is where the issue got away form the administration, as you appear to agree.

mra above seems to agree there is no reasonable doubt left at this point, so I am not sure he fits your "birther after the release" comment. (I don't care enuf at this point to research the "he spent money to conceal it" issue, s I can't comment on that)