RACHEL MARTIN, HOST:
Now, a story about social change, science and quite possibly serious academic fraud. Back in December, one of the most respected scientific journals in the world, Science magazine, published a groundbreaking political science study. In the study, same-sex marriage activists went door-to-door to have simple, polite conversations with voters who opposed gay marriage. The chats were like this one between an activist and a widower who started out firmly against a marriage. But by the end of the conversation, he had changed his mind.
(SOUNDBITE OF PODCAST, "THIS AMERICAN LIFE")
UNIDENTIFIED CANVASSER: So you know this issue is going to come up for a vote again in the future.
UNIDENTIFIED MAN: I would vote for it this time.
UNIDENTIFIED CANVASSER: Vote in favor of allowing gay and lesbian couples to marry? Why does that feel right to you?
UNIDENTIFIED MAN: Well, how would I say it? Let's see. I would hope that they would find the happiness that I had with mine.
UNIDENTIFIED CANVASSER: Yeah.
UNIDENTIFIED MAN: If you could have that kind of relationship with your partner, irrelevant of their sex, I would say you're a very lucky person because I know I had it.
MARTIN: The interviewees were supposedly surveyed about their opinions before and after these conversations. And many of their opinions did change, and they maintained that change a year later. The media, public radio included, took notice. The tape you just heard was from This American Life, which aired a segment on the study's finding. The research was also written up in The New York Times, The Washington Post, The Wall Street Journal, the list goes on. But this past week, the whole thing crumbled. One of the study's authors asked to have it retracted from the journal Science.
DONALD GREEN: My name is Donald Green. I'm a professor of political science at Columbia University.
MARTIN: Other researchers told Green they had found red flags in his study. Green says he came to believe his co-author, a grad student at UCLA named Michael LaCour, had faked the surveys. And that's critical because the surveys were the basis for the entire study. They proved that the gay marriage activists were indeed changing minds. Donald Green says even though his co-author LaCour was in charge of gathering the data, he himself bears some of the responsibility.
GREEN: Quite a bit of blame I think. You know, I should have been the one to have noticed this. And looking back on it, I wasn't suspicious enough.
MARTIN: We reached out to LaCour, who said he is still gathering evidence before giving any response. Meanwhile, many of the news organizations that reported on the original study have posted their own responses and retractions. Ivan Oransky runs a website called Retraction Watch, which monitors scientific misconduct. He says retractions are rare.
IVAN ORANSKY: It's about .02 percent. I mean, it's tiny.
MARTIN: Still, Oransky says the desire to be published in a big name journal can sometimes lead people to fudge the data.
ORANSKY: In order to get into those journals, you need to come up with something that isn't just slightly newer or, you know, incrementally more interesting than the last thing that was published. And so people are - perhaps they have more of an incentive to cut corners and maybe even commit fraud.
MARTIN: And he says even the peer review process isn't fool proof.
ORANSKY: It's something that academic researchers due for free. They're often under tremendous pressure to do it quickly because journals don't want to hold up publication.
MARTIN: Donald Green, the co-author of the controversial study, says this experience has changed how he will do research in the future.
GREEN: We're probably going to have to institute a new set of procedures whereby nobody gathers primary data alone. They always do it in teams. It's going to make things harder and more expensive.
MARTIN: But he says maybe that's the cost of getting it right.
NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.