TOO GOOD TO BE TRUE

Study That Showed Conversation Made People More Supportive Of Marriage Turns Out To Be Fraud

Michael LaCourLast December, the prestigious journal Science published a study that was really good news for marriage equality supporters. The study found that when canvassers spent 20 minutes talking to people, those people were more likely to be supporters of marriage equality. Moreover, that change of opinion was likely to be long-lasting.

“A lot of time we find in social science that most things don’t work, they don’t change people’s minds,” said Michael LaCour, the lead author of the study. “But we found that a single conversation was able to change voters’ minds up to a year later.”

The study was heralded as a major finding. As it turns out, it was also a major fraud.

When other researchers tried to replicate the findings, they failed. They also discovered that the data hadn’t been collected as described, the project wasn’t funded as described and in fact there probably wasn’t any survey of attitude change at all.

That led LaCour’s co-author, Donald Green, a professor of political science at Columbia University, to ask to have the paper retracted. “I am deeply embarrassed by this turn of events and apologize to the editors, reviewers, and readers of Science,” Green said.

LaCour, who is a doctoral student in political science at UCLA, hasn’t responded to the charge that he made it all up. He has tweeted that he’s “gathering evidence” and looking forward to “addressing the concerns raised.” In the meantime, he may want to consider looking for another career.

Photo credit: LaCour’s Twitter account