The Study That Found That Talking With A Gay Person Can Change The Mind Of Gay Marriage Opponent Was Completely Made Up

The paper has been retracted, but organizations that were using its findings to guide their campaigns say they’re pressing on.

The Study That Found That Talking With A Gay Person Can Change The Mind Of Gay Marriage Opponent Was Completely Made Up
[Illustrations: Markovka via Shutterstock]

With 36 states having legalized same-sex unions, the movement for marriage equality is winning over the nation. But in today’s polarized world, changing people’s hearts and minds has been a long, hard battle and the remaining 14 states are digging in their heels.


That’s one reason why a study published in the journal Science in 2014 was so heartening. It found that a 20-minute conversation with a gay canvasser actually changed many people’s minds about gay equality, even a year later. The work, conducted by UCLA graduate student Michael LaCour, was the biggest confirmation yet of what many researchers and political organizers wanted and have previously suspected to be true–that real human contact with “the other side” makes it harder to be polemical. The study received media coverage in This American Life, the Washington Post, the Wall Street Journal, and many other outlets (including, we’re pretty sure, us–though we can’t currently locate the link).

There was just a small problem. The data may have all been a fraud.

The problems surfaced when two outside researchers, David Broockman and Joshua Kalla, were so impressed with the results that they wanted to replicate and expand the study. But they surfaced a number of statistical irregularities. At one point, they contacted Qualtrics, the online survey firm that supposedly performed LaCour’s study, and learned that not only did the firm not know about the project and had never heard of the staffer who supposedly did the work, they didn’t even have the capabilities to perform the study as it was described.

Things unraveled from there. The researchers brought their concerns to Donald Green, the Columbia University political scientist who is the senior author on the study and supervised LaCour’s work. Green confronted LaCour, who could not produce his raw data and confessed to “falsely describing at least some of the details of the data collection.” On Tuesday, Green published a retraction note on his website and sent it to the journal Science. On Wednesday, as reported by BuzzFeed, the funding agencies said to fund the work also denied having involvement.

“I am deeply embarrassed by this turn of events and apologize to the editors, reviewers, and readers of Science,” Green told the blog Retraction Watch.

LaCour, for his part, hasn’t said much, but tells Co.Exist he intends to craft a reply on or before May 29. He says he’s gathering evidence and relevant information to provide a defense of his work and was given no notice before the allegations were posted online.


“I must note, however, that despite what many have printed, Science has not published a retraction of my article with Professor Green. I sent a statement to Science editor McNutt this evening, providing information as to why I stand by the findings in LaCour & Green (2014). I’ve requested that if Science editor McNutt publishes Professor’s Green’s retraction request, she publish my statement with it,” he wrote Co.Exist in an email. (Science indeed hasn’t published a retraction, but issued an “editorial expression of concern” while it investigates.)

The damage done by this sort of fraud is manyfold.

The public always has a tenuous relationship with science, and one false study can do plenty of permanent of damage. A 1998 study that showed a link between autism and childhood vaccines was later proved to be a fraud, but that hasn’t stopped the dangerous anti-vaccine movement from gaining momentum.

In any field, it’s hard to catch a willful liar (just ask the editors of Jayson Blair). In LaCour’s case, Green believes he fabricated an incredible amount of detail to make his work believable. Also, the data he “collected” seems to be patterned after existing data from a different survey, Broockman and Kalla found. “The data therefore look very much like real survey data because they are, in some sense, real,” Green told Co.Exist.

In this case, actually, the scientific process worked, in a way. It just took longer, until independent researchers came to replicate the results. Green, as LaCour’s supervisor, also initially questioned the early results, but managed to be convinced.

“Science is fundamentally about skepticism. It is about adducing evidence to overcome the objections of a determined skeptic,” he wrote in his email to us. “After [LaCour] showed me his first study, I suggested that he conduct a second study because I was skeptical, and I knew others would be, too. When the results came back confirming the first study, I was convinced. But I now see that I failed to consider another hypothesis: the findings were the same because the researcher was the same.”


The more lasting damage may be due to the setback to advocacy organizations who were bolstered by LaCour’s results. The Los Angeles LGBT Center was “shocked” by the news, according to NPR. It had worked with LaCour because it wanted an independent evaluation of the effects of its tactics, which involved sending canvassers out to talk about their personal experiences, especially as California was debating Proposition 8 in 2008. Door-to-door canvassing is hard work, and talking about difficult experiences with strangers who may not approve of you is certainly not fun–but if the center had more data to show that it worked, that would prove the tactic was worth it. Now, it still doesn’t have the randomized, controlled data that LaCour’s study gave them.

And because of the power of the study, LaCour was already working with Planned Parenthood and the Los Angeles LGBT Center again to conduct an independent study about whether their work sending out canvassers to talk openly about their abortions could change abortion opponents’ minds. The early results were reported to be promising. Planned Parenthood, however, is not going to be put down by the faked data. In an email to Co.Exist, the organization’s national, California-wide, and Los Angeles programs said that while “disturbed” to learn about the researcher’s apparent fabrication, they remain deeply committed to the program.

“We’ve seen the power and impact of this project with our own eyes in hundreds of door-to-door conversations,” they wrote in a joint statement.

This post was updated to include comments from Michael LaCour.


About the author

Jessica Leber is a staff editor and writer for Fast Company's Co.Exist. Previously, she was a business reporter for MIT’s Technology Review and an environmental reporter at ClimateWire.