The study, published by the journal Science in December, came under question this month when two graduate students trying to follow up on the work found evidence that the data had been misrepresented.

Share story

A leading scientific journal Thursday retracted a highly publicized study reporting that attitudes toward same-sex marriage could be altered by brief face-to-face conversations with people who have a stake in the issue.

The study, published by the journal Science in December, came under question this month when two graduate students trying to follow up on the work found evidence that the data had been misrepresented.

The study’s senior author, Donald Green, a prominent political scientist at Columbia University, asked that the study be retracted last week, after his co-author, Michael LaCour, a UCLA graduate student in political science, declined to furnish the raw data he had used to reach his conclusions.

LaCour did not agree to the retraction, the journal said. He could not be reached for comment.

Questions about the data were first reported by Retraction Watch, a blog covering scientific research.

The editors of Science cited three reasons for the retraction. First, incentives to participate in the survey had been misrepresented. The authors had claimed they had given people cash payments to enroll, to refer family and friends and to complete multiple surveys, but a letter last week from LaCour’s lawyer said no such payments were made, the retraction said.

Second, the sponsors of the study had been falsely identified; LaCour claimed funding from the Williams Institute, the Ford Foundation and the Evelyn and Walter Haas Jr. Fund, but according to LaCour’s lawyer, this statement is not true, the retraction said. Third, the authors could not produce the original data, necessary to resolve questions about the work.

In a letter to the journal last week, sent through his lawyer, LaCour acknowledged he had not paid study participants as he had claimed and said he had erased the raw data months ago, according to Marcia McNutt, the editor-in-chief of Science.

In emails to The New York Times, LaCour said he would provide a “definitive response” by the end of this week.

The students who flagged possible problems with the research, Joshua Kalla and David Broockman, then at the University of California, Berkeley, had tried to conduct their own version of the original study.

They asked canvassers with a personal stake in a contentious gay-rights issue to try to sway voters’ opinions. But the researchers could not get the same level of participation from voters that LaCour had reported.