The prestigious Science journal has officially pulled down a widely publicized study about changing attitudes toward same-sex marriage after its senior author admitted to lies. Published in December, the study reported that attitudes toward same-sex marriage can be changed by short one-on-one conversations with gay canvassers who have a stake in the issue.
LaCour accused of faking data
The study came under question when a pair of Stanford University graduate students tried to replicate the findings. They found evidence that data in that study was heavily misrepresented. Then Stanford students tipped off Donald Green, lead author of the December study, on misrepresentations. Donald Green, a political scientist at Columbia University, asked the journal last week to retract the study.
— Sol Messing (@SolomonMg) May 22, 2015
Green’s co-author Michael LaCour of the University of California, Los Angeles, refused to furnish the raw data he had used in the study, saying that he had erased the data months ago. LaCour is accused of faking the data, but he did not agree to the retraction. He told the New York Times that he would respond to allegations by the end of this week. LaCour’s lawyer told the Science journal that there were three issues with the study.
The survey company never heard of this same-sex marriage study
First, the authors said that they had given people “cash payments” to enroll, to refer to family and friends, and complete the surveys. LaCour’s lawyer said no such payments were made to conduct the same-sex marriage study. Second, LaCour had claimed that the study was funded by well-known non-profits such as the Ford Foundation and the Williams Institute. That is not true, said the lawyer. Third, authors of the study failed to produce the original data.
When the Stanford University graduate students asked canvassers with a personal stake in the contentious same-sex marriage issue, they did not see the same level of participation from voters that LaCour had claimed. Then they contacted the survey company that LaCour reportedly used to help conduct his study. The survey company said it had never heard of that project.