A UCLA graduate student accused of falsifying data in a study released a response Friday refuting a report by three researchers who found irregularities in the study.
In the original study, Michael LaCour and his co-author, Columbia University professor Donald Green, found evidence suggesting that conversations with gay canvassers can change voter attitudes on same-sex marriage.
The study was published in the journal Science in December, and was retracted by the journal Thursday after a request by Green.
Three researchers – David Broockman, a soon-to-be assistant professor at Stanford University, Joshua Kalla, a graduate student at UC Berkeley, and Peter Aronow, an assistant professor at Yale University – released the report last week, accusing LaCour of being responsible for “irregularities” in the data collection.
LaCour released a 23-page response to their criticisms Friday that included four claims against the researchers’ report. In the response, LaCour claimed his study met the replication standard and followed institutional policy in destroying the raw survey data.
“The request by Broockman et al. … for the raw data is inconsistent with the best practices in the field, and not permitted under the institutional rules at UCLA,” LaCour said in the response.
The report by Broockman, Kalla and Aronow included a timeline of events and comparisons to a 2012 survey, the Cooperative Campaign Analysis Project, in which LaCour’s adviser, political science professor Lynn Vavreck, was involved. The report claimed some of LaCour’s data was statistically indistinguishable from the survey data.
Brookman, Kalla and Aronow’s report included graphs comparing LaCour’s findings and CCAP data that appear to be almost identical. In his response, LaCour said the researchers used the incorrect variable for the CCAP data and further manipulated the variable to make it appear identical.
LaCour said the researchers failed to describe their sampling procedure when they replicated the study. He claimed that failing to present the sampling procedure in the replicated studies makes it impossible to evaluate the study’s scientific merit and thus has no bearing on his study.
LaCour also claimed the report did not accurately present the timeline of events and that the researchers failed to disclose their critique.
In his version of the timeline, LaCour said Vavreck did not disclose the identity of the accusers or the specifics of the accusations. During their meeting, LaCour said he was asked to submit his personal survey account information and refused.
LaCour further criticized Broockman, Kalla and Aranow for failing to put their report through the peer-review process.
In recent days, several news outlets have accused LaCour of falsifying other details of his study, including his funding sources.
In the acknowledgements of his study, LaCour listed the Williams Institute at UCLA School of Law as a funding source for his study. Lauren Jow, a spokeswoman for the Williams Institute, said the institute was not aware it was listed as a source and did not provide funding.
In his response, LaCour said he did receive a grant offer from the Williams Institute but did not accept the funds. LaCour admitted the Ford Foundation grant, which he also listed as a funding source, did not exist.
“I take full responsibility for errors in the design, implementation and data collection regarding the field experiments and panel survey reported in LaCour and Green,” LaCour said in the report. “I also take full responsibility and apologize for misrepresenting survey incentives and funding in LaCour and Green.”
In an interview with The New York Times, LaCour said he lied about his funding sources because some colleagues doubted his work and he wanted to give more credibility in his study.
Compiled by Alejandra Reyes-Velarde, Bruin reporter.