Columbia University political science professor Donald Green’s retraction this week of a popular article published in the December issue of the academic journal Science follows revelations that his co-author allegedly faked data for the study, “When contact changes minds: An experiment on transmission of support of gay marriage.”

According to the academic watchdog blog Retraction Watch, Green published a retraction of the paper Tuesday after confronting co-author Michael LaCour, a graduate assistant at UCLA.

The study received widespread coverage from The New York Times, Vox, The Huffington Post, The Washington Post, The Wall Street Journal and others when it was released in December.

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

In an email to POLITICO, Green said he spoke with LaCour by phone on Tuesday and that he “maintained that he did not fabricate the data but told me that he could not locate the Qualtrics source files for the surveys on the Qualtrics interface or on any of his drives.”

Qualtrics was the survey platform that was purportedly used, though a company spokesman clarified to POLITICO that it did not collaborate with LaCour or anyone else on the study.

“I asked him to write a retraction, and he indicated he would do so, but when it did not appear last night, I sent off my own retraction,” Green wrote.

The investigation into the paper began when graduate students at the University of California, Berkeley, were initially impressed with the work and wanted to do an extension of it, according to a timelineof their probe posted Tuesday. When the students started a similar study, they found they were not getting the large response rate that Green and LaCour received in theirs.

In another email to POLITICO, LaCour wrote that he has read the investigation and will respond: “I’m gathering evidence and relevant information so I can provide a single comprehensive response. I will do so at my earliest opportunity.”

The students’ report chronicles numerous irregularities of the study, which they said allegedly consisted of “repeated observations of the same 11,948 voters over a series of weeks,” including the use of the “feeling thermometer” survey technique, which they called “notoriously unreliable” in terms of measurement error.

“However, in both studies, respondents’ feeling thermometer values are extremely reliable — more so than nearly any other survey items of which we are aware,” the report states.

Qualtrics said it was not familiar with the project and “denied having the capabilities” to do some of what the survey described, according to Green, after UCLA’s political science department chair contacted the company. The graduate students also contacted a Yale political science professor to help look into the discrepancies.

After speaking with LaCour, Green told one of the graduate students and the Yale professor that the UCLA graduate assistant had confessed to “falsely describing at least some of the details of the data collection.”

“Looking back, the failure to verify the original Qualtrics data was a serious mistake,” Green told Retraction Watch, submitting a retraction letter as an appendix to the students’ report on Tuesday.

Sasha Issenberg, a journalist who wrote about Green and LaCour’s work before it was published in Science, said Wednesday he was initially impressed and excited about the study’s methodology and findings.

“Unfortunately, I don’t know any more than what I’ve read, so beyond shock I don’t have much of a reaction at the moment,” Issenberg, who also has a connection to UCLA’s political science department, said in an email to POLITICO.

A previous version of this article misstated Qualtrics’ role.