Guy behind discredited gay marriage study says he destroyed the data that proves him right

The author of a newly retracted gay marriage study claims that he destroyed the data that would prove his findings were legitimate.

On Thursday, the journal that initially published UCLA grad student Michael LaCour's high-profile study, which claimed that opponents of gay marriage could be convinced to support it after a single conversation with someone who identifies as gay, officially retracted it. Earlier this month, several researchers who followed up on LaCour's findings, revealed that data used in the study were essentially nonexistent.

On Friday, LaCour responded in a 23-page paper that attempted to refute the criticisms made by researchers David Broockman, Joshua Kalla, and Peter Aronow.

But it also failed to answer some pretty big questions.

• First, LaCour claimed that he destroyed the data set that skeptics noted is missing out of respect for the anonymity of the participants involved in his research. But as the New York Times notes, UCLA's policy only requires researchers to destroy data that can be used to identify someone in the study.
• Next, LaCour called the researchers' decision to publish their critique without peer review and without going to him with their findings first "unprecedented, unethical, and anomalous in the relevant literature" and implied that the researchers may have made a "possibly intentional error" in their rebuttal. Specifically, LaCour claims that the researchers flubbed an equation in their study that would prove LaCour's study wrong.
  
• Most importantly, LaCour also failed to address one of the biggest problems with the disputed study: Why a person that allegedly helped LaCour get data at a company called uStamp doesn't exist. As New York magazine points out, LaCour claimed to have gotten data for his study from an employee named Jason Peterson. When the researchers reached out to uStamp, the company said that not only did the data not exist, but the person who LaCour allegedly worked with didn't either.

LaCour did, however, admit to lying about grants that he never received.

On Friday, the three researchers said LaCour's response doesn't answer their questions.

"In our view, none of the claims made in LaCour's response meaningfully address the concerns articulated in our report, Professor retraction request, or the Science retraction," the researchers said in a statement.

Another study in question

Regardless of how the controversy regarding the current study pans out, LaCour is also still facing questions about fudging the numbers of yet another study.

In that study, called "The Echo Chambers Are Empty," LaCour cites data that allegedly shows that, contrary to popular belief, most politically partisan people consume centrist, rather than ideologically slanted, media.

Knowing about the controversy surrounding one of LaCour's previous studies, Emory University researcher Gregory Martin flagged this study as well. When Martin tried to replicate that study using LaCour's data, he found that many of the data sets LaCour used were either nonexistent or incomplete.