Study Featured On 'This American Life' Called Into Question Over Data Irregularities
May 21, 2015 at 4:25 AM (PT)
A study prominently featured in the prologue of a recent episode of CHICAGO PUBLIC MEDIA News-Talk WBEZ/CHICAGO and PRX's "THIS AMERICAN LIFE" that appeared to support the idea that opponents of same-sex marriage could have their opinions changed by gay canvassers has been called into question after researchers found irregularities in the data and the way it was collected. The study's co-author, COLUMBIA UNIVERSITY professor DONALD GREEN, has called for the report to be retracted because of problems with the data allegedly collected by his co-author, UCLA doctoral candidate MICHAEL LACOUR.
The LOS ANGELES TIMES reports that the study, a report on which figured prominently in the APRIL 25th episode of the popular syndicated public radio show and podcast, came under fire after UNIVERSITY OF CALIFORNIA-BERKELEY and STANFORD graduate students approached GREEN with concerns over the study's data, and SCIENCE, the journal which published the study, has published an "Expression of Concern" about the validity of the data. A list of irregularities in the dataset has been issued by YALE Assistant Professor PETER ARONOW, STANFORD incoming Assistant Professor DAVID BROOCKMAN, and BERKELEY grad student JOSHUA KALLA.
"THIS AMERICAN LIFE" host IRA GLASS wrote a response that explained the problems, noting that the BERKELEY and STANFORD replication study found far fewer responses and discovered that the survey firm LACOUR claimed to have used denied working on the study and did not have the capability to do some of the recruitment procedures indicated in LACOUR's data. “Maybe the thing to convey in your blogpost,” GREEN told GLASS, “would be something to the effect that, just because the data don’t exist to demonstrate the effectiveness of this method of changing minds, doesn’t mean the hypothesis is false. And now the real work begins.”