Back in May, it was revealed that Michael LaCour, then a graduate student in political science at UCLA, appeared to have fabricated the data behind what looked like a groundbreaking study on gay-marriage attitudes that he co-authored with the celebrated Columbia political scientist Donald Green. Science of Us told the story of how David Broockman, then a graduate student at UC Berkeley (and currently a professor at the Stanford Graduate School of Business), discovered the fraud and, alongside fellow student Josh Kalla and (eventually) Yale political scientist Peter Aronow, reported their explosive findings to the world. Now, Broockman, Kalla, and Aronow have received an award for their efforts. The Leamer-Rosenthal Prizes for Open Social Science, explains the Berkeley Initiative for Transparency in the Social Sciences, go to researchers whose work demonstrates “commitments to the basic scientific values of openness, integrity, and transparency.”
This is a pretty fitting way to end the year, as 2015 was marked by ever-increasing discussion of issues of replication, transparency, and fraud in the social sciences. First, the LaCour case brought a lot of these issues to the forefront; then, in August, the most comprehensive effort yet to replicate past psychological science findings was published, and it revealed an alarmingly low hit rate of under 50 percent, suggesting that many of the discoveries that have made headlines in recent years have a shaky underlying foundation.
While most cases of questionable results being published have less to do with spectacular instances of fraud and more to do with the sometimes-fuzzy boundaries between rigorous and sloppy research, there’s now an overwhelming case that the social sciences are going to have to make some serious improvements in these areas. But the first steps are diagnosing the problems and coming up with ways to solve them, and that’s what researchers seem to be doing.