When a study fails to replicate in political science, retractions are not very common. Often, the re-analysis that questions the original findings is published in the same or another journal; and then the original author sometimes writes a third article defending his/her findings (see here). Recently, a failed replication made it into the news, and suddenly there’s a ‘scandal.’ Here are the most important pieces you should read if you want to know the details.
Science has recently retracted a study on gay marriage, asking the research question: “Can a single conversation change minds on divisive social issues, such as same-sex marriage?”.
The retracted article failed to replicate, as political scientists David Broockman, Joshua Kalla and Peter Aronow reported in their critique of the article.
Michael LaCour, has responded to the retraction in the form of a 23-page essay . The New York Magazine wrote about this statement here, as did sciencemag.org.
Among other news media, the Washington Post and the New York Times and The Guardian in the UK reported about this. A good source is always Retraction Watch which publishes updates on the case here.
Thomas Leeper blogs about how in-house replication before journal submission can solve such issues.
Any other blogs I’ve missed? Please comment below!
Finally, I’d like to point out that I would like to see the discussion moving on towards providing tools and guidelines in the field of political science to ensure research transparency. We have our scandal, just like Psychology, Economics and Cancer Research had theirs. Now let’s start changing the field.
[…] The blogsite Political Science Replication summarizes the recent “scandal” in which a graduate student at UCLA (Michael LaCour) is accused of fabricating data for an article published in Science magazine, leading to its retraction. The summary includes a description of the discovery of the fraudulent data, LaCour’s response, coverage by the media, and blog posts on the subject. To read more, click here. […]
[…] a blog post by Jesse Singal at the New York Magazine, one of the students who tried to replicate LaCour‘s findings on gay marriage said that young researchers are often actively discouraged from […]
[…] available, scholars cannot evaluate and cross-check published work. Just have a look at the recent LaCour scandal in political […]
[…] There are many corrections in political science, but not many rejections (LaCour was an expeption). What determines the forgiveness in our field, and is it good or bad? [I’ll […]