Education research needs more retractions
My guest is cognitive scientist Daniel Willingham, professor at the University of Virginia and author of "Why Don't Students Like School""
By Daniel Willingham
You have probably heard about the retraction by the medical journal Lancet of a 1998 article linking a routine childhood vaccination with autism. This retraction made me think of the fact that such retractions are never seen in Education journals.
Retractions in medical journals are rare. According to a report in the Journal of Medical Ethics, during the decade 1995-2004, 0.0065% of articles were later retracted. That percentage has been increasing in the last few decades.
A search of the best-known education database, known as ERIC, showed just one retraction, and that was from an experimental psychology journal catalogued in ERIC.
In medical journals the most common reasons were scientific misconduct (17%) inability to replicate the results (20%), or the discovery of an error in the reported results (22%).
Educational research is not medical research. There are many more theoretical articles, to which these reasons would seldom apply. And even for the empirical articles there are so many variables beyond the experimenter’s control that it is rare that one can conduct a true replication.
But the other categories—scientific misconduct and simple error—surely occur some of the time. Why have there been no retractions for those reasons?
My goal here is not to chide education researchers or to paint them as less morally upright than medical researchers. But it might be worth considering what pressures—social, legal, financial, cultural—would encourage education researchers to retract articles they know to be in error, and what pressures discourage that practice.
The reputation of scientists took a beating over the release of emails stolen from East Anglia University that made some scientists seem—at least on occasion--less interested in scientific truth and more interested in advancing their position on scientific matters.
It is a good moment for each scientific field to take stock and consider the systems in place to ensure integrity. Educational research may be ripe for such an effort.
Follow my blog all day, every day by bookmarking http://voices.washingtonpost.com/answer-sheet/
Follow all the Post’s Education news & blogs on our Facebook fan page, the "PostSchools" feed on Twitter or our Education home page at http://washingtonpost.com/education.
| February 8, 2010; 1:25 PM ET
Categories: Daniel Willingham, Guest Bloggers, Research | Tags: Daniel Willingham
Save & Share: Previous: Up to Montgomery County principals to challenge fliers
Next: Any merit to National Merit program?
Posted by: efavorite | February 8, 2010 7:41 PM | Report abuse
The comments to this entry are closed.