Cerebral and visceral reactions »« The Daily Show on the fighting nuns

Another psychology researcher resigns amidst data massaging flap

I wrote recently about University of Pennsylvania researcher Uri Simonsohn who has developed a new, but yet unpublished, statistical technique to detect whether raw data have been improperly massaged to get positive, publishable results.

One faculty member of a Dutch university has already resigned because of suspicions raised by this method and now Ed Yong, who has been covering this story is great detail, has news of a second psychologist, an American this time, resigning under a similar cloud.

Who knows, this idea of ‘data detectives’ scrutinizing the work of others may catch on. Journals constitute one set of potential users but another could be funding agencies.

Comments

  1. 'Tis Himself says

    This has become a hot topic in the economics sphere. We’re anxiously waiting for Simonsohn to publish his paper so we can look at it and apply his methods to our own discipline.

  2. kurt says

    Not exactly related, but I read today that the science writer Jonah Lehrer has just had his book “Imagine” pulled from stores because he had made up or embellished some of the quotes in it. It’s really hard to understand why someone would do that, because the book without the embellishments would still be just as marketable (unlike the case of a scientific study where altered data might make the results more likely to be published).

  3. Mano Singham says

    Lehrer has also resigned from the New Yorker and issued an apology.

    I have not understood that mentality to make up quotes myself. But it appears that if you are a professional writer selling to magazines, having killer original quotes tend to sway editors.

    This quite different from academic writing where we quote from published works.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>