Friday, January 29, 2010

A Quick Note

As I write this, I'm at the 17th Annual CARD Conference in Lake Mary, Florida, waiting for the poster sessions to open. I brought a copy of Ioannidis, 2008 for leisure reading, and was struck by the following quote:

On the other hand, research in a field with small studies, strong conflicts of interest, intense competition for generating 'positive' results and prior documentation of publication bias should have high prior odds of bias before doing the meta-analysis [25]. Even if no signal is shown in statistical tests for bias, the odds of bias remain high.

This description seems... awfully familiar. <sarcasm>I wonder just what field I'm thinking of.</sarcasm>

No comments:

Post a Comment