Are your tax dollars funding scientific research that can’t be repeated? Elizabeth Tracey reports


The majority of cancer studies researchers attempted to repeat in a recent analysis were not reproducible, a recent paper found, with 59% of 50 attempts failing. William Nelson, director of the Kimmel Cancer Center at Johns Hopkins, says trying to validate results is a laudable goal, but it’s very difficult to execute. A major issue is pressure on researchers to publish their findings.

Nelson: I do worry about publishing. If you look at the number of articles published, it’s really been continuing to go up. Things that are clearly inappropriate, plagiarism, it’s a little harder to do with the search tools that are present where you can go search out borrowed text. Occasionally you’ll see things published about the level and degree of fraud and misleading publications. There’s room for improvement, and I think one thing that this group should be lauded for was trying to take this on.   :28

Biomedical research is largely funded in the US and many other countries using public funds, so attempting some sort of oversight is certainly indicated. At Johns Hopkins, I’m Elizabeth Tracey.