Checking your work 27 June, 2011Posted by Simon Nickerson in Uncategorized.
Carl Zimmer has an excellent piece in the NY Times about science’s failure to correct its own mistakes, which is often touted as one of its most important qualities. He points out numerous recent examples (including ESP and arsenic-based life) of controversial research where other scientists don’t actually attempt to replicate the research (or journals refuse to publish their results when they do). Scientists quoted in the story point out that they don’t want to put their own research programs on hold to check someone else’s findings, especially when they’re sure that the initial research was wrong.
When the only published research on an important issue reaches conclusion A, but the majority of the scientific community believes in conclusion B, it seems to me that we’re devaluing the importance of research journals. Zimmer concludes his piece with an appeal for those in science to put more resources into replicating important experimental results. This strikes me as slightly naive – what are the incentives for people to act this way? – but I don’t know what else might work.