blog

You might have seen that the UK government has released its Select Committee’s report on peer review in science. The chair of the committee, Andrew Miller MP, says that the “general oversight of research integrity in the UK [is] unsatisfactory and complacent.”

Note that he doesn’t say that the research is unsatisfactory–simply the oversight. I think that’s probably fair; there is little in the way of a structured anti-fraud mechanism at ground level to prevent fraudulent and wasteful research. Peer review of articles goes some way towards this; post-publication checking (it happens all the time, as scientists read papers and think “hang on a minute…”) of articles picks up more; and of course the ultimate gold standard, if you like, is experimental reproducibility (which has killed many non-fraudulent results, too).

I haven’t read the full report but there’s a fair bit of reaction knocking around, as you might expect. Brian Deer, writing in the Grauniad, seems to be channelling Simon Jenkins in his response. You’re better off reading what Sense about Science MD Tracey Brown has to say, particularly about patronage:

This is what the peer review system is for, selecting what is worthy of publication and further study. There is only one alternative to this – patronage.

Tracey is concerned about a recommendation in the report, that raw data should be published openly. Now, we at F1000 have a vested interest in the publication of data, and we’ve been talking to several of you about our plans in that direction. However, we’re not necessarily interested in data for the sake of “research integrity”; rather we want the stuff to be used more easily. Tracey says that “most misunderstandings from scientific research come from an absence of meaning and context” rather than an inability to see the data, and is not convinced that the committee has got this one right. I have some sympathy for this point of view–after all, it’s bad enough trying to get the mainstream media to understand the correct message from a piece of research when the message is controlled, let alone what might happen if they have unfettered access to everything.

This one’s going to be interesting to watch, and could affect you, as a practicing scientist, in unforeseen ways. I’d welcome your feedback.