12 June 2013

Sharing responsibility for bad papers

When you read a bad paper, whose fault is it?

Robert Horvitz (quoted in Box 1 here; hat tip to Mike Taylor) and Fred Schram (who I quoted a couple of years back) both put the burden squarely on the authors. This bothers me, as I started to articulate yesterday.

Journals routinely take credit when things go right. Journals and publishers love to talk about how they “add value” to papers. Journals talk about “attracting” high quality submissions. They have “rigorous” peer review. When impact factors go up, it must be due to the good job the journal’s editorial board is doing.

But when things go wrong, and a journal is faced with charges that a paper should not have been published, it is all to easy for editors to wash their hands of the whole thing and let the authors twist in the wind. For instance:

Peer review does not end with publication. In the event that an accepted manuscript is questioned by the scientific community on the basis that the authors acted unethically, plagiarized, or where there are queries relating to the data or interpretation of the data, the editors will contact the authors to investigate unethical/fraudulent/plagiarized works or the journal editor will invite or accept letters to the editors.

Notice anything there about acknowledging a bad decision? About apologizing? About ensuring that reviewers are not asked to review for a journal again? Anything about investigating the process that led to a bad decision? I’m not even saying it was necessary in that case, but it exemplifies the “we followed procedure, we did nothing wrong” arguments that are routinely pulled out in response to criticism.

Some journals are quite good at critical self-appraisal of their overall processes. For instance, Nature showed some good reflection about sexism in their editorial process. But it is much rarer for a journal to show the same reflection over any single paper, except perhaps for clerical errors. Again, I struggle to think of cases.

I know it’s human nature to want to take credit and avoid blame, but mature people suck it up and take the heat as well as the glory. Editors and journals should share some of the load when poor papers get through their editorial decision making. I would also like it if reviewers would admit some culpability, but am not holding my breath, given the reviewing procedures at most journals protect anonymity to the extent they do.

Related posts

The Crustacean Society 2011: Day 3
Back room science

External links

Caustic volleys and the sting of peer review: what’s the solution?

Photo by !anaughty! on Flickr; used under a Creative Commons license.

1 comment:

Mike Taylor said...

To clarify ... when I quoted (approvingly) "what is in the paper is fundamentally the responsibility of the authors, not of the reviewers", I wasn't particularly meaning that reviewers should be absolved of blame if bad work is published. The point I thought the quote was making, and the one I was endorsing, is that it's no part of the reviewer's job to tell the author what paper to write -- only to help him improve the one he wrote. "Your paper should do X instead of Y" has no place in a good review.