« The best sentence I read this morning | Main | The most obnoxious blogger in the world »
Are lead papers in a journal of higher quality?
Maybe not, maybe you just think they are:
Leading papers in a journal’s issue attract, on average, more citations than those that follow. It is, however, difficult to assess whether they are of better quality (as is often suggested), or whether this happens just because they appear first in an issue. We make use of a natural experiment that was carried out by a journal in which papers are randomly ordered in some issues, while this order is not random in others. We show that leading papers in randomly ordered issues also attract more citations, which casts some doubt on whether, in general, leading papers are of higher quality.
Here is the full paper, courtesy of Pluralist Economics Review.
Posted by Tyler Cowen on August 7, 2008 at 06:09 AM in Science | Permalink
Comments
I know that when I need to pull some quotes from a book I didn't read, I use the random number generator on the ti-83. If you quote from the first section of a book or first few pages of a chapter the prof will know that you didn't read! Maybe these professional economists should do the same?
Posted by: person at Aug 7, 2008 8:49:00 AM
The same mechanism, of course, should be true of journals. The fact that "top tier" journals get more cites does not necessarily mean that the editors are doing a great job and, to my mind at least, means that you have to take a very long view in assessing journal "quality".
Posted by: journalsceptic at Aug 7, 2008 9:08:49 AM
I find this to be true about alphabetical ordering. That's one reason I like my blog's name (aguanomics). The other reason is that I can now claim to cover A-Z :)
Posted by: David Zetland at Aug 7, 2008 1:09:45 PM
How do you measure a paper's quality, if not by the number of citations it gets?
Posted by: Potrero John at Aug 7, 2008 4:32:11 PM
As a journal editor, I confirm that most of us,
I do in any case, try to put a paper as lead
of an issue that is more interesting than the
others, or most of others. I find it interesting
to see how often my guesses are correct. It
would seem that those leads do tend to get
cited more than others, but there are exceptions,
in both directions, leaders that end up as losers
and papers far down the order that end up getting
cited a whole lot.
Posted by: Barkley Rosser at Aug 11, 2008 2:29:20 PM
I should add to this that this does not imply that the rest of the ordering somehow indicates some ranking. What tends to go on, at least at my journal, is that the subsequent ordering tends to involve subjects, with papers about a similar topic being put together. This means that an issue is often set in topics a bit, especially at the front end, by which paper does go first.
Posted by: Barkley Rosser at Aug 12, 2008 2:18:50 AM
