Comments on: The ordering of authors’ names in academic publications http://www.overcomingbias.com/2008/01/the-ordering-of-authors%e2%80%99-names-in-academic-publications.html Overcoming Bias is economist Robin Hanson’s blog, on honesty, signaling, disagreement, forecasting, and the far future. Sun, 12 Feb 2012 04:20:14 +0000 hourly 1 http://wordpress.org/?v=3.1.3 By: Adrian Tschoegl http://www.overcomingbias.com/2008/01/the-ordering-of-authors%e2%80%99-names-in-academic-publications.html#comment-410182 Adrian Tschoegl Fri, 04 Jan 2008 20:44:15 +0000 http://prod.ob.trike.com.au/2008/01/the-ordering-of-authors%e2%80%99-names-in-academic-publications.html#comment-410182 Eddie: problem is specifying property rights first. If the default is alpha, then I think there is no problem. However, if one author was clearly the lead author, then what happens? Eddie: problem is specifying property rights first. If the default is alpha, then I think there is no problem. However, if one author was clearly the lead author, then what happens?

]]> By: eddie http://www.overcomingbias.com/2008/01/the-ordering-of-authors%e2%80%99-names-in-academic-publications.html#comment-410181 eddie Thu, 03 Jan 2008 19:34:55 +0000 http://prod.ob.trike.com.au/2008/01/the-ordering-of-authors%e2%80%99-names-in-academic-publications.html#comment-410181 Surely we're missing an obvious solution, at least for economics papers? Why don't the authors simply bid for placement? Surely we’re missing an obvious solution, at least for economics papers? Why don’t the authors simply bid for placement?

]]> By: Barkley Rosser http://www.overcomingbias.com/2008/01/the-ordering-of-authors%e2%80%99-names-in-academic-publications.html#comment-410180 Barkley Rosser Wed, 02 Jan 2008 20:45:43 +0000 http://prod.ob.trike.com.au/2008/01/the-ordering-of-authors%e2%80%99-names-in-academic-publications.html#comment-410180 Anon, Yes, Anatol (now deceased). Thank you. BTW, do you know what relation they had, if any? I know the Galbraiths and Baldwins were and are father and son. Anon,

Yes, Anatol (now deceased). Thank you.
BTW, do you know what relation they had, if any?
I know the Galbraiths and Baldwins were and are
father and son.

]]> By: Stephen Simmons http://www.overcomingbias.com/2008/01/the-ordering-of-authors%e2%80%99-names-in-academic-publications.html#comment-410179 Stephen Simmons Wed, 02 Jan 2008 20:04:55 +0000 http://prod.ob.trike.com.au/2008/01/the-ordering-of-authors%e2%80%99-names-in-academic-publications.html#comment-410179 I encountered another form of alphabetical discrimination while studying my MBA. The careers department produced a book of CVs, arranged alphabetically by surname, to help prospective employers invite students to on-campus interviews. Each employer (mainly the top strategy consulting firms, investments banks, etc) has a limited number of interview slots to fill by invitation. When the lists for on-campus interview times were posted, those with CVs near the start of the book received 30% more invitations than those unlucky to be near the end. Did the recruiters have rational reasons for preferring the earlier surnames? Perhaps S-Z contained more non-native-English speakers. Or maybe the recruiters simply got tired before finishing the hundreds of similar CVs. If so, a smart recruiter would get better results working back from Z. As far as I know, noone checked which strategy performed better at picking the students who went on to graduate with distinction... I encountered another form of alphabetical discrimination while studying my MBA. The careers department produced a book of CVs, arranged alphabetically by surname, to help prospective employers invite students to on-campus interviews. Each employer (mainly the top strategy consulting firms, investments banks, etc) has a limited number of interview slots to fill by invitation.

When the lists for on-campus interview times were posted, those with CVs near the start of the book received 30% more invitations than those unlucky to be near the end.

Did the recruiters have rational reasons for preferring the earlier surnames? Perhaps S-Z contained more non-native-English speakers. Or maybe the recruiters simply got tired before finishing the hundreds of similar CVs. If so, a smart recruiter would get better results working back from Z.

As far as I know, noone checked which strategy performed better at picking the students who went on to graduate with distinction…

]]> By: Adrian Tschoegl http://www.overcomingbias.com/2008/01/the-ordering-of-authors%e2%80%99-names-in-academic-publications.html#comment-410178 Adrian Tschoegl Wed, 02 Jan 2008 18:20:21 +0000 http://prod.ob.trike.com.au/2008/01/the-ordering-of-authors%e2%80%99-names-in-academic-publications.html#comment-410178 Barkley: and apologies for misspelling your first name. I am so inured to misspellings of my names (first and last, though especially the last), that I am not as careful as I should be to double check the spellings of rare names. Barkley: and apologies for misspelling your first name. I am so inured to misspellings of my names (first and last, though especially the last), that I am not as careful as I should be to double check the spellings of rare names.

]]> By: Anonymous http://www.overcomingbias.com/2008/01/the-ordering-of-authors%e2%80%99-names-in-academic-publications.html#comment-410177 Anonymous Wed, 02 Jan 2008 18:15:23 +0000 http://prod.ob.trike.com.au/2008/01/the-ordering-of-authors%e2%80%99-names-in-academic-publications.html#comment-410177 Barkely: Thanks for the long comment. I must admit I never gave the first name initial any thought; there are so few Tschoegls out there the issue has never come up for me. You have a valid gripe and I will amend my lazy habits in the future. Barkely: Thanks for the long comment. I must admit I never gave the first name initial any thought; there are so few Tschoegls out there the issue has never come up for me. You have a valid gripe and I will amend my lazy habits in the future.

]]> By: anon http://www.overcomingbias.com/2008/01/the-ordering-of-authors%e2%80%99-names-in-academic-publications.html#comment-410176 anon Wed, 02 Jan 2008 18:05:40 +0000 http://prod.ob.trike.com.au/2008/01/the-ordering-of-authors%e2%80%99-names-in-academic-publications.html#comment-410176 Barkley, <a href="http://en.wikipedia.org/wiki/Anatol_Rapoport" rel="nofollow">Anatol Rapoport</a> Barkley,

Anatol Rapoport

]]> By: Barkley Rosser http://www.overcomingbias.com/2008/01/the-ordering-of-authors%e2%80%99-names-in-academic-publications.html#comment-410175 Barkley Rosser Wed, 02 Jan 2008 17:29:26 +0000 http://prod.ob.trike.com.au/2008/01/the-ordering-of-authors%e2%80%99-names-in-academic-publications.html#comment-410175 As a journal editor and someone who has published papers in multiple disciplines, let me note two problems, not fully articulated so far. One is that this is partly a matter of discipline conventions, while the second is that it is partly a matter of journal conventions. I also note that sometimes authors do explicitly say that their order is randomized, or whatever, although this does not necessarily overcome the problems associated with "et al." unless the same set of coauthors publish multiple papers with each other that receive roughly equal rates of citation. Regarding disciplines, it appears to me that there is a tendency for there to be a natural versus social science split in conventions. Papers in natural sciences, especially for research coming out of labs, tend to have more coauthors, and the convention seems to be more that the first coauthor is somehow the lead, either in terms of being the real leader of the research in question, and main author, or sometimes simply the leader of the lab/most prestigious, although often these will coincide. Social sciences tend to have fewer coauthors, and tend more to go with the alphabetical norm as the standard for author order. Regarding journals there is very much the issue of "et al.," which I personally dislike. Of course in Reference sections, most will have all authors listed, so the issue becomes how the authors are referred to in the text. Some journals eschew "et al." and allow all to be listed. However, a widespread norm is to list all if there are just two, but to impose "et al." for more than two. I see no justification for this, and personally oppose it (it is imposed on my journal from above by our publisher), but clearly it does create a problem. It is not clear that announced randomization necessarily helps much in this case. While I am at it, I shall also note a bugaboo of mine for Referencing, which involves something that is standard in the natural sciences and seems to have spread from them to many of the social science journals as well. It is the use of initials with a last name. I can name several pairs of economists who have different first names but identical initials who do, or have, published within the same sub-areas of economics: There are two (were) two J.K. Galbraiths, John and James, two R.E. Baldwins, Robert and Richard, two R. Gordons, Robert and Roger, and two A. Rapaports, Ariel and Amnon, and there is another one that I am forgetting at the moment... As a journal editor and someone who has published papers in multiple disciplines,
let me note two problems, not fully articulated so far. One is that this is partly
a matter of discipline conventions, while the second is that it is partly a matter
of journal conventions. I also note that sometimes authors do explicitly say that
their order is randomized, or whatever, although this does not necessarily overcome
the problems associated with “et al.” unless the same set of coauthors publish
multiple papers with each other that receive roughly equal rates of citation.

Regarding disciplines, it appears to me that there is a tendency for there to be a
natural versus social science split in conventions. Papers in natural sciences,
especially for research coming out of labs, tend to have more coauthors, and the
convention seems to be more that the first coauthor is somehow the lead, either in
terms of being the real leader of the research in question, and main author, or
sometimes simply the leader of the lab/most prestigious, although often these will
coincide. Social sciences tend to have fewer coauthors, and tend more to go with
the alphabetical norm as the standard for author order.

Regarding journals there is very much the issue of “et al.,” which I personally
dislike. Of course in Reference sections, most will have all authors listed, so
the issue becomes how the authors are referred to in the text. Some journals eschew
“et al.” and allow all to be listed. However, a widespread norm is to list all if
there are just two, but to impose “et al.” for more than two. I see no justification
for this, and personally oppose it (it is imposed on my journal from above by our
publisher), but clearly it does create a problem. It is not clear that announced
randomization necessarily helps much in this case.

While I am at it, I shall also note a bugaboo of mine for Referencing, which involves
something that is standard in the natural sciences and seems to have spread from them
to many of the social science journals as well. It is the use of initials with a last
name. I can name several pairs of economists who have different first names but identical
initials who do, or have, published within the same sub-areas of economics: There are two
(were) two J.K. Galbraiths, John and James, two R.E. Baldwins, Robert and Richard, two
R. Gordons, Robert and Roger, and two A. Rapaports, Ariel and Amnon, and there is another
one that I am forgetting at the moment…

]]> By: Adrian Tschoegl http://www.overcomingbias.com/2008/01/the-ordering-of-authors%e2%80%99-names-in-academic-publications.html#comment-410174 Adrian Tschoegl Wed, 02 Jan 2008 16:30:39 +0000 http://prod.ob.trike.com.au/2008/01/the-ordering-of-authors%e2%80%99-names-in-academic-publications.html#comment-410174 Denis: That's the dilemma. Any unambiguous pecking order results in bias, and so loss. Removing the bias introduces conflict, and so loss. Steve: cute. Still, I would have thought that a "paper, scissors, rock" procedure would have worked reasonably well for Kohfeld and Likens. rjs: Journals' style manuals impose the "et al." to reduce tedious repetition and waste of space. Even then, though, the problem seems to be that salience rests with being the first mentioned. If the citations in the text are represented by a number that links to a sequence in the bibliography, that reduces the effect of the repetition of the name of the first author, but it doesn't remove the salience of the first name in the sequence. In the case of some papers that I co-authored with Cliff Ball (now at Duke) and Walt Torous (now at UCLA), the conversational reference will generally be, "That paper about X that Cliff and those other guys wrote." Denis: That’s the dilemma. Any unambiguous pecking order results in bias, and so loss. Removing the bias introduces conflict, and so loss.
Steve: cute. Still, I would have thought that a “paper, scissors, rock” procedure would have worked reasonably well for Kohfeld and Likens.
rjs: Journals’ style manuals impose the “et al.” to reduce tedious repetition and waste of space. Even then, though, the problem seems to be that salience rests with being the first mentioned. If the citations in the text are represented by a number that links to a sequence in the bibliography, that reduces the effect of the repetition of the name of the first author, but it doesn’t remove the salience of the first name in the sequence. In the case of some papers that I co-authored with Cliff Ball (now at Duke) and Walt Torous (now at UCLA), the conversational reference will generally be, “That paper about X that Cliff and those other guys wrote.”

]]> By: rjs http://www.overcomingbias.com/2008/01/the-ordering-of-authors%e2%80%99-names-in-academic-publications.html#comment-410173 rjs Wed, 02 Jan 2008 03:48:09 +0000 http://prod.ob.trike.com.au/2008/01/the-ordering-of-authors%e2%80%99-names-in-academic-publications.html#comment-410173 It seems like you're only looking at one half of the problem? Why does the position matter? If it's purely because of names being lost in the "et al", then scrap the use of "et al". If there's more to it than that, you need to explore what it is and work from there! It seems like you’re only looking at one half of the problem? Why does the position matter? If it’s purely because of names being lost in the “et al”, then scrap the use of “et al”. If there’s more to it than that, you need to explore what it is and work from there!

]]>