On Tue, 13 Oct 2009 14:58:40 -0400, Scott Lilienfeld wrote:
> 
> Hi Mike - I think (?) we're saying the same thing.  

Sort of.  What I've tried to do is focus on the different traditions
that areas may have which will lead to certain expectations of
how to interpret sequence of authors:

(1) Rugged Inidividualist Research:  typically a small number
of authors and the authors are listed in decreasing contribution
to the manuscript and/or research.  Let's call this the "PsycInfo
view" or the "PsycInfo heuristic" because unless one knows 
the circumstances under which author sequence was determined
a reader or a tenure-promotion committe will rely upon this heuristic
to evaluate a person's publications (e.g., 1 point for first authorship,
0.5 points for second author on a 2 author paper, 0.25 if second
author or later on multiple author paper).  This heuristic is easy
to use and may be some sort of valid.

(2) Team Approach to Research:  typically this will include a largish
number of people because this is usually grant funded psychiatric
or medical research that is expected to take several years to complete.
There will be a Principal Investigator or two who wrote up the grant
(usually with others who may be involved in the research if it is funding),
a project dirctor, and then a hierarchial structure of Ph.D.s, perhaps
nurses and M.D. (I still find the title "junior research psychiatrist" funny,
especially when they act that way ;-), M.A.s, B.A.s, and undergraduate
research assistant.  If people do a good job and stay in their positions
for the long haul, then their contributions will be acknowledge as a
co-author on resulting publications (not all publications but probably
those considered "archival", that is, the article researchers will be directed
to as the primary source(s) among the slew of papers that would be
generated in such a situation.  The sequence of authors will depend
upon (a) whether the PI was heavily involved in day-to-day operations
of the research in contrast to writing new grants, (b) the project director
managing the day-to-day activities did well enough so that the PI just
has to be check in ocassionally, and (c) relative importance of the services
delivered by the associated M.D.s, Ph.D.s, R.N.s, M.A.s, B.A.s, and
undergraduates.  For "archival" publications, the first author will either 
be the project director or PI depending upon (a) who did most of the
actual research (oversight) and writing and (b) how much of megalomaniac
the PI is, and, otherwise, the last author would be the PI  (Note: if a lab 
or facilitiy head may be listed last as compensation for use of the facility, 
which would then put the PI back up to first position).  The intermediate
order of authors would reflect some measure of the author's contribution
though this will usually be subjectively determined. This gives rise to
what I would call the "PubMed heuristics" because much of this type of
research will be published in PubMed/Medline and when one is 
examining references from such a database with, say 10-12 authors,
assigning "authorship" points to a particular person will often depend 
upon where that author's name is located (front or end; ignore the people 
in the middle).  Given that most research psychiatrists in an area are
familiar with each other and their work (NOTE:  I continue to be
shocked when I hear one psychiatrist, usually a senior well known one
proclaim "I don't believe anything that comes out of that guy's lab or
department" -- the exact reasons why this is so is better not reviewed
because it may lead to legally actionable consequences).

Anyway, what I want to say is that a person using the PsycInfo
heuristic is likely to do the following:
(A)  Papers with  up to 3-4 authors, assume that authorship is based
on amount of contribution to the research/writing.
(B)  Papers with more than 4 authors (especially around 10 or more),
just focus on the first author because if everyone is listed in decreasing
order of contribution, they are of no interest. Suspicion about the
researcher's ability are raised because he/she needed so much help
to do the research.

For people working in the "small science" work of psychology (i.e.,
not managing million dollar grants for their research or not having
any grant funding at all) then the PsycInfo heuristic will seem natural
particularly since there will only be a couple of "key personnel" who
will write the paper (or the senior faculty will write it and decide if
anyone is listed as co-author; there are a number of researchers who
will take sole authorship and list the other research personnel in a
footnote).  If we teach the PsycInfo heuristic to our students, then
they too will (a) expect to see papers with few authors and (b)
papers based on team research with a little suspicion because there
the questions of why were so many people needed to do the research,
why did the leader researcher do it him/herself?).

For people using the PubMed heuristic they are likely to do the following:
(A) Papers with up to 3-4 papers, assume that the authorship is based
on the amount of contribution of authors (grant writing researchers will only
write papers with such few authors usually as summary or review papers,
or opinion/theoretical statements).
(B) Papers with more than 4 authors, assume that the first author is the
project director (if not known to be the PI) who was given the responsibility
to write up the research, the PI is listed last, and the "little people" fall
between these two.

The PubMed heuristic breaksdown when the research reported is not
"big science", that is, big science assumes one has assembled a large
team to handle all of the aspects of a complex research project and all
of these people will be paid to make a contribution.  If one doesn't have
a grant to fund a large staff, then the additonal authors may be people that
the author/researcher was able to manage to provide some free service
in exchange for a co-authorship. The PsycInfo heuristic would be more
appropriate here but within the psychiatric/medical research community,
the PubMed heuristic may be inapprorpiately applied.

In summary, yes, let's state the obvious:  knowing more about the actual
research situation will allow one to better understand why the author
sequence is what it is.  However, if you are looking at a reference in
PsycInfo or PubMed, you are unlikely to have this knowledge unless you
know the researcher involved.  Tenure and promotions committee may
also find themselves in such situations.  I suspect that psychology departments
will use the PsycInfo heuristic in evaluating a psychogist's publications
even for team-based research.  I suspect that academic psychiatry departments
will use the PubMed heuristic  (i.e., first author is junior to the last author
who is senior and supervising; everyone else listed as co-author presumably
did a competent if not better job).

>Last authorships are viewed with skepticism in many academic 
>quarters because, as you note, they can mean a number of different 
>things depending on the situation, tradition in the lab, etc.  That is, 
>if someone is a last author on a paper, it's often hard to decipher 
>what it means unless one has a better handle on the lab tradition, 
>implicit authorship model, and so on.  Best....Scott

I guess one could set up a research program to study the different ways
that authorships are sequenced and what position in sequence means.
Then again, maybe one would also want to do research on how and why  
such consideration are important to researchers and tenure and promotion
committees as well as to others.  Ultimately it comes down to assigning
merit but, really, how hard do most people really want to work on
figuring out what the appropriate level of merit to assign?  Couldn't
they just follow the heuristic "First Authorship is Best"?

-Mike Palij
New York University
[email protected]





---
To make changes to your subscription contact:

Bill Southerly ([email protected])

Reply via email to