RE: Open access and impact factor

["David Goodman" <David.Goodman@liu.edu>]

I agree with Brian, Sue M, and others, that the effect of open access on
impact is not likely to be a major one, especially for important papers. I
would expect it to be of more significance for minor work, that people
will read if convenient but not bother otherwise. In a year or two we
shall not have to use terms such as "likely" or "expect", because we shall
have extensive data.
 
Heather raises the recurring question of whether impact is equivalent to
merit. I agree that it is not. Without resorting to her valid but dramatic
examples, impact also has a great deal to do with fashion, with available
research funding, with perceived practical interest, with useful
methodology. I do not think that anyone has a way to measure merit,
certainly not in the short term. One need only look at the work
represented by Nobel prizes over the past century, ranging from just plain
wrong to paradigm-shifting.  The utility of measuring impact is that is a
a measurable quantity, and at least it correlates with perceived merit
under some conditions. The limitations have never been discussed better
than by Garfield, and an introduction can be conveniently found in the
help pages for WoS or the introduction to the print SCI. There are those
who think the correlation good enough to be used without regard to the
exceptions and limitations; I am not among them.
 
David Goodman
dgoodman@liu.edu