Lionel Robson, INTERNET:Lionel.Robson@sam.comms.unsw.EDU.AU claims:

> === It is fine and dandy trying to justify serial costs.  But the
> simple fact is that the readership level or utilisation rate of
> serial articles is extraordinarily low.
>
> If academics want to pay to have material published that says
> something about their understanding of economics.  Who would want to
> pay to publish material that the punters never read?  It would be
> like buying tickets to the movies but never going to see the action.

It is unfortunate that myths such as this can survive in institutions that
pride themselves on excellence in research. Major attempts to cook up data
in support of a conclusion that library collections are "never read," such
as the infamous Pitt study, have been discredited by serious library
researchers and repudiated by library users.

Other studies on science communications shows that most articles are well
read for many years. We also can cite a number of research studies that
indicate that research is often approved, carried out, and published on
the basis of inadequate information. All science research starts, so to
speak, in the library. If the library is poor and researchers / reviewers
are poorly informed, chances are the research will be trivial, wrong, or
will repeat work already done.

The real questions, therefore, are:

(A) What are the costs to the productivity of research and to the
effectiveness of education of decimated library collections?

(B) Why isn't there more library research supported by better library
collections?

(C) Who has benefited from the budget share cut from libraries and
instruction?

Albert Henderson, Editor, PUBLISHING RESEARCH QUARTERLY
70244.1532@compuserve.com