Date: Tue, 16 Mar 1999 14:34:07 -0600
From: "Alfred B. Kraemer" <akraemer@POST.ITS.MCW.EDU>
Subject: Re: Core journals (ISI J. Cit. Reports)

There may be an application for core lists of journals for some types of
libraries. For our library, a bio-medical research library, a core list
based on criteria which are not derived from our institution, does not
seem to work.
One often-proposed approach involves the use of the ISI Journal Citation
Reports to produce a listing of high-impact journals by suject. There
are several flaws with this approach:

- subject grouping in the ISI Journal Citation Reports often does not
match the subject coverage of the journal very well. In many instances
the subject coverage of a journal is much narrower than the subject term
assigned to the subject grouping. This stacks the 'cards' in favor of
journals with a broader coverage, e.g. "Journal of biological
Chemistry"  and against journals with a narrower focus.
- the top-ranking journals -using the ISI Journal Citation Reports- are
in many instances those one would expect to be there while most of the
journals which cause us headaches -very specialized, high-priced
journals with above average annual price increases- could probably not
be evaluated accurately using a 'core list' derived from the above
journal citation reports.
- the predominant research areas at our institution often require access
at significant levels to journals that publish research in those
specialities but may not be ranked in the top 50% of an ISI subject
ranking by journal citation numbers.

Instead of relying on external core lists, we decided on a different
approach:

About five years ago, we decided to barcode every journal volume and
issue so as to be able to scan for use before reshelving. In our case
the costs were warranted in light of our journal budget. Some uses are
certainly missed but there is no indication that those 'misses' are not
evenly distributed. Cuts and adds of journals are made annually based on
use data for active subscriptions and ILL data for journals often
requested but not held.
That way we have in essence a 'core list' of journals which are used at
our institutions at levels that make a subscription worthwhile.
I feel this is the only kind of core list that works for an institution
like ours, and presumably for other research libraries.

Core list based on general data too often miss important or include
irrelevant journals.
Another consideration is the scope of core lists: the Brandon-Hlll core
list of medical journals may be useful for hospital libraries but is far
to short for our purposes.

If I have overlooked a crucial argument in favor of using external core
lists, I would like to hear about it. Certainly, our method is not an
easy one, however, I do not believe that core lists workable for
research libraries.

--
Alfred B. Kraemer
Head, Technical Services
Medical College of Wisconsin Libraries
8701 Watertown Plank Road
Milwaukee, WI 53226
414-456-4273
E-mail: akraemer@mcw.edu