---------- Forwarded message ----------
Date: Fri, 5 Jul 1996 16:13:20 EDT
From: Albert Henderson <70244.1532@COMPUSERVE.COM>
Subject: Journal use

I am glad see the following and to have an opportunity to clarify my
position on use and usage studies. - Al Henderson

Steve Black <blacks@ROSNET.STROSE.EDU> wrote:

>   Mr. Henderson seems to be saying that if a library has no print
> subscription, it has inadequate access to research.  That point should be
> challenged on several grounds.

I have tried to be specific about the mission of a library collection in a
research university, which is considerably different from a liberal arts
school, a voc-ag, a teachers college, etc. I would also indicate that
geography and opportunities for local resource sharing play a great role
in these matters. There are only 125 research universities, I believe.

>  First, if the goal is to provide current research to the people who
> need it, a printed journal may not be the best means of delivery.  The
> rapid increase in the exchange of scholarly information on the internet,
> especially in the sciences, suggests that alternatives to print are
> sometimes superior.  The forte of print journals is to record and store a
> fixed record of research, not to rapidly disseminate research.

This is correct, of course. However, in terms of usage Donald W. King's
research has indicated that scientists and engineers use of library
collections has been increasing, often to read or photocopy articles that
appeared in journals they once subscribed to but discarded. If my own
experience is any indication, such searches often lack precise citations
and browsability of the collection is essential. "Current research" in
science generally includes roots that may go back several decades as well
as immediate laboratory and/or field work. Lines of research that have
been dropped, as the study of steriods was in the 1940s, may be resumed.

>   Second, a system of ready access to indexing combined with reasonably
> efficient document delivery *is* a workable alternative to having a
> journal in house.

To the extent that indexes were comprehensive, this might work pretty
well. However, most most indexes offer limited coverage and, as a result
of canceled subscriptions, they have been reducing their coverage. See the
article by R. Kaser in PUBLISHING RESEARCH QUARTERLY (11,3 1995 pp 10-24).
Document delivery is not without its problems of lost time, high cost, and
failure.

> True, there is a loss in browsing and serendipitous
> finds when journals are not on the shelves, but economic reality prevents
> libraries from having all journals of interest to our users.

One of the points I have been trying to make in terms of the "economic
reality" has been that the university bureaucracy has drained financial
resources that used to belong to the libraries. Spending on libraries has
not kept up with spending on research. Spending on administrative
expansions that contribute little or nothing to research, instruction, or
public service has outpaced all other categories in the 1980s.  (I will be
very happy to share extensive details by mail.) I consider this very
unfair to library users and to librarians.

I can understand why university employees would not wish to publicly
oppose or criticize your administration. However, hundreds of researchers
have signed an ad hoc petition asking the Federal government to reform the
way it provides financial support to libraries. For reasons too complex to
recite here, the major professional associations of scientists, scholars,
and librarians have kept to the sidelines on this issue in spite of their
obligation to represent the interests of their members..

>   Third, if free market economics apply to journals, libraries *should*
> behave in the best economic interest of the communities they serve, and
> let the invisible hand slap us all around ;-).

The productivity of research, which depends on excellence in preparing
proposals and reviews, is difficult to measure. However, a reading of
published research 5 years after publication provides 20-20 evaluation. In
the late 60s and early 70s studies rejected 25 to 50 percent of published
research. The most recent study rejected 80 per cent. In other words, the
invisible hand has been at work and I think it's only a matter of time
before the body politic catches on. They have already caught university
administrations spending research money on alcoholic parties and other
extravagances unrelated to science.

You really must take cost-benefit analysis beyond the library and apply it
to the university mission. The absence of a journal may deny a scientist
the opportunity to do breakthrough work. IBM scientists Bednorz and
Mueller based their Nobel Prize-winning research in superconductivity on a
report published in a French journal that you would probably consider used
too little to keep around. I called the University of Houston (center for
superconductivity research in the US) serials librarian who said they had
never taken it.

So, in terms of a "library" cost-benefit, what's a Nobel Prize worth?
Nothing. No library, of course, ever gets one.

> A recent article by Bruce
> Kingma, "The Economics of access vs. ownership:  The costs and benefits
> of access to scholarly articles via Interlibrary Loan and journal
> subscriptions", Journal of Interlibrary Loan, document delivery &
> information supply 6(3) 1996, concludes that for the costs incurred at a
> university library (SUNY-Albany), the decision-maker may subtract $63
> (the fixed cost of having a journal in the library) from the subscription
> price, and divide by total uses of all subscription years.  If the
> resulting figure is less than the cost of ILL/document delivery, the
> subscription should be retained.  If it is more, it is economically more
> efficient to cancel the subscription and substitute it with document
> delivery.

I have no problem with this insofar as it appears to be a very limited
analysis (I haven't read it), although it does ignore the value of benefit
to the library patron. My complaint with the usage studies performed for
this purpose is that they are methodologically sloppy and so intent on
blind obendience to a budget. Results are likely to be unreliable and
unfair to the library patrons, particularly those involved in esoteric
research. There are certainly examples of publications so narrow that they
would be of interest, and might be essential, to only one research
program. That program may delegate one researcher to go through that
journal and photocopy contents and articles for the entire group. This is
common practice yet I have never seen it dealt with in a usage study
published in a library management journal. I would prefer to see
collection development decisions made in negotiation with the relative
handful of research-level users who use the most esoteric journals rather
than to use figures that mix senior scientists and undergraduates usage.

>   If all libraries actually did such a cost-benefit analysis and acted on
> it, I wonder what would happen.  Perhaps we'd see lots of journals priced
> at $62 ;-).

Cancellations result in higher prices because the front-end costs of
preparation and administration are paid by fewer participants. For
example, the chilling of the research climate in the late 1960s resulted
in sharp cancellations; PHYSICAL REVIEW decided to subdivide and, in the
process, raised its per-page prices substantially. Its library
subscriptions are now half what they were in 1968 balanced by much higher
pricing. It also produces many more pages representing increased research
activity. The economic model that is appropriate is probably a cooperative
or club rather than classic Adam Smith agricultural supply and demand. The
libraries represent a pool of intellectual resources. If financial support
drops, the pool dries up thus limiting the activities of members; vendors
raise prices to stay in business or quit.

>   The problem with free market economics, of course, is that research is
> not widget production, and the advancement of knowledge requires a
> committment to subsidize inquiry in areas that may have no immediate
> financial return.  But even if that committment is in full force,
> subscribing to print journals may not be the best way to meet the needs
> of researchers.

By all means, take advantage of all the options. However, examples of new
media replacing old media are rare. The journal was an improvement on the
printed letter, which is absolutely thriving in the form of email and
preprints!

Thanks for taking notice and responding. I appreciate your interest.

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