Re: "Discoverability" Problem or Non-Existence Problem?

From: Alma Swan <a.swan_at_TALK21.COM>
Date: Sat, 21 Jul 2007 20:14:23 +0100

Frank McCown asked:
> Can anyone point me to research on the topic of how
> researchers go about finding research papers on a particular
> topic? Do most researchers start with Google Scholar, etc.
> or do they first go to OAIster or ScienceDirect? When do
> they stop searching? I'm looking for something addressing
> computer scientists, perhaps a large survey.

One of our recent studies, based on surveys about researcher behaviour,
indicated that over 70% of researchers use Google or Google Scholar as their
first-choice tool for finding information on a topic
n%20author%20study.pdf). Hardly any use OAIster (they haven't heard of it)
and ScienceDirect is further down the line, if available. We split the
responses by major disciplines (physical sciences, life sciences, social
sciences, arts & humanities) so computer scientists are subsumed into
physical sciences in the report.

Lee van Orsdel and Katherine Born also did a study recently on this topic
and reported that Google is used by over 60% of researchers (Van Orsdel L C
and Born K (2006) Journals in the time of Google. , April
15, 2006. )

When researchers do not have access to articles via their library then they
set off on a particular path of discovery. Here is an excerpt of our report
from an even more recent study
( or, based on surveys and discussions
with researchers, on what researchers do when their library holdings don't
provide access to articles they know they want to see:

"It is commonly said nowadays, with reference to scholars'
information-seeking and use behaviour, that unless something is available in
digital form it is invisible. What we learned from this study did not
entirely support that notion, but there is nevertheless a substantial degree
of truth in it. Researchers told us that they would ideally like to be able
to find everything they need for their work in digital form. The ideal is
unlikely to be fully achieved in some areas, but in most it is not an
unimaginable concept. Scholarly publishers continue with their
archive-digitisation programmes; books are now increasingly available in
electronic form; libraries are hard at work digitising legacy material and
special collections where appropriate. Tomorrow's scholars, especially in
the sciences, will find that the bulk of what they need - or think they need
- to access is accessible digitally, and for many this is already the case.

That is not necessarily to say it is available digitally. There is still a
cost for getting at much of the material not purchased by a researcher's
institution, and those extra costs must be paid for. Sometimes this is a
dissuasive factor: researchers say that if they run up against a toll
barrier whilst following up an article or book they 'give up and find
something else I can get for nothing'. This is not good for scholarship.

There is another barrier, too, that is bad for scholarship, and one that is
largely self-imposed by scholars and their (rather new-found) attitudes to
obtaining information for their work. This is that if an article or book
will take some time to obtain - and that time varies between disciplines -
then the scholar will, as in the case of toll-barriers, abandon the chase
and go without.

The usual pathway followed to obtain an article not instantly available via
a library subscription is much the same in sciences, social sciences, and
arts and humanities:
n the scholar seeks access to the full-text, anticipating immediate
and free electronic availability via the local library; if fails
n tries Google to see if an Open Access version is available; if
n emails a friend in another institution with better library provision
to see if they can email the article; if fails
n emails the author; if fails
n orders article via inter-library loan
n consults a subject librarian for expert help

The difference between the disciplines is that some will press on through
all steps while others only try some of them. Arts and humanities scholars
have an information-pursuit timeframe that is longer than that of
scientists, and told us that if a book is genuinely considered to be very
important to their work they will wait months if necessary to lay their
hands on it (and in general this still means it will arrive in printed
form). Scientists, on the other hand, and especially some kinds of
scientists who work in the very fast-moving fields, want information
immediately, on their computer screen. The two days that the British
Library can take to supply an article to a mathematician is 'bearable but
not ideal', we were told, and is viewed in that way by scientists in fields
that move reasonably fast but not excessively so (the scientists we spoke to
appeared unaware of the BL's ability to deliver within an hour, or unwilling
to consider this option because of its relatively high cost). Cell and
molecular biologists, in contrast, told us that "now means in seconds; I
want it on my screen immediately". If the article cannot be supplied within
such a very short timeframe, one that allows for a very few click-through
steps, then the pursuit is abandoned.

For scientists this is a problem, but one about which they are pragmatic: if
the article cannot be read - or at least located and requested - within
minutes, it remains invisible. For libraries trying their best to service
the needs of their users, it poses a real dilemma. No library can offer the
level of provision that such scientists say they want (indeed, say they
need). Yet despite their need, researchers go without rather than spend time
and effort and patience chasing an article deemed necessary. The result is
information-deficit, with researchers proceeding on the basis of not having
taken that information into account in their work. Does this have important

Alma Swan
Key Perspectives Ltd
Truro, UK
Received on Sat Jul 21 2007 - 21:08:56 BST

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:49:01 GMT