Gaga over Google? Scholar in the Social Sciences
| Date | 01 September 2005 |
| DOI | https://doi.org/10.1108/07419050510633952 |
| Pages | 42-45 |
| Published date | 01 September 2005 |
| Author | Susan Gardner,Susanna Eng |
| Subject Matter | Library & information science |
Gaga over Google?
Scholar in the Social Sciences
Susan Gardner and Susanna Eng
42 LIBRARY HITECH NEWS Number 8 2005, pp. 42-45, #Emerald Group Publishing Limited, 0741-9058, DOI 10.1108/07419050510633952
What is Google Scholar?
The appearance of Google Scholar
(http://scholar.google.com) in November
2004, still currently in beta testing, had
people proclaiming the possibility of
``the world's most exhaustive academic
library'' (Rupley, 2005). Google Scholar
(i.e. ``Scholar'') is a free search engine
that retrieves academic, ``scholarly''
literature when its robot software crawls
the invisible or ``deep'' web of
participating publishers, unearthing
items not accessed by other search
engines. The invisible web is 500 times
larger than the surface web and 2,000
times greater in quality content, so the
potential of Scholar significant (Myhill,
2005). Scholar also includes sources it
discovers when its own items cite them,
even if the actual document is
unavailable online. It has the potential to
be a one-stop shopping place for
scholars, but is currently strongest in
scientific, technical, and medical
disciplines (Abram, 2005). However,
Google's vagueness about Scholar's
actual source list and publisher
partnerships is troubling, and the lack of
clarity over what constitutes ``scholarly''
within the database should make
researchers cautious (Mayhill, 2005).
Google Scholar: usurping traditional
library resources?
Already, despite much excitement
and praise of the initiative as a potential
``boon for research'' (Young, 2004),
comparisons of Scholar to traditional
library resources have been swift and
unflattering. In particular, there has
been criticism of the Scholar interface.
Peter Jacso, in his monthly review
column Peter's Digital Reference Shelf,
points out that Scholar offers no
controlled language indexing like most
proprietary databases do, meaning there
can be numerous variants of author and
journal names in the database. He also
points out that some of the advanced
search features present in Scholar are
flawed. The publication search
restrictor only allows one variant at a
time, no truncation, and no quotation
searching. Jacso's searches utilizing the
``year'' restrictor actually retrieved
more results than not using it. The use
of the Boolean ``OR'' operator
(traditionally an expander) resulted in
fewer, not expanded, results. Finally,
Jacso noted that the output of results
allows no sorting, downloading, or
emailing like many subscription-based
article index databases do (Jacso,
2005).
How does the content compare with
the mostly fee-based information
retrieval systems currently used by
libraries? Will it put subscription-based
services out of business? There have
been limited tests done comparing these
types of sources, and Scholar usually
compares poorly. Scholar compared
negatively in informal tests against the
fee-based, multidisciplinary citation
databases Web of Science (ISI) and
Scopus (Elsevier), as well as against
Innovative Interfaces Inc.'s metasearch
engine Metafind and the discipline-
specific databases PsycINFO and
ATLA (Jacso and Goodacre, 2005).
Scholar also was found inadequate
when compared with native search
engines of publishers' archives and
with Elsevier's free Scirus search
engine for the sciences (Jacso and
Felter, 2005).
Rationale
Given the incongruity between the
purported capabilities of Google
Scholar and its actual delivery of
services, we sought to conduct an
investigation to determine whether it
would be appropriate to incorporate
Google Scholar into reference and
instruction at our undergraduate library.
How would it compare with
traditionally taught resources?
Anecdotal evidence already indicates
that undergraduate students at USC
already use Google Scholar, since it is
an interface they easily recognize.
Therefore, we believed it was critical
for us to become familiar with the
system and to maximize its potential.
Students will most likely bypass our
reference assistance for this product, so
we needed to see if the information they
receive will be adequate for their
research. If Google Scholar lives up to
its promise, institutions of higher
learning may no longer need to
subscribe to and pay for the fee-based
databases that are so pervasive today.
Going into this experiment, we
expected Google Scholar to be far
inferior to the subscription databases
based on the reviews we had read.
Methodology
Because of our own research
backgrounds, we wanted to test Google
Scholar against some of the larger and
most respected subscription databases
in the social sciences to determine how
it fared. We chose PsycINFO, ERIC,
and ISI's Social Science Citation Index
(SSCI) as our representative databases,
and we chose ``homeschooling'' as our
subject area since this is a topic with
both an educational and a psychological
point of view. We did not intend to
compare the results of the three fee-
based databases against one another,
but only against Google Scholar. On
June 2, 2005, we did searches on
homeschooling in these three databases
and Google Scholar. We typed in
``homeschooling'' or ``home schooling''
To continue reading
Request your trial