Imperfect Tools: Google Scholar vs.
Traditional Commercial Library Databases
by Julie Arendt (Morris Library, Southern Illinois University) <email@example.com>
ike every other resource that a library might cabulary or find the special materials, they cannot portunities with
offer, Google Scholar has strengths and experience these supposed advantages. For there two questions,
limitations. Instead of rejecting Google to be a clear advantage of a subscription database “What are the
Scholar because it does not do everything that the over Google Scholar, novice users should be most important
library or librarians do, Google Scholar should be able to complete their work more easily with the things for them
accepted or rejected based on how well it assists subscription database than they can with Google to learn from
in a particular step in information seeking. That Scholar. Many subscription databases provide a my presenta-
step traditionally has been assisted by indexing clear advantage by simplifying access to special tion?” and,
and abstracting resources. In some circumstances materials or by leveraging their controlled vo- “What can I
Google Scholar is a better tool than the indexing cabularies. The interface designs that highlight teach them that will
and abstracting resources; in other circumstances subject terms next to results sets, such as those help them the most on their work for this course?”
it is not. This article examines the strengths and in EBSCOhost and Engineering Village, should My answers to these questions are always more
weaknesses of Google Scholar compared to be commended for their effort to guide novices to than I can fit into a fifty-minute session. I have to
subscription indexing and abstracting databases. controlled vocabularies without interrupting users’ jettison the material that is less essential.
It critiques college and university libraries’ con- searches. Some databases and interfaces simplify If the best reason I can find for teaching a
tinued use of subscription databases that fail to users’ work in other ways. For example, Web of particular subscription database to undergradu-
provide a clear advantage over Google Scholar. Knowledge provides citation assistance through ates is simply to expose them to the database of
When Google Scholar was introduced, it EndNote Web, and full-text resources like JSTOR a particular discipline, it is a topic that I consider
initially met with some praise and a fair amount provide easy access to complete documents. less essential. After they graduate, most students
of criticism from the library world. Both the It has been argued that the subscription database will no longer be affiliated with a university and
praise and criticism generally were deserved. is better than Google Scholar after a user learns may no longer have easy access to university-level
Unfortunately, early responses sometimes com- how to use it.8 For most students, especially un- subscription databases. Although it may be pos-
pared Google Scholar to the library as a whole1 dergraduates, this amount of database knowledge sible for graduates to travel to the nearest public
or to an idealized vision of library databases2 is unrealistic. Often the end users of the library’s university library or to find a way to purchase
rather than to the real, imperfect indexing and databases have not had any formal training.9 If they short-term access to a database, the time, effort,
abstracting databases offered through the library. receive training, it often is a one-time guest lecture and expense involved are substantial barriers that
Some of the faults that early commentators found by a librarian or informal instruction at the refer- should not be ignored. Doesn’t it make sense to
in Google Scholar included lack of a controlled ence desk. Because one-shot instruction sessions introduce students to appropriate free resources
vocabulary, lack of authority control, incomplete and brief instruction during reference encounters rather than expecting them to find a way to
or uneven coverage depending on discipline, and are the norm, there rarely is time for most users to get access to subscription resources? In many
time lags between publication and appearance in thoroughly learn how to use a database. Typically cases, the appropriate free resource is Google
the database. These same faults could be pointed there is only enough time to transform a complete Scholar, although it could be ERIC, PubMed,
out for Web of Science, a venerable subscription novice database user into a slightly-less-than-com- AGRICOLA, or another conventional library
database. Another criticism of Google Scholar plete novice. Even when there is more time, the resource that does not require payment. Even for
was that its definition of “scholarly” includes time spent teaching a database reduces the time graduate students, where familiarity with the most
materials that have not undergone peer review, so available to teach information literacy skills. The important databases in their field should be a part
it may lead users to this unvetted material. Again, Information Literacy Competency Standards for of students’ education, Google Scholar has value.
this criticism also could be leveled against a sub- Higher Education from the Association of Col- Many graduate students will go on to be faculty,
scription database. For example, book reviews, lege and Research Libraries lists five capacities and even though they will still conduct research,
editorials and commentaries regularly appear in of an information literate person.10 Only one of they may not have the same library resources. At
search results from Academic Search Premier, those capacities deals directly with searching colleges with small budgets, the premier database
even when the search is limited to scholarly (peer techniques. for a discipline may be too expensive. As Yvonne
reviewed) journals. Instead of comparing Google As suggested by Diane Zabel, perhaps it Jones described, alternatives for faculty in this
Scholar to the ideal resource, a fairer comparison would be better for librarians to have regular, situation can be to search multiple subscription
would be to actual subscription databases. ongoing collaboration with faculty to integrate databases to get about half the coverage of the
Some evaluations have explored whether a information literacy throughout students’ disci- premier database or to search Google Scholar
subscription database produces better results than plinary studies.11 Perhaps it would be better to to get about half the coverage of the premier
Google Scholar. When librarians conduct test teach the broader information literacy concepts in database.14 With those options, searching Google
searches using advanced search features in library a separate, mandatory course and to use one-shot Scholar is a reasonable choice.
databases, they get somewhat better results with instruction sessions for discipline-specific biblio- Another reason to teach a subscription da-
the database than with Google Scholar.3-5 When graphic and database instruction.12 In colleges and tabase is to present general tactics for database
college students conduct the searches, the advan- universities that manage to successfully implement searching. Students can apply skills, such as
tage for the subscription database evaporates. either model, librarians would have the luxury of selecting keywords, leveraging controlled vo-
The sources students find from Google Scholar approaching reference and one-shot instruction cabularies, using Boolean logic, and broadening
are as good as or better than those found through sessions with the knowledge that students will or narrowing a search, to other situations. Some
the library’s databases.6,7 For these novice users, cover the other important ideas somewhere else. of these skills are possible to teach within Google
often subscription databases do not provide a clear I do not have that luxury, and many of my col- Scholar, and some are not. The trouble is that the
advantage over Google Scholar. leagues at other institutions also work without that skills are taught at the same time as the arbitrary
Librarians may be able to use controlled vo- luxury. I go to classes where students’ exposure to mechanics of where to click to get a particular
cabularies to produce more precise results from information literacy is as varied as the courses and database to work. Even for databases with the
a database than from Google Scholar or to find instructors they have experienced up to that point. best interfaces, it takes several steps of naviga-
special materials that could not be found through I am not the first to suggest that in a world with tion through the library Website just to get to the
Google Scholar, but library patrons are not librar- Google Scholar, it is time to move away from database. When the database requires several
ians. Simply having a controlled vocabulary or teaching the mechanics of searching databases additional clicks, I wonder if the core message
special materials is not good enough for a novice to teaching more of the whole of information will get buried in the procedures. Every minute
user. If users cannot figure out the controlled vo- seeking.13 I try to approach these teaching op- continued on page 28
26 April 2008
Against the Grain / <http://www.against-the-grain.com>
Imperfect Tools ...
from page 26
spent teaching these mechanics is a minute less
against the grain profile
spent on teaching general concepts in database
searching. Reference Librarian, Sciences/Assistant Professor
Sometimes those extra minutes on database Morris Library; Southern Illinois University Carbondale
navigation are worthwhile. In some subjects, the 605 Agriculture Drive; Mail Code 6632; Carbondale, IL 62901
appropriate disciplinary database may produce Phone: (618) 453 2779 • Fax (618) 453 5706
better results with less effort for students despite <firstname.lastname@example.org> • http://www.lib.siu.edu/index.html
the extra navigation. Google Scholar is weaker
in the social sciences and humanities than it is
in the sciences.15 Some disciplinary databases Born & lived: I was born and grew up in Neenah, Wisconsin. I have lived in
have useful search features that are unavailable Madison, Wisconsin; Bochum, Germany; Tucson, Arizona; Ann Arbor, Michigan;
in Google Scholar. When students know how and Carbondale, Illinois.
to use these features, they appreciate them.16 On in my spare time i like to: Bicycle, run, read and
the other hand, when the interface is hard to use cook vegetarian food.
and the advantages over Google Scholar are
small, those extra minutes spent on navigation
Goal i hope to achieve five years from now:
pale in comparison to the other things that could I hope to have successfully earned tenure!
be taught. how/where do i see the industry in five
Although an hour is too short to build “an years: In a lot of ways, the industry will be like it is
intellectual framework for understanding, finding, today. Online access to materials will still be growing.
evaluating, and using information,”17 it is enough The end of print will still be far from reality. Libraries
time to encourage students to think critically about will still be struggling to afford all the materials and
the information they find and to think about the services that they would like to provide.
legal and social issues involved. Knowing why
it matters that there are differences between a
white paper, newspaper, magazine, or scholarly database is no longer the vital tool for discovery the alternatives. Two basic questions worth
journal article, or some other type of source will it once was. Money not spent on a hard-to-use considering when evaluating subscription and
serve students a lot longer than knowing where indexing and abstracting database can instead be instruction choices: 1. How is this database better
to click in a particular database interface to find spent to supply the full text information itself. than Google Scholar? 2. Assuming the subscrip-
its advanced search tools. Knowing why it is For some indexing and abstracting databases, it tion product is better, is the advantage worth
important to cite sources should be useful after is time to reexamine their value. the money and resources that would have to be
graduation, unlike knowing where to click on the I am not arguing that subscription index- devoted to it? These questions remain valid, but
college library’s Website. ing and abstracting databases should all be the answers will depend on the library’s patrons,
One objection that may be raised to teaching abandoned, but they should be compared with budget and philosophy.
Google Scholar is that it will direct students
away from subscription databases that the library Endnotes
spends so much to have. In writing this article,
1. Cathcart, Rachael, and Amanda Roberts. “Evaluating Google Scholar as a Tool for Information
I felt apprehensive that I would be accused of Literacy.” Internet Reference Services Quarterly 10, no. 3/4 (2005): 167-76.
disloyalty to the library and to the profession for 2. York, Maurice C. “Calling the Scholars Home: Google Scholar as a Tool for Rediscovering the
directing students to a non-library resource like Academic Library.” Internet Reference Services Quarterly 10, no. 3/4 (2005): 117-33.
Google Scholar. I believe that presenting arcane 3. Shultz, Mary. “Comparing Test Searches in PubMed and Google Scholar.” Journal of the Medical
or confusing databases with no clear advantage Library Association 95, no. 4 (2007): 442-45.
over Google Scholar will do more to drive users 4. Levine-Clark, Michael, and Joseph Kraus. “Finding Chemistry Information Using Google
away than directing them to Google Scholar will. Scholar: A Comparison with Chemical Abstracts Service.” Science & Technology Libraries 27, no.
As part of the library profession, my goal is to 4 (2007): 3-17.
guide patrons toward what I believe are the best 5. Callicott, Burton, and Debbie Vaughn. “Google Scholar vs. Library Scholar: Testing the Perfor-
resources for their research. Sometimes those mance of Schoogle.” Internet Reference Services Quarterly 10, no. 3/4 (2005).
resources are within the library, and sometimes 6. Helms-Park, Rena, Pavlina Radia, and Paul Stapleton. “A Preliminary Assessment of Google
they are not. From the student’s perspective, the Scholar as a Source of EAP Students’ Research Materials.” Internet and Higher Education 10, no.
value of the database is not in the dollars that 1 (2007): 65-76.
the library paid for it but in the usefulness of the 7. Jung, Seikyung, Jonathan L. Herlocker, Janet Webster, Margaret Mellinger, and Jeremy
information it provides. For them, the database Frumkin. “Libraryfind: System Design and Usability Testing of Academic Metasearch System.”
Journal of the American Society for Information Science and Technology 59, no. 3 (2008): 375-89.
that can lead to the best resources for the task
8. Callicott, & Vaughn.
with the least effort is the one that is worth the
9. Elsevier. “Usability Drives Value of Bibliographic Databases.” (2003): http://web.archive.org/
most. It does patrons a disservice to direct them to web/20040716112223/http://www.info.sciencedirect.com/content_coverage/databases/sd_bdwhitepaper.pdf.
library-paid resources out of tradition or because 10. Association of College & Research Libraries. “Information Literacy Competency Standards for Higher
they are expensive. Education.” (2007): http://www.ala.org/ala/acrl/acrlstandards/informationliteracycompetency.cfm.
Libraries perennially have had the problem 11. Zabel, Diane. “A Reaction to ‘Information Literacy and Higher Education.” Journal of Academic
that more information exists than any one library Librarianship 30, no. 1 (2004): 17-21.
can afford to possess. At one time, a library’s in- 12. Owusu-Ansah, Edward K. “Information Literacy and Higher Education: Placing the Academic
dexing and abstracting databases were vital for pa- Library in the Center of a Comprehensive Solution.” Journal of Academic Librarianship 30, no. 1
trons to discover information. Libraries willingly (2004): 3-16.
sacrificed the ability to possess some materials to 13. Williams, Genevieve. “Unclear on the Context: Refocusing on Information Literacy’s Evaluative
pay for indexes and abstracts. Librarians knew Component in the Age of Google.” Library Philosophy and Practice 2007, June (2007).
that the information hidden in journals and books 14. Jones, D. Yvonne. “Biology Article Retrieval from Various Databases: Making Good Choices
with Limited Resources.” Issues in Science and Technology Librarianship 44 (2005).
would stay hidden if their contents were too hard
15. Neuhaus, Chris, Ellen Neuhaus, Alan Asher, and Clint Wrede. “The Depth and Breadth of
to find. Today libraries still deal with the problem Google Scholar: An Empirical Study.” portal: Libraries and the Academy 6, no. 2 (2006): 127-41.
that there is more information than any library can 16. Jung, Herlocker, Webster, Mellinger, & Frumkin.
afford. Because Google Scholar offers an alter- 17. Association of College & Research Libraries.
native, the subscription indexing and abstracting
28 April 2008
Against the Grain / <http://www.against-the-grain.com>