Google Scholar as an Alternative to Scholarly Databases

By , February 18, 2011 12:08 pm

by Kerry Fitz-Gerald

A recent article, “Google Scholar’s Dramatic Coverage Improvement Five Years After Debut,” by Xiaotian Chen in Serials Review reports on the results of an empirical study comparing Google Scholar’s coverage of scholarly journals with the coverage of commercial services. Though I don’t agree with Chen’s conclusion that the dramatic improvements in Google Scholar’s coverage means libraries can drop subscriptions to commercial database providers, I do think that his work emphasizes how valuable Google Scholar can be as a cost-effective research tool.

In short, Chen explored how well Google Scholar performed at retrieving scholarly journal article records. He first identified 8 commercial databases that Google Scholar partners with and then chose 400 articles at random that appeared in those databases. He then checked to see what percentage of those articles Google Scholar could find. His findings indicated that Scholar has a 98 to 100% success rate. He concludes that “Google Scholar is able to retrieve all scholarly publications from databases and Web sites that are open to Google Scholar.” (p. 226)

For researchers without access to commercial databases and indices, this is terrific news. Even though Google Scholar won’t retrieve the full text of the article, researchers can nonetheless identify potentially useful sources without worrying that they’re missing something by not using a commercial service’s own search engine. And since Hein Online, with its broad coverage of law reviews, is a Google Scholar partner, legal researchers have a viable alternative to expensive journal searches on Lexis and Westlaw. Articles identified through Google Scholar may need to be retrieved from Lexis and Westlaw, but this could be done using a “get a document” type retrieval request, rather than a more expensive search.

Even researchers with access to a variety of commercial databases and indexes may wish to use Google Scholar. Given this highly accurate retrieval rate, Google Scholar can serve as an affordable federated search tool. Rather than search HeinOnline, JSTOR, Emerald, Eric, and Project Muse individually, a researcher can conduct a single search and retrieve items indexed in any of these databases, plus many more.

One aspect of his research that Chen glosses over, however, is his discovery that “Google Scholar excludes some non-scholarly writings in the databases it is allowed to crawl.” (p. 225) Items like reports and surveys were not locatable using Google Scholar (though they were retrievable using Google), even though these items were in databases spidered by Google Scholar. When Chen determined there was a nearly 100% successful retrieval rate, he was focusing on scholarly publications only.

This limitation, coupled with the need to be able to retrieve full text sources, cuts against his suggestion that “libraries can seriously consider cancelling a large number of subscription-based abstracts and indexes since their unique contents and value are rapidly evaporating” (p. 226). But that debate’s for another day. Today, I’m glad to know that Google Scholar works well as a free, federated search tool.

LLOPS is powered by WordPress • Panorama Theme by ThemocracyLog In