Google Scholar

Abstract

Google Scholar is searching for words in a website, not vocabulary terms that have been vetted and associated with citations, as in the other databases. Because Google Scholar retrieves data from a variety of resources, it can be used to answer the basic questions.

Google Scholar. Google, 1600 Amphitheatre Parkway, Mountain View, CA 94043; 650.253.0000; fax, 650.253.0001; http://www.scholar .google.com. Free website

Google has certainly had an impact on the way searches are conducted: easy to use, many results. In November 2004, Google Scholar was born. Google Scholar was designed, like its predecessor Google, as a plain, simple, easy-touse interface for finding information. Search results can include scholarly literature citations, peerreviewed publications, theses, books, abstracts, other literature, articles from academic publishers, professional organizations, preprint repositories, universities, and other scholarly organizations [I]. This is far more comprehensive than many of the databases that are listed in most academic health sciences libraries' websites. Google Scholar is responsible for bringing more users to biomedical journal websites, such as BMJ, than PubMed [2, 3]. It even crawls the OCLC Worldcat <http://www .worldcat.org> website for books.

The obvious main feature of Google Scholar is its simplistic approach to searching for scientific, medical, and legal information. The contrasting databases are PubMed, Web of Science, and Scopus [4]. In almost every test case, Google Scholar was shown to have consistently higher retrieval than PubMed and Web of Science. Higher retrieval may not always be beneficial to the user. It depends on the purpose of the search and the ability to control retrieval.

Google offers an Advanced Searching tool that only partially aids in controlling retrieval; however, it is not as specialized as having the ability to search using a vocabulary. PubMed has a specific well-publicized list of biomedical journals that it indexes and has a thesaurus, Medical Subject Headings (MeSH), that is recommended over keyword searching for retrieving the most appropriate literature [1, 4, 5]. When using keywords instead of MeSH in PubMed, retrieval is higher; however, retrieval is always limited to the journals that PubMed indexes [I]. Google Scholar does not have this limitation. The purpose of PubMed is to provide vetted medical literature to the users - not just any website that happens to have the keywords that were entered in the search box. Given Google Scholar's search algorithms, a possible use of Google Scholar at the professional level might be as a supportive tool for a systematic review to ascertain that all possible articles have been retrieved.

Since the last review [6], Google Scholar added "cited by" linking, a feature available in Web of Science and Scopus. Web of Science is well known for its ability to locate how often an article has been cited. Google Scholar populates the "cited by" link by crawling the web for scholarly literature, and the algorithms find the citing references. Searching in Google Scholar's Advanced Searching mode for "N Cecchino" in the author field gave eight results. The top-two results were cited by other articles, once each. This indicates that "cited by" dictates the ranking of each link. Performing the same author search in Web of Science did not find the citation at all, which implies that Web of Science's indexing can also be faulty, especially because one of the journals is indexed in the Web of Science. In light of the "cited by" functionality in Google Scholar, in this case, Google Scholar outperformed Web of Science, despite the breadth of what it searches.

Google Scholar's retrieval may present older primary literature first in the list of links because of the number of times the links have been clicked and the number of times the articles were cited, whereas the PubMed default is to present the most current citations first [I].

At the time of the last Google Scholar review [6], Elsevier, Highwire, American Chemical Society, and Ingenta were not included in the agreements for full- text access. Since then, Google Scholar has enabled access to the full text of these publishers. Because Google does not publish a list of accessible journal titles, the ability to say which titles or articles are available would require searching all journal titles.

The lack of control over what is and is not accessible via Google Scholar makes it less trustworthy, especially when performing searches for specific information, and thus second best to the other search tools. Of the databases tested - PubMed, Scopus, Web of Science, and Google Scholar - PubMed stood out as being the most currently updated of the five. Google Scholar's retrieval was inconsistent but could access obscure resources. Scopus had a broader journal base but lacked recent journal articles, when compared to Web of Science. PubMed continues to be the "optimal tool" for biomedical research [7].

The Advanced Searching screen in Google Scholar is an attempt to improve search options by specifying author searching, searching terms separately, using inclusive terms in phrases, and using a variety of other checks. Searching without specificity makes for a larger retrieval. One of the advantages of the Advanced Searching option in Google Scholar is that it helps hone the search results; however, the retrieval is not controlled by vocabulary indexing but by keywords in the web pages, citation information, and the number of times a link has been clicked. Because of this retrieval method, newly published articles or websites will not rank at the top of a Google Scholar search results, and retrieval may not have anything to do with what the searcher is trying to find. Google Scholar is searching for words in a website, not vocabulary terms that have been vetted and associated with citations, as in the other databases.

Because Google Scholar retrieves data from a variety of resources, it can be used to answer the basic questions. Google Scholar may also help high school students or college students who are interested in writing a paper on a health-related topic or legal issue. In biomedical or clinical searching, because of the nature of Google Scholar's indexing versus that of PubMed, PubMed continues to be the premier searching tool for biomedical literature; Google Scholar may be considered for a back-up search tool. Typically, users just want their information and do not want to implement strategies to access information. Google Scholar could be a powerful tool for searching if users combine the hedges and hacks to get at information; however, all words are still treated like keywords [8]. This crawling algorithm is Google Scholar's drawback when it comes to doing a thorough, well-vetted search. While PubMed's interface is not like Google Scholar's, it still offers the logical approach to finding the appropriate literature. So, "no serious researcher interested in current medical information or practice excellence should rely on Google Scholar for up to date information" [6]. The question remains: "How can librarians succeed in getting the message out that easy" (more) is not always better [6]?

References

References

1. Freeman MK, Lauderdale SA, Kendrach MG, Woolley TW. Google Scholar versus PubMed in locating primary literature to answer drug-related questions. Ann Pharmacother. 2009 Mar; 43(3):478-84. Epub 2009 Mar 3.

2. Kulkarni AV, Aziz B, Shams I, Busse JW. Comparisons of citations in Web of Science, Scopus, and Google Scholar for articles published in general medical journals. JAMA. 2009 Sep 9;302(10):1092-6.

3. Giustini D. How Google is changing medicine. BMJ. 2005 Dec 24;331(7531): 1487-8.

4. Shultz M. Comparing test searches in PubMed and Google Scholar. J Med Libr Assoc. 2007 Oct;95(4):442-5.

5. Henderson J. Google Scholar: a source for clinicians? CMAJ. 2005 Jun 7;172(12):1549-50.

6. Vine R. Google Scholar [electronic resources review]. J Med Libr Assoc. 2006 Jan;94(l):97-9.

7. Falagas ME, Pitsouni EI, Malietzis GA, Pappas G. Comparison of PubMed, Scopus, Web of Science, and Google Scholar: strengths and weaknesses. FASEB. 2008 Feb;22(2):338-42. Epub 2007 Sep 20.

8. Ripple AS. Expert Googling: best practice and advanced strategies for using Google in health sciences libraries. Med Ref Serv Q. 2006 Summer; 25(2):97-107.

DOI: 10.3163/1536-5050.98.4.016