Biomedical Digital Libraries Volume 3

Viewing options:  Abstract  Full text  PDF (1.4MB) Associated material:  Readers' comments  Pre-publication history  PubMed record Related literature:  Articles citing this article on Google Scholar on PubMed Central  Other articles by authors on Google Scholar Burnham JF on PubMed Burnham JF  Related articles/pages on Google on Google Scholar on PubMed Tools:  Download citation(s)  Download XML  Email to a friend  Order reprints Post to:  Citeulike  Connotea  Del.icio.us  Digg  Facebook

Journal coverage comment and question

Victoria Mitchell   (27 March 2006)  University of Oregon

First, thank you to the author for this review. Such reviews are always helpful to other librarians. Comparisons can be tricky, I know, but a couple of things popped out at me in the comparison of WoS and Scopus:

1) The comparison of the number of journals indexed is comparing only the Science Citation Index (SCI) portion of WoS with Scopus. Apparently Scopus includes social sciences, so it seems to me that it would be a fairer comparison to include the number of journals indexed in both SCI and the Social Sciences Citation Index (SSCI) in the WoS number. I believe WoS's number would still be less than Scopus', but the difference would not be as great.

2) I am mystified by the large difference in number of search results between Scopus and WoS for the same journal title for the same period of time. (American Journal of Cardiology, 1992-date.) I would love to know if the author has obtained any information on why this might be the case. Is Scopus indexing items that WoS doesn't? Since WoS allegedly indexes reviews, editorials, letters, notes, corrections, and more, it's rather surprising. Or could these search result numbers somehow be misleading? (E.g., is it possible to get more than one "hit" for the same article?)

Competing interests

Although I have pointed out an area where the comparison may not be completely fair between the two products, I do not have any personal, economic or other interest in either company or product. My institution does subscribe to Web of Science (SCI and SSCI), but not to Scopus, and I haven't had the opportunity to run any comparisons similar to the author's.

top

Response to posted comment by Victoria Mitchell

Judy Burnham   (17 April 2006)  University of South Alabama Biomedical Library

In answer to the first question, because the topics searched were health related, I don't think there would have been a significant difference in the number of citations retrieved if SSCI had been included. Because this article was targeted for biomedical libraries, I choose health related topics for the searches.

When trying to determine an answer to the second question concerning the discrepancy between WOS and Scopus for the journal title search, I discovered that with Scopus, unless you put quotes around the journal title, it searches for those words in the title, not the exact title. Also, information about the Scopus database notes completeness only from 1996 forward. When the search is re-run the results are as follows:

• Scopus - "American Journal of Cardiology" 1996-date - 9213

• WOS - American Journal of Cardiology, 1996-date - 13,152

I wanted to investigate further to find a reason for the differences, so I searched the databases for citations from the title for 1996 only. These were the results:

• Scopus - "American Journal of Cardiology", 1996 - 1431

• WOS - American Journal of Cardiology, 1996 - 816

To look further, citations from both databases were put into EndNote, duplicates removed and unique titles from each database were reviewed. Many of the Scopus unique titles had an alpha notation in the issue or pagination, i.e., issue 5A or pg 31c, making me suspect a supplement. When WOS was contacted about the citations not being included, their response was that "those issues would be claimed from the publisher."

But as to an explanation between the higher number in WOS for '96-date and the higher number in Scopus for 1996 alone, I don't have an answer!

Competing interests

None

top