We had 8 attendees at today’s meeting.
Microsoft Academic Search vs. Google Scholar
Louise Klusek lead a discussion of the ins and out of these two services, how they stacked up against each other, and how they compared to Bearcat Search and Web of Science. Before today’s meeting, we had taken a look at this article from Science:
Bohannon, John. “Google Scholar Wins Raves—But Can It Be Trusted?.”Science 343.6166 (2014): 14-14. full text available
Here’s a summary of sorts of what we talked about (please add any comments to this post if I forgot something important).
Microsoft Academic Search
- Visualization of publication histories, author networks, citation networks
- Keywords that are given their own pages in the service where you get definitions, display of related keywords, publication history for that word, and more (check out this example for “information need”)
- Citation metrics for articles (for example, see this record for an article by Brenda Dervin and Patricia Dewdney)
- Links to PDFs and publisher’s record (the PDF links will only work if you are on campus or you are off campus and have authenticated yourself by using a library resource earlier AND we happen to have access to that publication)
- Browse top authors, journals, keywords, and organizations (i.e., institutional affiliations of authors) for any discipline (e.g., library science)
- Nice author profile pages (e.g., Brenda Dervin)
- Theoretically more transparent than Google Scholar about what is indexed, but we had still had lots of questions
- No connection to our SFX /Find It service that allows off campus users to gain access to content we have licenses for (Google Scholar has this in the form of “Find Full Text at Baruch” links next to items on the search results pages)
- Limited subject metadata
- Ease of use
- Interdisciplinarity (this is true of Microsoft Academic Search, Bearcat Search, and, to a lesser extent, Web of Science)
- Items in search results page feature “Find Full Text at Baruch” links that connect to our SFX service
- Article-level metrics
- “Find Full Text at Baruch” links only work if you connect to Google Scholar from our databases page, or if you are on campus, or if you have first authenticated by connecting some other library resource earlier in your browsing session
- Students have a hard time figuring out the type of source from the search results page (is it a book, a book chapter, an article, something else?)
- Lack of subject metadata
- Author profile pages aren’t automatically created (e.g., none for Brenda Dervin)
We talked also about the problem of article-level and journal-level metrics in these products, noting that the numbers rarely agree. Although we didn’t look at an example during the meeting, consider this difference in the way that Brenda Dervin/Patricia Dewdney article is counted:
Louise shared this Northwestern University Libraries guide to citation analysis in case anyone wants to delve into the topic more deeply.
On the topic of bibliometrics, we talked a bit about the popularity here at Baruch of SSRN, which provides data at the author level and the article level.
There was a lot of interest in having another Tech Sharecase in which we answered each other’s questions about how to do things in Excel. If you have anything you’d like to be able to do in Excel, just post it here as a comment so we can look into it before our next meeting.