Elsevier's SciVal

Wed Jul 1 11:58:55 EDT 2009

Another use of the UCSD map/data - work by Kevin Boyack and Dick Klavans.
See also http://www.info.spotlight.scival.com.
k
>  
>
>
> http://chronicle.com/weekly/v55/i40/40research_analytics.htm
>
> New Tool Compares Scholars' Research Strengths
> Big publisher offers software to track performance in 80,000 areas
>
> By DAVID GLENN
>
> Just as anxious novelists can check their sales rankings on Amazon 24
> times a day, academic researchers have a host of online tools for
> monitoring their citation stats - and those of their rivals.
>
> Google Scholar, Reuters Thomson's citation indices, and Springer's
> AuthorMapper - those are just a few of the products that claim to reveal
> which scholars and departments are having the most impact on their
> fields.
>
> Now status-conscious researchers (and their department chairs and deans)
> have a new tool to obsess over. This week the scholarly publisher
> Elsevier unveiled SciVal Spotlight, an online service that attempts to
> uncover universities' strengths and weaknesses in no fewer than 80,000
> areas of research.
>
> "We can do this at a university level as well as at a national level,"
> says Jay Katzen, Elsevier's managing director for academic and
> government products. "What are your strengths? Who are the researchers
> in your university who are driving this core competency? Who are the
> researchers at competing universities who are performing well, in case
> you want to recruit them?"
>
> But while the product boasts a new methodology, it is unlikely to
> silence the familiar criticisms of bibliometric research measurements,
> including the fear that scholars will game the system by over-citing
> their friends' work and the complaint that journal citations cannot do
> justice to the humanities and other monograph-heavy fields.
>
> Still, the company hopes to sell the service to university
> administrators and federal education departments, especially in Europe
> and East Asia, where governments are increasingly relying on
> quantitative measurements of research productivity. The price tag varies
> by institutional size but could range into six figures.
>
> Because it focuses on narrow subfields, including cross-disciplinary
> topics, SciVal Spotlight might be embraced by departments and
> universities that feel that their distinctive strengths are overlooked
> in some of the cruder research-ranking systems.
>
> Mr. Katzen says that SciVal Spotlight's chief virtue is its fine-grained
> level of detail. "With Spotlight, we've taken a step back and taken the
> journal, per se, out of the equation," Mr. Katzen says. Instead of
> relying on journal-level citation impacts, the Spotlight database
> analyzes citation patterns for roughly two million individual articles
> in Elsevier's archive.
>
> Analyzing citations at the level of individual articles offers a much
> more precise picture of influential work in emerging fields. In
> nanotechnology, for example, scholars might cite work from journals in
> physics, chemistry, and engineering - but because the work cuts across
> several disciplines, the importance of the articles might not be picked
> up in traditional journal-level citation analyses.
>
> After crunching the cross-citation numbers, Elsevier defined
> approximately 80,000 clusters, or "distinctive research competencies."
> (The company did not try to name those thousands of clusters; that would
> have kept a team of interns busy for a long time. Instead, the system
> uses references like "competency number 4, whose distinctive key words
> are 'amino acids' and 'fatty acids.'") The database then takes the
> research produced by a university and maps it onto those 80,000
> clusters.
>
> Beyond Journal Citations
>
> Diana M. Hicks, a professor of public policy at the Georgia Institute of
> Technology who often writes about university ranking systems, says that
> the new Elsevier product's value will depend heavily on how
> intelligently it has defined those 80,000 research areas. (Ms. Hicks
> spoke to The Chronicle last week, before SciVal Spotlight was unveiled,
> so she could not directly assess the product.)
>
> "It will all depend on how well that part of the engine works," Ms.
> Hicks says. If the system has correctly identified small subfields that
> are genuinely of interest to researchers and grant makers - for example,
> emerging specialized areas in nanotechnology or cell biology - then it
> could be very useful, she says, because the commonly cited measures
> produced by the National Science Foundation are not fine-grained enough.
>
> "There is huge value in being able to target narrow areas - to see how
> well you're doing in optimal electronics, for example," Ms. Hicks says.
> "You can't just pick five journals and get it right."
>
> Mr. Katzen says that he hopes the product will be of interest not only
> to administrators, but also to individual scholars who want to keep up
> with what colleagues are doing in their subfields.
>
> Elsevier's rival Springer has similar hopes for its AuthorMapper
> citation system, which went live in January. (Unlike the new Elsevier
> product, AuthorMapper is free.)
>
> "Our journal editors and authors have the task of trying to keep up with
> these fields," says Brian Bishop, Springer's director of e-product
> development and innovation. "In scientific communication, you hear a
> lot, 'Oh, I know everyone in my field. I know the best people.' And that
> may be true. But everyone's social network can benefit from just a
> little bit of innovation. It's good to double-check." He notes that at
> least a few scholars have added AuthorMapper widgets to their blogs, so
> they can share real-time visual updates of new papers that have appeared
> in their subfields.
>
> AuthorMapper relies primarily on Springer's own journal database, but
> Mr. Bishop says that he would love to add metadata from other
> publishers.
>
> Ms. Hicks, of Georgia Tech, says that research-ranking tools like these
> are here to stay. But she warns that it is difficult, and probably
> foolish, to use journal or article-based citation measures outside the
> hard sciences. "When you get into the humanities, they do a lot of
> books, and books aren't in the databases," she says. "And their
> references go back to Aristotle. They don't cite their colleagues; they
> cite Aristotle. So you don't get the same dynamics with citations. It's
> not like chemistry."
>
> Following the Money
>
> In the coming weeks, Elsevier also plans to release a second product:
> SciVal Funding, a database that alerts researchers about new publicand
> private-sector grant opportunities in their fields. Much of the
> information is electronically aggregated, but some of it is gathered by
> hand by Elsevier staff members.
>
> "The idea," Mr. Katzen says, "is to reduce the amount of time finding
> opportunities, writing proposals, and so on. We want to switch the
> equation, so you can spend more time working on your research." Because
> the system automatically meshes with Elsevier's database of researchers,
> he says, scholars need not waste time teaching the database about their
> scholarly interests. "We'd like to think this is a much better workflow
> tool than many of the databases that are out there," Mr. Katzen says.
>
> Do these new data products indicate that Elsevier, Springer, and other
> publishers foresee a day when their core journal-publishing businesses
> will no longer be so profitable?
>
> "This is the new world order that we're living in," says Mr. Bishop, of
> Springer. "Suddenly Springer is a software vendor, among other things."
>
> http://chronicle.com
> Section: The Faculty
>
>
>
> ------------------------------------------------------------------------
> --------
> Copyright (c) 2009 by The Chronicle of Higher Education
>
> Subscribe | About The Chronicle | Contact us | Terms of use | Privacy
> policy | Help 
>
>   

-- 
Katy Borner
Victor H. Yngve Associate Professor of Information Science
Director, CI for Network Science Center, http://cns.slis.indiana.edu
Curator, Mapping Science exhibit, http://scimaps.org 

School of Library and Information Science, Indiana University
Wells Library 021, 1320 E. Tenth Street, Bloomington, IN 47405, USA
Phone: (812) 855-3256  Fax: -6166 