Thursday, July 06, 2006

Elsevier's Scopus versus ISI Web of Science and Google Scholar

With a title like this I should attract a lot of hits -;) But to call me a Black Hat spammer right away, I don't think so. I believe this reseach deserves a lot of attention. I wonder if Jeroen Bosman will publish his report "Scopus doorgelicht en vergeleken : de dekking van de citatiedatabase Scopus inclusief vergelijkingen met Web of Science and Google Scholar" on this subject also in English. (He just confirmed to me he will, so be patient!) In the ongoing debate about the language of blogging/publication, I am convinced that in this case the language of publication is a negative impuls. Making it available in a open repository is on the other hand a very positive point. There is certainly international interest for this! In a free translation it would be: "Scopus examined and compared: the coverage of citationdatabase Scopus with Web of Science and Google Scholar." There is also a powerpoint available. This presentation is also in Dutch, but will give you a very nice extra insight and overviews of coverage with example searches and hits. This sure is interesting reading and an English summary would be great. You will have to wait for that i am afraid and be satisfied with a short preliminary one based on highlights of the summary. Highlights of the summary:
  • Changing student search habits (less focus on speciality databases, using links rather than searching systematically) and the appearance of "competitors" like Google Scholar and perhaps Windows Live Academic make research into multidisciplinary databases necessary
  • Content: Scopus has almost 28 million record and WoS -including backfiles- 37 million. Scopus includes 15.000 journal titles against almost 9000 in WoS.
  • Coverage: •preliminary conclusions: –Scopus has after 1996 about 30% more, and béfore that about 10% less coverage than WoS –Scopus coverage has a whimsical, hard to explain course through time –Coverage in bio/earth/geog/enviro/engin/comp very good –Coverage health also very good, but doubt over most recent years –Coverage psych/antropo/econ/chem acceptable, but before 1996 there are several gaps –Coverage phys/astro/math/socio less good, or reasons for doubt –Hardly coverage philo/theo/arts/lang
  • Up-to-dateness: there is hardly any difference between WoS-Scopus considering the speed of updating records
  • Citation details: again the difference is rather small, there is a large overlap between WoS-Scopus, but the difference with Google Scholar is larger and it finds more ánd more unique publications.
  • Speed: Google Scholar is faster than WoS or Scopus. WoS has the disadvantage of being a little slower than Scopus.
  • Opinions: most "power users" like the Scopus interface and value the Refine and Citation Tracker, but it can not replace -for the time being- the Journal Citation Reports, according to them. Citation Bosman, J., I.v. Mourik, M. Rasch, E. Sieverts & H. Verhoeff (2006). Scopus doorgelicht en vergeleken: de dekking en functionaliteit van de citatiedatabase Scopus, inclusief vergelijkingen met Web of Science en Google Scholar.

Tags: , , , ,


Wow!ter said...

From a reliable source, I know they are working on an English language version, an article, to be submitted to an appropriate LIS journal.
I'm not sure in what stage that articke is at this moment. I know Jeroen is reading your blog, so perhaps, this is an open invitation for a comments from his side.

Jeroen Bosman said...

Indeed we are working on an article in English, based on the report and with some updates and new insights of course. It will be submitted to an open access LIS journal. Meanwhile just be patient or try to understand the graphs and tables in the Dutch version. That must not be too hard...

Guus van den Brekel a.k.a. Namro Orman said...

The negative impuls I mentioned is just to add an argument in the earlier discussion about blogging/publishing language. It is certainly not referring to the quality of the Dutch content which is high, almost without comparison.
And together with the powerpoint this is very understandable, even for non-Dutch speaking readers.

Anonymous said...

RT@MatAbraz: New useful search engine that returns full PDF scientific articles not subject to access fees