[Sigmetrics] Getting accurate number of publication counts for benchmarking

[Pikas, Christina K.]

Hi All,
I posted this to my blog: http://christinaslisrant.scientopia.org/2015/12/02/bibliometrics-getting-an-accurate-count-of-articles-published-by-an-organization-in-a-year/

But I figured I would  post here, too. (Waving at co-authors who are anticipating a draft from me!)

I'm looking at scholarly output from peer institutions as part of a benchmarking exercise and I ran across several strange things in the results this year. As I caveated in the blog post, I do know that typically we limit to journal articles and use a single database to avoid some issues.

The particular thing I wanted to bring to everyone's attention is that OSA provided 6 different records for the same article to Scopus. Because these 6 conferences were collocated, the article was listed as having been presented at each (I confirmed with OSA). Scopus had this as 6 articles. If I hadn't been de-duping across two databases I wouldn't have noticed.

Christina

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.asis.org/pipermail/sigmetrics/attachments/20151202/6ff058c4/attachment.html>