Great example of data quality issues
Look at the affiliations in Scopus:
Here's what the article pdf from the publisher has:
Now see the record in Thomson Reuters ISI Web of Science (perfect, in fact, more readable than the original):
THIS is what I'm talking about when I say that Scopus has data quality issues. I am of course not affiliated and this doesn't represent anyone I know!
Update: 11/21 - on top of all of this, apparently searches in Scopus this week have had wildly different retrieved sets - from like 900 to 20,000 when the saved search was repeated...