Even in c. 2007 there were many inaccuracies. Here, for example, Portugal is missing its refereed journal; Ireland may be over-counted. This is one of my favorite graphics to show in lectures, because each time something on it is found to be in error. (In the Courtauld, someone questioned the count for Slovakia; in Singapore, someone questioned several Asian countries…) But each time, the plausible corrections have been close to the figures here, so I think it might be safe to say they are roughly correct.
Incidentally: the Netherlands “wins” with the most art history per capita. When I gave this material in Leiden, some people in the audience applauded, but in Utrecht there was some amusement: why, someone asked, did the Netherlands need quite so much art history?
The data indicates the relatively small proportion of the Chinese public that is involved with art: the size of the art scene there, as John Clark has observed, is related to the overall population. Yet the data cannot be uniformly interpreted, because reviewing is culturally relative. Some countries, like China, have adopted it mainly in the sciences, where international collaborations and comparisons are more common.
This kind of data can help suggest patterns of emulation and resistance: places where institutions are more likely to form departments that identify themselves as art history, or that practice peer reviewing. These statistical questions are developed as insitutional histories in the essay, and in The End of Diversity in Art Historical Writing.