It's grad recruiting time.. As part of an effort to look at web resources for prospective graduate students and how program rankings (NRC, US News etc.) might be used on the internet by prospective students, my department was looking recently at the rankings section of the website phds.org. This site has a personalized system for giving an individual a list of grad schools that might be a good match based on specified interests and priorities.
Some of the categories that can be selected are straightforward: Do you want to be in a large department or small? Do you care if the program has faculty who publish highly cited papers? Options are: no importance, low, medium etc., with various gradations up to high.
Some of the categories are kind of strange: e.g., How important is it to you that "The distribution of publications per faculty member is uniform"? It is unspecified whether that distribution will be uniformly high, low, or mediocre. There is a similar question about the uniformity of citations.
The part that I found disappointing is called "Program Composition Measures". There is no category for selecting the importance you place on faculty diversity. There are categories related to the number of female graduate students and to the number of Ph.D.'s granted to women and minorities, but nothing about faculty. These data exist, and it is curious to exclude them from the selection criteria.
As part of a recent NRC questionnaire, faculty at PhD-granting institutions were asked what is most important to them about program composition measures, with a list of 7-8 items (I don't remember the exact number), including number of publications per faculty, citation index of faculty, number of external grants, and faculty diversity; ethnic diversity and gender diversity are itemized separately. You can only choose 4 items at most, so it is too bad that diversity is split into separate items. To me, it seemed like I had to choose between diversity (one or both types) and other items I thought were important as well.
1 month ago