Read Burris's "The Academic Caste System: Prestige Hierarchies in PhD Exchange Networks" in American Sociological Review. He argues that departmental prestige is most correlated to amount of social capital, i.e. level of "well-connectedness" to other institutions in the field. He measures this by looking at who employs whose PhD students.
1) Wow. Guess my oft-spoken analogy of US PhD students being "children of their departments" is spot on. But Burris takes it one step further--they're brides. Ugh. What's the bride-price? Number of publications and external grants acquired during grad school? Not sure if I like this rhetorical territory, but it's abundantly clear to me now why there is so much pressure on students to "move on" even when taking a new degree.
2) What about international networks? Burris looks only at doctoral-granting programs in the United States, but the US isn't the only place with research universities. Does this methodological myopia indicate an intuitive sense of other countries' lack of importance in accruing social capital in the US university context? Would the degree of internationalization of different departments change his results in any way? Or are top departments also the best networked abroad? I can't help but think that perhaps departments' international social capital might be at least partially dependent upon fluke of geographic location--Is there a big airport serving their area?
**3) If Burris is right, you can see why there is an "incest prohibition" in American academe--why departments are typically adverse to hiring their own degree holders. However, this prohibition does not exist in other countries. How does prestige work in, say, the UK, where no such prohibition exists?
4) And finally, the obvious selfish question: What does this mean for my situation? T_T