Are U.S. Students a Bellwether of Quality?

See allHide authors and affiliations

Science  28 May 2004:
Vol. 304, Issue 5675, pp. 1280
DOI: 10.1126/science.304.5675.1280

At the same time university administrators raise concerns that the inflow of foreign talent is drying up, the notion persists that the quality of a graduate program can be measured by the number of foreign students enrolled. In particular, the higher the percentage, the lower the quality of the program. Although most academics admit that it's based at best on anecdotal evidence from site visits and casual conversation, they seem to be convinced of its validity.

The biggest assessment of U.S. graduate departments—the National Research Council's decennial ratings—says otherwise. The NRC assessment asks scientists to rate the quality of programs in their field, based on the reputation of their faculty members. The 1995 assessment correlated that rating with various demographic information, including the percentage of Ph.D.s awarded to U.S. citizens. The higher the score, the stronger the connection between program quality and domestic enrollment.


The NRC's analysis found a negative correlation for programs in 10 of the 26 scientific areas represented, meaning that the quality of the program is positively linked to the number of foreign graduates. The negative correlation is highest for psychology, with a rating of −0.35, and aerospace engineering, with −0.17. Even on the plus side, only two fields topped 0.30 (chemical engineering, at 0.39, and biomedical engineering, at 0.31). As NRC's James Voytuk notes, “that's a pretty weak correlation. I'd tend to disregard anything smaller than 0.30.”

Navigate This Article