NRC Releases Methodology for Rating Doctoral Programs

by Daniel L. Bennett

Three years after collecting the data to rate doctoral programs in the US, the National Research Council has finally released its methodology that will be used in its ratings. The statistical methods employed are quite technical in nature (Inside Higher Ed has a good summary). One very cool thing about the NRC methodology is the differential weighting of variables by discipline, according to faculty in the field's perception. This makes sense because different fields of study value different characteristics. The variables to be used in the ratings include:

Publications per Alloted Faculty (Non-Humanities)
Number of Published Books and Articles er Alloted Faculty (Humanities)
Average Citations per Publication (Non-Humanities)
Percent of Faculty with Grants
Percent Interdisciplinary
Percent Non-Asian Minority Faculty of Core and New Faculty
Percent Female Faculty of Core and New Faculty
Awards per Allocated Faculty
Average GRE
Percent Students Receiving Full Support in 1st Year
Percent 1st Year Students with External Funding
Percent Non-Asian Minority Students
Percent Female Students
Percent International Students
Average Annual PhD's Graduated
Average Completions (8-year for humanities, 6 years for other fields)
Time to Degree (Full and Part-Time)
Percent PhD's with a Definite Plan for an Academic Position
Student Work Space
Health Insurance
Student Activities

What is not cool about the rankings is the outdated data to be included in the rankings. Our friend Bob Morse at U.S. News expressed some valid concerns:

Are the data that will be used in the rankings losing their analytical validity since they will be from the 2005-2006 academic year?

Why wasn't the NRC able to produce its rankings more quickly, using more up-to-date information?

How many faculty members have switched institutions and departments since the NRC first started collecting data in fall 2006?

My other concern with the methodology is the inclusion of race and gender diversity as a factor of quality. I concede the fact that some applicants may consider the racial/ethnic community at a university in their decision of where to attend, but incentivizing programs to recruit and admit persons based on their skin color is a biased attempt at social engineering that will not improve the quality of a program (unless of course these happen to be the most promising scholars). One commenter over at IHE said it best:

diversity should be about ideas, not skin