I was recently asked on my advice regarding how to choose a PhD program, so I decide to write a full post about it and share my perspective with my readers. There are of course two different issues: (1) choosing which PhD programs to apply to, (2) choosing which one to join, when accepted. For today, I'll focus on identifying which programs to apply to.
If you are the best student your whole university has had in recent memory, you will apply to the top schools in your discipline, which the "Graduate Schools" issue of US News helpfully ranks for you. But the magazine on newsstands only provides the rankings for the Top 10 programs in each field, which leaves many other students in the dark as to where they should apply.
Judging the quality of a PhD program is difficult for everyone but in particular for undergraduate students, who usually don't know what to look for; besides, a university with less name recognition than the one they're currently enrolled at can have stellar PhD programs in specific departments. If no one tells potential PhD students they should apply to this or that school, they will not figure it out by themselves, since they are most aware of the (often quite different) undergraduate-level rankings.
As an example, the university where I work - Lehigh University - is currently ranked 37th by US News among national research universities for its undergraduate programs, but my department is ranked 17th among all industrial engineering departments in the country at the graduate level. (Departments are not ranked at the undergraduate level.) I doubt our own undergraduates are aware of that.
Although full rankings are not published in US News, faculty members still end up knowing where their department stands compared to their peer group, so an option would be to enlist the help of professors - specifically, those who will write recommendation letters - in the student's department to refine a list of possible schools. The emphasis is on refine: you don't want your professors to do all the work for you. If you haven't done any research on your side, they might just mumble the first name that goes through their mind and leave it at that - not optimal.
So here is my most important piece of advice. I would highly recommend that undergraduate students interested in graduate programs download the free Excel file posted on the National Academies Press website; this file provides data on "more than 5,000 doctoral programs at 212 universities, cover such characteristics as faculty publications, grants, and awards; student GRE scores, financial support, and employment outcomes; and program size, time to degree, and faculty composition. Measures of faculty and student diversity are also included." (Registration is required, but information to provide is minimal.)
Because the data was collected in 2005-2006, it might no longer be accurate and therefore should be used cautiously to generate departmental rankings, but represents a fine starting point to compile a list of potential PhD programs. The file is quite large (35MB), but incorporates macros that lets the user only see the data he cares about, such as Engineering -> Operations Research, Systems Engineering and Industrial Engineering. This dramatically narrows down the list of programs. Then the file can be ranked according to just about any criterion used in the analysis.
The methodology provides a range of rankings (obtained by computing 5th and 95th percentiles) using two different methodologies, called R (for regression) and S (for survey), which differ in the way they weigh metrics such as program size, i.e., average number of PhDs granted over the past five years. The idea of range helps emphasize the difficulty of computing a precise rank to rate a given program, while the spreadsheet provides a trove of information for prospective graduate students, such as median time to degree and percentage of students who complete their degree in six years or less.
Once undergraduate students have compiled a list of possible schools using this information, they should also ask where students from their department have gone when they have enrolled to graduate school. This is good to know because it means the admissions committee at those universities knows the students' department and values the training they receive; unfortunately, some universities send few of their students to graduate school so there might not be a lot of data points.
Preliminary list in hand, students should spend time perusing the departments' webpages to eliminate programs, if any, that are not a good fit given their interests. Finally, they should ask faculty members for their input - especially faculty members they plan to ask recommendation letters from. Here, the conversation will shift from "what are good PhD programs?" to "where do you think I have a chance?" This discussion is particularly valuable because (a) many faculty members won't write a strong letter to admissions committees if they feel the student has little chance to do well in those programs (they don't want to weaken their credibility, as they will still have to recommend other students in later years; most professors politely decline to serve as a reference, e.g., citing time constraints, to give students the opportunity to find a better advocate), and (b) application fees are expensive, so it will save students a lot of time, money and energy if they use their professors' feedback to decide where to apply.
The topics of choosing a PhD program (when accepted to several) and of selecting an advisor will be the focus of future posts this month.