This page has links to several sources of data, gathered in various ways and covering different time frames. It is critical to read the footnotes and headings carefully to avoid misinterpretation.
1. NRC Completion Data and Time to Degree
The Completion data and Time to Degree data in this table are from the National Research Council Survey. See http://nrc.arizona.edu for more details. Note that these data are for PhD students who were fulltime throughout their graduate careers. The NRC will compare programs within the same discipline — e.g., psychology programs with other psychology programs; history programs with other history programs. We expect results later this fall, but at this time we do not know program rankings or have data from other universities. When the data become available, we will post comparison data by discipline and for peers.
It is NOT valid to compare very different programs to each other using either completion rates or time to degree as quality measures. Disciplines differ in terms of their requirements and culture, so that we will see large differences by discipline. Also, some programs accept students directly into the PhD while others admit only those holding a master's. Higher completions rates may not indicate higher academic quality.
NRC (as best we know) will use SIX year rates to calculate completion; EIGHT years for humanities. While students continue to graduate after year six, they do not show in these results. NRC uses 6/8 year rate NOT because it's the norm, but because it's the goal — NRC hopes the data will encourage shorter time to degree.
The completion data are cohort data: cohorts are followed as they move through the program. Completion data are based on cohorts entering PhD programs in Fall 1996-2000 (Fall 1996-1998 for Humanities.) Time to Degree data are retrospective. Time to degree looks at fulltime students who graduated in academic years 2003-2005 and indicates how long they took to receive their degrees from first admission into the PhD program.
The new NRC has very different methodology including no reputational component, so the new NRC will not be directly comparable to the old survey.
In the same table, we have included data on selectivity and yield for Fall 2007 and 2008.
2. National Rankings
• Prior NRC rankings from the 1995 study
• Rankings analysis from the 2011 edition of U.S. News & World Report's America's Best Graduate Schools
• Rankings from the 2013 edition of U.S. News & World Report's America's Best Graduate Schools
3. CGS Completion Data
The Council of Graduate Schools has also been collecting completion data, using a different methodology. Their data includes part-time students and those who 'stopped out'. Therefore, their reported completion rates tend to be lower and their time to degree longer than what we expect from the NRC data. They report that about 57% of PhD students complete within 10 years; about 36% complete within 6 years; about 38% of Humanities students complete within 8 years. We include the Cumulative Ten-Year data from the CGS study.
4. UA Non-NRC Programs' Completion Rates and Time to Degree
We have also compiled similar data for those PhD programs that were NOT in the NRC study; there we used ALL students, including part timers, to avoid very small numbers.
5. Ethnicity by Broad Field
Ethnicity by Broad Field shows UA data from the NRC survey. These data are helpful benchmarks.
6. Selected Data from CGS Graduate Enrollment and Degrees: 1997 to 2007
7. NSF Survey of Earned Doctorates (SED)
The SED collects data on the number and characteristics of individuals receiving research doctoral degrees from all accredited U.S. institutions.
The results of this annual survey are used to assess national characteristics and trends in doctorate education and degrees.
8. UA Masters Programs Completion Data and Time to Degree
In this data set, completions use a six-year rate, based on 5 entering cohorts, Academic Years 1998-2002, of all Masters students. Median Time to Degree is based on graduation data from the last 10 years. Programs with fewer than 10 students in the 5 entering cohorts are excluded.
Other data will be added soon, including data from Survey of Earned Doctorates.