Research-Doctorate Programs in the United States: Continuity and Change

Many changes have taken place in size and structure of the research-doctorate enterprise in this country since 1982, when the National Research Council (NRC) issued its first report on the status of research-doctorate programs in the Sciences (including the broad fields of Biological Sciences, Physical Sciences and Mathematics, and Social and Behavioral Sciences), Engineering, and Arts and Humanities in the United States (Jones, Lindzey, and Coggeshall, 1982). From 1980 to 1992, for example, the number of institutions awarding a Ph.D. grew from 325 to 364, an increase of more than ten percent. In 1993, the number of doctoral degree recipients in all fields in the United States reached an all-time high of 39,754. Aware of these changes and of the academic community's interest in the earlier assessment of research-doctorate programs, the Conference Board of Associated Research Councils in 1990 asked the NRC, as a member of the board, to update the 1982 study.

After a planning phase in 1991, the NRC appointed the Committee for the Study of Research-Doctorate Programs in the United States and asked that they undertake a four-year study, taking the 1982 assessment as their starting point. This report represents an effort to build upon and update the information collected for the 1982 study, to collect new information, to analyze key components of the new database, and to make that data base available to interested researchers and scholars for further analysis. It focuses on "research training programs" although we recognized that doctoral education has a range of purposes, and graduates follow a variety of career paths in academia, industry, and government. The study examines programs in the following 41 fields:

Arts and Humanities: Art History, Classics, Comparative Literature, English Language and Literature, French Language and Literature, German Language and Literature, Linguistics, Music, Philosophy, Religion, Spanish and Portuguese Language and Literature.

Biological Sciences: Biochemistry and Molecular Biology; Cell and Developmental Biology; Ecology, Evolution, and Behavior; Molecular and General Genetics; Neurosciences; Pharmacology; Physiology.

Engineering: Aerospace Engineering, Biomedical Engineering, Chemical Engineering, Civil Engineering, Electrical Engineering, Industrial Engineering, Materials Science, Mechanical Engineering.

Physical Sciences and Mathematics: Astrophysics and Astronomy, Chemistry, Computer Sciences, Geosciences, Mathematics, Oceanography, Physics, Statistics and Biostatistics.

Social and Behavioral Sciences: Anthropology, Economics, Geography, History, Political Science, Psychology, Sociology.

·The number of Ph.D.s produced nationally;
·The number of programs training Ph.D.s within a particular field; and
·The average number of Ph.D.s produced per program.

Fields included in the study also have met a criterion of "robustness," that is, they have awarded a minimum of about 500 degrees in about 50 programs for the years 1986 to 1990.

The 41 fields covered in this report consist of:

· All fields in the 1982 report, although the Biological Sciences are represented differently;
· Eight new fields: Comparative Literature, Religion, Aerospace Engineering, Biomedical Engineering, Industrial Engineering, Materials Sciences, Astronomy and Astrophysics, and Oceanography; and
· Some new fields in the broad area of Biological Sciences.

Based on the analysis of degree production patterns and on reports from "Institutional Coordinators" (ICs) who compiled and submitted information about programs at their institutions, the committee identified 3,634 research-doctorate programs at 274 U.S. universities--105 private and 169 public institutions--which met the criteria and are included in the study. This sample represents about 35 percent more programs than the number included in the 1982 study. Taken together, these programs involved about 78,000 faculty members and trained about 90 percent of the total number of Ph.D.s produced in these fields between 1986 and 1992. Of the 228 institutions in the 1982 study, 214 participated in this one and many added more programs for review.

The committee used diverse strategies for collecting the two primary types of data contained in this report.

To generate reputational measures--faculty opinion of program quality--the committee conducted the National Survey of Graduate Faculty in Spring 1993. The survey instrument was a questionnaire designed to elicit ratings on the scholarly quality of the program faculty, the effectiveness of each program in educating research scholars and scientists, and the relative change in program quality over the years. The questionnaire replicated key questions appearing on the 1982 survey form thus permitting the calculation of "change" measures for the 1,916 programs appearing in both studies.

To collect data on the characteristics of the 3,634 programs included in this study, the committee decided to update some statistics from the 1982 study (such as number of faculty and number of graduates) and include, exclude, or improve upon other 1982 data depending on whether the data sets were still available and/or relevant. In many cases, a careful matching of faculty lists with various sources of information occurred. In other cases data were drawn from the Doctorate Records File (DRF) on a program by program basis. Among the new data included in this report are statistics related to the participation of women in research-doctorate education. Appendix G describes the chief data sets used in generating the descriptive statistics found in this report.

Educators and policymakers agree that certain distinctive features of the doctoral training environment facilitate the preparation of research scholars and scientists. These include a blend of well-prepared graduate students, talented faculty, and sufficient institutional resources to permit the independent exploration of promising new research directions.

Survey forms were sent to a sample of faculty raters chosen from lists provided by ICs in all 41 fields included in the study. Each rater received a questionnaire with approximately 50 programs in their field selected at random from the roster of participating programs. For each institution they were asked to rate, raters were given a faculty roster provided by the ICs. The committee set as its goal a total of at least 100 ratings per program. Raters were asked to comment on two dimensions of program quality: (1) "scholarly quality of program faculty," and (2) "effectiveness in educating research scholars/scientists." Ratings for "scholarly quality of program faculty" were pooled and an average rating calculated using a five-point scale ranging from 0 to 5, with 0 signifying "not sufficient for doctoral education" and 5 signifying "distinguished." Of the 3,634 programs included in the study, about 62 percent were rated as "distinguished," "strong," or "good," although this varied by field:

Each rater was also asked to comment on the effectiveness of a program in "educating research scholars/scientists." Mean ratings were calculated using a five-point scale with 0 representing "not effective" and 5 representing "extremely effective." About two-thirds of the programs were considered to be "extremely effective" or "reasonably effective." Fewer than 10 percent were considered to be "not effective" in this regard.

Responding to comments that the presentation of study results by alphabetical listing of programs in the 1982 report created some difficulties for readers, the committee decided that providing a rank ordering of programs within fields is a more convenient way for readers to review and interpret the 1993 information.

The committee chose the mean rating of the "scholarly quality of program faculty" as the dimension along which to array program information. Thus, replies from respondents to the National Survey of Graduate Faculty were pooled, a mean rating was calculated for each program, and a rank ordering was produced within each of the 41 fields in the study.

Rank ordered information requires careful interpretation, of course. A program may be ranked first with respect to "scholarly quality of program faculty," but well down the list with respect to another dimension. As a result the committee created a separate appendix that illustrates the relative standing of programs with respect to a number of variables.

Given the large number of programs within a field, and to facilitate a broad understanding of the data and findings, the committee organized institutions within each field into four groups or "quarters" based on the mean rating of the "scholarly quality of program faculty." Admittedly these are arbitrary groupings. However, these quality groupings represent an efficient way to highlight differences in program characteristics within a field or across fields.

What follows is a brief overview of some of the more interesting observations that can be made from the data collected by the committee and organized in the numerous companion appendix tables.

A strong positive correlation between the number of faculty and its reputational standing has been demonstrated in the past but has not been explored thoroughly. From data collected by the committee, the size-"quality grouping" relationship was found to be the strongest in the Biological Sciences and weakest in the Arts and Humanities. By and large, however, top-rated programs in most fields tended to have a larger number of faculty and more graduate students than lower-rated programs.

Another factor thought to be associated with the relative rating of the "scholarly quality of program faculty" is faculty involvement in research and scholarship. Owing to differences in patterns of scholarship across participating fields, the committee developed three measures in these area: (1) patterns of federal grant support for the period 1986 -1992; (2) publication and citation patterns for the period 1988-1992; and (3) selected "awards and honors" among faculty in the Arts and Humanities.

As would be expected from the important role academia has come to play in conducting research in the national interest, the vast majority of research-doctorate programs included in the study had faculty who received some type of federal grant support between 1986 and 1992. A large fraction of top-rated programs in most fields had faculty who had received federal support during that period, although the relationship between "quality grouping" and grant support was weaker in the Arts and Humanities and a number of disciplines comprising the Social and Behavioral Sciences.

Analysis of publication/citation patterns in the Sciences and Engineering showed a similar pattern. However, the clearest relationship between ratings of the "scholarly quality of program faculty" and these productivity measures occurred with respect to "citation"with faculty in top-rated programs cited much more often than faculty in lower-rated programs who published.

To explore the relationship between "quality" scores in the Arts and Humanities and "scholarship," the committee compiled a list of awards and honors using a variety of sources. The list was matched against a list of faculty members provided by the ICs. From this analysis, the committee observed that a larger share of faculty in top-rated programs in the Arts and Humanities were likely to have received a prestigious award than faculty in lower-rated programs. This relationship was most evident in the fields of Classics, Comparative Literature, Philosophy, and English Language and Literature, reflecting in part the sources of information that were used to compile this listing of awards and honors. There is a need to extend the analysis begun by the committee to include other types of awards and honors to explore institutional differences across all fields of the Arts and Humanities.

Patterns of stability and change were analyzed across each of the 27 fields, where it was found, overall, that somewhat fewer programs rated in the top quarter in the Arts and Humanities in 1982 remained in the top quarter in 1993 (80 percent) than the fraction observed in some of the other broad fields (e.g. 89 percent in the Social and Behavioral Sciences). (See Table ES-3.)

The committee also considered the relative distribution by "quality grouping" for programs appearing for the first time in the 1993 study in one of these 27 fields. They found that these programs received a mix of high, medium, and low ratings, although the chances were much higher that these newly participating programs would appear at the bottom half of the quality distribution in 1993. (See Figure ES-1.)

Figure ES-1Relative distribution of research-doctorate programs appearing in the 1993 study for the first time in fields included in both the 1982 and 1993 study, by broad field and 1993 quality grouping. Based on "Scholarly Quality of Program Faculty." See Appendixes J—N.

Changes in selected characteristics of the 1,916 programs since 1982 were analyzed in three areas: (1) average number of faculty (Fall 1980 versus Fall 1992); (2) average number of graduates (1975-1980 versus 1987-1992); and (3) median time to degree. With the exception of most fields in the Arts and Humanities and in the Social and Behavioral Sciences, the number of faculty and of graduates increased for programs appearing in the 1982 study regardless of quality grouping.

In the Arts and Humanities and in the Social and Behavioral Sciences, faculty rosters were essentially the same size in 1992, but the relative number of program graduates dropped considerably especially from programs rated in the top-quarter in 1982.

It took graduates in the 1980s longer to earn a degree on average than graduates of these programs took 10 years earlier. The longer time to degree was more pronounced for graduates from programs rated in the bottom quarter in 1982 for most fields.