Top Producers of Women Mathematics Doctorates (from Notices)

Recently, Science magazine reported on two surveys among astronomers that pointed to the ``chilly environment'' women have found in that discipline (Science, 21 June 1991, pages 1604-1606). In June, surgery professor Frances Conley resigned from her position at Stanford University, which she had held for 25 years, to protest the appointment to chair of a colleague who she says harrassed her for years with demeaning remarks and attitudes (New York Times, 12 June 1991). Mathematician Jenny Harrison has brought a suit against the University of California at Berkeley, in which she claims that she was denied tenure while men with equal or lesser qualifications were granted tenure (Science, 28 June 1991, pages 1781-1783). A recent New York Times article looked into the reasons why, despite their increasing numbers in science and mathematics, women rarely reach the top of their fields (New York Times, 21 May 1991).

These national news stories point up the still uneasy nature of women's involvement in science and mathematics research. Why aren't there more women in mathematics? Despite the feminist movement, the percentage that women make up of the new crop of Ph.D.s in mathematics each year has hovered around the 20% mark for around ten years. Some claim that there are biological reasons why men tend to outperform women in mathematics, but that debate -- which typically centers on data from the precollege level and may have little bearing on talent for mathematical research -- is far from being settled. What does seem clear, however, is that women find the social environment of mathematics and science to be something less than welcoming. The common wisdom is that women drop out of graduate school in mathematics in greater proportions than do men. Therefore, graduate school is a good place to look for ways to keep more women in mathematics.

An examination of data from the Annual AMS-MAA Surveys over the last ten years provides some thought-provoking information. These data are supplied to the AMS by mathematical sciences departments in the U.S. and Canada. Finding the appropriate data to answer the question ``Which schools produce the most women doctorates in mathematics?'' is not a simple matter. Table 1 shows the top sixteen departments in terms of the total number of women doctorates. Table 2 shows a ranking according to percentage. Care is needed in the interpretation of both tables. For example, in Table 1, MIT and Berkeley top the list primarily because those two departments were overall the largest producers of doctorates in that ten-year period; but in terms of percentage of women doctorates in that period, those two departments fall below the 17% average for all departments of mathematics with doctoral programs. However, looking at straight percentages is also problematic: the small numbers of doctorates at some of the departments mean that the percentages are unstable. (There are other problems as well; for example, Illinois State, the school with the highest percentage of women doctorates over the past ten years, awards only the doctorate of arts degree in mathematics, which does not require a mathematics research thesis.)

To try to balance the percentages with the total number of doctorates produced, schools having more than ten doctorates between 1980-1981 and 1989-1990 were divided into sets according to the size of their doctoral programs (where size is the total number of Ph.D.s produced over the last ten years). For each set of schools, Table 3 lists the three with the highest percentages of women over the last ten years. Because the majority of mathematics doctorates come from Group I departments, Table 5 lists comparable data for all departments in Group I (the definition of groups used in the survey are described in the accompanying box). Table 4 shows the average percentage of women doctorates from Groups I, II, and III. A higher concentration of women is found in Group 11 and III departments than in Group I.

Looking at fall 1990 graduate school enrollments, one also finds a lower concentration of women in Group I departments: 23% of the first-year students are women, compared to 34% and 37%, respectively, for Groups II and III. (Among all students, not just first-year students, the analogous figures are 21%, 29%, and 33%.) A similar pattern can be found in the composition of mathematics faculties. The latest AMS-MAA Survey Report (Notices, May/June 1991) shows that women make up 6.0% of those Group I faculty members holding a Ph.D., 6.5% in Group II, and 8.8% in Group III. In fact, the highest concentration of women in mathematics faculties is found among master's and bachelor's degree-granting departments, where women make up 14% and 17%, respectively, of the faculty holding doctoral degrees. The analogous figure for Group I, II, and III combined departments is only 7%. (It is interesting to note, too, that the highest percentages of women among mathematics faculties occurs among non tenure-eligible appointments.)

Notices queried faculty from a number of departments that, based on this data, were among the higher producers of women doctorates in mathematics, and asked about the reasons for their success. Most faculty expressed surprise that their department was doing a good job, and many had no idea why. Most of the departments had no organized programs for recruitment or retention of women students, and data on women students was not easily at hand. Some faculty seemed surprised that the issue was even raised; one chair remarked that this is not the kind of thing a mathematics department usually pays attention to. Despite the lack of clarity of the factors at play, two general themes emerge.

First, most of the departments reported efforts to create a good atmosphere for all students. For example, Bruce Palka, who has been the graduate advisor at the University of Texas at Austin for a number of years, points to the ``warm, friendly'' atmosphere of the department. A very high percentage of the students who come to visit the department end up choosing it for graduate school, he says. Faculty member Martha K. Smith reports that each semester she holds a reception for women graduate students and faculty. With forty-two women out of a total of 163 graduate students, Smith says there is a ``critical mass'' of women that tends to mitigate feelings of isolation. At the University of Maryland, faculty member Rebecca Herb reports that she has made efforts to bring the women graduate students and faculty together. There is a women's discussion group that meets monthly and a mentorship program that brings new women graduate students together with those who have been in the department a while.

The University of Massachusetts at Amherst has actively recruited students over the past ten years, many of them women. According to faculty member Tim Cook, who has been instrumental in the recruitment drive, the faculty personally contacted students showing interest in and potential for doctoral work. Although no special effort was made to attract female students, Cook believes that the personal touch made a difference. ``We tried to make the department as comfortable and as friendly a place as possible,'' he declares. Prospective female students meet female faculty members, and in the department there is a great deal of personal attention paid to the students. In addition, the department tries to insure that its rules are flexible enough to facilitate the students' chances at success. In a number of instances, for example, the department made allowances for coursetaking schedules and teaching loads for women students who had families or who had long commutes.

Andy Magid, chair of the mathematics department at the University of Oklahoma, also points to a ``warm, welcoming department'' as a key factor in retaining women students. ``In a relatively small, relatively young program like ours, doctoral students are like junior colleagues, and we try to lead each one to completing the degree,'' he says. ``If we've had unusual success with women students, I'd say it's because we've been pretty good about eliminating barriers for all students.'' The department makes no special recruitment efforts to bring women into the Ph.D. program, but such national programs as the Department of Education's Patricia Roberts Harris Fellowships have supported women students in his department.

The second important factor seems to be having women on the faculty. The departments having a high percentage of women doctorates pointed to the visibility of women on the faculty as having a positive effect on the women students. For example, Rutgers University is an institution known for having an unusually large number of productive women researchers on the faculty. According to Rutgers faculty member Amy Cohen, these women ``provide an existence proof and a happier climate'' for the women students. A graduate student association, a prelim-exam preparation seminar, and a T.A.-training program help to bring students together and create a community atmosphere. In addition, Cohen says that, in part because of Douglas College, a women's college on the Rutgers campus, there is generally a high degree of awareness of encouraging the aspirations of women students.

Mary W. Gray of American University attributes the large percentage of women doctorates from her department to having nearly one-quarter women on the faculty. ``Seeing women who are succesful, and our efforts to be supportive, make a difference,'' she says. She also notes that they get many of their women students through ``word of mouth'' -- faculty at other institutions know that American has women on the faculty and is successful at producing women Ph.D.s. And finally, Gray reports, ``We don't lose women.'' She says that most of the women who come with the intention of getting a Ph.D. make it through the program and do not drop out at the master's degree level.

(Sometimes perceptions about the effect of women on the faculty are mixed: In one department with a faculty of about fifty, a male professor said that a major factor in their success at producing female doctorates was having three tenured women on the faculty, while a female professor said the number of women probably was not a major factor, because there were only three.)

It can sometimes be difficult to find a direct connection between having tenured women on the faculty and producing women doctorates. According to information provided by departments for the latest AMS-MAA Survey, for example, Camegie Mellon University has no tenured women faculty, but, among Group I departments, that institution produced the highest percentage of women doctorates over the past ten years (see Table 5). Similarly, the largest overall producer of women doctorates, MIT, has no tenured women faculty. Indeed, the Group I departments having no tenured women faculty produced 30% of the the women doctorates coming from Group I departments. Nonetheless, it is sobering to note that, among those Group I departments that produced less than 10% women doctorates over the last ten years, the number of women on their faculties is just four out of a total of 197, or 2%. Among faculty in all Group I departments, 5% are women. (For all of the previous statements concerning faculty, it must be noted that the figures leave out four Group I departments that did not respond to the 1990 Departmental Profile Survey.)

Some institutions whose departments of mathematics produced low percentages of women doctorates in mathe- matics show up high on the list for applied mathematics and statistics departments. For example, Harvard University, which had one of the lowest percentages of women mathe- matics doctorates among Group I departments, produced the largest percentage, 17%, of women in applied mathematics over the past ten years. Similarly, Rice University, which also had a low percentage of women in mathematics, pro- duced 24% in applied mathematics. Some have speculated that women tend to prefer applied mathematics over pure because of the wider range of employment opportunities available. (Because of inconsistent reporting from applied mathematics departments, the data for applied mathematics are incomplete, and it is difficult to draw many conclusions.)

In the Annual Survey data on doctorates in probability and statistics, women have a high representation. Since 1985-1986, the percentage of women among those receiving doctorates in the fields of probability and statistics varied from 29% to 38%. As with applied mathematics, some of the schools having low percentages of women mathe- matics doctorates showed higher numbers in statistics and probability. Data supplied by doctorate-granting departments of statistics are incomplete because of nonresponding de- partments; however, it appears that the top producers of women doctorates in statistics, biostatistics, and biometrics departments are University of North Carolina at Chapel Hill, Comell University, Harvard University, University of California at Berkeley, University of Michigan, University of Washington, Iowa State University, Ohio State University, University of California at Los Angeles, and University of Wisconsin. In addition, women make up 13% of faculty members holding the Ph.D. in statistics, biostatistics, and biometrics doctorate-granting departments, a much higher figure than the 7% for Groups I-III combined.

It is difficult to discern from statistics alone why certain departments have been more successful than others at producing women mathematics doctorates. Many of the factors at play are complex, involving the ``climate'' in the department and social interactions that can elude quantitative analysis. A study of a number of exemplary departments--including interviews with faculty, students, and administrators--could bring successful ideas and new thinking to other departments that are trying to improve the climate for their women students. Such improvements would, in the end, benefit all students.

The data in the tables below [ web editor's note: 1 and 2] were reported by departments responding to Annual AMS-MAA Surveys of New Doctorates from 1981 to 1990. In a few cases of nonresponses, attempts were made to contact departments for missing data. Names and thesis titles of new doctorates are published annually in the November issue of Notices of the AMS. For Annual AMS- MAA Survey reports, departments are divided into groups according to the highest degree offered in the mathematical sciences. Groups referred to in the following tables:

Groups I and II include the leading departments of mathematics in the U.S. according to the 1982 assessment of Research-Doctorate Programs conducted by the Conference Board of Associated Research Councils in which departments were rated according to the quality of their graduate faculty.¹

Group III contains the remaining U.S. departments of mathematics reporting a doctoral program.

**Table 1:** Leading U.S. doctorate-granting departments of mathematics by number of women doctorates, from academic year 1980-1981 to 1989-1990
	Total Women Doctorates	Total Doctorates	% Women
Massachusetts Institute of Technology	32	209	15.3
California, University of (Berkeley)	30	295	10.2
Maryland, University of	26	120	21.7
NYU-Courant Institute	19	155	12.3
Rutgers University, New Brunswick	17	77	22.1
Wisconsin, University of	16	152	10.5
California, University of (Los Angeles)	15	108	13.9
Illinois, University of (Urbana-Champaign)	15	105	14.3
Massachusetts, University of (Amherst)	15	44	34.1
Michigan, University of	14	115	12.2
Notre Dame, University of	14	53	26.4
Texas, University of (Austin)	14	60	23.3
California, University of (San Diego)	13	79	16.5
Carnegie Mellon University	13	46	28.3
Pennsylvania State University	13	64	20.3
Pittsburgh, University of	13	57	22.8

**Table 2:** Leading U.S. doctorate-granting departments of mathematics* by percentage of women doctorates, from academic year 1980-1981 to 1989-1990
	% Women	Total Doctorates	Total Women
Illinois State University	54.5	11	6
Oklahoma, University of	41.7	12	5
Memphis State University	38.9	18	7
Adelphi University	38.5	13	5
Missouri, University of (Rolla)	38.5	13	5
American University	36.0	25	9
Florida State University	35.3	17	6
Aubum University	34.8	23	8
Massachusetts, University of (Amherst)	34.1	44	15
Lehigh University	33.3	21	7
South Carolina, University of	29.2	24	7
Dartmouth College	28.6	21	6
Carnegie Mellon University	28.3	46	13
Kansas State University	27.3	22	6
Texas, University of (Arlington)	27.3	33	9
* Departments granting at least ten doctorates during the ten-year period.

For this table [web editor's note: 3], departments of mathematics were divided into categories of comparably-sized doctoral programs, where size is defined to be the number of doctorates awarded by the department between 1980-1981 and 1989-1990. The size categories are given in the leftmost column. For each size category, the table lists the three departments having the highest percentage of women doctorates in that ten-year period. The rightmost column gives additional information about the departments in each size category.

**Table 3:** Percentage of women among doctoral programs of comparable size
Size of Department	Top three departments by % women doctorates	% women doctorates	Average % women doctorates for departments in this group
Depts granting 100 doctorates and above	Maryland, University of	21.7	13.3% for 8 departments (167 women out of a total 1259 doctorates)
	Massachusetts Institute of Technology	15.3
	Illinois, University of (Urbana-Champaign)	15.0
Depts granting 80-99 doctorates	Ohio State University	12.6	10.5% for 5 departments (46 women out of a total 436 doctorates)
	Chicago, University of	12.5
	Princeton University	9.8
Depts granting 60-79 doctorates	Texas, University of (Austin)	23.3	17.4% for 8 departments (96 women out of a total 551 doctorates)
	Rutgers University, New Brunswick	21.1
	Pennsylvania State University	20.3
Depts granting 40-59 doctorates	Massachusetts, University of (Amherst)	34.1	17.2% for 21 departments (175 women out of a total 1019 doctorates)
	Carnegie Mellon University	28.3
	Notre Dame, University of	26.4
Depts granting 20-39 doctorates	American University	36.0	20.3% for 50 departments (278 women out of a total 1367 doctorates)
	Auburn University	34.8
	Lehigh University	33.4
Depts granting 10-19 doctorates	Illinois State University	54.5	19.0% for 41 departments (109 women out of a total 573 doctorates)
	Oklahoma, University of	41.7
	Drexel University	40.0

**Table 5:** Group I departments of mathematics: women doctorates from 1980-1981 to 1989-1990
Department	%Women Doctorates	Number of Women Doctorates	Total Number of Doctorates
Brandeis University	20.0	8	40
Brown University	14.6	7	48
California Institute of Technology	9.4	3	32
California, University of (Berkeley)	10.2	30	295
California, University of (Los Angeles)	13.9	15	108
California, University of (San Diego)	16.5	13	79
Carnegie Mellon University	28.3	13	46
Chicago, University of	12.5	11	88
Columbia University	12.1	7	58
Comell University	21.0	12	57
CUNY Graduate School	24.1	7	29
Harvard University	8.0	7	87
Illinois, University of (Chicago)	18.5	10	54
Illinois, University of (Urbana-Champaign)	14.3	15	105
Indiana University at Bloomington	16.9	10	59
Johns Hopkins University	22.2	6	27
Maryland, University of	21.7	26	120
Massachusetts Institute of Technology	15.3	32	209
Michigan, University of	12.2	14	115
Minnesota, University of	9.8	8	82
New York University-Courant Institute	12.3	19	155
North Carolina, University of (Chapel Hill)	25.0	6	24
Northwestern University	15.5	7	45
Ohio State University	12.6	11	87
Pennsylvania State University	20.3	13	64
Pennsylvania, University of	22.6	7	31
Princeton University	9.8	9	92
Purdue University	11.8	9	76
Rice University	14.3	2	14
Rutgers University (New Brunswick)	22.1	17	77
Stanford University	13.6	8	59
SUNY at Stony Brook	18.7	12	64
Texas, University of (Austin)	23.3	14	60
Utah, University of	13.3	6	45
Virginia, University of	20.7	6	29
Washington University	14.6	6	41
Washington, University of	8.8	5	57
Wisconsin, University of (Madison)	10.5	16	152
Yale University	12.9	8	62

Top Producers of Women Mathematics Doctorates

Allyn Jackson