Next: 3. Careers Up: The Past, Present, and Previous: 1. Introduction

2. Education

Women perform significantly better than men at all levels of education from high school to graduate school, but they do not perform better in general on standardized tests such as the quantitative SAT and GRE tests. Chipman and Thomas (1985) show that gender differences in mathematics performance do not emerge until high school where more men enroll in trigonometry and calculus (26% and 8%, respectively) than do women (20% and 6%, respectively) with the same percent enrollment in algebra (66% men, 69% women) and geometry (54%). The lower standardized test scores appear to be explained by high school course selection and by demographic and socioeconomic factors and not by genetic differences (Hornig (1987)). However, high school grades in mathematics for women are not less than those for men, though men have tended to have more years of preparation in mathematics at the high school level.

Of entering freshman, essentially the same percentage of men as of women intend to major in either mathematics or statistics (see Table 1). Thus, in 1971, 2.6% of the men and 2.9% of the women declared this major, while in 1980 the corresponding percentages were 0.7% and 0.6% respectively, with the figures remaining steady throughout the 1980s. Since in 1972 and in 1986, 44% and 46%, respectively, of the bachelor degrees were awarded to women, these represent reasonable approximate percentages for both men and women. Throughout this entire period, the same proportion (0.1%) of both men and women sought statistics as a career goal.

**Table 1:** Education Trends since 1971.
	Freshman major in math./stat.		Bachelors to women in		Doctorates to women in Math. Sciences
	Men	Women	Math.	Stat.
1971	2.6%	2.9%	138.2%	25.3%	7.8%
1973	1.8	1.6	40.3	34.8	9.7
1975	1.1	1.1	42.1	31.9	9.6
1978	1.1	0.8	41.4	39.2	13.7
1980	0.7	0.6	42.3	42.1	12.0
1982	0.6	0.7	43.2	41.1	13.8
1983	0.8	0.8	45.7	48.7	16.1
1988	0.7	0.6	46.5*		16.2
* 1986 figure; Source: Debold (1987), Vetter (1981), NSF (1990)

Table 1 also shows the proportion of bachelor's degrees with a major in mathematics (statistics) awarded to women, as well as the proportion of doctorates in the mathematical sciences that were earned by women. In particular, we note that the proportion of bachelor degrees in mathematics to women has increased from 38% in 1971 to 46 1983, and the proportion of doctorates in the mathematical sciences awarded to women has increased from 8% in 1971 to 16% in 1983. In each case there has been a levelling off over the period 1983-1988. These changes are partly due to declines in the numbers of men in these categories. For example, there were 9,259 women and 14,454 men in 1972, and 5,006 women and 6,593 men in 1982 receiving the bachelors degree in mathematics; while there were 89 women and 1,039 men in 1972, and 94 women and 587 men in 1982 earning the doctoral degree in the mathematical sciences (see Weis (1985)). Thus, at least at the undergraduate level, parity seems to have been achieved, suggesting that whatever deterrants that may have existed previously may have evaporated. Presumably we can hope for a continued growth in numbers at the graduate school level, though the stability of the 1983-1988 figures may temper such expectations. Throughout this period, an approximately constant one third of the masters degrees in mathematics were awarded to women.

The drop in proportion of graduate degrees to women signals a higher attrition rate in mathematics for women between the bachelors degree and a doctoral degree (where, for example, the National Science Foundation (1986) data of parity indices show a loss from 35% of the bachelors degrees in 1977 to 15% of the doctorates in 1984). The Office of Technology Assessment (1985) report states that in 1982, 37% of the bachelors degrees but only 27% of graduate enrollments in mathematics were women. Thus, part of the decrease is due perhaps to admission procedures where standardized test scores (which underestimate women's performance) and recommendations by faculty (presumably male) weigh more heavily than does academic performance as an undergraduate.

However, a large source of this attrition is attributed to a failure to complete the stated degree objective. A major factor influencing graduate students' persistence with a degree is the availability of financial aid. Haven and Horch (1972) and the 1981 Survey of Doctorate Recipients showed that women receive less aid than men and Harris (1972) found that women tended to come from wealthier families, presumably because, in the absence of other aid, only these women could afford this education. For mathematics graduate students, a significantly higher proportion of teaching assistantships went to women and a significantly higher proportion of research assistantships supported men. However, for those graduating with doctorates in 1988 (of whom 16% were women), 17% of teaching and of research assistantships were awarded to women suggesting that some progress has been achieved. Solomon (1976) observed that the actual level of support was generally lower for women than for men. On the other hand, women tend to gain a higher proportion of fellowships (where their better scholastic performance is no doubt a major contributing factor).

While financial aid is clearly an important indicator of persistence to degree completion, it is not the only factor. In a study at the University of Illinois, Berg and Ferber (1983) found that, despite equality of financial inducements, the attrition rate for women was still higher than that for men. They suggested that women students have a lower involvement with faculty and were less likely to be treated as colleagues. This need for greater mentoring by faculty will be hard to meet until there are more women faculty. Berg and Ferber (1983) also found that women tended to have less confidence in themselves despite their superior performances.

Next: 3. Careers Up: The Past, Present, and Previous: 1. Introduction