Math and science add up for women
Harvard President Lawrence H. Summers created an uproar at an academic conference earlier this month when he said that innate differences between men and women might be one reason fewer women succeed in science and math careers.
Arguing that fewer girls than boys have top scores on science and math tests in late high school years, he posited this as one factor in the disproportionately low participation of women in the highest ranks of science and math at the most prestigious academic institutions.
Summers’ statement is problematic for many reasons. In fact, it is uninformed statements such as these that help perpetuate an environment that keeps women out. Traditional and incorrect assumptions about “natural” ability and “inherent” interest in math and science have long been recognized by social scientists as a primary roadblock in women’s efforts to succeed in these fields. A wealth of social science data supports socialization factors that influence both men and women from birth, and unexamined cultural assumptions about the appropriate roles for women and men as the most influential determinants in the choice of career paths.
As it relates to academic performance, gender stereotypes and biases promote attitudes that undervalue girls and women and undermine their confidence in their innate abilities. Even today, research indicates that these biases are reflected in subtle ways not only in the classroom, but also in the standardized tests that students take.
In fact, the difference in math SAT scores between young men and women is small and has been decreasing, a fact that doesn’t support an innate differences argument. Researchers who have studied the SAT score gap posit several reasons why it may exist, concluding generally that the difference reflects the use of different test-taking strategies by girls and boys, not different abilities in math or science.
In countries such as England, Australia and the Netherlands, where the tests require solutions to long problems, there are no differences in the average scores of men and women. Actually, when we compare average scores across different countries, we find that in countries such as Hungary and Japan the scores of both genders are higher than the scores of both genders in the U.S.
It is also relevant to note that in all other performance measurements, in all areas of academics, including math and science, from class grades to honors received to advanced placements, from kindergarten through college, girl students outperform boy students. Moreover, we know that although young men score slightly higher on standardized tests than young women overall, this difference reflects a small percentage of students at either end of the scale. The exaggerated emphasis on this difference ignores the vast overlap and the wide within-group variations in ability for both men and women. Even, for the sake of argument, if there is some innate difference between men and women’s abilities, we are focused on exaggerating those differences and using them as rationalizations for unequal opportunities.
The second critical factor to consider is the impact of social expectations. Researchers have found that parents and teachers tend to underestimate girls’ aptitude and performance in math courses, and to overestimate boys’, even at the early age of three. Such expectations can influence children in profound ways. Experiments have conclusively demonstrated that children who are expected to do well outperform children who are expected to do poorly, although they may be equal in innate ability. In addition, research on stereotype threat has shown that people will perform more poorly when they believe they are not supposed to do well; for example, it has been shown that girls will score significantly higher on a math test when they take it with other girls than when they take the test with a mixed group of boys and girls.
We do know that girls and women tend to drop out of the math and science pipeline over time. The reasons are complex and involve many factors. A few examples: we know that girls often drop math and science courses or stop before advanced level courses when they get grades of B, while boys persist even with grades of C. It has also been shown that women must work harder and perform to a higher standard to achieve the same evaluations as men in academic science and math environments. Finally, in these same environments, women report significantly more feelings of isolation, exclusion, and difficulty balancing work and family than do men. These few examples reflect processes far a field from innate ability.
Finally, in attempting to explain the under-representation of women in science, President Summers’ noted that few married women with children are willing to accept the sacrifice of extraordinary commitments of time and energy that top positions at universities require. There is an unstated assumption in this position that only women do (and should do) family work that family work is private and needs no support from society, and therefore, that work and family are incompatible. There are many possible arrangements of family work that can be made to facilitate women’s fuller participation in the workplace. If President Summers, as a leader from one of the most prestigious universities in the world, is committed to creating an equitable work environment for women, there must be a recognition at the most fundamental levels that ideas about work and family must change.
President Summers’ impolitic statement, widely broadcast, has done damage to the cause he claims to represent, and contributes to a climate that undermines women’s advancement in math and science careers.
Signed by the following people at the University of Rhode Island:
Karen Stein, chair of the Women Studies Program
Helen Mederer, chair of the Sociology Department
Janett Trubatch, vice provost, research, and principal investigator of the URI ADVANCE Program. The program is funded by a $3.5 million National Science Foundation grant to promote the careers of women in science and engineering.
Barbara Silver, program director of the URI ADVANCE Program
Joan Peckham, professor computer science
Lisa Harlow, professor of experimental psychology
Karen Wishner, professor of oceanography
Faye Boudreaux-Bartels, professor of electrical engineering
Cathy Roheim, professor of environment and natural resource economics
Harry Knickle, professor of chemical engineering
January 26, 2005