For example, both sexes find computation items easier than word problems and the differences across kinds of items are larger than gender differences (Marshall). Girls do better than boys on some kinds of word problems including data sufficiency items and logical puzzles (Becker). On multiple choice tests girls may have poorer test-taking strategies rather than poorer problem solving skills. In particular, girls tend more than boys to omit items, especially difficult ones, whereas boys are more likely to guess (Becker).

If even one of the possibilities can be eliminated as incorrect on a multiple choice format, guessing is a better strategy than leaving items blank even on a test that includes a correction for guessing. That rapid intelligent guessing is a better strategy than applying and solving the correct formulas is demonstrated by the success of special courses in the U.S. that train students to increase their overall SAT (college admission test) performance by as much as 150 points (Linn & Hyde).

To illustrate the culturally constructed nature of these assumptions, I propose a thought experiment: How could a reversed pattern of performance be interpreted to the advantage of males? Imagine that males do better as young students on computations but there is no difference among older students, and that females do better on tests of problem solving after they become teenagers. Imagine that boys get better classroom grades although they do more poorly on standardized tests, and that the differences favouring girls are greatest in highly precocious samples.

It is possible that educators and researchers would worry why males lose their early advantage in computation and explore what happens in classrooms that might relate to this loss. Accuracy in computation would be seen as a "concern with, attention to, and appreciation of numerical detail or competence in handling numerical systems and their operators" (Damarin 1993, p.8), whereas a relative advantage on problem solving would be seen as fooling around or playing with math instead of really doing it. Much would be made of boys' better grades as more realistic measures of mathematics achievement and girls would be labeled underachievers with their pattern of higher scores on standardized tests and lower classroom performance. Precocious samples would be seen as highly unrepresentative and much more would be made of the importance of studying mathematics achievement in representative samples in order to fill the growing need for mathematically competent people in a wide range of technical and scientific professions.

Gender and Mathematics: Where to
go From Here?

Given the discrepancy between the nonexistent or very small and very limited gender differences in mathematics achievement and the belief that mathematics is a male domain, what will bring about effective social change? Clearly a demonstration of empirical similarity is not enough. One must also work to change the symbolic masculinization of mathematics and reflect this change in the classroom. Effective social change requires both equality and equity of mathematics education. Walter Secada describes equality as a quantitative concept and equity as a qualitative one. Equality is determined by the absence of a difference among demographic groups in opportunities to learn, access to educational resources, or educational outcomes.

Clearly, an educational context that supports inequality is inequitable. However, equity includes and goes beyond measurable inequalities. Equality within a system that is symbolically masculine is both difficult to achieve and insufficient to ensure equity. Equity involves fairness and requires a focus not only on the distribution of existing resources, but also on the inclusiveness of what is being taught. Is what is being distributed worth having? (Secada). Are measures of what is being learned culture-fair?

A number of innovative courses have been designed to make the content of mathematics courses gender inclusive. Sometimes this involves the adaptation of traditional feminine activities to the classroom such as the use of embroidery to teach border symmetry (Verhage). In other cases this is done by including examples familiar and interesting to a wide range of students. Thus, in a calculus course, problems of the rate of temperature change in a dead body, or measuring the area of deforestation from an aerial map (Barnes) can be used instead of the typical abstract problems that are of very little interest to the majority of students, both male and female. Even issues of equality could be made directly relevant to learning by bringing in sets of data that examine gender, race or class differences on various achievement measures and using them to illustrate statistical and mathematical concepts, including a critique of how they may be distorted.