Copyright © 1999 by Begell House, Inc. Reprinted by permission
Why Aren't There More Women Engineers?
Women Focused Explanations
Women's Mathematic and Visual-Spatial Reasoning Abilities
One factor cited as contributing to the very slow progress of women's employment in engineering fields is a belief that women do not perform as well as men on mathematic and visual-spatial tasks. These skills are considered to be necessary skills to succeed in engineering careers. Typically, research has investigated gender differences in either math skills or visual-spatial. The results of these studies have been mixed.
Mathematical abilities. There are literally hundreds of studies comparing women's and men's math performance. Several studies in the 1960s and 1970s reported that men outperform women especially on standardized math achievement tests (see Maccoby & Jacklin, 1974 for a summary). However, findings that men perform better than women on mathematical tasks are not always replicated. Hyde, Fennama, and Lamon (1990) conducted a meta-analysis examining the gender differences in math performance scores from 254 independent studies involving over three million subjects and found the average performance difference to be negligible. (Meta-analysis uses statistical methods to combine results of several different studies of the same behavior so that the overall pattern can be evaluated.) Overall, girls appear to have a slight advantage over boys in math performance scores during elementary grades. This small difference in math performance reverses by high school grades and appears to increase slightly through college and into adulthood with men tending to outperform women. Hyde, et al., (1990) found that the average effect size in math performance differences reported for men and women in high school and beyond was 0.15 standard deviations. This very small difference can be statistically significant with very large sample sizes (e.g. SAT scores from 10,000 high school students) but it often translates into a one or two percentage point difference on actual exam scores.
Meta-analytic studies also have examined gender differences in specific components of math abilities including computational skills, comprehension of math concepts, and problem-solving (Hyde et al., 1990; Kimball, 1989; Linn and Petersen, 1986; Lips, 1993). Many studies have shown women consistently performing as well as or better than men in computational skills and comprehension of math concepts. Men consistently performed better than women on measures of mathematical problem-solving skills. Finally, some research has shown that among high school students scoring in the top 3% of standardized math achievement measures (e.g. SAT), men outnumber women by approximately 13 to 1 (Benbow & Stanly, 1983). Among students in the remaining 97% of math ability, measures of gender differences are small and inconsistent from study to study.
Taken as a whole, research does not support a belief that there are substantive differences between men's and women's mathematic skills or that such differences adequately explain why there are so few women in engineering careers. Generally, men do not perform computational tasks or understand math concepts better than women. It does appear that men gain a small advantage over women when performing mathematical problem-solving tasks but this advantage does not emerge before high school years. Furthermore, there are plausible explanations for men gaining a slight advantage over women in mathematical problem-solving tasks. Specifically, girls take fewer math classes in high school and college than do boys, and this is particularly evident in advanced level math classes such as calculus (Kimball, 1989). These data suggest to us, that there are many more women who have the requisite math skills for earning an engineering degree than actually do so.
Visual-spatial abilities. Research on gender differences in visual-spatial task performance often shows men outperforming women. When performance differences on three component skills (mental rotation, spatial perception, and visualization tasks) of visual-spatial abilities are analyzed, men tend to outperform women on mental rotation and spatial perception tasks. Women appear to perform visualization tasks as well as men (Lips, 1993). The meaning of these performance differences is equivocal. Many visual-spatial performance tasks include a measure of speed and men's superior performance reflects their ability to perform the task faster than women, not more accurately. If we could translate this difference in speed to engineering jobs, one might argue that men could do more engineering tasks faster than women. However, whether men make efficient use of this time savings on the job remains an unanswered empirical question. Furthermore, as computers are used increasingly for design tasks and other engineering activities that require mental rotation and spatial perception skills, a difference in speed may be irrelevant.
Finally, if there are true gender differences in visual-spatial abilities then changes in teaching methods should not effect this difference. Several educational programs have been designed specifically to teach visual-spatial skills to girls so their performance would be equivalent to boys (Linn & Petersen, 1986; Unger & Crawford, 1996; Vasta, Knott & Gaze, 1996). Programs have included all-girl high school science classes, training in imagery and mental rotation skills, computer learning opportunities, or self-discovery training in visual-spatial reasoning for co-ed classes. Outcome evaluations of such programs often show improvement in girls' performance on visual-spatial tasks so that gender differences are reduced or eliminated on some tasks. Apparently, women's visual-spatial skill levels are malleable with changes in educational methods. Perhaps some comparable changes in engineering education would similarly increase women's presence and performance in these career fields.
Women's Interests, Attitudes and Self-Efficacy Beliefs
