Dr. Correll first became interested in gender differences in self-assessment when she taught chemistry to high school students. She realized that no matter how well the girls in her classes did, she had trouble convincing them that they had any scientific ability. At the same time, she found that no matter how poorly the boys in her classes did, they continued to believe that they were very good at chemistry.
Once she went to graduate school, she delved into this issue, analyzing a dataset from more than 16,000 high school students, and found that, in fact, among students with similar past math achievement girls assessed their mathematical abilities lower than boys did. In a lab experiment on gender differences in self-assessment, Dr. Correll found that women assess themselves as less competent in “male” fields, even when the “male” field is fictitious.
The figure above shows an example from this experiment. Subjects are asked if this rectangle is more black or more white. As it turns out, it is not actually important how much black or white there is (there are equal amounts). Dr. Correll describes this fictitious ability to detect correct proportions of black and white as “contrast-sensitivity ability.” When participants in Group A were told that men were more likely to have high levels of “contrast-sensitivity ability,” women assessed their contrast-sensitivity ability lower than men did. When this ability was described as equally strong in men and women to participants in Group B, gender differences in self-assessment were not found. Dr. Correll’s results support the idea that cultural beliefs about gender and not actual gender differences can influence self-assessments and lead to girls’ and women’s lower assessment of their abilities in fields labeled “male,” like STEM.
Not only do women rate their abilities lower in so-called “male” fields, they also hold themselves to a higher standard compared with men in these so-called “male” fields. In the same fictitious skill of “contrast sensitivity,” students were asked, “How high would you have to score to be convinced that you have high ability in this task?” In the group where students were told that “men are better at this task,” women indicated that they would have to earn at least 89% to think they had high ability in that area. On the other hand, men thought that a score of 79% would indicate high ability. That is a difference of 10 points! However, in the group where students were told, “There is no gender difference in performance on this task,” men and women had a much more similar idea of what score would indicate high ability.
If you think about this finding as it relates to math and science, fields in which men are considered to excel, it suggests that girls believe that they have to be better in math and science than boys believe they have to be in order to think of themselves as good in these fields.
There are many elements to choosing a career, but researchers agree that one element is believing that you can be successful at it. Girls’ lower self-assessment of their math ability, even in the face of good grades and test scores, along with their higher standard for performance in “masculine” fields, helps to explain why fewer girls than boys aspire to science and engineering careers.
So what can we do to help girls to more accurately assess their abilities in math and science? Make performance standards and expectations clear. When students have clear information from teachers about what grade or score signifies good performance, they are less likely to rely on stereotypes to assess their abilities. Also, girls are less likely than boys are to interpret their academic successes in math and science as an indication that they have the skills necessary to become a successful engineer or computer scientist. Encourage girls to see their success in high school math and science for what it is: not just a requirement for going to college but also an indication that they have the skills to succeed in a whole range of science and engineering professions.
Note: much of this text is from the AAUW ppt describing highlights of the Why So Few? report.