Although the past several generations have seen declining gender inequalities in educational attainment, gender-based differences in the fields of study we choose seem to persist (see here). For example, the percentage of women obtaining degrees in the science, technology, engineering, and mathematics (STEM) fields has remained exceedingly static in the last few decades (see here).

In trying to explain this persistent trend, some conclude that (1) women are not as interested in these fields, and/or that (2) women just aren’t as good as men in these domains. But how would one tell whether these explanations are right or wrong?

One problem is that people share specific, culturally based ideas about what men and women are and should be. Numerous studies demonstrate that the dearth of women in STEM fields can be directly linked to negative associations regarding girls and the sciences, and especially girls and math ability (see here and here).

While it is true that we tend to draw generalizations about groups of people based on what we observe, it‘s also the case that – as the cartoon suggests – we pay more attention to evidence that confirms our preexisting beliefs. In addition, stereotypes can have independent effects on people’s attitudes and behaviors, often becoming self-fulfilling prophecies. So, whether or not women actually perform worse at math-related tasks is somewhat confounding since awareness of this very idea can actually cause women to do worse on such tasks – see a recent study here.

There is also evidence to suggest that stereotypes start to matter very early. In a recent study published in the March/April issue of Child Development, Cvencek, Meltzoff, and Greenwald found that, as early as second grade, children have internalized the cultural stereotype that math is for boys. The researchers also found that elementary school boys identified with math more strongly than did girls on both implicit and self-report measures. As the researchers point out, these findings are especially significant because they show that cultural stereotypes about math are actually absorbed prior to the ages at which gender differences in math achievement begin to occur.

This and other evidence suggest that the classroom is a key environment for effective and timely interventions. So, what could be done at/by schools?

First of all, stereotypes are situational. Increasing girls’ interest in math and science may depend less on boosting their general confidence than on broadening our understandings and depictions of these fields. As others have pointed out, we need to avoid the idea that there’s something wrong with girls that needs to be “fixed” and focus on fixing organizations instead – their rules, assumptions, messages etc.

For example, in Unlocking the Clubhouse, researchers Margolis and Fisher conducted a five-year study of computer science majors at Carnegie Mellon University. Their in-depth interviews established that one recurring view of computer science, held by both men and women, was that people in this field were geeks and hackers. When researchers asked students if they thought they fit that image, about 3/4 of the males said yes, as opposed to only 1/3 of the females. In response, researchers set out to change the image of computer science, focusing on developing a more accurate portrayal of the program and the profession, in the hope that this would attract more women.

They were right. When the study began in 1995, female enrollment was about 7 percent; by 2000, it was about 42 percent. The researchers concluded that they did not need to change stereotypes about women, so much as stereotypes about computer science.

Similarly, researchers Murphy, Steele and Gross (2007) conducted an experiment where women (all math majors at Stanford University) were shown videos of an upcoming science conference at Stanford. Half of the women were shown a gender-balanced video while the other half watched a male-dominated video. Afterwards, researchers asked women how interested they would be in attending; they also measured their physiological responses as they watched the videos. In addition to being less likely to attend, women in the unbalanced video condition showed heightened signs of stress and vigilance.

These two examples illustrate how important it is to portray images that suggest that people fit in.

"Children have their antennae up and are assimilating the stereotypes exhibited by parents, educators, peers, games and the media," said Meltzoff – cited earlier. "Perhaps if we can depict math as being equally for boys and girls, we can help broaden the interests and aspirations of all our children."

Why is this important? One reason is of course because STEM disciplines tend to lead to rewarding jobs where women are still a minority. Perhaps more important, it’s central to the type of society we want to create for our children. Don’t we want them to feel free to challenge occupational norms in pursuit of their true wishes and aspirations? Or should we just be satisfied with more male scientists and more female kindergarten teachers? I hope not. A truly emancipated society is one where invisible barriers have been effectively eliminated. This is not where we are today. When, even from the earliest ages, we feel compelled just to act out our prescribed parts, neither we – nor the society as a whole – will ever know exactly what we’ve been missing.

- Esther Quintero