Sunday, 12 July 2009

Understanding why girls underperform at science

Like a sand-castle crumbling away in the rising tide, the once-popular idea that there are innate gender differences in science and maths aptitude is being undermined by a succession of new research findings.

Earlier this year, for example, Stephen Ceci and colleagues sifted through more than 400 relevant journal articles and concluded that far from there being any evidence for a sex-linked difference in science aptitude, the principal factor affecting the relative absence of women in science is their life choices, especially in relation to having children.

Now a new study, by Brian Nosek and colleagues, has looked at the number of people in different countries who hold implicit gender-science stereotypes, and compared this with gender differences in international school science test scores. The researchers' finding is that the two are mutually reinforcing - a culture's implicit belief that females are not associated with science can actually harm girls' and women's science performance, they argue.

To recap, if something about being female really does predispose a person to be weaker at science, then across the world, girls should under-perform, on average, relative to boys. However, whilst this is true in some countries, other countries actually show the opposite pattern, with girls outperforming boys.

Inspired by such observations, Nosek's team wondered whether cultural beliefs about gender and science might negatively affect girls' science performance. They used a version of the implicit association test, hosted on a website, to record implicit beliefs about gender and science among more than half a million people from 34 countries. By allocating categories (e.g. "male-related words" and "science-related words") to either the same or different response keys, the test shows how easily people associate those categories in their mind. Around the world, 70 per cent of participants exhibited an implicit stereotype - associating science with males more than females.

The researchers then looked at international science test scores recorded in 1999 and 2003 for children aged about 12 years. They found a correlation with the implicit stereotype scores, so that in those countries where more people held stereotyped beliefs about gender and science, girls tended to under-perform at science relative to boys.

Taken in the context of previous research showing that awareness of gender stereotypes can harm people's performance (e.g. girls perform worse at maths after being reminded of the stereotype that females are inferior at maths), the researchers said their correlational findings support the idea that a culture's implicit beliefs about science stereotypes can affect girls' science performance in a mutually reinforcing fashion.

"National policy initiatives addressing both factors simultaneously stand the best chance to maximise national scientific achievement," they said. "Education campaigns attempting to bolster women's participation and performance must overcome the pervasive implicit stereotypes that are already embodied in individual minds."
_________________________________

ResearchBlogging.orgNosek, B., Smyth, F., Sriram, N., Lindner, N., Devos, T., Ayala, A., Bar-Anan, Y., Bergh, R., Cai, H., Gonsalkorale, K., Kesebir, S., Maliszewski, N., Neto, F., Olli, E., Park, J., Schnabel, K., Shiomura, K., Tulbure, B., Wiers, R., Somogyi, M., Akrami, N., Ekehammar, B., Vianello, M., Banaji, M., & Greenwald, A. (2009). National differences in gender-science stereotypes predict national sex differences in science and math achievement. Proceedings of the National Academy of Sciences, 106 (26), 10593-10597 DOI: 10.1073/pnas.0809921106

Link to earlier, related Digest item.
Link to another, earlier related Digest item.
Link to one further related Digest item.

Post written by Christian Jarrett (@psych_writer) for the BPS Research Digest.

5 comments:

  1. Anonymous9:50 am

    Correlation isn't causation though, is it? There could have been a million other factors which caused girls to perform worse in these tests.

    Also, I don't believe that this comment can be generalised from:

    "Taken in the context of previous research showing that awareness of gender stereotypes can harm people's performance (e.g. girls perform worse at maths after being reminded of the stereotype that females are inferior at maths)"

    If I was told that white people were worse than black people at maths and I was white; or short people were worse at PE than tall people; and then had to do an exam in that subject, I'd perform worse. Being told you're in a group which isn't good at something, and then having to do that something, you're bound to do worse.

    Come on Nosek et al - you can do better than that. What do they teach them in Virginia these days?

    ReplyDelete
  2. Anonymous2:19 pm

    In the 1980s I heard the Admissions Tutor of Imperial College's Engineering Department say, in an informal covmersation, "If I took the best applicants there would be at least 90% girls. I don't, of course, I take equal numbers of boys and girls"

    ReplyDelete
    Replies
    1. Anonymous5:47 am

      Of course that this Admissions Tutor of Imperial College´s Engineering Department was a sexist, and there are so many men sexist against their own gender. Thas anyone believes that such a thing can be true ? Neither 90% , neither 40%. For some persons, and nowadays for the majority, boys must be put out of Universities or only tolerated at most of 10% of the students. Empowering women doesn´t mean that we have to reject men, men´s must have qualities or not ?

      Delete
  3. @Anonymous(1): Not sure I get your complaint. I think you missed the basic premise of the study. Sure there could be lots of reasons for females to perform poorly, but in fact, they don't all do so, and there is an quantifiable difference between cultures. So the question here is not "why do they perform poorly" but "why do them perform poorly in *some* countries, and better in others?" If poor performance was due to a large set of unrelated factors, why would the numbers differ so much from one country to the next?

    You clearly understand and validate the idea that being exposed to a negative stereotype about yourself would lower your performance. But you don't seem to see how, in countries where more people hold negative stereotypes, this fact might produce overall lower scores for women, while in countries with lower negative stereotypes, more women do better? What's not to understand? The correlation to negative stereotypes provides a strong possible explanation.

    ReplyDelete
  4. Anonymous5:42 am

    In the countries that this study says are without gender stereotypes, girls outperform boys why ? Because in reality girls are better thatn boys in science ? Is that your point ? Girls must be better un all fields than boys, because if not there is a cultural problem ? Or in those countries that girls outperform boys that is because are boys that are suffering those sterotypes that girls are innate better at schools, what means that boys don´t try harder ? Countries like Sweden, etc, in schools the mensage is that girls are better than boys, so boys don´t need to even try be better.

    ReplyDelete

Note: only a member of this blog may post a comment.

Google+