Understanding the “sex gap”in science and math
Study focuses on relationship between implicit stereotypes and academic achievementBY GARY BOASCONTRIBUTING EDITOR
The assumption that boys are natu- rally drawn to – and inherently skilled at – math and science whilegirls are predisposed to the more liberalarts is deeply embedded in our culture.
And it is continually reinforced and perpetuated by observations that boys do infact do better in these areas.
There is evidence, however, that this so-called “sex gap” is shaped by socioculturalfactors. A recent report found, for example, that differences in math performancehave been declining over time. Another revealed correlations between the size of thegap and national indicators of gender egalitarianism. If aptitude for math and science were somehow intrinsic to boys andnot to girls, we probably would not seesuch variability across time and place.
Brian A. Nosek, an associate professorof psychology at the University of Virginiain Charlottesville, and colleagues decidedto explore the influence of socioculturalfactors, looking specifically at the role ofimplicit, or unconscious, stereotypes aboutgender and science. “Stereotypes haveoften been implicated in contributing tothe sex gap, but the evidence for self-re-ported stereotypes predicting such outcomes is mixed. We thought that implicitstereotype measures might be more effective predictors because they do not requireself-awareness of possessing them, andthey can exist in people’s minds even ifthey are consciously rejected,” he explained.
Nosek serves as director of Project Im-
plicit, which seeks to uncover the differ-
ences between conscious and unconscious
attitudes through administration of Im-
plicit Association Tests (IATs). Visitors to
the Project Implicit Web site (http://im-
plicit.harvard.edu/) can complete tests
covering a range of topics: measuring as-
sociation strengths, for example, between
gender (male, female) and academics (sci-
ence, liberal arts). Those who participated
in the gender-science component com-
pleted the test, a short questionnaire that
measured beliefs and attitudes, and math
and science and demographics question-
For a study published in the June 30issue of PNAS, Nosek and a number ofcolleagues from across the globe looked atIAT data collected between May 2000 andJuly 2008. More than half a million IATswere completed during this time. The researchers focused specifically on thenearly 300,000 tests completed by citizensof the 34 countries covered by the 2003Trends in International Mathematics andScience Study (TIMSS). This facilitatedcomparison with the results of that study,in which standardized exams of math andscience achievement were administered tosamples of eighth-graders.
They reported three main findings: (1)The implicit association tests confirmedthe existence of implicit stereotypes associating males with science much more sothan females. (2) The investigators foundthat nation-level implicit stereotypes predict nation-level sex differences inachievement in eighth-grade science andmath. And finally, ( 3) they noted that self-reported (that is, conscious) stereotypes donot predict differences in achievement.
So what does this tell us? First, on some
unconscious level, many people still as-
sume that males have greater aptitude for
math and science than females – even if
they have convinced themselves that they
believe otherwise. The PNAS study shows
that more than 70 percent of the 500,000+
IAT respondents were more apt to associ-
ate males with science and females with
liberal arts than the reverse. That said, the
extent to which people make such assump-
tions varies considerably, both across indi-
viduals and across cultures.
Which brings us to the second take-home lesson: There is a strong correlationbetween how well people think males andfemales will do in math and science andhow well they actually do, as recorded bythe TIMSS. Nosek and colleagues avoidthe obvious and unanswerable chicken-and-egg question here, but they note thatimplicit gender stereotypes and sex gapsare mutually reinforcing. Because they areexposed to the stereotypes pretty muchfrom birth, girls often show less interestthan boys in science and math, and as a result they may not perform as well in theseareas. This serves to reinforce the beliefthat boys are more inherently skilled inmath and science. And so on.
While governments around the worldare working to close the sex gap in scienceand math, it is not yet clear which interventions have been the most or the leasteffective. Researchers and policy makerswill want to know, however. “Those questions are highly important and surelyamong the next issues to investigate in thisresearch,” Nosek said.