Aarna goes shopping for toys
Aarna is a bright little kid. Despite her young age, she shows special interest in mathematics and physics. She is fascinated by numbers and squeals with pleasure on learning new things. Her parents revel in her enthusiasm. They take a trip with their lovely kid to a supermarket to buy some toys, and this is how the toys section looks like:
Boy’s section is painted blue, full of Legos, motor and motion toys, electronic equipment replicas etc. While the Girl’s section is painted pink with dolls and housemaking equipment replicas. Aarna picks a talking Barbie doll and brings it home. Do you want to know what the Barbie tells her?
- “I’ll always be there to help you”
- “Party dresses are fun!”
- “Do you have a crush on anyone?”
- “Maths class is tough!”
This is not just a story! As late as 1992, a Teen-Talk Barbie was produced, which uttered words like those [Teen-Talk Barbie]. It would be great if only toys uttered such words for girls though! But this is a much more widespread cancer for our society, which renders all of us reeling due to unhealthy social dynamics it propagates. We face dire consequences by losing a great talent pool to dangerous myths.
Let us analyze main claims peddled by proponents of the view that the lack of women in high power STEM (Science Technology Engineering and Mathematics) jobs is because of an issue of “innate talent” [Lawrence Summers][Google Memo] [A Very good counter to Google Memo]. We are mainly going to follow analysis from Harvard cognitive psychologist Elizabeth S. Spelke.
Aarna wanted to compete for a science quiz and was quite confident of acing it. Some were skeptical and said “Men are endowed with better mental rotation and visualization skills while women are endowed with better verbal skills. Hence men possess better maths and science abilities” This is just how girls and boys are hardwired!
Glittering generalities, clear results, and binary outcomes, though crowd-pleasing and a beacon for popular science and media, are never the norm for sound science! There is always nuance. Attention to details is needed and one has to be very careful while drawing conclusions. Rather than a simple “women are verbal and men are visuospatial” conclusion, research has found that women tend to do well in verbal fluency, algebra and memory for spatial location of objects while men tend to do well in verbal analogies, mathematical word problems and geometric configuration of objects. Meta-analyses (a meta analysis itself is a super-analysis of multiple research works with a huge amount of data) have revealed that these effect sizes are quite small. [Halpern 2000] [Hyde 2005]. Also since these studies are done at quite later stages of life, one can never know whether these causes are biological or environmental or which factor plays a larger role.
However, even if we take these differences as given, no matter the cause, even then the leap from specific cognitive skills to overall maths and science abilities is a big one! One thing is clear that men and women might use different strategies to solve a problem. Here are a few examples:
- In a navigation task, men might prefer to employ geometrical information while women might prefer to employ landmarks. [Choi 1996] However men and women perform equally when only one source of information is present [Wang 2002]
- In an visual comparison task, men might prefer to rotate the object to compare while women might prefer to look at the features and compare them. One can outperform another based on which component is critical for a given problem [Hyde 2005]
- In mathematical reasoning, men prefer to use spatial visualization and perform better on problems based on that. Women prefer to use computations and perform better on problems based on that. However the gap seems to reduce when the different nature of problems is corrected for. [Geary 1996]
Based on these observations, most careful researchers in the field conclude that
- men and women have similar cognitive abilities but different cognitive profiles. [Halpern 2005] [Pinker 2002] (Though as we mentioned, the reasons can’t be just attributed to biology, but socio-cultural factors also play a role in these differences.)
- Research deals with averages, not individuals. An individual male or female can have completely different cognitive profiles from whatever is mentioned above.
- Besides these differences are not hardwired and training has been shown to mitigate spatial skill differences for both men and women. [Sorby 2010]
Aarna was perplexed why most top positions in STEM belong to men. She was told “Men show greater variability in their maths skills, hence at extreme skill levels, there are more males than females”. This is just a reflection of reality!
This line of argument became popular with initial results of long-term SMPY (study of mathematically precocious youth) program. [Lawrence Summers] Boys and girls were selected based on their mathematical skills. More boys than girls entered the program. There were 12 boys for every girl at high score end in SAT-M (scholastic aptitude test- maths) exam. [Benbow 1983] People latched onto these result to claim the preponderance of boys over girls at extreme skill levels.
As time would have it, the study was continued for a long time and eventually this study itself cropped up results which counter the claim of male preponderance at extreme skill levels. In early samples more boys took demanding courses in high school and entered maths programs. But in later samples, the gaps reduced drastically! Similar number of boys and girls were selected in SMPY. Similar numbers took demanding courses in high school. Similar numbers graduated with mathematics degrees. For example in one SMPY batch, 10.3% men and 9.7% of women received bachelor’s in mathematics and 2.2% of men and 2.1% of women went on to receive the masters degree in mathematics. [Benbow 2000] The number of boys at high end of SAT-M scores has fallen to 3 boys to every girl.
Let us talk a little bit about SAT-M or other standard test scores. 3 boys per girl at extreme end of scoring is still a fairly high number but there are complex socio-cultural and cognitive reasons for that. One important reason is preponderance of questions which favor certain strategies favoring boys. Research has shown the SAT-M score underpredicts the future performance for girls [Spelke 2005]: Among boys and girls with of same SAT-M score, girls go on to earn higher grades in future!
Stereotype Threat: If people will tell Aarna that girls can’t do well in maths and sciences; she will not do well, irrespective of her abilities. Widespread TV ads showing stereotypes also hamper her!
Stereotyping of girls sometimes leads to added worry and pressure on them. They have to devote working memory in not worrying and hence their performance on high skill maths tests suffers. The stereotype that girls can’t do maths, hence becomes a self fulfilling prophecy! Imagine how frustrating and confidence draining that can be!
In a study at University of Michigan, top 10-15% girls and boys as per their SAT-M score were chosen and given a difficult mathematics test. There were negligible sex differences in usual test, but when the students were told that girls perform worse in the test, the performance of girls dropped! [Spencer 1999] In another more alarming study, more sex differences emerged in a maths test with high performing students when girls were shown gender stereotyping ads like beauty product ads (which is what most media does anyways) while negligible differences were present when ads showed women in empowered roles like engineers! [Davies 2002]
Aarna wants to become an engineer, but someone tells her “Men like working with objects and their mechanical relations. Women like working with people and emotions” You’ll do great as a nurse or teacher!
Teasing out cause from effect is very difficult in any scientific endeavor. So a good time to judge whether above claim is true is to look at infants, when few external parameters can affect the results of an experiment! Three decades of research on this topic with thousands of studies on infants, in fact, does not provide any evidence of male advantage in object perception and reasoning [Spelke 2005] An older literature review conducted in 1974 [Macoby 1974] also concludes that women are people oriented and men are object oriented as first of many “unfounded beliefs about sex differences” (p. 349)
Some other proponents of this claim use evolutionary psychology to argue that men have evolved to work outside homes and deal with tools etc. while women are evolved to take care of children and stay home. Hence men’s object oriented skills are better and women’s social skills are better. With due respect our hunter-gatherer ancestors no one questions evolution, but trying to explain all human traits 5 million years later is not trivial! These arguments are good for mating preferences and other very basic human traits which were present in hunter-gatherer society. But application of evolutionary arguments in high-level maths and science skills, which is a very recent human achievement, is fraught with dangers! [Geary 2007] Hence this line of reasoning, in the face of other evidence, also does not hold much ground.
Why There Aren’t Enough Women in Top STEM Positions?
It is a matter of choice, not necessarily ability. Also not to mention the chronic and horrible harassment culture prevalent in male dominated fields. #MeToo campaign has brought many such stories to light and we encourage you to look them up and read them. Also let’s face it, there is severe gender discrimination! We won’t talk much about these very important issues as they are raging topics on their own. But we will tell you about an outrageous incident from Sweden [Wenneras 1997] to give you an idea (Sweden is considered to be leading country in terms of equality, imagine situation in other countries!).
Two investigators wanted to learn about gender discrimination in peer-review process in awarding post-doctoral fellowships. They were denied access to documents (National Science Foundation in US has also rejected such proposals saying that the data is not in a form which allows comparison of men and women). But Swedish Freedom of Press act granted them access and the results were startling! The top female researchers (more than 100 impact points) were rated comparable to the worst male researchers (less than 20 impact points). All other female researchers were rated below all male researchers!
Social improvements with time and across cultures already show evidence of performance gap between men and women closing. Hope we can build a world, in time, for kids like Aarna, where there is none!
If there indeed was a biological advantage for males, it should appear across cultures and the difference remain constant for a long time, unless the brain evolves further. But preponderance of high scoring males has dropped in US itself in a matter of 10 years. In 1970, women constituted 5% law school and 8% medical school students, this number is close to 50% now! [Letter in Science] Also, the preponderance of high scoring males reduces by a good amount in some countries [Deary 2003] and becomes completely absent in others [Feingold 1994]. In the data from Program for International Student Assessment (PISA), where 276,000 fifteen year students appeared from different countries, strong correlation is found in the emancipation status of women and the difference in scores of boys and girls. The differences are negligible in countries like Norway while large in countries like Turkey.
Summary and Conclusions
If you came this far, please take these away!
- Research on cognitive abilities of males and females does not support the claim that men have higher “innate ability” for maths and sciences.
- As infants and in younger age, the aptitudes of male and females are the same and they show similar core cognitive abilities. [Spelke 2005]
- Older age shows differences which
- are usually quite small
- stem primarily from differing problem solving strategies: does not mean one is better in overall maths and science abilities than another.
- Hence for one of the most important problems facing humanities: The loss of big talent pool, one needs to look at reasons beyond “innate abilities” and deal with discrimination and harassment at workplaces for a start!
Note: We conceived of this article when our friend Neha was blessed with a daughter, Aarna. We hope that she grows up in a world where no one has such misconceptions about abilities of girls!
[Lawrence Summers] https://www.harvard.edu/president/speeches/summers_2005/nber.php
[A very good counter to Google Memo] https://www.wired.com/story/the-pernicious-science-of-james-damores-google-memo/
[Halpern 2000] Halpern, D. (2000). Sex differences in cognitive abilities (3rd ed.). Mahwah, NJ: Erlbaum.
[Hyde 2005] The Gender Similarities Hypothesis, American Psychologist, Vol. 60, No. 6, 581–592, DOI: 10.1037/0003-066X.60.6.581
[Choi 1996] Choi, J., & Silverman, I. (1996). Sexual dimorphism in spatial behaviors:
Applications to route learning. Evolution & Cognition, 2, 165–171.
[Wang 2002] Wang, R. F., & Spelke, E. S. (2002). Human spatial representation:
Insights from animals. Trends in Cognitive Sciences, 6, 376–382
[Geary 1996] Geary, D. C. (1996). Sexual selection and sex differences in mathematical
abilities. Behavioral & Brain Sciences, 19, 229–284.
[Halpern 2005] Halpern, D., Wai, J., & Saw, A. (2005). A psychobiosocial model: Why
females are sometimes greater than and sometimes less than males in
math achievement. In A. M. Gallagher & J. C. Kaufman (Eds.), Gender
differences in mathematics (pp. 48–72). New York: Cambridge University
[Pinker 2002] Pinker, S. (2002). The blank slate: The modern denial of human nature
[Halpern 2007] Halpern, Diane F. et al. “The Science of Sex Differences in Science and Mathematics.” Psychological science in the public interest : a journal of the American Psychological Society 8.1 (2007): 1–51. PMC. Web.
[Sorby 2010] Sorby SA, Baartmans BJ. The development and assessment of a course for enhancing the 3-D spatial visualization skills of first year engineering students. Journal of Engineering Education. 2000;89:301–307.
[Benbow 1983] Benbow, C. P., & Stanley, J. C. (1983, December 2). Sex differences in
mathematical reasoning ability: More facts. Science, 222, 1029–1030.
[Benbow 2000] Benbow, C. P., Lubinski, D., Shea, D. L., & Eftekhari-Sanjani, H. (2000).
Sex differences in mathematical reasoning ability at age 13: Their status
20 years later. Psychological Science, 11, 474–480.
[Spencer 1999] S. J. Spencer, C. M. Steele and D. M. Quinn, ‘Stereotype threat and women’s maths performance’,Journal of Experimental Social Psychology, 35 (1999), 428
[Davies 2002] Davies PG, Spencer SJ, Quinn DM, Gerhardstein R. Consuming images: How television commercials that elicit stereotype threat can restrain women academically and professionally. Personality and Social Psychology Bulletin. 2002; 28:1615–1628
[Connellan 2000] Connellan, J., Baron-Cohen, S., Wheelwright, S., Batki, A., & Ahluwalia,
J. (2000). Sex differences in human neonatal social perception. Infant
Behavior & Development, 23, 113–118.
[Spelke 2005] Sex Differences in Intrinsic Aptitude for Mathematics
and Science?A Critical Review. American Psychologist, Vol. 60, No. 9, 950–958 DOI: 10.1037/0003-066X.60.9.950
[Macoby 1974] Maccoby, E. E., & Jacklin, C. N. (1974). Psychology of sex differences.
Stanford, CA: Stanford University Press.
[Geary 2007] Geary, DC. Educating the evolved mind: Conceptual foundations for an evolutionary educational psychology. In: Carlson, JS.; Levin, JR., editors. Educating the evolved mind. Greenwich, CT: Information Age; 2007. p. 1-100
[Badgett 2003] Badgett M, Folbre N. Job gendering: Occupational choice and the marriage market. Industrial Relations. 2003;42:270–298.
[Sandberg 2001] Sandberg JF, Hofferth SL, Changes in children’s time with parents: United States, 1981-1997. Demography. 2001 Aug; 38(3):423-36
[Letter in Science] Multiple authors, Science 18 Feb 2005: Vol. 307, Issue 5712, pp. 1043
[Deary 2003] Deary, I. J., Thorpe, G., Wilson, V., Starr, J. M., & Whalley, L. J. (2003).
Population sex differences in IQ at age 11: The Scottish mental survey
1932. Intelligence, 31, 533–542
[Feingold 1994] Feingold, A. (1992). Sex differences in variability in intellectual abilities:
A new look at an old controversy. Review of Educational Research, 62,
[Wenneras 1997] Wenneras C, Wold A. Nepotism and sexism in peer-review. Nature. 1997; 387:341–343