May 29, 2008 -- For decades, researchers and educators have debated why boys tend to perform better than girls in math. Are men naturally more logical creatures and thus better at scientific endeavors? Are girls not encouraged by their families, their friends or society at large to pursue scientific careers?
Researchers believe they may have found at least one answer: where girls live. Girls living in countries where there is more gender equality perform better in math, sometimes outpacing boys, than girls who live in countries with more male-dominated societies.
"In societies which are more gender equal, there is a lower gender gap in mathematics," said Paola Sapienza, an associate finance professor at Northwestern University's Kellogg School of Management and co-author of the study published Thursday in the journal Science. Also, "there is a much higher gender gap in reading. Girls become much better in reading" in these countries.
To conduct the study, Sapienza, along with Luigi Guiso at the Instituto Universitario Europeo and Ferdinando Monte and Luigi Zingales at the University of Chicago, examined boys' and girls' test scores worldwide on the same test, using data from the Programme for International Test Assessment. The program provided its 2003 test data from more than 276,000 students from 40 countries around the world.
According to that data, girls worldwide scored on average 10.5 points lower than boys (or 2 percent lower than boys) in math. In reading, on average, girls outscored boys by 32.7 points (or 6.6 percent higher than boys).
The United States fell in line with the worldwide average. In the United States, American girls scored 9.8 points lower than boys in math.
"These are what we called the gender gaps in math and in reading," Sapienza said.
Generally, there are two explanations for the gender gap in math, according to Sapienza: biological and environmental. The biological reasoning says that boys are naturally better in math based on research involving spatial tests.
"It's not very strong evidence because we don't have strong correlation between spatial abilities and math scores," she said.
The environmental-based reasoning posits that girls don't perform as well in math because there aren't enough societal incentives to go into science.
Sapienza and her colleagues then parsed the test data by country; in certain countries, the gender gap didn't hold up and the team wondered why.
"The gender gap in mathematics disappeared [in some countries]. Basically, boys and girls become similar in terms of statistical evidence," she said. "In the same countries where math difference disappeared, the reading gap in girls got better."
Because the culture in each country was different, the researchers then began to look at the gender equality in each country. To do this, they took stats from the World Economic Forum's Gender Gap Index, which measures how women score in access to health and professional resources, and from a World Values Survey, which measures residents' opinions on topics such as women in the work force, school or politics.
"The interesting thing is these two measures correlate," Sapienza said. "It's reassuring. They should be measuring pretty much the same thing."
According to Sapienza, she and her colleagues were somewhat surprised by the findings.
"At some level, we were looking for it," she said. "We were surprised at how strong the correlation was."
The highest performer in math was Iceland, which also has high gender equality index. Girls in that country outscored boys by more than 14 points. In Turkey, which scored low in gender gap equality, girls performed the worst: on average, 22.6 fewer points than boys.
Not every country proved the theory, however. In Indonesia and Thailand, both of which scored low in gender equality, girls outscored boys in math.
Although the study points to at least some environmental influence on girls' math performance, Sapienza stopped short of saying societies' gender equality caused the uptick in math performance, an important distinction for a scientist.
"What we established is the correlation. It's very hard to claim that the moment we change gender quality the math and reading gaps are going to change," she said.
As a woman, she said the results made her happy.
"There's at least a hint this is driven by cultures. So it may be possibly to modify [gender gap] in some societies and that's good news for everybody," she said.