New study shows boys outnumber girls 13 to 1 as top math scorers
Differences between the natural capabilities of the sexes can become an emotionally charged issue. But such differences do exist. And an aptitude for mathematics at the highest levels of achievement may be one of them.
This is strongly implied by the latest findings of Camilla Persson Benbow and Julian C. Stanley of Johns Hopkins University. In a study of students selected for high intellectual ability, they find ''that by age 13 a large sex difference in mathematical reasoning ability exists and that it is especially pronounced at the high end of the distribution (of achievement).'' Their data show that, among students who scored 700 or more in the College Board Scholastic Aptitude Test (SAT) for math, ''boys outnumbered girls 13 to 1.''
One should not make too much of these findings. They should not be allowed to become a ''battle-of-the-sexes'' issue. They do not imply that, in some sense, males are smarter or intellectually more gifted than females. They do not imply that females are grossly inferior in mathematical ability. And they certainly do not say anything at all about the ability of a individual boy or girl, man or woman.
But, as the two researchers explain in reporting their work in Science, the usual explanations of environmental influences and social pressures cannot account for the differences in mathematical aptitude revealed here. Something else - perhaps something more basic - is involved.
As Dr. Stanley confirmed in a telephone interview, they purposely did not speculate as to what this something might be in their paper. They did not want such an emotionally-charged subject to interfere with careful consideration of their study. They merely note that ''reasons for this sex difference are unclear.''
Nevertheless, society at large should face squarely two challenges raised by studies such as this.
First, if careful research suggests inherent differences between the sexes in certain intellectual aptitudes, this should not be dismissed out of hand because it is emotionally unacceptable to some people.
Secondly, such findings should not be perceived as implying some kind of inferiority of one sex versus the other. Carefully pursued and integrated with other basic research on mankind's nature, the new findings may one day help elucidate the special talents which each sex brings to the human enterprise.
In this perspective, it is important to understand exactly what Benbow and Stanley say they have found.
Their present study covers two student groups. One of these includes seventh-graders selected in 1980, 1981, and 1982 by the Johns Hopkins regional talent searches from the Middle Atlantic region. Those selected were then given the SAT. A separate nationwide talent search was made in which any student under age 13 and willing to take the SAT was eligible.
The regional talent search turned up 19,883 boys and 19,937 girls - 39,830 in all - who ranked in the upper 3 percent of mathematical, verbal, or overall intellectual ability as determined by the SAT. Benbow and Stanley report that, within this group, no important difference in verbal ability was found between males and females. But, among the high scorers in mathematical ability, there were more boys than girls.
To express it in numbers, among those above the mean (average) in the SAT-M (math) scores, the boy/girl ratio was 1.5 to 1. Among those scoring 600 or better - the upper 21 percent - the ratio was 4.1 to 1. Benbow and Stanley note that this confirms the result of an earlier study of 9,927 mathematically-talented students found by the Johns Hopkins regional talent search. And it confirms it from a broader and much larger data base.
The second group of high math achievers located in the national search is much smaller. These are students under age 13 who achieve 700 or better in the SAT-M. The researchers point out that such children are extremely rare, representing the top one in 10,000 of their age group.
As of last September, an intensive search had located 260 boys and only 20 girls in this category. That is a ratio of 13 to 1, which Benbow and Stanley say they believe to be significant because ''the available evidence suggests there was essentially equal participation of boys and girls in the (national) talent searches.''
They then note that preliminary reports of the 1983 Middle Atlantic talent search yield results for a group of 15,000 students that are similar to what the present study reveals. Including samples gathered by annual Middle Atlantic talent searches going back to 1972, Benbow and Stanley have tested some 65,000 students.
They say, ''It is abundantly clear that far more boys than girls (chiefly 12 -year-olds) scored in the highest ranges on the SAT-M, even though girls were matched with boys by intellectual ability, age, grade, and voluntary participation.''
Such, then, are the researchers' findings. The important points about them are that significant differences in mathematical capability show up only at the highest levels of achievement and that these differences cannot be explained away by environmental influences.
Stanley says that their researches show that the girls were not socially discouraged from studying math. They were not intimidated by social pressure to do poorly when they did study math. Within these select groups, they have plunged into math courses with an enthusiasm and tenacity comparable to that of the boys. Their mathematical ability would outshine that of most men and women in the US population. But there nonetheless is a significant difference between the sexes at this high intellectual level.