Wednesday, July 02, 2008

Why Admitting Women Raised Test Scores

Several men's colleges found that their test scores went up significantly when they admitted women. The readiest explanation is that women are just better students than men. In many respects this is true.

Joseph Soares, in The Power of Privilege: Yale and America's Elite Colleges, treats the story of the supposed displacement of the old boys' network by meritocracy at Yale and other elite colleges. I will consider the major conclusion of his work tomorrow. Today I want to lift up an illuminating detail of that process.

Yale has always admitted men of the northeastern Protestant establishment who showed promise of turning into leaders. In the early 20th century increasing numbers of smart, promising men who were not of the northeastern Protestant establishment also wanted in. Most challenging to Yale and the other elite colleges were the Jews, known in Yale euphemism of the '30s as "Brooklyn boys." The Ivies and other elite colleges invented the College Board and its standardized tests as a way of screening all applicants in a common way. At first they expected that the northeastern Protestant rich boys would win on that test, too, as they had on other selection criteria. It soon became clear that non-WASPs, especially Jews, were doing quite well on the standardized tests. It also became clear that the SAT and other standardized tests didn't predict school performance very well, anyway.

Thus, Yale and the other elite colleges scaled back the weight they put on standardized tests in the '50s and '60s. They developed many other measures of leadership potential for the young men they considered for admission. Test scores counted to some extent, but the "character" and "culture" measures, though less quantitative, counted for a great deal.

When Yale decided, in a hurry, in the late '60s to admit women, the men who had always run admissions felt at a loss to know how to weigh the character, culture, and leadership potential of women. So for the pioneering classes of "co-eds," Yale relied almost exclusively on test scores. As a result, 1970, the first full year of women's admission to Yale College, Yale posted its highest average test scores.

The median SAT verbal score for the class admitted
in 1952: 634
in 1970: 699
in 1992: 660

By the end of the '70s Yale had figured out how to weigh women's character, too -- which meant they admitted women with lower average test scores, just as they had always done for men.

Admitting women to Yale -- and probably at other former men's colleges -- temporarily raised standardized test scores significantly because that was all they had to go on. When the rudiments of an old girls' network began to be constructed, Yale women could have lower test scores, "gentlewoman's Cs," and leadership potential. Just like the men.


Olivia said...

Dear Gruntled,
I like the post and agree with the findings, but find it hard to believe that the admissions board at Yale wouldn't admit women based on their family connections. I would think that as single-sex schools open up, they would try their best to prevent "strays" from infilrating the university. Maybe the daughters and sisters of Yale alumni were just smarter.

Jillian said...

How can you compare scores from 1952 to 1992? Haven't the test been substantially changed by the test makers in that period of time?

Gruntled said...

They did, in fact, think that they would be looking for women with Yale connections in that first class. The alumni admissions interviewers also had their Seven Sisters wives participate in the women's (but not men's) interviews. In the end, Soares reports, they went with scores and grades for those first few classes.

The SAT did change from '52 to '92, though the big renormings came after that. You can already see the fall-off in scores by the end of the '70s from the 1970 peak. The plateau was higher than all-male Yale had been, but lower than during the more fully meritocratic moment of women's admissions.