Not the language of scientists

Today, the New York times quoted an expert — a psychologist. Either that, or they reached for some random person halfway across the country to offer a viewpoint they really liked. But it seems like they wanted to have comments by a scientist. Here’s what the psychologist had to say about the fact that New York City admits more boys than girls to its top elite schools (which admission is apparently only by exam):

“It is very suspect that you don’t have as many girls as boys in New York City’s specialized schools,” said Janet S. Hyde, a psychology professor at the University of Wisconsin who has published research on girls’ performance in math and science from elementary school through college. Individual girls might be losing opportunities, she said, “but it is also bad for society as a whole because in a global economy we need to identify the best scientists and mathematicians.”

When a scientist says that something is “suspect”, what they are supposed to mean is that it might not be true.

“I discovered cold fusion,” I might say.

“That seems suspect,” the scientist might reply.

But Dr. Hyde (that was cheap — eds.) is not using the language of scientists. She’s saying that it’s morally suspect, and opining about what is good and bad for society. Which is fine, I suppose — people can opine about these issues, and people in one of my fields (Philosophy) make it part of their job. But if Dr. Hyde isn’t speaking as a scientist, then we’re really back to “some random person in Wisconsin thinks admitting students to a school based solely on an examination is a bad idea because they apparently think the tests don’t identify the best scientists and mathematicians.”

In which case, why do I care?

The question of whether schools should be allowed to use a single examination for admissions is an interesting one. I don’t think that the answer is obvious, and I encourage a lively debate in the comments. But I’m also pretty sure that the mere fact that these examinations yield male majorities in the students body doesn’t make them any more suspect as tools for identifying mathematical ability than nearly every college admission system in the country is made suspect as an indicator of academic excellence by the fact that they seem to admit more females. Different admissions systems measure different things — and as one of my former professors was fond of saying, “[evaluations] don’t measure what they want to measure, they measure what they measure.”

In any case, the question is certainly not going to be settled by the random musings of some person in Wisconsin who thinks that the tests are “suspect”.

(Of course, it’s not going to be settled by the random musings of an attorney-philosopher, either. But I’m neither trying to settle it nor being quoted for my expertise in the New York Times.)


  1. Some years ago I came across a group picture of the twelve finalists selected from the US high school population to represent the US in the International Math Olympiads. All twelve finalists were male and six were East Asian (Chinese and/or Korean judging by the names).

    This simply the reality of the racial/gender distribution of mathematical abilities.

  2. Stacy in NJ says:

    And you’re surprised that the NY Times would cherry pick an “expert” who reinforced their own preference bias?

    Or, you’re surprised that said expert is so obviously full of shite?

    Michael, you really need to take off your rose colored glasses. Most of what’s offered as deep thoughts by Right Thinking People is complete and total shite.

    • Stacy in NJ says:

      The question of whether NYC can use a single test to determine acceptance to a SHS is a settled legal matter. Please see the Hecht-Calandra Act.

  3. I think Hyde meant “statistically suspect.” Meaning that the distribution of boy girls’ and boys’ performance is about about the same on these subjects, so there is no reason to expect different levels of admission to top schools. Hence, she suspects bias, either in admission, in whose encouraged to apply, something.

    • Stacy in NJ says:

      It depends on your definition of performance. If performance is classroom grades, then it is statistically suspect. If it’s examination scores, then no.

      • Stacy in NJ says:

        Males outscore females on the math portion of the SAT:

        “Even though female high school students are better prepared academically on many different measures than their male classmates, both overall and for mathematics specifically, female high school students score significantly lower on the SAT math test, and the +30-point differences in test scores favoring males has persisted for generations. At the high end of math performance, high school males significantly outperformed their female counterparts on the 2012 SAT math test by a ratio of 2-1 for perfect and near-perfect score”

    • Mark Roulo says:

      But the distribution is not the same. The girls perform better, on average, than the boys in high school math classes, but:
      (a) Have a slightly worse average on the math portion of the SAT, and
      (b) Are wildly under-represented compared to the boys on both extremes of said SAT math test.


      In other words, on tests where things like class participation don’t count, the girls have a smaller spread of results. And these eight schools are picking from one extreme end.


      So the performance is *NOT* the same, unless Dr. Janet Hyde doesn’t understand statistical concepts like standard deviation.

  4. cranberry says:

    I recommend the graphic accompanying the article.

    Why are people worked up about those 8 high schools? Seven of the eight are more evenly balanced, in gender terms, than any of the “other specialized high schools,” which use criteria beyond the admission test. Of course, the “other” high schools enroll far more girls than boys.

    The test-only high schools seem fairer, to me, in terms of gender balance. The selective schools which use other criteria seem to be biased in favor of girls.

    • Mark Roulo says:

      “Why are people worked up about those 8 high schools?”


      Let me help 🙂

      Why are people worked up about those 8 high schools? Because the “other” high schools enroll far more girls than boys. These eight are an outlier, and clearly must be doing something wrong.

  5. Crimson Wife says:

    What is the gender breakdown of students applying to the exam schools? I would suspect that the STEM schools skew male in their applicant pool while the performing arts school featured in the movie and TV show Fame skews female. This in and of itself is not problematic unless there is some sort of active discouragement going on from school personnel.

    • cranberry says:

      If the pattern’s anything like college patterns. MIT, Georgia Tech, etc. receive more applications from male applicants than female applicants. In general, those colleges admit a higher percentage of female applicants from the smaller pool of female applicants, in order to enroll equal numbers of men and women. It still doesn’t work, in that MIT is currently 55% male, according to College Navigator.

      As far as I can tell, girls have an equal shot at admission to the SHSAT high schools. If they’re interested in applying. If they do well on the exam.

      It is fair to the individual applicant to predicate admission on the results of a single exam. It would not be fair to exclude high-scoring boys in favor of girls who did not score as well.