The paths of our lives -- marked in seventh grade?

Also in the Science Times for June 8, 2010, John Tierney discusses the possible differences between men's and women's ability in science. He argues that although men and women on average may have equal capabilities, the crucial issue for their contributions to science, and in particular for their representation on academic science faculties, may be that more men than women are exceptionally good at science. Tierney cites a forthcoming study finding this inequality in “gifted seventh graders,” and it will certainly be interesting to see what future studies show or suggest on this score.

Even if this asserted difference between men and women (or boys and girls) exists, and is actually the product of something innate, it wouldn’t automatically follow that it was the source of the underrepresentation of women among top scientists – unless such fine distinctions in ability do shape later achievement. But perhaps they do: Tierney says that in fact “[o]ther studies have shown that these differences in extreme test scores correlate with later achievements in science and academia.”

The particular studies he refers to are studies of cohorts of "intellectually precocious youths" who were identified in the 1970s and early 1980s and have been followed since. It does indeed appear that boys whose SAT – Math scores were in the top quarter of this highly talented group are, for example, between 3 and 11 times as likely to have subsequently earned tenure at a top-50 university as are their counterparts with SAT – Math scores in the bottom quarter of the group. (Interestingly, among women the effects are less clear.)

Tierney goes on to say that “of course, a high score on a test is hardly the only factor important for a successful career in science, and no one claims that the right-tail disparity [the disparity between boys and girls at the far-right, or highest-achieving, end of the math test spectrum] is the sole reason for the relatively low number of female professors in math-oriented sciences.” Certainly the present ratio of men to women on science faculties must also reflect past discrimination, both conscious and unconscious, the effects of which are locked in by tenure for a generation or more.

But when those effects are someday put to one side, how far will these fine differences among the very talented explain any gender disparities in outcome that persist? Tierney says that “Even when you consider only members of an elite group like the top percentile of the seventh graders on the SAT Math test, someone at the 99.9 level is more likely than someone at the 99.1 level to get a doctorate in science or to win tenure at a top university.”

But if this is true for precocious seventh graders, is it also true for the population in general? Perhaps – but it may also be the case that many of the talented adults we respect were not people who showed “precocious” talent in seventh grade. For these talented later bloomers, the relationship between any of their test scores and their later achievements may be quite different than it is for the identifiably precocious.

The studies Tierney cites appear to show that some people begin their progress on the path to adult achievement early and advance steadily along it. In one sense, that’s self-evident, but it is striking to realize that seventh-grade SAT scores can capture clear signs of this progress. But these studies don’t show what might be called the converse: that people who ultimately attained great achievements as adults predominantly began as high achievers in those very areas as children. Until that’s shown, the significance of early test scores, and of the difference between the highest and the high among those early test scores, won’t be fully known.

I hope that our progress through life will not be fully predictable – but it’s hard to deny that studies like this are making the course of our lives more predictable than once was the case. As individuals, most of us see the progress of only a few lives, and the folk wisdom we evolve from what we see is likely far from perfect. As systematic data are accumulated, however, the saving uncertainties in our knowledge may become fewer and fewer.