Middle school students taking the SAT? Are some of the best better than the rest?

During grad school I taught for the Johns Hopkins Center for Talented Youth which offers advanced summer classes to 7th and 8th grade students who score above the national average on the SAT.  In a short three week session, these youngsters gobbled up the Harvard undergraduate Intro to Psych course.  It was fun to work with such bright minds, and I often wonder what became of the students I met.

There are plenty of data on the overall success of kids who participate in these talent search programs.  Last week Aimee Groth at Business Insider shared a table from the Duke Talent Search showing measurable differences in lifetime achievement based on the rank ordering of kids within the talent search pool.  Kids in the top quartile had earned more doctorates, secured more patents, and were more likely to have tenured professorships than kids in the lowest quartile.

The Business Insider folks ran the title “This Table Proves Just How Much SAT Scores Predict Future Success”.  We get it that dramatic titles are good for web traffic. And yes, data like these (and tons more of it in the published literature) should be helpful to people like Keld Jensen who declared in a very poorly reasoned article on Forbes.com that “intelligence is overrated”.

Although her title was an oversell, Groth’s table does challenge the “threshold hypothesis” which claims that after a certain point, individual differences in intelligence don’t matter anymore.  The 7th and 8th grade talent searches that use the SAT are doing something very obvious and effective.  They take talented middle school kids and challenge them to a high school test.  The effect is like zooming in with a microscope – resolution increases, and individual differences become apparent in a group that previously looked the same because they all scored at the top of the national middle school standards.  Is it just random in this selective group some kids score 400 on the SAT math section and some score 700? Of course not.  Score differences like that represent real ability differences, even among the elite, and research study after research study has shown that the relative advantage persists to some degree.

This is not to subscribe to some kind of IQ determinism. So what if there is a correlation between 8th grade test scores and life success? The young kids are not simply given their doctorates and patents.  They need to stay in school, make good choices, find something they love, and work hard at it.  Personality factors like grit, perseverance, inquisitiveness, motivation, integrity, resilience (not to mention external factors like social support and plain old good luck) play a huge role in determining success, and many an IQ hare is passed by a tortoise in the race of life. But there’s no denying that well designed tests can reveal meaningful ability differences.  In the pre-employment setting, cognitive tests have consistently proven themselves as the most effective predictors of future success.