3 ms·

I would question the statistics here without more details. As a simple matter of course, you would expect very few valedictorians to do anything impressive beca

by gwern 9y ago

I would question the statistics here without more details. As a simple matter of course, you would expect very few valedictorians to do anything impressive because few people ever do anything impressive; in a population of 320 million people (more if you consider competition with foreign countries), 'impressive' typically means something like being in the top 0.001% or higher. For something like a Nobel, winning it means you are literally 1 in a few million people. (Quick, what's the name of the 200th best chess player or tennis player or mathematician?) While on the other hand, every year of every high school in the USA produces a valedictorian (or more than one); if a million kids graduate high school every year with an average class size of 200, then there would be 5000+ valedictorians every year. Are there 5000+ famous impressive people every single year without fail with 100% turnover? No, not really. It gets worse because high schools aren't equally selective. There's probably plenty of tiny little high schools out in the Midwest or in other rural areas where 'valedictorian' means being the best-graded out of 20 or 50 kids, while on the other hand, at a magnet high school like Stuyvesant, half the school might be valedictorian level if they had gone elsewhere (but they can't all be the lone Stuy valedictorian). Unless you've taken these into account, the rest of the discussion of grades selecting for Conscientiousness and conformity etc, while plausible, sound like just so many Just So stories. (I have a similar issue with analyses of Hunter College Elementary School which speculate at great length about why its alumnis appear to be so disappointing, when as far as I can tell, the underperformance is exactly what one would expect from the unreliability of early-childhood IQ tests plus base rates: https://www.gwern.net/Statistical%20notes#genius-revisited-on-the-value-of-high-iq-elementary-schools https://www.gwern.net/Statistical%20notes#genius-revisited-o... )

Fezzik 9y ago
I would just say the metric of "changing the world" the author is using is simply a bad metric. As you say, almost no one changes the world. The interesting part, to me, is that Valedictorians end up doing pretty darn well: "Of the 95 percent who went on to graduate college, their average GPA was 3.6, and by 1994, 60 percent had received a graduate degree. There was little debate that high school success predicted college success. Nearly 90 percent are now in professional careers with 40 percent in the highest tier jobs." In short, being a Valedictorian in high-school is a great predictor of future success. The only perspective from which Valedictorians are not successful, given the data set, is from a holier-than-though mindset. Garbling that message by saying not every Valedictorian has moved mountains is... silly.

by gwern 9y ago

Fezzik 9y ago

I would just say the metric of "changing the world" the author is using is simply a bad metric. As you say, almost no one changes the world. The interesting part, to me, is that Valedictorians end up doing pretty darn well: "Of the 95 percent who went on to graduate college, their average GPA was 3.6, and by 1994, 60 percent had received a graduate degree. There was little debate that high school success predicted college success. Nearly 90 percent are now in professional careers with 40 percent in the highest tier jobs." In short, being a Valedictorian in high-school is a great predictor of future success. The only perspective from which Valedictorians are not successful, given the data set, is from a holier-than-though mindset. Garbling that message by saying not every Valedictorian has moved mountains is... silly.