I think test prep helps a ton for standardized tests. It familiarizes you with the subjects and testing format. Which means when it's time to do the real thing, it's all old hat.
The article only mentions the average score increase, which is useless without a distribution. Because a great many kids have no interest in improving and are just being forced to do test prep by their school and parents. It won't help kids that don't care.
Interesting link. Goes to prove that test prep in the easily measurable metrics (enrollment in prep courses, tutors) has negligible effects. Somehow the writer manages to waive away all unequal environmental factors due to income with that one proof, (even turns it on its head: "black people actually do more test prep!") which speaks volumes about his agenda.
I didn't mean different classes of prep, but different quality. You would be surprised how much test prep gives detrimental guidance. I had to unteach a lot of poor or dated suggestions.
And I should note I did this for free... Not party of a program or company.
> Test prep generates very small improvements in scores.
> If test prep did work, it would reduce racial gaps since blacks do more test prep than average.
You have some evidence for either of these propositions vis-a-vis NYC specialized school admissions tests, or did you just want to grind your axe regardless of relevance?
Do you believe that this particular narrowly specialized test with little data available is wildly different from the mainstream tests on which we have data? If so, why?
Is it really your claim that SAT prep rates by race are identical to prep rates for every narrow specialized test out there? No confounding factors having to do with how well known the SATs are in various communities as compared to other tests? None to do with public and charitable funding for SAT prep that doesn't exist for other tests? Nothing like that? It's a standardized test so let's extrapolate from whatever data happens to be handy?
Do you have data on any narrow, specialized exam to posit this correlation with the SAT?
It's people like you with your ablesist bias that are the problem. You conflated lower test scores with lesser moral worth. I really hope that was just an arguing tactic, i.e. a lie because if your estimation of people's moral worth is based on their IQ…
This subthread is going down the toilet, which is unfortunate because it had the ingredients for an interesting conversation but then devolved into moralizing.
Something I've learned in 15 years of marriage: it's hard enough to maintain comity and charity in debates with someone who shares almost all of your values and principles, and with whose background you are thoroughly acquainted. A big part of that is that we use slightly different words and framings to convey ideas, but (I think perhaps due to combinatoric explosions) we can hear dramatically different things from each other as the little mismatches stack up. For me, this seems unsurprisingly to be especially the case in abstract discussions.
So I think there's little hope in sustaining a moral debate 3 comments deep on a thread with a stranger on HN, and I think we'd all be a little better off if we stopped pretending we could.
I regret the tone and the resulting loss of an opportunity to persuade bystanders who aren't mostly of the same opinion as me already. In future I shall aim to be more Scott Alexander and less Charles Clymer/Arthur Chu.
Of course it was an arguing tactic, it was an attempt to respond to yummyfajitas' race-baiting. If you want to be morally outraged, try reading the comment section of his marginalrevolution link.
http://marginalrevolution.com/marginalrevolution/2014/03/the...
If test prep did work, it would reduce racial gaps since blacks do more test prep than average.