[lbo-talk] Elite Institutions/SATs

andie nachgeborenen andie_nachgeborenen at yahoo.com
Wed Jul 30 07:58:09 PDT 2003


The notion
> that SATs measure anything except
> the degree to which one has been schooled in a
> particular culture is laughable.

Actually it is worse than that. It is _demonstrable_ that SAT type tests measure nothing but the ability to perform on SAT type tests. First, ETS (the Educational Testing Service, which created and administers these tests) no longer claims that the tests measure a quantity called "aptitude," whatever that is. Its stromngest claim is that the tests are milfly predictive of first year college or grad school grades. That is ALL, nothing more, fin, end of claim. However, the evidence for this is that SAT scores plus HS (or college in the case of LSAT, GRE, etc.) grades, combined, predict about 20% of the variance in 1st yr grades, and -- here's the kicker -- controlling for SAT scores, HS/college grades alone predict about 22-24% of the variance in college/grad school grades (respectively). That is to say, SAT-type scores added to the mix make the resulting score less accurately predictive of the only thing that ETS even claims that its test scores predict!

Moreover, an important and devastating fact about how ETS normalizes its results. When it has a question for which high scorers on an ETS test largely get the question "wrong," that is, don't choose what ETS scorers think is the "best" answer antecedently, they recalibrate the "best" answer to be theone chosen by the highs scorers! No kidding.

These results in part from David Owen'w brilliant and measured indictment of ETS, None of the Above.

The real explanation of the role and point of SAT-type tests is given in the profound discussion of "The Examination" in Foucault's _Discipline and Punish_: basically, to control large populations in institutions by giving a pseudo-meaningful impersonal objective criterion for sorting and ranking them for purposes of bureaucratic management.


>
> But, I do think there are two real advantages to
> elite institutions:
>
> 1. encourages experiment>
> 2. Discourges passivity


>
> But, is this worth paying 8 times the tuition? Or,
> conversely, are there other ways
> to foster these qualities: critical/selfless enquiry
> and creativity?

Worth it for what purpose? Surethere are other ways to promote these goals, amny of them probably much better. In our society, they mostly just as difficult as going to an elite school, or unavailable.And there is a further question of "worth it for what"?

Joannea leaves out the further and mnost fundamental reason to go to sucha school, namely the cachet it carries in the eyes of others, which, in many circles, translates into better jobs, more opportunities, more money, etc. These are selish reasons but real ones.

If you think you can do as well with the $ you'd save by investing your Princeton tuition and going to Ohio State instead, understanding "doing just as well" to mean promoting your life goals, then it makes sense to do that. It probably does not make economic sense to go $100,000 into debt for an undergrad liberal arts degree. (A law, med, or biz grad degree will have the prospect of paying you back in fairly short order -- but some of my colleagues are carring $1500/month debt to pay for, e,g., Northwestern Law School. My LS debt, from Ohio State, is about $350/mo.) Harvard, Princeton, and Yale, recognizing this, have replaced loans with outright grants -- they can afford to. But (1) if you won't save and invest (most won't), (2) or you won't do just as well given what you want, and (3) you don't need to go that deeply into debt either because you can get grants or you have affluent parents or a big inheritance, then an elite undergrad degree makes a lot of sense.

jks

__________________________________ Do you Yahoo!? Yahoo! SiteBuilder - Free, easy-to-use web site design software http://sitebuilder.yahoo.com



More information about the lbo-talk mailing list