Tuesday, April 27, 2010

Income effect on SAT scores


A very interesting set of results on the effect of family income and ethnicity on SAT scores in the US [Hat tip to Droolrag for the pointer]:

... [We] looked at the College Board’s 2009 California demographic research study and parsed out the data showing scores by income and ethnicity, and then added an aggregate mean for additional review. This data shows that test-takers with family incomes of less than $20,000 a year had a mean score of 1310 while test-takers with family incomes of over $200,000 had a mean score of 1715, a difference of 405 points. Further calculations showed a 40-point average score increase for every additional $20,000 in income.

A look at the details reveals that the income advantage is even higher at the low end: For income under $100,000, the advantage is nearly 60 points for every additional $20,000 in income.

3 Comments:

  1. Rainbow Scientist said...

    That's interesting! I am reading "The Genius in All of us" by David Shenk, and this resonates with his arguments that environment and training is more important than genetics in all of us.

  2. Lekhni said...

    I wonder if the study took into account the fact that the higher the family income, the more likely that the test taker had SAT coaching.

    Btw, I'd appreciate it if you also allow comments with Name/Url option. The Google account option is a pretty useless one, but it's the only one available to me on your blog now :(

  3. Abi said...

    @Lekhni: I don't know any more about the study than what's there in that link. Yes, access to SAT coaching could be one of the explanations for the observed income effect; but access to better schools -- made possible by life in wealthier neighborhoods -- is probably a better explanation.

    As for comments, I'm sorry that Blogger does not allow me to use other means of non-anonymous commenting system. However, isn't it possible for your WordPress installation to run an OpenID server -- using this plugin, for example?