Geoffrey MillerVerified account

Yes, I agree. But what kinds of achievement IVs do you mean?

I talked about a few of them in the thread I wrote a couple days ago in response to Steve. Some good work on teachers’ achievement value added vs. behavior value added (both seem to affect outcomes, in different configurations and magnitudes).

Also some admissions regression discontinuities provide evidence that test scores in and of themselves impact attainment. If you’re willing to make the stretch that achievement->test scores, I think you should be willing to reject the strong position I outlined above.

Ps, I agree with the much weaker and much-less-likely-to-go-viral claim that maybe IQ tests can be replaced with achievement tests for most purposes.

Except having replaced IQ tests with achievement tests in the '60s, in 2020 achievement tests are being done away with (e.g. UC, Chicago). Dear old Binet would be saddened, as he developed his ability test specifically to avoid to measure ability in those lacking a supportive env

Right. My Twitter feed the last week or so seems to be full of "throw out all standardized tests" sentiments and any suggestion that tests, which were, at least, *intended* to "level the playing field" can still be helpful for that purpose are met with silence or disdain.

Opinions differ. Personally, I see a lot of value in achievement tests and potential usefulness in identifying ppl who've overcome adversity and might be overlooked. IQ tests are different though in their promise to measure something more general and independent of ability.

But isn't 'g' supposed to be a measure of ability though. A measure of general learning ability? So, what makes that different from an "ability" test in your definition?