Bryan Caplan  

Another Thought on Martorell and Clark

Sumner on the Big Macro Debate... Hummel on Krugman, Keynes, and...
From my last post on Martorell and Clark:
M&C have the best data set ever constructed for detecting ability bias.  After all, they've got measures of years of education, earnings, and initial test scores - i.e. test scores before high school has had much time to work its cognitive magic (if any). 

The upshot: I'd really like to see M&C write another paper where they simply estimate the return to education with and without the initial test score as a control.  Indeed, I'd consider this a more credible lower bound on ability bias than the whole IV literature has managed to produce.
A further thought: M&C know the exact testing dates.  So they could actually estimate earnings as a function of initial test score and continuously measured subsequent educational attainment.  Anyone want to bet on how much this will slash a naive estimate of private return to education?

Comments and Sharing

COMMENTS (3 to date)
Clint Kennedy writes:

The return on education based on testing is more than salaries, but that's a given. The bias of standard testing is very significant because the state and federal standard tests usually don't adequately account for the student's background, ie the tests are not real world enough for them. This is especially true for poor urban students.

Andy writes:

Pretty sure it's a big no-no econometrically to through in intermediate outcomes like that.

Andy writes:

Ugh, "throw" not "through".

Comments for this entry have been closed
Return to top