Eric A. Hanushek and Margaret E. Raymond, Journal of Policy Analysis and Management, Vol. 24, No. 2
2005
This short, dense article by famed education data gurus Hanushek and Raymond answers the question posed in the title with a "yes": in NAEP scores, they find a "3.5-point gain that came with accountability," a statistically significant amount. As they take pains to show, such analysis is tricky. Accountability systems are imposed statewide, leaving no "control" schools from which to draw comparisons within states. And as in all education analyses, demographics, spending, and any number of other policies muddy the waters - how can one be sure that accountability, not some other factor, led to improved student performance? They acknowledge these challenges and then do a satisfactory job of working through them. For example, they show few discernable differences between early-adopter states and those that implemented their accountability systems more recently - to ease the concern that some other phenomena might have led to both the creation of the accountability systems and the test gains themselves. They then break down the results by race and, disappointingly, find the least gains among black students, with more progress for whites and the most for Hispanics. And they go further. They suggest that introducing accountability systems is associated with only a very slight increase in the number of kids classified as special ed. They also correlate their results with the various Fordham, Education Week, and Carnoy/Loeb ratings of accountability systems to show that stronger systems generally predict larger achievement gains (though "these findings should be treated with caution," given that judgments help drive the ratings and the three methods produce very different grades). They even dig into the nature of accountability systems themselves and find that school "report cards," which inform but don't reward or punish, have little effect. It's a challenging but rewarding article and you can find it here.
