November 21, 2006

One more study of accountability using aggregate NAEP scores

Yesterday, EPAA published Relationships between High-Stakes Testing Policies and Student Achievement after Controlling for Demographic Factors in Aggregated Data, by Gregory J. Marchant, Sharon E. Paulson, and Adam Shunk of Ball State University. Its conclusions:

The few relationships between high-stakes testing and achievement or improvement in reading, writing, or science tended to appear only when demographic data were missing; and the minimal relationships with math achievement were consistent with findings in previous research. Considering the cost and potential unintended negative consequences, high-stakes testing policies seem to provide a questionable means of improving student learning.

Marchant et al. worked very hard on this study (go read it!), and it tends to reinforce what we've read elsewhere (e.g., Nichols, Glass, and Berliner's work earlier this year). I think, though, that given the availability of individual-level NAEP data, it would be wonderful to see people start to use that, and so today, I have my first editorial, No More Aggregate NAEP Studies? It is not a criticism of the work that people have already done using aggregate data to want researchers to use the finer detail now available. The board members were wonderful in providing honest feedback when I circulated a draft a few weeks ago, and the short editorial is better for their input.

Listen to this article
Posted in EPAA on November 21, 2006 2:06 PM |