Left Behind By Design: Proficiency Counts and Test-Based Accountability

Christina Hentges

Derek Neal and Diane Whitmore Schanzenbach
National Bureau of Economic Research
July 2007

A new study from University of Chicago economists Derek Neal and Diane Whitmore Schanzenbach claims that rewarding or penalizing schools based on their number of proficient students adversely affects both high- and low-performing youngsters. The authors examine data from Chicago Public Schools (which introduced accountability reforms in both 1996 and, following enactment of NCLB, 2002). They find that students who scored in the middle deciles before the reforms showed significant gains on tests taken after each reform was implemented. But the highest- and lowest-achieving students barely improved, even though schools were held accountable. The authors explain their findings thus: school officials figure that academic laggards are, well, simply unable to do well on tests. Therefore, rather than embark on the Sisyphean task of educating them, teachers redirect their efforts toward those students who are just "on the bubble" and thus seem more likely to attain proficiency on state tests--i.e., youngsters whose academic achievement isn't great but isn't awful, either. This paper calls for accountability systems that measure schools based on any improvement by kids, not just on the attainment of specific benchmarks. Although Neal and Schanzenbach make a compelling case based on the 1996 changes, their analysis of NCLB's effects is shallow, starting with the fact that their data ends in 2002, the very year the law first went into effect. Find the paper here.