Measure for Measure: On the Use of Endogenous Ability Data in School-Process Research

The appropriate use of ability test performance in evaluating schools has been a subject of considerable debate. The present study evaluates three strategies for controlling input-level ability differences in such research: longitudinal, predetermined ability measures; cross-sectional, endogenous ability measures; and SES measures as proxies for ability. In general, endogenous measures of ability behave much like predetermined measures, while SES measures are, for many outcomes, seriously deficient as proxies forability measures. We conclude that research that has relied upon cross-sectional data to controlfor student ability is not seriously flawed because of this, and that the omission of ability controls altogether is a much more serious failing.