What is a Good School, and Can Parents Tell? Evidence on the Multidimensionality of School Output

Is a school's impact on high-stakes test scores a good measure of its overall impact on students? Do parents value school impacts on tests, longer-run outcomes, or both? To answer the first question, we exploit quasi-random school assignments and data from Trinidad and Tobago. We construct exogenous instruments for each individual school and estimate the causal impacts of individual schools on several short- and longer-run outcomes. Schools' impacts on high-stakes tests are weakly related to impacts on low-stakes tests, dropout, crime, teen motherhood, and formal labor market participation. To answer the second question, we link estimated school impacts to parents' ranked lists of schools. We propose a modified mulitnomial logit model that allows one to infer preferences for school attributes even in some settings where choices are strategic. Parents of higher-achieving students value schools that improve high-stakes test scores conditional on average outcomes, proximity, and even peer quality. Parents also value schools that reduce crime and increase formal labor market participation. Most parents' preferences for school impacts on labor-market and crime outcomes are, as strong, or stronger than those for test scores. These results provide a potential explanation for recent findings that parent preferences are not strongly related to test-score impacts. They also suggest that evaluations based solely on test scores may be very misleading about the welfare effects of school choice