Using meta-analysis we find that, on average, sought-after schools do not improve student test scores. A potential explanation for this result is that parents value schools that improve outcomes not well-measured by test scores. We explore this notion using both administrative and survey data from Barbados. Using a regression discontinuity design, preferred schools have better peers but do not improve short-run test scores. Consistent with the proposed explanation, the same students at the same schools have more post-secondary school completion and

improved adult well-being (based on an index of educational attainment, occupational rank, earnings, and health). These long-run benefits are larger for females who also experience reduced teen motherhood. Mechanisms are explored.