Abstract
Kazakhstan's secondary education system encompasses a remarkably diverse array of school types, ranging from highly selective exam-based institutions to small rural schools. This study examines how school type is associated with academic performance and post-secondary destinations using administrative data on the full population of 499,150 high school graduates from 2021 to 2023. We classify schools into 19 mutually exclusive categories and estimate ordinary least squares regressions for log-transformed Unified National Test scores and logistic regressions with average marginal effects for five post-secondary outcomes: university enrollment, vocational education, study abroad, employment, and not in education, employment, or training status. Graduates of the most selective schools score 24 to 33 percent higher on the Unified National Test and are 12 to 25 percentage points more likely to enroll in university than regular urban school graduates. At the other end of the spectrum, sport school graduates score 24 percent lower and are channeled primarily into vocational education, while military school graduates are 45 percentage points more likely to be not in education, employment, or training, likely reflecting post-graduation service obligations. We argue that these large associations between school type and outcomes primarily reflect selection: families with greater resources sort their children into higher-performing school types, and the observed school type premiums capture the combined effects of student selection, peer composition, and institutional quality that cannot be disentangled in an observational framework.