The estimated effect of Catholic schooling on educational outcomes using propensity score matching
Attendance at Catholic high schools is estimated to improve math test scores and to increase high school graduation rates and enrolment in 4-year college. Propensity score matching methods are used to obtain these estimated effects, based on data from the National Educational Longitudinal Study. Since selection into Catholic schools is non-random, matching methods help to overcome the problem of choosing instruments for identifying the Catholic school effect on educational outcomes. The difference-in-differences approach is used on test score data in order to control for fixed unobservable influences on outcomes.