cover image: Research Group on Human Capital Working Paper Series

20.500.12592/1vhhsg3

Research Group on Human Capital Working Paper Series

12 Jan 2024

The controls are i) how much farther the top-scoring high school is to student i’s middle school compared to the closest high school (DistanceTop−Closest), ii) how much higher the top-scoring high school’s mean admission score is compared to to the closest high school (Admission ScoreTop−Closest) and iii) a constant. [...] Conversely, if the highest-admission score school is not the nearest to the student’s middle school, then the likelihood of the student attending it decreases by 6.8 pp for each additional kilometer that separates the high school from the middle school relative to the distance between the middle school and the nearest high school. [...] Roughly 94% of the variation in the number of high schools across towns is explained by the town population measured immediately after the fall of the communist regime in the 1992 census. [...] Column 1 shows estimates of a regression of admission scores on the number of high schools and columns 2-4 show estimates of regressions of indicator variables for student admission score quartiles on the number of high schools in the town students attend high school. [...] The mean peer admission score is instrumented by the interac- tion between the student admission score decile and the number of high schools in their town.
Pages
66
Published in
Canada