Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/272322 
Year of Publication: 
2022
Series/Report no.: 
Serie Documentos de Trabajo No. 830
Publisher: 
Universidad del Centro de Estudios Macroeconómicos de Argentina (UCEMA), Buenos Aires
Abstract: 
Using several data sources from Chile, we study the impact of the size of the school choice set at the time of starting primary school. With that purpose, we exploit multiple cutoffs defining the minimum age at entry, which not only define when a student can start elementary school, but also the set of schools from which she/he can choose. Moreover, differences across municipalities in the composition of the schools according to these cutoffs, allow us not only to account for municipality fixed factors (educational markets) but also for differences in the characteristics between schools choosing different deadlines. That is, we compare the difference in outcomes for children living in the same municipality around the different cutoffs with those for children in other municipalities that experience a different change in the available set of schools across cutoffs (double difference in RD). We show that a larger set of schools increases the probability of starting in a better school, measured by a non-high-stakes examination. Moreover, this quasi-experimental variation reveals an important reduction in the likelihood of dropping out and a reduction in the probability that a child would switch schools during her/his school life. Secondly, for a subsample of students who have completed high school, we observe that a larger school choice set at the start of primary school increases students' chances of taking the national examination required for higher education and the likelihood of being enrolled in college.
Subjects: 
Latin America
Chile
School choice set
School achievements
JEL: 
A21
I24
I25
I28
Document Type: 
Working Paper

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.