Abstract
In Latin American countries such as Mexico or Colombia, for instance, there are severe capacity constraints in the public schooling system that disproportionately affect poor children.
The objective of this impact evaluation is to document using a rigorous experimental research design how a comprehensive private schooling alternative affects early-grade student outcomes among disadvantaged populations in Mexico City. These outcomes include traditional indicators such as enrollment, attendance and performance on national math and language tests, as well as other outcomes that include second language (English) acquisition, technological proficiency and socio-emotional wellbeing. The second objective is to document behavioral impacts on parents that include involvement in children's education, parenting practices and socio-emotional wellbeing. These outcomes are relevant in the context of this evaluation because a strong component of the program analyzed is its parenting school. Finally, the third objective is to document conditions that might make the program scalable.
The study design incorporates Christel House Mexico lottery application process. Christel House Mexico (CHM) is a non-profit philanthropic organization that provides free or heavily subsidized access to a private school in Mexico City to 412 low-income children in grades 1 through 9. Every year, there are more eligible applicants than available first-grade slots at Christel House Mexico. CHM has determined that, as of 2013, a lottery assignment mechanism is the fairest way to allocate oversubscribed first-grade positions among eligible applicants.
The treatment/control allocation rule for the study was a publicly held lottery in June 2013 for the 2013 first-grade preselected applicants that do not have sibling priority. Similarly, a lottery among first-grade applicants in the 2014 cohort that do not have sibling/orphan priority will randomly allocate students to treatment and control conditions.
There are three main data sources for this impact evaluation. The first data source is baseline administrative data that CHM collects prior to the lottery from all lottery applicants, which includes student socio-demographic information, performance on the baseline screening tests that include, among others, the Wechsler Intelligence Scale for Children - Fourth Edition (WISC-IV), and basic literacy tests.
The second data source is follow-up performance data on the Early Grade Reading Assessment (EGRA) and Early Grade Math Assessment (EGMA).
The third data source is cost data from CHM and public schools that lottery losers attend.
Data will be collected annually from FY 2013 to FY 2016. Data collection will be conducted annually from the baseline year (that is, the year before students are admitted into CHM) to the end of 2nd grade. Researchers plan to collect data during May of each evaluation year.
Baseline data for FY 2013 is documented here.
