Social Connections and Peer Effects: An Experiment in Selective Enrollment High Schools in Peru

How to cite

Vasquez, Roman Andres Zarate. 2017. "Social Connections and Peer Effects: An Experiment in Selective Enrollment High Schools in Peru." AEA RCT Registry. AEARCTR-0002600. https://doi.org/10.1257/rct.2600-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Roman Andres Zarate Vasquez*

* Primary investigator

Registry ID
AEARCTR-0002600
DOI
https://doi.org/10.1257/rct.2600-1.0
First registered
Last updated
Status
On going
Trial dates
2016-12-10 – 2018-03-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/2600

Abstract

This project aims to quantify the causal effects of being exposed to peers with better cognitive and social skills. To identify these effects, I conduct an RCT in Selective Enrollment High Schools in Peru. Students were classified into four types based on admission test scores and centrality measures of social networks: (i) high-cognitive high-social, (ii) high-cognitive low-social, (iii) low-cognitive high-social, (iv) low-cognitive low-social. Conditional on the student’s type, they were randomly assigned to groups with different types of peers. These groups were used for the assignment to dormitories (in 3rd, 4th and 5th grade) and classrooms (in 3rd grade). I can use this variation to identify causal peer effects on social networks, standardized test scores, and non-cognitive skills. An important component of the latter are social skills, which will be measured by a group of psychologists through interviews and group activities. The final purpose of the project is to develop a theoretical model that uses this variation and results to identify the optimal group composition given an objective function of a policy-maker.

Intervention

Students in Selective Enrollment High Schools (COAR) in Peru were assigned to groups that vary peers' baseline test scores and social skills.

Experimental design

Students were randomly allocated to groups that vary peer characteristics.

Design details

Randomization method
The randomization was done by a computer in STATA. It was stratified at the school-grade-sex level.
Randomization unit
Individual.
Clusters
There are no clusters.
Observations
For 3rd grade, the sample is composed of 2,700 students. For 4th and 5th grade, the sample is composed of 3,700 students.
Arms
One-fourth of the students in each school are assigned to each type of peers.
Primary outcomes
Test scores Non-cognitive skills Social Networks College Outcomes

Cited by

No citation data available for this registration yet.

Keywords