Efficiency in Organizational Structure: Experimental Evidence from Ugandan Schools
How to cite
Pelnik, Carolyn; Bartkus, Viva; Brooks, Wyatt; Kaboski, Joseph; Rubey, Kelly; Sikenyi, Maurice. 2023. "Efficiency in Organizational Structure: Experimental Evidence from Ugandan Schools." AEA RCT Registry. AEARCTR-0012445. https://doi.org/10.1257/rct.12445-1.0
Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.
* Primary investigator
Abstract
We experimentally vary the organizational structure of schools, and we assess the effect on resource allocation within schools and school efficiency. We do this by measuring head teacher and classroom teacher time allocation and student test scores, before and after the organizational change. We consider two organizational changes. First, in all treated schools we initiate a community volunteers’ program that supplements classroom instruction and enrolls out-of-school students. Second, in a subset of treated schools we also provide a highly trained and educated teaching fellow who assists in school-wide management generally. Schools are randomly assigned to one of two treatment groups or to a control group receiving no intervention.
Intervention
In all treated schools, we initiate a community volunteers program that supplements classroom instruction with additional literacy and numeracy programming, following the principles of Teaching at the Right Level (TaRL), and works to (re-)enroll out of school students. In a subset of treated schools, we also pay a highly trained and educated teaching fellow to assist in school-wide management generally. This results in one treatment group with volunteers only and another treatment group with volunteers and a fellow.
Experimental design
We randomly assign 112 schools (in 112 distinct subcounties) across Uganda to a pure control group or one of two treatment arms: (1) volunteers only and (2) volunteers plus a fellow, stratifying on district. Within each school, we measure enrollment, attendance, literacy, and numeracy for P4 ("primary four") students. When a P4 class is too large for all students to be surveyed, we randomly draw 20 students for participation in the study.
Design details
Cited by
No citation data available for this registration yet.