Promoting Reading Proficiency Among Children in Zambia: Evidence From a Randomized Experiment of a Teacher-Assisted Reading Program
How to cite
Saasa, Sherinah; Sambo, Cleopas. 2025. "Promoting Reading Proficiency Among Children in Zambia: Evidence From a Randomized Experiment of a Teacher-Assisted Reading Program." AEA RCT Registry. AEARCTR-0014251. https://doi.org/10.1257/rct.14251-1.0
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
This study investigates the effectiveness of an after-school teacher-assisted reading intervention among 5th grade children in improving reading proficiency (comprehension and reading speed), math and English exam performance. We also assess whether gender, hope and self-efficacy moderate intervention effects. Findings show no significant differences in outcomes between experimental and control groups immediately post-intervention. However, our findings show delayed effects of the intervention for the experimental group with significant gains in reading proficiency, math and English performance from post-treatment to 3-month follow-up. In addition, the waitlist control group showed improvement in reading proficiency, math and English performance immediately post-treatment, with gains maintained from post-treatment to 3-month follow-up. We find significant moderation effects of hope and self-efficacy on English and math exam outcomes but not for reading proficiency.
Intervention
The intervention was assisted-reading where a proficient reader helps children struggling with reading to build reading fluency. It included added instructional time and more one-on-on time where teachers spent time with about 8-10 children to assist them with reading skills. Assisted reading sessions were held after school for about 1.5 hours a session, twice a week for 6 weeks. A total of 12 sessions with 18 hours of instructional time.
Experimental design
The study employed a randomized cluster sampling design. Four schools from low income neighborhoods were randomly selected from a list of schools in Lusaka District using data from the Zambia Subnational Poverty Mapping report. Within each schools, a universal recruitment approach was used to identify 5th grade children that met eligibility for study participation (i.e. very poor reading levels per teacher report). Schools were the unit of randomization into the waitlist control or experimental arm.
Design details
Cited by
No citation data available for this registration yet.