Literacy Boost in Rwanda: A Randomized Control Trial

Last registered on May 23, 2017

Pre-Trial

Trial Information

General Information

Title
Literacy Boost in Rwanda: A Randomized Control Trial
RCT ID
AEARCTR-0002214
Initial registration date
May 23, 2017

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
May 23, 2017, 2:34 PM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Primary Investigator

Affiliation
Stanford University

Other Primary Investigator(s)

PI Affiliation
Stanford University

Additional Trial Information

Status
Completed
Start date
2013-08-01
End date
2016-07-01
Secondary IDs
Abstract
This randomized control trial tested the impact of a school-only (SO) approach versus a life-wide learning (LWL) approach to supporting early primary grade learning in rural Rwanda. Schools in SO and LWL conditions received reading materials and teacher training. In addition, LWL villages received training and materials to enrich home and community literacy ecologies. Reading assessments administered to 1668 Primary 1 students in 2013 and again two years later showed that both treatments had positive impacts on learning; LWL produced greater impact. However, nearly one-third of students across all groups could not read at the end of the study, indicating that there is still significant work to do to ensure learning for all.

Registration Citation

Citation
Friedlander, Elliott and Claude Goldenberg. 2017. "Literacy Boost in Rwanda: A Randomized Control Trial." AEA RCT Registry. May 23. https://doi.org/10.1257/rct.2214-1.0
Former Citation
Friedlander, Elliott and Claude Goldenberg. 2017. "Literacy Boost in Rwanda: A Randomized Control Trial." AEA RCT Registry. May 23. https://www.socialscienceregistry.org/trials/2214/history/17914
Sponsors & Partners

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information
Experimental Details

Interventions

Intervention(s)
School-Only Treatment:
a) Regular in-service teacher training for early primary teachers to improve reading pedagogy.
b) Provision of local language reading materials to the school

Life-wide Learning Treatment:
a) Regular in-service teacher training for early primary teachers to improve reading pedagogy.
b) Provision of local language reading materials to the schools.
c) Provision of local language reading materials to communities
d) Reading awareness workshops to educate families and communities on how to support children's learning during the entire waking day and year.
e) Students paired up into 'Reading Buddies' to borrow books and read them together.
f) Led by a local community volunteer, children gather on a weekly basis to engage in fun games and activities that support their learning.
g) Reading Festivals and competitions
Intervention Start Date
2014-01-01
Intervention End Date
2015-12-31

Primary Outcomes

Primary Outcomes (end points)
Promotion / Retention rates (e.g. Was a student who was assessed in Grade 1 in 2013 enrolled in Grade 3 in 2015?)
Student Reading Skills
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
This RCT randomly assigned all sectors within one of 30 districts in Rwanda to one of two treatment groups or a Control group. The first treatment group supported children's learning in schools, while the other treatment group supported children learning in both schools and in their homes and communities.

As the intervention worked in both communities and schools, the sectors were assigned to treatment to limit contamination of treatment groups.
Experimental Design Details
Sectors were randomly assigned to treatment or control conditions using a cluster stratified randomization process. First, the reading assessment data was analyzed for students who reported speaking a language other than Kinyarwanda at home . The research team used statistics from the baseline reading assessment (described below) as they could not locate any official statistics regarding languages spoken at home and disaggregated by sector. We used the percentage of students who reported speaking a language other than Kinyarwanda at home as representative of the actual population. Four sectors with sizeable percentages of students who reported speaking a language other than regular Kinyarwanda at home were assigned to treatment groups or control . Four sectors do not divide evenly among three groups. Therefore, to conduct random assignment, the research team assigned a rank at random to the four sectors within each assignment block. The first of these sectors was assigned to X, the second to Y, and the third to Z. The fourth was moved to a new assignment block for the additional fourth sector from each group. After conducting the initial round of random assignment, there were three sectors in the additional block. One of each of these three sectors was assigned at random to each of the study conditions.

Following this, the remaining sectors were categorized as low, medium or high achieving sectors based on Primary school leaving examination results for P.6 students, provided to the research team by the district education office for 2012. These were the only data available for assessing average achievement in schools. Within each of these three groups (low, medium, and high achieving), sectors were randomly assigned to treatment or control groups. Following this procedure, each treatment and control group was made up of seven randomly assigned sectors.

For transparency, the actual assignment of sectors to control or treatment conditions was done in the presence of and with the participation of Sector Education Officers in October 2013, and was video recorded (available from the authors upon request).
Randomization Method
Randomization was conducted with the participation of District Education Officials within the district office. The randomization process was filmed.
Randomization Unit
All 21 sectors that make up the project district were randomized into one of the 3 groups. Each sector contains an average of 4.5 schools
Was the treatment clustered?
Yes

Experiment Characteristics

Sample size: planned number of clusters
21 clusters containing 102 schools and 635 communities
Sample size: planned number of observations
2041 students at baseline; 1668 at endline (longitudinal observations)
Sample size (or number of clusters) by treatment arms
7 Control Sectors
7 School-Only Sectors
7 Life-wide Learning Sectors
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB

Institutional Review Boards (IRBs)

IRB Name
Rwandan National Ethics Committee
IRB Approval Date
2013-06-25
IRB Approval Number
634/RNEC/2013
IRB Name
Stanford University
IRB Approval Date
2013-06-23
IRB Approval Number
26580

Post-Trial

Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials