NEW UPDATE: Completed trials may now upload and register supplementary documents (e.g. null results reports, populated pre-analysis plans, or post-trial results reports) in the Post Trial section under Reports, Papers, & Other Materials.
Smoothing the Cost of Education: Primary School Saving in Uganda
Last registered on January 08, 2014


Trial Information
General Information
Smoothing the Cost of Education: Primary School Saving in Uganda
Initial registration date
Not yet registered
Last updated
January 08, 2014 1:45 PM EST
Primary Investigator
Northwestern University
Other Primary Investigator(s)
PI Affiliation
University of Texas, Austin
Additional Trial Information
Start date
End date
Secondary IDs
Can a financial service be used to improve educational outcomes? To answer this question, Innovations for Poverty Action (IPA) partnered with the Private Education Development Network (PEDN) to design and implement the Super Savers Program. A school based savings program, the Super Savers Program aimed to connect savings and education in Uganda. Along with program implementation, IPA directed a research study to assess the program’s impact.

The study used a randomized control design to measure the impact of different variations of the Super Savers Program. In one variation, students who participated received their savings in cash at school, with some encouragement to invest their savings towards school expenditures. In the second variation, the savings payout was provided in a more restrictive form: vouchers requiring an education-related purchase. Students at a third group of schools served as a comparison.

The program variation which included a savings payout in the form of cash generated higher savings in the program accounts. When combined with a parent outreach program, the program led to an increase in the percentage of children in the classroom with basic scholastic materials and an 8% increase in test scores on an exam covering language and math skills. We find no effect for the fully-committed account, and we find no effect for either account on attendance, enrollment, or non-cognitive skills.

The results from this study contribute to existing research in two ways: First, while we know that savings programs are particularly effective when tied to a specific goal—such as education—putting too many restrictions on payouts can be counteractive. Second, reaching out to parents and households can be an important way to improve primary education.
Registration Citation
Karlan, Dean and Leigh Linden. 2014. "Smoothing the Cost of Education: Primary School Saving in Uganda." AEA RCT Registry. January 08. https://doi.org/10.1257/rct.81-1.0.
Sponsors & Partners

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

Request Information
Experimental Details
Working with a local nonprofit organization Private Education Development Network (PEDN) in the Busoga sub-region of the Eastern region of Uganda, we randomly assigned 136 primary schools to one of three groups: a strong commitment savings account, a weak commitment savings account, or control. For both treatments Innovations for Poverty Action (IPA) developed a brief training component for the teachers, and also coordinated the transfer of money from a savings box held at the school to a local bank for safekeeping. One year into the implementation, we implemented one sub-treatment, a parental involvement workshop, in half of the treatment schools. We measured school inputs, test scores, and attendance, both before and after the intervention.
Intervention Start Date
Intervention End Date
Primary Outcomes
Primary Outcomes (end points)
Savings levels (by school and by student), familiarity with the program, use of the program, possession of school supplies, parental involvement, savings attitudes, payment of school fees, test scores
Primary Outcomes (explanation)
Parental Involvement Index consists of: (1) Student thinks parents are responsible for children's education (2) Has your parent come to your school in the past year? (3) Has your parent seen a report of yours from school in the past year?
Savings Attitude Index consists of 7 statements, each of which the student evaluated on a Likert scale, 1-5. All scales were converted after the fact so that higher on the scale meant more positive attitude toward saving. (1) Saving money is not necessary if you live at home with your family. (2) Saving is a good thing to do. (3) Saving is for adults only. (4) My parents or relatives would be proud of me for saving. (5) Managing to save makes me feel happy with myself. (6) It's better to spend money today than to save it for use in the future. (7) Every time I get money I put away some money for saving.
Secondary Outcomes
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
We selected 136 primary schools from the Jinja, Iganga, Mayuge and Luuka districts of the Busoga Region because they predominantly comprised poor rural and peri-urban schools. We then administered a baseline survey and test during the final trimester of 2009. We then randomly assigned schools to receive either the cash treatment, voucher treatment, or no treatment. Each treatment group comprised 39 schools, and the remaining 58 schools became the control group.

Following the first randomization, school outreach began. It took two trimesters to recruit the majority of schools, but by the beginning of the third trimester of 2010, 95 percent of the treatment schools had agreed to participate.

In 2011, we conducted a second randomization for the parent sensitization program, to isolate the effect of the program while still treating all of the schools. Half of the treatment schools were assigned to receive the treatment in the first term of 2011 while the rest received the treatment during the second term.

Finally, we conducted the follow-up survey and exam during the beginning of the third trimester of 2011.
Experimental Design Details
Randomization Method
In office by a computer. We stratified by the total normalized score on the baseline exam and by geographic regions call sub-counties.

In 2011, we conducted a second randomization for the parent sensitization program. Half of the schools in each treatment were assigned to each group. We stratified assignment by the schools’ initial treatment group and sub-county, and checked for balance using the demeaned savings rates from 2010.
Randomization Unit
Was the treatment clustered?
Experiment Characteristics
Sample size: planned number of clusters
136 schools
Sample size: planned number of observations
4,716 students from the 136 schools
Sample size (or number of clusters) by treatment arms
39 cash treatment, 39 voucher treatment, 58 control schools
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB Name
Innovations for Poverty Action
IRB Approval Date
IRB Approval Number
Post Trial Information
Study Withdrawal
Is the intervention completed?
Is data collection complete?
Data Publication
Data Publication
Is public data available?
Program Files
Program Files
Reports, Papers & Other Materials
Relevant Paper(s)