Voice, Exit, and Awareness: A Randomized Evaluation of an Information Campaign in South African Primary Schools

October 3, 2016

Photo: Trevor Samson / World Bank

Providing parents with information about their children’s education and empowering them to make good decisions about their future is an important step in helping poor communities make their school systems accountable. Across the developing world, new initiatives are giving parents both “voice” and “exit”—namely, making them aware of low test scores in their children’s schools, enabling communication between parents and teachers, and giving them the tools to switch to different schools if necessary.  Standardized test results—which can help authorities identify schools that are excelling or in need of assistance, and can empower parents and teachers with useful information—are an important first step. This evaluation will contribute to an understanding of the power of information on parental involvement in their children’s schooling and the impact on learning.


SIEF Cluster: 



South Africa

Study Sample: 

200 schools


2014 - 2017

Policy Tools

Information campaign



South Africa recently instituted national standardized tests for public primary and ninth grade students, but educators have not widely disseminated test results. The government is working with researchers to explore how to best present and distribute test performance data, including comparative school data. The evaluation will examine how different distribution approaches, along with opportunities for parent-teacher dialogue, affect parent, teacher and student actions, as well as learning outcomes.


The evaluation will study the effects of providing parents with information about school outcomes, relative to other schools in the country of similar socio-economic status; and relative to nearby schools at which parents could enroll their children. It will also study the effect of facilitating dialogue between parents and teachers about collaborative actions that could improve school outcomes.

The intervention will be targeted at low-income black communities in South Africa’s poorer provinces. The 200 primary schools in the sample will be randomly assigned to one of four groups: a control group that will receive no new information about school performance; an information intervention group, where parents will be supplied with information about school performance relative to national benchmarks; an information and exit intervention group, which provides information about school performance relative to national benchmarks and neighboring schools; and an information and voice group, which will give parents information about the school relative to national benchmarks, as well as  information about “success story” schools in similar socio-economic circumstances. The fourth group will also participate in a facilitated dialogue between parents and teachers.

The evaluation will capture changes in parents’ beliefs and activities, teachers’ activities, and children’s activities and learning outcomes.


  • Robert Garlick, Duke University
  • Owen Ozier, World Bank
  • Jacobus Cilliers, World Bank
  • Stephen Taylor, South African Department of Basic Education
  • Andrew Zeitlin, Georgetown University