Improving Student Participation
Which programs most effectively get children into school?
This page summarizes J-PAL's policy bulletin on student participation, "Roll Call: Getting Children Into School."
Despite a dramatic rise in the number of children enrolled in school, pockets of low enrollment remain, particularly in remote or conflict-affected areas. As of 2015, 61 million children of primary school age were out of school and over 202 million adolescents of secondary school age were out of school. Additionally, millions of children who are enrolled in school are not attending regularly.
Education is an investment of time, money, and effort with many of the benefits coming far in the future. Across many different countries and contexts, the studies in this review show a clear and reasonably consistent picture: parents’ and students’ investment in education is quite sensitive to the costs and perceived benefits of schooling.
Shortening travel time to school
In areas where few schools exist, creating new local schools is a very effective way to increase enrollment and attendance.
Introducing “village-based schools” in remote areas of Afghanistan increased enrollment rates from 27 to 69 percentage points (a 156 percent increase). In areas in Pakistan where no school existed within 1.5 kilometers, granting per-student subsidies to local entrepreneurs to establish new private schools led to an increase in enrollment rates from 30 to 80 percent in the first year. In the second year, the program’s impact was 30 percentage points due to comparison group enrollment rising from 30 to 50 percent.
Reducing distance to school can be particularly helpful for girls.
The “village-based schools” program in Afghanistan improved enrollment rates among girls by 17 percentage points more than it did for boys, eliminating the gender gap when schools became closer. In Pakistan, creating local schools raised girls’ enrollment by 5 percentage points more than boys’ during the first year. In the second year, rising enrollment for both boys and girls across the province eliminated the gender gap in the comparison group, resulting in an equal two-year impact of the intervention for both genders.
Subsidies and in-kind transfers
Where school fees exist, eliminating them can lead to large increases in participation.
A study in Ghana tested the impact of providing full secondary school scholarships to low-income, academically qualified students. Eight years on, girls and boys who had received a scholarship were, respectively, 29 percentage points (60 percent) and 31 percentage points (50 percent) more likely to have ever enrolled in senior high school than students who did not receive a scholarship.
Conditional Cash Transfers (CCTs) have been widely tested and are consistently effective at increasing school participation.
Results from randomized evaluations of CCTs in Burkina Faso, Cambodia, China, Colombia, Ecuador, Honduras, Malawi, Mexico, Morocco, Nepal, Nicaragua, and Tanzania have been highly consistent across countries, with all eighteen RCTs finding positive impacts on school participation.
Even small incentives, or removing small costs, can have large impacts.
In Cambodia, a cash transfer equivalent to just US$20 a year, or 3.3 percent of average per capita spending, substantially increased enrollment and attendance. Sixth grade girls who received free uniforms in Kenya for two years were 3.1 percentage points (16 percent) less likely to drop out after three years than their peers who did not receive uniforms (19 percent of whom dropped out). Boys who received uniforms were 2.4 percentage points (19 percent) less likely to drop out.
CCT design details matter.
A new generation of evaluations have shown that small changes in the design of traditional CCT programs can make them more effective. Examples include timing the payments to coincide with deadlines for school fees, and designing the transfers to incorporate incentives for student achievement.
There are mixed results on the importance of conditionality.
Of two studies that randomized the type of conditionality attached to transfers, one found that CCTs were more effective than UCTs at boosting attendance while the other found similar effects for labeled cash transfers and CCTs.
Reducing the burden of school by reducing child morbidity
Addressing anemia and worm infections increases school attendance.
When the Indian NGO Pratham provided preschoolers with iron and vitamin A supplementation and deworming medication, weight increased among participating children by roughly 1.1 pounds, and preschool participation rates increased by 5.8 percentage points (an 8 percent increase from a baseline attendance rate of 71 percent).
Improving the quality of education
Improving education quality can—but does not always—increase student attendance by increasing the perceived benefits of education, at least in the short term.
One reason that quality improvements may not translate into higher participation may be that parents and students find it hard to judge the quality of education in the short run.
Programs that address perception gaps or make the benefits of education more salient can change behavior at low cost.
A study in the Dominican Republic showed that informing boys of the average wages earned by people in their area based on education levels raised their own perceived returns to education, and that boys who received this information completed an additional 0.20 years of schooling.
Examples can be powerful in changing perceptions.
In India, researchers tested the impact of sending recruiters to hold information sessions for women that included information about jobs in the business process outsourcing (BPO) industry, the compensation levels, necessary qualifications, and how to apply. Women in villages where recruiters held information sessions expressed a greater interest in working throughout their lives—even after marriage and childbirth—indicating shifting aspirations toward work as a longer-term career.
Involving communities in school management
Greater involvement of communities in school management has mixed results on participation.
Out of the seven studies testing community monitoring or school-based management interventions included in this review, three programs improved both participation and learning and two programs improved neither. A program in Indonesia improved learning but had no impact on dropout, and a program in The Gambia improved participation but had no impact on learning.
Adding school supplies
Increased spending on inputs, such as adding computers to classes, does not appear to increase participation through increased perceived benefits.
Programs in Kenya and Sierra Leone provided textbooks to primary schools and found no impact on student dropout, grade repetition, or daily attendance.
Increasing student motivation
Students’ perceived costs and benefits are important for boosting enrollment and attendance.
A program in Kenya that offered merit-based scholarships worth US$19.20 per year for two years to sixth-grade girls who scored in the top 15 percent on district-wide exams increased attendance in the year prior to the final awards by 3.2 percentage points for girls, a one-quarter reduction in absenteeism given baseline attendance rates of 87 percent.
Disaggregating results of studies included in this review by gender shows that most programs that improved school participation overall were at least as effective—if not more effective—for girls as they were for boys. In Nicaragua and Colombia, boys had significantly lower attendance rates at baseline than girls, and in these cases, impacts of CCT programs for boys were larger than for girls. In other words, programs aimed at increasing participation tended to help the most disadvantaged gender best.
Cost-effectiveness analysis (CEA) summarizes complex programs in terms of a ratio of impacts to costs. It allows us to compare programs that seek to achieve the same goal, yet are evaluated in different countries and years. While we provide relative rankings, CEA does not, by itself, provide sufficient information to determine any policy or investment decisions. For example, the relative rankings may be different if relative input costs vary by country (for example, the cost of teachers’ salaries versus computers). However, CEA can serve as a useful starting point in the decision-making process, including by highlighting the types of programs that tend to be the most cost-effective and by establishing cost-effectiveness benchmarks.
J-PAL’s cost-effectiveness analyses focus on the perspective of policymakers considering which program to implement (Dhaliwal et. al, 2012). This determines the selection of discount rates, exchange rates etc. Evaluation costs are excluded from “program costs.”
When interpreting cost-effectiveness, it is important to bear in mind that some programs, particularly CCTs, achieve other objectives than improving student participation. Programs will also tend to be more expensive in richer countries, not least because attendance rates tend to be higher to begin with. Additionally, it may be relatively harder to achieve impacts on participation in secondary school than in primary because older children typically have opportunities to earn higher wages outside of school.
Performing the same analysis from the first cost-effectiveness figure above using “at scale” costs changed the exact cost-effectiveness estimate for some programs, but not which programs are found to be cost-effective overall.
Even excluding the costs of the transfers themselves, CCTs are still not a particularly cost-effective way to increase children’s time in school.
- Conditional and unconditional cash transfers can increase school enrollment and attendance, but are expensive to implement. CCTs and UCTs have been consistently shown to increase attendance at school.
- If CCTs and UCTs are being implemented for social protection reasons, restructuring payments may improve their effectiveness.
- Eliminating small costs can have substantial impacts on school participation.
- Programs that address perception gaps can change behavior at low cost.
- Even when misperceptions do not exist, the benefits of education can seem distant.
- Changes in education quality can be difficult to perceive and may not affect participation.
- Adding school supplies and infrastructure does not appear to increase enrollment or attendance.