Project

Luminos – Impact evaluation of an accelerated learning program for out-of-school children in Liberia

Kenna Mokobi , Maadhav Anand Kumar , Mico Rudasingwa , Jeff McManus 13 November 2024

Students taking part in a Second Chance literacy class in Bomi County, Liberia ©John Healey/IDinsight

Decision-maker’s challenge

Between 2000 and 2023, the out-of-school rate for primary school-aged children in sub-Saharan Africa dropped from 37% to 19%. However, in Liberia, the figures worsened, with the out-of-school rate increasing from 27% to 31% over the same period (UNESCO Institute for Statistics n.d.). As in other countries, COVID-19 and associated school closures likely slowed down years of progress toward attaining universal enrollment.

For children who enroll in school in Liberia, learning levels are low and learning gains are very slow. This may contribute to families’ decisions to unenroll their children to pursue income-generating activities or to marry at a young age (Kaffenberger, Sobol, and Spindelman, 2021). In our assessments of children enrolled in grades 1, 2, and 3 in government schools in Liberia, the average child can read only 7 words per minute (wpm) on a word reading task at the beginning of the 2022/2023 school year, below the threshold for reading fluency of at least 60 wpm. This figure only improved to 14 wpm by the end of the school year.

The Ministry of Education in Liberia has proposed several interventions to improve learning outcomes and the quality of instruction in government schools. However, recent studies have shown that although large sums have been directed towards education interventions in low- and middle-income countries between 2000 and 2015, there has been little improvement in learning outcomes (Angrist et al., 2023). Few educational interventions have transformative impacts on learning.

Impact opportunity

The Luminos program is a proposed model to address the dual problems of low access to education and low rates of learning. The Luminos program is an accelerated learning program for out-of-school children (OOSC) run by the Luminos Fund in Liberia. The program is a 10-month program that teaches OOSC basic reading and numeracy skills and supports children’s socio-emotional development. The Luminos Program curriculum covers the first three grades of school to help OOSC catch up to grade level and enroll in government schools. In addition, Luminos supports program graduates with their transition into government school by paying enrollment fees (typically 5-10 USD per child).

Our approach

IDinsight partnered with Luminos to conduct a randomized controlled trial (RCT) to assess the impact of the Luminos program on children’s learning and enrollment in government schools after the program. The RCT consisted of 3 rounds of data collection: a baseline, first endline immediately after the program, and a second endline one year after the end of the program. The sample consisted of a treatment group of 49 communities that received the Luminos program in the 2022-23 school year, and a control group of 49 comparable communities that did not receive the Luminos program. At each round, we collected data on a sample of government school children from every community to serve as a benchmark for treatment effects. We assessed 1,745 OOSC at baseline. At the first and second endline, we reassessed 1,502 and 1,372 of the OOSC assessed at baseline, respectively.

The results

At the first endline, we found large and statistically significant effects of the program on children’s reading and numeracy. OOSC in treatment communities were able to read 29 words per minute compared to 7 words per minute for control OOSC on the passage reading subtask. OOSC in treatment communities correctly answered twice as many addition questions and twice as many subtraction questions than control. Effects were similar across subgroups – such as gender, age, and children who started with lower baseline learning levels versus higher baseline learning levels. Compared with government schoolchildren in the same communities, children in the Luminos program started the 2022/2023 school year with much lower literacy and numeracy scores but ended the school year with similar numeracy scores and substantially higher literacy scores than their peers in school. This blogpost allows the reader to explore the first endline results through interactive visualizations.

Luminos Fund RCT results and interactive visualizations

Read the blog

At the second endline, we found that the Luminos program had a positive and statistically significant effect on enrollment rates. OOSC in treatment communities were 15 percentage points more likely to be enrolled in school one year after the program compared to control OOSC. However, we also observed that there was a large growth in enrollment in the control group between the first and second endline, from 32% to 60%.

Treatment effects persisted one year after the program across all the foundational reading and numeracy skills assessed. However, the size of the treatment effects is 10-15% smaller at the second endline compared to the first endline. This is primarily explained by growth in scores in the control group, particularly for control OOSC who enrolled in school. Treatment OOSC who successfully enrolled in school had significantly higher reading and numeracy scores at the second endline than at the first endline. In contrast, treatment OOSC who did not enroll in school regressed in their reading and numeracy levels.

Figure: Average oral reading fluency over time

To compare these results to other education interventions, we calculated the first endline treatment effects in terms of learning-adjusted years of schooling (LAYS). Our study found that Luminos graduates gained 3x as many words per minute as their peers in government school. Applying a ‘learning adjustment’ to these gains of 0.52, which indicates that a year of schooling in Liberia is equivalent to 0.52 years of schooling in the benchmark country of Singapore, according to the World Bank in 2021 – we find that the Luminos Program generates 1.59 LAYS after 10 months. This estimate would place the Luminos Program among the most effective interventions included in a review of 150 interventions in LMICs using the LAYS metric (Angrist et al., 2020).