Luminos Program Impact Evaluation – Second Endline

Executive Summary

This report describes the second endline results of an impact evaluation of the Luminos program in Liberia. The second endline was conducted in May-June 2024, which was one year after children in treatment communities graduated from the program.

The Luminos program is an accelerated learning program that teaches children basic reading and numeric skills and supports their socioemotional development. Luminos focuses on children who have never been to school or have been out of school for several years. The Luminos program aims to help out-of-school children (OOSC) rapidly acquire foundational literacy and numeracy skills so that they are prepared to enroll in government schools and effectively engage with the curriculum.

During the first endline, we assessed the literacy, numeracy, and socio-emotional skills of 1,502 out-of-school children (OOSC) across 49 treatment communities that received the Luminos Liberia program in the 2022-23 school year and 49 control communities that did not receive the program. Additionally, we collected data on the same measures for 348 government schoolchildren (GSC) to serve as a benchmark for learning gains. Results from the first endline of the RCT showed large, significant gains across all EGRA and EGMA subtasks  (IDinsight, 2023).

For this study, we reassessed 1,372 OOSC and 340 GSC from the original RCT one year after the end of the Luminos program. We recorded the enrollment status of these children to measure the program’s effect on enrollment rates and determine if the program was successful in helping children to transition to mainstream school. Additionally, we administered another round of EGRA and EGMA assessments to determine if learning gains persisted one year after the program and to compare treatment effect sizes to the first endline’s result.

This study found that 75% of OOSC in treatment communities reported that they were enrolled in government or private schools, indicating that the Luminos program led to a 15 percentage point increase in enrollment rates among OOSC. Large numbers of control OOSC also enrolled between the first and second endline: enrollment rates in the control group went from 32% at the first endline to 60% at the second endline. Although we cannot identify all the reasons for the large increase in enrollment rates in the control group, we believe that control group enrollment in government schools is likely responsible for comparatively larger growth in learning outcomes in the control group.

The study also found that the Luminos program leads to large improvements in reading and numeracy for OOSC that persist for at least a year after children graduate. Compared to the first endline, effect sizes are 10-15% smaller this year for most EGRA and EGMA subtasks, and effect sizes declined further for the most basic subtasks (letter recognition and number recognition). The slight decline in treatment effect sizes is primarily driven by larger growth in control group scores over the past year compared to the treatment group, likely due to high enrollment rates of control children by the end of year 2. We continue to find that treatment effects on reading subtasks are similar in size for younger vs older children and children who were previously enrolled in school vs dropouts. When comparing boys vs girls, we find that treatment effects are similar across most subtasks, though boys outperform girls across some subtasks.

When we compare government school children’s and treatment OOSC’s scores, we find that treatment OOSC continue to have higher endline averages for most EGRA subtasks, though GSC growth in EGMA subtasks outpaced treatment OOSC growth between the first and second endline. These results were corroborated by teachers, who reported that Luminos graduates generally compared favorably with other students who did not graduate from the Luminos program, particularly with regards to reading ability and attendance.

The positive and encouraging results from the second endline suggest that the program helps Luminos graduates to enroll and succeed in mainstream schools. Our results point to a few tweaks that Luminos could implement to strengthen post-program support for graduates and further improve long-term learning outcomes, including:

• Ensuring that transition support is reaching link schools and considering options for making this support more flexible so that it follows children who move to other communities.
• Continuing to carefully track program graduates and, if possible, monitoring post-program attendance in addition to enrollment.
• Providing communication or sensitization with government schoolteachers on the grade placement criteria used by the Implementing Partners (IPs) and Luminos.

Read the full report here.