Pilot evaluation of Early Childhood Education in Berbera, Somaliland

Research and evaluation team: Abdiaziz Ali, Silvia Kahihu, Ann Wanjiku, and Ken Lee

Sector: Education

Project type: Randomised Controlled Trial (RCT)

Location: Berbera, Somaliland

Sample: 92

Target group: Children aged 5 years

 

Project Overview

Background

Research consistently shows the positive impacts of early childhood education (ECE) on both short- and long-term cognitive and educational development outcomes. Studies by Ganimian et al. (2013) show improvements in cognitive skills while Martinez et al., (2012) finds improvements in school enrollment, fine motor skills, and problem solving. Currie (2000) highlights the importance of considering broader aspects like social and emotional development. Rossin-Slater & Wüst (2014) explores the lasting benefits of preschool, and Karoly (2016) emphasises the need for economic evaluations that capture both short-term and long-term benefits, particularly relevant for resource-constrained settings. Several studies emphasise programme quality; Heckman et al., (2010); Blimpo et al. (2019a); but little work has revealed what constitutes a quality programme.

Despite the growing body of research on ECD programmes, it remains unclear to what extent these findings can be extrapolated to the Horn of Africa, where opportunities for young children to access pre-primary education are limited. Furthermore, comprehensive data on ECE in Somaliland remains scarce. This study aims to address this knowledge gap by investigating the effectiveness of a specific ECE initiative implemented in Berbera, Somaliland, by Pharo Foundation. It is especially important given the lack of data on how ECE interventions impact young children in this region.

In 2022, Pharo Foundation launched the first-ever ECE programme in the city of Berbera, located along the nation’s northern coast, with the goal of expanding access to quality education for young children from low- to middle-income families. Berbera is a new Early Childhood Education (ECE) school with space for 90 children. Registrations typically exceed the number of seats available, which requires us to select who is admitted into the programme and who is not, using a “first-come, first-served” approach. Due to its novelty, Berbera offers an excellent opportunity to introduce a new admissions procedure, specifically a lottery system. 

In many countries, lotteries are used to determine placements in highly desirable programmes, such as French Immersion in Canada. To ensure a fairer selection process and enable a rigorous evaluation design, the Berbera pilot programme introduces a lottery-based admissions procedure. This sets the stage for a robust randomised controlled trial (RCT) to assess the programme’s impact on children’s development.

This trial is only the second documented social science experiment in Somaliland, marking a significant milestone in the region’s education sector.

Our research employs a seven-month RCT (October 2022-April 2023,) in Berbera, Somaliland. We measure the effects of ECE on key dimensions of children’s development and school readiness including the cognitive (emergent numeracy), linguistic (emergent literacy), social (socio-emotional)  and physical (gross and fine motor skills) domains. Additionally, since ECE  might free up caregivers’ time at home, we explore potential spill-over effects of the programme on parental well-being and maternal outcomes.

We have three primary questions we want to answer:

  1. What is the impact of ECE on childhood development (captured by IDELA)?​
  2. What is the impact of free childcare on the outcomes of mothers at home?​
  3. What kinds of kids benefit the most from ECE (i.e., heterogeneous treatment effects)?​

This study will provide evidence for the Foundation on how the ECE programme benefits children. In addition, should the findings show that there are certain kids (e.g., kids with low IDELA scores at baseline, kids with no older siblings, etc.) who stand to gain more than others, then the Foundation could increase its impact by altering its admissions protocols (e.g., extra outreach, incentives, etc.) to better target these children. This report provides endline findings for the pilot evaluation. The endline assessment took place at the end of the year just as children were transitioning from ECE to Grade 1.

 

Context 

The ECE landscape in Somaliland is characterised by a complex interaction between traditional Quranic schools and formal educational systems. Quranic schools, which form the backbone of early education in the region, typically enrol children as young as four, focusing on the memorization of the Qur’an and the Arabic alphabet. While these schools play a crucial role in preserving cultural and religious traditions, their pedagogical approach is often limited to rote learning without comprehension, leaving critical gaps in cognitive and educational development. 

In contrast, formal ECE programmes are much less prevalent. Where they do exist, they often suffer from inadequate quality and accessibility, particularly for children from lower-income families and those living in remote areas. Additionally, the broader education system in Somaliland faces substantial challenges, including low funding, a shortage of qualified teachers, and the absence of standardised curricula, which further exacerbates these issues. 

To help address these challenges, the Foundation began implementing ECE programmes in areas that lack them. The Foundation’s ECE programme has enabled the establishment of an ECE unit at the Ministry of Education Services (MoES), the development of a provisional ECE curriculum, and the training of the first group of ECE teachers. 

 

Intervention: The Early Childhood Education (ECE)  programme

The preschool operated 5 days a week for 2-4 hours per day, following daily structured learning sessions designed to stimulate child development through play and centre based activities. Each classroom held up to 30 children and was staffed by one instructor. While enrollment was limited to children between 5 years, classrooms were mixed by gender. The primary language of instruction was Somali, the local language. Additionally, the programme included a 5-week comprehensive teacher training conducted off-site after school hours, the provision of educational materials and daily nutrition support.

 

Research design and methodology

This study is a randomised control trial (RCT) of the Berbera ECE programme  to determine its impacts on children’s learning and development.  In 2022, we mobilised parents to register their children for the ECE programme, leveraging outreach efforts from school principals, community education committees, and the Ministry of Science and Education. Prior to registration, parents were invited to the school to receive comprehensive information about the ECE programme, including  the eligibility criteria for children’s enrollment, and the lottery admission process. 

To isolate a causal effect, the school introduced a lottery-based admission system and carried out both baseline (nT = 92; nC = 107) and endline surveys (nT = 85; nC = 80) to track child- and family-level outcomes. For each sampled child in both the treatment and control groups, we collected anthropometric measurements and gathered detailed baseline and endline data using the International Development and Early Learning Assessment (IDELA) tool, which evaluates children’s readiness for primary school across four domains: early literacy, early numeracy, social-emotional skills, and motor skills. We also conducted interviews with the child’s primary caregiver/head of the household to gather demographic and socio-economic information about parents, caregivers, and other household members. 

Baseline characteristics are balanced between treatment and control kids, with no significant differences for child characteristics (sex, age), household characteristics (infrastructure, no. siblings), asset ownership and child IDELA test scores. 

During the school year, an intensive tracking effort was made to locate the control children and minimise sample attrition. If the child had moved from his or her original place of residence we attempted to interview the child (and their current household) so long as he or she maintained residence in Berbera city.  Overall, we successfully located 83% of the baseline sample. 

 

Results

We estimate a large and statistically significant impact of the Berbera ECE programme on IDELA measures. Relative to the comparison group, enrolled children demonstrated a 15 p.p. increase in average scores (equivalent to a standard deviation), with the largest gains occurring along the emergent literacy dimension. Although this effect is derived from a small sample in a single school (in its first year of operation), its magnitude is larger than those documented in much of the education research literature.

Notably, only 4.7% of control kids participated in a preschool environment during the ECE term, while 96.1% of treatment kids participated in the Berbera ECE programme. This significant difference underscores that children in our programme are substantially better prepared for formal schooling than their peers in the control group.

We find no significant impact on key maternal outcomes like employment status, income, and months worked. However, stress related to providing and preparing healthy meals for children decreased from 70% to 43.5%, likely due to the ECE programme offering meals to students.

Additionally, we find no statistically significant interaction effects, likely due to our small sample size. We had hypothesised that certain children (e.g., girls, first-born, poorer children) might benefit more from the ECE programme, potentially justifying targeted efforts. Future research may explore these questions further.

While interpreting the findings, it’s important to consider some caveats: the small sample size and the single school setting limit the generalizability of the results. This pilot study, however, paved the way for a larger study. Building on the success of the Berbera ECE programme, the initiative has been scaled to Hargeisa, where a more extensive study is currently underway. The results from this expanded study are forthcoming and are expected to provide deeper insights into the programme’s impact on early childhood education and development across a broader context.