Research and Evaluation team: Doreen Umwari, Christine Osae, Ann Wanjiku and Ken Lee
Sector: Education
Location: Kigali, Rwanda
Sample: 122
Target group: Teachers
Project overview
The Teacher Capacity Building project in Rwanda, led by Pharo Foundation in collaboration with TEACH Rwanda (TR), is a pilot initiative designed to improve the quality of education at the preschool and elementary school tiers. The project focuses on enhancing teacher competence through targeted training and mentorship programs, with the ultimate goal of increasing primary school completion rates and ensuring smoother transitions to secondary education.
Background Information
The quality of basic education in sub-Saharan Africa, including Rwanda, is closely linked to the competence of its teachers. However, existing pre-service and in-service training programs in Rwanda are inadequate and often misaligned with classroom realities. This has led to subpar educational outcomes, with delayed entry into primary school, high repetition and dropout rates, and low completion and transition rates.
In Rwanda, only 43.53% of preschool teachers have received adequate training, leaving many Early Childhood Education (ECE) centres reliant on untrained staff. Consequently, children entering primary school lack foundational skills in literacy and numeracy. This perpetuates poor learning outcomes and hampers national education goals. Research highlights low readiness levels among preschool learners, with nearly half of children entering Grade 1 having never attended preschool, and many who do attend still lacking required competencies.
To address these challenges, the project has adopted a strategic approach focusing on teacher capacity building, leveraging the evidence that improved teacher performance directly enhances the quality of instruction and student outcomes.
Intervention details
The teacher capacity-building program was delivered to preschool and elementary school teachers through a partnership with TEACH Rwanda (TR), in 16 private, public and community based schools in Kigali over 11 months. Teachers participated in ongoing, tailored training sessions conducted on-site that emphasised effective, student-centred teaching approaches focused on Learning Through Play (LTP) and Teaching at the Right Level (TaRL) methodologies.
During the first three months, teachers received training in 15 tailored modules, covering topics such as the Competence-Based Curriculum (CBC), play-based learning principles, curriculum planning, classroom management etc. The training sessions were conducted on-site by school-based mentors through 2-hour modules. In addition, teachers received weekly ongoing coaching and support by mentors who spent four days in the field working directly with them. On the fifth day, mentors collaborated with the Head of Education to address observed challenges and plan upcoming activities.
Regular classroom observations were conducted to assess the application of new techniques, followed by detailed, targeted feedback to support continuous improvement. Each teacher was observed once a week. Collaborative teaching sessions were also organised to foster peer learning, allowing teachers to exchange ideas and refine their skills in real-time.
Five months later, participants attended a five-day joint refresher training session, which included a specialized day for training school leaders. The training emphasized age-appropriate play-based instructional techniques, fostering positive classroom and work environments. To support the implementation, each beneficiary school received a computer and relevant training materials from the Foundation.
To monitor progress and provide additional support, mentors conducted follow-up evaluations three, six, and nine months into the program. Based on emerging needs, school-based mentors also offered customised professional development sessions, addressing specific challenges that teachers encountered during the implementation of LTP and TaRL practices.
Methodology
We evaluated the effectiveness of an in-service teacher training and coaching program to understand its effects on teaching practices and to generate evidence to inform potential scaling efforts.
The study employed a quasi-experimental design, comparing teachers who participated in the program (treatment group) with those who did not (comparison group). The study sample included 174 teachers from 16 treatment schools, and 152 teachers from 15 comparison schools encompassing public, private, and community-based institutions. These schools were carefully chosen by the education team to ensure contextual similarities with the treatment schools, allowing for meaningful comparisons.
Our primary method of data collection at baseline and endline was through classroom observations. A standardised observation tool was utilised to assess teaching practices across three domains: pedagogy, learning environment and curriculum planning and implementation. The observations were conducted by independent, trained mentors who served as enumerators. These mentors underwent rigorous training on the observation protocol to ensure consistency and reliability in data collection.
Data collection involved assigning scores for specific elements within each observation. Each domain consisted of 15 elements, contributing to a maximum total score of 45 for the entire observation tool. Observers rated each element on a binary scale, with "yes" scored as 1 and "no" as 0.
For this analysis, we calculated scores for each teacher across the three domains by averaging the ratings assigned during the observations. We first calculated the overall score, which is simply the teacher’s average score across all three domains. This captured the general importance of the full set of teaching practices measured by the evaluation.
We also calculated separate average scores for each domain to examine differences in specific teaching practices. The pedagogy score measured differences in teachers’ ratings on standards related to the use of instructional practices, student-centred approaches, and interactive teaching methods. The learning environment score captured differences in teachers’ ratings on standards that assessed classroom organisation, the creation of a supportive and inclusive atmosphere, and the use of play-based learning materials. The curriculum planning and implementation score focused on differences in teachers’ ratings regarding their ability to design and deliver lessons effectively, integrate play-based activities, and employ reflective techniques such as KWLS (Know, Want to Know, Learned, Still Wondering) to monitor children’s engagement and learning progression. Our primary analysis compared these scores across the treatment (n=122) and comparison (n=152) groups at the endline. This approach allowed us to examine changes in teaching practices over time and evaluate the program’s impact on the participating teachers.
Results
Data analysis is ongoing, and the results will be shared in subsequent reporting. These findings will contribute to understanding the program's efficacy and its potential for broader implementation.