Building on the successes we saw in 2023, new midline results from the evaluation of our large-scale “FasterMath” programme in our Freetown City Council (FCC) partnership showed big impacts on teachers’ knowledge and improvements in student learning. Overall student math outcomes in schools receiving FasterMath improved by 0.16 SD relative to control schools at the mid-year point. This would already be considered a medium-sized effect in the literature and a cost-effective one (given the low per student unit cost). To understand why these results are so exciting, it's important to understand the contributing factors to the programme’s success and how these can be replicated to scale FasterMath. We explain this below.
The Mayor’s #TransformFreetown Initiative was the springboard for improved learner outcomes
In January 2021, Rising and EducAid partnered with Freetown City Council (FCC) to support the Mayor’s #TransformFreetown Initiative to achieve the following education target: “Ensure that by 2022, 100% of schools for which FCC has responsibility are covered by a monitoring framework that includes teaching, learning and safety standards.” Mayor Yvonne Aki-Sawyerr OBE’s initial vision focused on developing and adopting a framework of minimum school standards, training school leaders to build knowledge and capacity, and providing additional in-school mentorship for ongoing monitoring and support. Following the successful engagement with school leaders in the first phase of the partnership, the focus deepened into classrooms, to improve the quality of teaching and learning of foundational numeracy.
Why FasterMath is the perfect fit for FCC
In Freetown City schools, Rising’s FasterMath programme is implemented by government teachers with fortnightly in-school coaching and support from Cluster Mentors. In FasterMath, teachers assess students and select the level of content that is appropriate for the learning level of their class. Students participate in one FasterMath lesson (45 minutes) per day. Teachers are supported by detailed lesson guides and are also trained on foundational numeracy instruction. The combination of ability-level instruction, with high quality resources and training, alongside routine coaching and monitoring from Cluster Mentors, has led to an increase in students' proficiency in math.
The FCC FasterMath partnership was designed as a randomised controlled trial (RCT). At the start of the 2022-2023 academic year, 241 schools were identified for inclusion. Schools were randomly assigned to treatment (FasterMath programme) or control after a baseline student learning assessment. A total of 123 schools were randomly selected to participate in the FasterMath programme, which included approximately 22,000 students and over 750 teachers and school leaders.
The partnership’s objectives were clear:
Improve student learning outcomes in foundational math;
Increase teacher pedagogical knowledge and confidence in teaching math.
In the first year of implementation, a rigorous research design and strong focus on capacity building contributed to early successes
From September 2022, the focus was first on setting up the research evaluation and investing in staff capacity building. There was an intensive period of training and professional development for Cluster Mentors, who were all new to the FasterMath programme, which included video-based observations and feedback. Cluster Mentors then led the roll out of FasterMath by leading teacher and school leader training sessions, with in-class teaching starting in late January 2023. Video and audio were used to gain insights into classrooms and coaching conversations, and provided essential feedback for iteration and improvement.
744 teachers and school leaders were trained on Growth Mindset, Safeguarding and FasterMath (May 2023)
Due to the national elections in Sierra Leone, teaching stopped by early June 2023. FasterMath was therefore implemented for approximately 14-16 weeks rather than the longer time scale as intended.
Results from the first period of FasterMath delivery demonstrated a positive effect on the nearly 750 teachers, with a significant improvement in teachers’ understanding of the content and their pedagogical knowledge. Although teachers made positive gains, the results did not yet indicate a significant effect of FasterMath on student learning.
In the second year of implementation, we made adjustments based on evidence and feedback, which helped us reach our goals
The second year of FasterMath delivery expanded from upper primary grades to include lower primary grades where offered, to test the potential benefits with younger students. The number of schools remained the same but the number of students increased to approximately 34,000.
Mid-year results indicated a positive effect on teacher knowledge, continuing the positive gains from the prior year. Results also indicated that FasterMath improved overall student math outcomes by 0.16 SD. This effect was driven primarily by improvements in 1-digit and 2-digit addition and subtraction, and improvements in word problem solving (0.14, 0.14 and 0.31 SD respectively). These are the skills where students had the lowest learning levels at the start of the academic year. Similarly, the FasterMath programme reduced the percentage of zero scores in 1-digit and 2-digit addition and subtraction by 3.5 to 5.6 p.p.
Results varied by grade, with students in grades 3, 5 and 6 showing larger learning gains. Additionally, students who participated in FasterMath also became more confident in their ability to solve math problems and reported that the assessment was easier than those in the control group.
A scalable, cost effective model
The partnership took an iterative approach to improve FasterMath design and delivery. The first year of implementation focused on building the capacity of Cluster Mentors, teachers and school leaders to deliver high quality lessons, and included the use of innovative, video-based strategies to identify areas for improvement. This led to improvements in the FasterMath content as well as the approaches to training and coaching. The second year simplified in-school implementation to increase teaching and learning time, and continued to strengthen the approaches to high quality lesson preparation and delivery.
Results demonstrate the potential of Rising’s FasterMath program to improve foundational numeracy skills for primary grade students at scale. The first year of FasterMath implementation provided evidence for the positive effect on teachers, with an improvement in their content knowledge and self-confidence in teaching math. In the second year of implementation, improved teaching has started to translate into improved student learning, particularly on the foundational numeracy skills that students struggle with the most. Importantly, the cost of implementing FasterMath is relatively low, at approximately $15 per student per year at current scale.
The positive effects on teachers and students suggest the benefits of ability-level instruction combined with high quality resources, training and coaching is a feasible, scalable method.
What’s next for the FCC Partnership?
In the next phase, we will continue to prioritise a collaborative and transparent, data-informed approach to our partnership with FCC. We will also:
Explore opportunities to expand FasterMath into the control group schools, reaching students who have not yet benefited from the programme;
Test the potential to progress towards a school-based model, where school leaders are equipped with the tools and training they need to drive high-quality FasterMath implementation in their schools;
Continue to strengthen our data systems for large-scale FasterMath implementation;
Pilot Rori, Rising’s multi-award winning AI-powered math tutor chatbot, particularly with students in JSS.
*FasterMath is an accelerated learning programme designed to catch up foundational numeracy skills in primary grades.
The analysis for this evaluation was supported and reviewed by Prevail Fund.