Implementing differentiated learning interventions in Ghana: The case of STARS

• In Event: Core components of effective implementation: Unpacking successful educational interventions

Proposal

This paper presents an overview of the implementation experiences and lessons learned from a collaboration between Innovations for Poverty Action (IPA) and Ghana's Ministry of Education (MoE) on differentiated learning interventions. For over a decade, IPA has collaborated with the MoE on the design, implementation, evaluation, and scale of differentiated learning interventions. In its initial phase, this work represented the first time a targeted instruction intervention had been implemented by a government in Africa. In its current phase, the MoE is scaling differentiated learning to over 10,000 public schools. Across all phases, government staff have led and conducted all implemented activities without direct support from NGOs on implementation and activities.
IPA’s differentiated learning partnership with MoE and its key agency, Ghana Education Service began with a randomized control trial (RCT) in 2010, the Teacher Community Assistant Initiative. This initial RCT assessed four targeted instruction interventions across a sample of 500 schools. The interventions, implemented either with community assistants or with classroom teachers, increased student learning, but all interventions were defined by significant implementation challenges (Duflo et al., 2024). Inadequate managerial capacity and support for teachers likely constrained full implementation.
We conducted a second RCT called Strengthening Accountability to Reach All Students (STARS) to assess whether additional managerial support would solve these challenges. STARS aimed to improve implementation by enhancing headteachers’ and circuit supervisors’ managerial and coaching roles. In the base intervention, both headteachers and circuit supervisors were trained to visit classrooms and observe teachers using a checklist to monitor implementation of the program. In a second treatment variation, in addition to the base intervention, headteachers and circuit supervisors received additional training resources focused on how to be effective managers. Both treatment variations improved learning equally. A year after the conclusion of the intervention, 40 percent of schools were still grouping children by ability level, representing a level of implementation about six times higher than what was observed during the intervention in the first RCT (Beg et al., 2023).
STARS was developed by working along all levels of the education system to build on existing lessons from Ghana and contextualize international best-practices. It combines targeted instruction, teacher training and management support within the existing government system to improve student learning. The programme includes assessment to categorize students by proficiency levels for targeted instruction, dedicated instructional time, contextualised teaching and learning materials with scripted guides, and a system of monitoring and ongoing support for teachers.
Based on the success of STARS, as part of the Ghana Accountability for Learning Outcomes Project (GALOP), the base intervention of STARS is being scaled to over 10,000 public schools, and plans are in place to scale even further. IPA has been providing technical support to ensure fidelity of implementation and program rollout and scaling decisions are data-driven, strengthening of MEL systems and building government capacity for sustainability of the program beyond the funding years.

Authors

• Kanan Dubal, Innovations for Poverty Action

• Henry Atimone, Innovations For Poverty Action