Blueprint for Use of Evidence in Language of instructions Decisions: Framework for a comparison of LOI evidence & policy in Sub-Saharan Africa

• In Event: Multilingual Education Can Do Wonders: Improve equitable learning, promote social cohesion, and bridge divides

Proposal

Millions of children across low- and middle-income countries (LMICs) are nowhere near acquiring foundational reading and math skills (World Bank, 2022). While this problem stems from many issues, language plays a central role (Nakamura et al., 2023). At least half, and in some countries over 90%, of all children in LMICs are being taught in languages they do not use and understand (World Bank, 2021; Nakamura et al., 2024). At the same time, education stakeholders increasingly demand access to education in post-colonial languages earlier in children’s educational trajectory, given the strong relationship between skills in these languages and socioeconomic mobility (Azam et al., 2013). In addition, varying political agendas cause the sociopolitical milieu affecting language of instruction choices to constantly shift. Therefore, decisions related to which language(s) to teach in, for how long, and to what end must be both politically palatable and empirically effective to improve foundational literacy and numeracy (FLN) outcomes. Decision makers need a systematic framework that politically and empirically reconcile this tension between the need to teach children in a familiar language as long as possible and the need to teach children in an unfamiliar language as early as possible.
AIR, in collaboration with the Gates Foundation, has developed the Blueprint for the Use of evidence In Language of instruction (LoI) Decisions (BUILD) to provide an empirically grounded and politically workable framework to guide decision makers to improve bilingual and multilingual foundational literacy outcomes across LMICs.
Our approach to this work includes the development of three components: a BUILD framework, a progression scale, and a theoretical model for the science of bilingual reading. The BUILD framework encompasses the components of the educational system that will influence the relationship between language and learning. The progression scale helps track a country’s progress on the path towards effective integration of language considerations into solving learning poverty. The theoretical model outlines the scientific underpinning of reading acquisition in bilingual contexts, which is central to all other decisions. Bringing these together, the BUILD project provides stakeholders with a blueprint for achieving the desired foundational literacy outcomes in bilingual and multilingual contexts.
In this presentation, we discuss the underpinnings of this model, along with overarching takeaways from evidence reviews that we conducted across four African countries based on the model. Based on the BUILD theoretical model, we reviewed the alignment of key components in country education systems with evidence on best practices related to language of instruction. These components include the overall FLN goals and policies, the teacher workforce, the curriculum and materials, and the monitoring and evaluation system, among others. Finally, we will discuss an implementation framework – Foundational Learning Improvement Program (FLIP) - that integrates lessons from BUILD, offering a practical synthesis of evidence and implementation in the context of resource constraints that LMICs are currently facing. AIR’s FLIP is an evidence-based program that works directly with governments in low- and middle-income countries to improve their capacity to implement foundational learning programs at the classroom and community level.

Author

• Pooja Nakamura, American Institutes for Research