This article is adapted from a blog post published on the Centre of Excellence for Development Impact and Learning (CEDIL) website. The original article contains references to other research papers, with links.
When schools teach pupils in their mother tongues, it helps their reading in their own languages and may also improve their reading in other languages. Pooja Nakamura, Adria Molotsky, Rosa Castra Zarzur, Varsha Ranjit, Yasmina Haddad and Thomas de Hoop discuss the findings of a CEDIL-funded systematic review that examines the impact of language of instruction policies on literacy outcomes.
A shocking 70% of all 10-year-old students in low- and middle-income countries (LMICs) cannot read a simple text. An average of 40% of those students are educated in languages they do not use or understand, and most of them will then need to transition to education in a new language of instruction.
There is now substantial evidence that mother tongue-based multilingual education (i.e., the child is taught in their “own” language first) has multiple benefits. At the same time, there is a well-established link between post-colonial languages and socioeconomic mobility, leading to high demand for earlier and earlier introduction to these languages in schools. Other factors that complicate language of instruction choices include linguistically heterogenous classrooms, lack of teaching and learning materials, limited trained teachers, and lack of political or community will.
This leads to a situation in which decisionmakers must reconcile the well-documented benefits of mother-tongue instruction and the quest for socioeconomic mobility through a post-colonial or international (later acquired) language at earlier grades.
We conducted a systematic review focused on reconciling evidence gaps by examining the impact of language of instruction choices – especially mother tongue education and the timing of transition to a new language of instruction – on primary school literacy outcomes in LMICs.
Specifically, we examined the effects mother tongue-based education policies and language of instruction transition policies on students’ literacy, biliteracy, and multilingual literacy skill development.
The review summarises evidence from 45 high-quality studies conducted in LMICs, primarily in Sub-Saharan Africa and South and Southeast Asia. The studies include education in a range of native languages, lingua francas like Kiswahili and Amharic, as well as later acquired post-colonial languages, such as English, French and Portuguese.
Mother tongue reading programmes lead to significant impacts on mother-tongue reading scores. There is also some evidence that mother tongue reading programmes have significant impacts on English (or later acquired language) reading scores. However, the results were not conclusive in terms of impact of mother tongue instruction on learning outcomes after the transition to a new language of instruction.
There is a lack of high-quality evidence on the trade-off between putting resources into high-quality English language of instruction versus high-quality mother tongue education and then transitioning to English language of instruction. There is also a large gap in the evidence determining when students should transition to a new language of instruction and how such decisions should differ by context.
Recommendations for next steps
Based on our systematic review, we make the following policy recommendations:
- Continue to invest in teaching students in their “own” languages first.
- Investigate with high-quality studies (that meet rigorous quality standards against a variety of biases) whether – and if so, in what ways and why – mother tongue education programmes impact literacy and other academic outcomes in not only the mother tongue, but also in the later-acquired languages.
- Conduct cost-effectiveness analyses to help provide decisionmakers with the trade-offs that are inherent in determining how much time and resources should be spent teaching in one language versus another.
Our review contributes to the mounting evidence that the initial language of instruction must be in a language children understand best. However, there is still a dire lack of evidence on how best to transition from one language of instruction to a new one – even though most students across LMICs are being educated in multilingual environments in which transition will occur. Our team is examining several dimensions of language of instruction transitions across seven countries, and we plan to update this systematic review as new evidence comes to light on language of instruction transition.
This blog post relates to a systematic review published in Campbell Systematic Reviews.