Mother-Tongue Instruction and Student Achievement: A Secondary Data Synthesis of Longitudinal and Global Evidence

Mother-Tongue Instruction and Student Achievement: A Secondary Data Synthesis of Longitudinal and Global Evidence

1. Introduction

The question “Does teaching children in their mother tongue improve student achievement?” has become central to education policy debates worldwide. Policymakers face tradeoffs between the short-term practicality of using national or global languages for instruction and the pedagogical advantages of early schooling in a child’s first language. A growing body of secondary-data evidence and large longitudinal studies suggests that mother-tongue-based instruction — particularly in foundational years — is associated with improved learning outcomes, reduced repetition and dropout rates, and stronger long-term academic attainment (World Bank, 2026; UNESCO, 2026). (World Bank)

This paper assembles and analyzes secondary data and published findings to (a) summarize the theoretical basis for why mother-tongue instruction affects achievement, (b) synthesize empirical evidence from large studies and international reports, (c) propose an analytic framework for secondary-data work on the topic, and (d) provide policy recommendations informed by the evidence.

2. Theoretical background

Two interlinked theoretical ideas undergird the argument for mother-tongue instruction.

2.1 Linguistic interdependence and the threshold hypothesis. Cummins’s theory of linguistic interdependence posits that cognitive and academic skills transfer across languages: proficiency developed in the first language (L1) facilitates second-language (L2) learning and higher cognitive functions (Cummins, 1979). The “threshold hypothesis” further suggests that a minimum competence in both languages is necessary before bilingualism confers cognitive advantages; poorly developed competence in both languages can be detrimental. This theory explains why a strong foundation in L1 supports later achievement in other languages and subjects. (coli.uni-saarland.de)

2.2 Pathways to literacy and content learning. Early literacy and numeracy are mediated by comprehension of instruction. When instruction uses the language the child understands best, learners acquire foundational skills faster, which then generalize to higher-order learning (World Bank, 2005). Delays in comprehension create cumulative learning deficits, contributing to “learning poverty” where children fail to read with comprehension by grade 3 (World Bank, 2021). (openknowledge.worldbank.org)

3. Evidence from secondary data and large-scale studies

This section synthesizes high-impact secondary data sources and large longitudinal research that have shaped current consensus.

3.1 Longitudinal evidence from U.S. school systems (Thomas & Collier). Thomas and Collier’s multi-year, large-sample studies of language-minority students in U.S. public schools demonstrate that students in well-implemented bilingual or dual-language programs outperform peers in English-immersion programs in long-term academic achievement (Thomas & Collier, 1997; 2002). Their analyses, based on extensive administrative records and achievement data, show cumulative academic gains for students whose early schooling included instruction in their home language alongside systematic development of English. (Squarespace)

3.2 Global syntheses and policy reports (World Bank, UNESCO). The World Bank’s policy synthesis on language of instruction highlights that poor language choices in early education are a key contributor to low learning outcomes and that mother-tongue instruction can reduce learning poverty when accompanied by quality teaching materials and teacher preparation (World Bank, 2021). UNESCO’s technical briefs and campaigns similarly advocate mother-tongue and multilingual education as central to inclusion and quality learning, emphasizing evidence that early L1 instruction improves comprehension, attendance, and progression. (World Bank)

3.3 Research from Africa, Asia, and other contexts. Country studies and regional research — for example, Heugh’s reviews on Sub-Saharan Africa and policy analyses of mother-tongue programs — document improvements in early literacy, reduced repetition, and better conceptual understanding when children begin learning in their home language (Heugh, 2006; World Bank case studies). These findings hold in contexts where curricular resources and teacher training support L1 instruction and provide a planned transition to national or international languages. (unesdoc.unesco.org)

Synthesis of empirical pattern. Across multiple contexts, the empirical pattern is consistent: (1) mother-tongue instruction accelerates early literacy and numeracy acquisition; (2) it reduces dropouts and repetition; and (3) with an appropriate bilingual transition strategy, it does not compromise later L2 proficiency — and often enhances it over the long run.

4. Methodological approach for secondary-data analysis

This study constructs a meta-analytic synthesis using secondary sources (published reports, longitudinal datasets, and peer-reviewed studies). The analytic steps for researchers wanting to replicate or extend this work are:

1. Define inclusion criteria. Include longitudinal or large cross-sectional studies comparing cohorts receiving L1-based instruction versus L2-immersion or mixed approaches; include randomized controlled trials (where available), quasi-experimental designs, and national program evaluations.

2. Data sources. Use administrative assessment datasets (e.g., large national early grade reading assessments), international databases (World Bank Education Statistics, UIS), and program evaluations (UNICEF, UNESCO, World Bank project reports). When raw datasets are unavailable, extract standardized effect sizes or achievement differentials from published reports.

3. Outcomes of interest. Primary: early grade reading with comprehension, basic numeracy. Secondary: progression rates, repetition, dropout rates, later standardized test scores, and L2 proficiency.

4. Analytic methods. Where raw microdata exist, apply hierarchical linear models (HLM) to account for student, classroom, and school effects. For aggregated reported effects, compute meta-analytic effect sizes (Hedges’ g) and assess heterogeneity (I²). For policy synthesis, conduct subgroup analyses by program design (mother-tongue only, mother-tongue with transitional bilingual model, and dual-language immersion), region, and availability of L1 resources.

5. Bias and robustness checks. Use risk-of-bias assessment for non-randomized studies (e.g., ROBINS-I approach), sensitivity analyses excluding high-risk studies, and publication-bias diagnostics (Egger’s test, funnel plots).

5. Findings

Because this paper is based on secondary data, the findings summarize consistent patterns reported across high-quality studies and international reports:

• Effect on foundational literacy: Programs that begin instruction in L1 show higher early grade reading comprehension and faster acquisition of decoding and oral language skills than immediate L2–medium cohorts (Thomas & Collier, 2002; World Bank, 2021). (Squarespace)

• Retention and progression: Mother-tongue provision correlates with lower repetition and dropout rates in primary education, improving the likelihood of completing basic education (World Bank, 2005; UNESCO, 2022). (openknowledge.worldbank.org)

• Long-term academic attainment and L2 outcomes: Carefully structured bilingual/dual-language programs that include systematic L2 development produce equal or better L2 proficiency in later grades compared with students who experienced early immersion, while also maintaining stronger content learning (Thomas & Collier, 2002). (Squarespace)

• Context and implementation matter: The benefits are conditional on implementation quality — availability of L1 teaching materials, teacher training in multilingual pedagogy, community engagement, and a planned transition model to L2. Where these are missing, mother-tongue instruction alone is insufficient. (World Bank)

6. Policy and practice implications

Based on the evidence, the following actionable recommendations emerge for policymakers:

1. Adopt mother-tongue-first strategies in early grades. Begin foundational literacy and numeracy instruction in the child’s home language while simultaneously developing L2 as a subject (not the medium) until early literacy benchmarks are achieved.

2. Invest in L1 materials and teacher preparation. Allocate resources to create curricula, textbooks, and teacher guides in local languages and to train teachers in multilingual pedagogy.

3. Design explicit transition pathways. Implement planned, evidence-based models for gradual L2 medium integration (transitional bilingual programs or dual-language models) rather than abrupt switches.

4. Monitor learning with language-sensitive assessments. Use early grade assessments in L1 and L2 to track comprehension and skill acquisition, enabling program adjustments.

5. Contextualize policy. Tailor mother-tongue approaches to local linguistic ecologies and labor-market considerations; ensure flexibility where languages of wider communication are essential for later economic opportunities.

International organizations (e.g., UNESCO, World Bank) recommend embedding such strategies in national education plans and funding teacher development accordingly. (UNESCO)

7. Limitations

This paper synthesizes secondary evidence and therefore cannot provide new causal estimates beyond what existing studies report. Heterogeneity in program designs, measurement tools, and contextual factors complicates meta-analytic aggregation. Some regions lack large, high-quality longitudinal data, which constrains generalizability. Finally, political and sociolinguistic dynamics (e.g., language prestige, migration) influence implementation and outcomes in ways that secondary data may not fully capture.

8. Conclusion

Secondary evidence from longitudinal studies and international policy reports shows that mother-tongue instruction in early schooling substantially benefits student achievement when implemented with complementary supports (materials, teacher training, and planned transition to additional languages). Policymakers aiming to reduce learning poverty and improve foundational skills should consider mother-tongue-first strategies as an evidence-based component of education reform, combined with robust monitoring and context-sensitive program design.

References

Cummins, J. (1979). Linguistic interdependence and the educational development of bilingual children. Review of Educational Research, 49(2), 222–251. (coli.uni-saarland.de)

Heugh, K. (2006). Mother tongue and bilingual education: Policy and practice in southern Africa. In UNESCO (Ed.), Mother tongue and bilingual education (INSIGHTS series). UNESCO. (unesdoc.unesco.org)

Thomas, W. P., & Collier, V. P. (1997). School effectiveness for language minority students. National Clearinghouse for Bilingual Education. (Summary of longitudinal research). (ncela.ed.gov)

Thomas, W. P., & Collier, V. P. (2002). A national study of school effectiveness for language-minority students’ long-term academic achievement. (Final report). Center for Research on Education, Diversity & Excellence. (Squarespace)

World Bank. (2005). In their own language…Education for All. World Bank Brief / Education Notes. (Discusses multilingual education and outcomes). (openknowledge.worldbank.org)

World Bank. (2021). Effective language of instruction policies for learning. Crawford, M. (ed.). World Bank. (Policy brief summarizing evidence on language of instruction and learning poverty). (World Bank)

World Bank. (2021, July 14). Teaching young children in the language they speak at home is essential to eliminate learning poverty [Press release]. (World Bank)

UNESCO. (2022). Why mother language-based education is essential. UNESCO Education Article. Retrieved from UNESCO website. (UNESCO)

World Bank. (n.d.). Mother Tongue as Bridge Language of Instruction (case studies and guidance). World Bank Education. (World Bank)