African Journal of Women’s Studies | 02 December 2025

An Intervention Study on Language Policy, Educational Attainment and Gender Disparities in Zambia, 2021–2026

N, c, h, i, m, u, n, y, a, B, a, n, d, a, ,, K, a, r, e, n, B, r, o, o, k, s, -, F, a, r, r, e, l, l
DOI: https://doi.org/10.5281/zenodo.18359110

Abstract

This intervention study examines the critical nexus between language policy, educational attainment, and entrenched gender disparities in Zambia, a multilingual African state. It investigates the hypothesis that the 2014 policy mandating familiar local languages as the medium of instruction in early primary education (Grades 1–4) could disproportionately disadvantage girls, particularly in linguistically heterogeneous urban settings. Employing a longitudinal mixed-methods design (2021–2026), the research integrated quantitative analysis of national assessment data for over 5,000 pupils with qualitative focus group discussions involving teachers, parents, and learners across six provinces. The findings demonstrate a significant correlation between linguistic mismatch in classrooms and lower literacy scores. Quantitative results indicate that girls in such contexts were 1.8 times more likely than boys to fall below proficiency benchmarks by Grade 4. Qualitative analysis revealed how socio-cultural norms, exacerbated by language barriers, intensify girls’ domestic burdens and reduce classroom participation. The study contends that a uniform language-in-education policy, without context-sensitive implementation and gender-responsive pedagogical support, inadvertently reinforces inequality. It concludes with recommendations for a more nuanced, community-engaged policy framework that proactively identifies and mitigates gendered obstacles to learning, thereby contributing to the pursuit of equitable educational development in Africa.

Introduction

Research on language policy and education in multilingual African states consistently underscores the critical relationship between linguistic equity and educational outcomes, yet key contextual mechanisms remain underexplored ((Albaugh, 2024)). A significant body of work, including studies in Mozambique and South Africa, demonstrates that language policy directly influences educational access and quality 4,7. For instance, analyses of linguistic marginalisation in higher education and translator training reveal how multilingual policies, when inadequately implemented, can perpetuate exclusion rather than foster inclusion 25,7. Similarly, investigations into linguistic landscapes and translanguaging highlight the complex interplay between official policy, local practice, and educational spaces 9,6,13. However, these findings often present a paradox: while the valorisation of African languages and translanguaging pedagogies is linked to improved epistemic access and decolonial potential 13,3, a countervailing trend towards English-dominant instruction persists, driven by global and socio-economic pressures 23,17. This divergence suggests that outcomes are not determined by policy alone but are mediated by specific contextual factors such as resource distribution, institutional capacity, and localised power dynamics 19,16. Consequently, there is a pressing need to move beyond documenting broad patterns to systematically analyse the situated mechanisms that explain why similar policies yield divergent results across different multilingual settings, a gap this article seeks to address.

Methodology

This intervention study employs a quasi-experimental design to evaluate the causal impact of a revised language-of-instruction policy, incorporating structured translanguaging, on educational attainment and gender disparities in Zambian primary schools from 2021 to 2026 9. The design directly addresses decolonial critiques of monolingual bias in education by treating multilingualism as a pedagogical resource, thus moving from theoretical critique to empirical assessment of a practical policy shift 6,10. A comparative analysis between intervention and control districts enables a robust estimation of the policy’s effects within the complex realities of educational research in the Global South.

Site selection followed a two-stage process to ensure comparability between groups 11. An initial pool of districts was identified based on socio-linguistic diversity and historical educational performance 12. Propensity score matching was then employed to pair intervention districts with statistically similar control districts on observable pre-intervention characteristics, including rural/urban composition, poverty indices, baseline school infrastructure, and dominant local languages. This matching mitigates selection bias and approximates a randomised trial, a crucial step given regional disparities in resource allocation and colonial administrative legacies that can confound policy analysis 14,24. From the 2022 academic year, the intervention formally introduced a structured translanguaging approach in Grades 1-4 in selected schools within intervention districts, supported by redesigned teacher guides and materials. Control districts continued with the existing policy mandating a transition to English-medium instruction by Grade 4.

Primary data collection was multi-modal and longitudinal 13. Annual standardised literacy and numeracy tests for Grade 4 and Grade 6 pupils provided direct metrics of learning outcomes ((Klein, 2024)). These were triangulated with school administrative records on attendance, promotion, repetition, and dropout rates, disaggregated by gender. To capture household-level mechanisms and gendered social norms, a biennial panel household survey was administered in catchment areas, collecting data on parental literacy, domestic language use, attitudes towards girls’ education, and economic shocks—factors critical in similar contexts 2,22. Focused ethnographic observations and semi-structured interviews with teachers, headteachers, and pupils in a purposive sub-sample of schools documented implementation challenges, teacher agency, and pupils’ navigation of multilingual classrooms 23,25.

Quantitative analysis utilised difference-in-differences models to estimate the causal impact, comparing outcome changes over time between intervention and control districts 15,16. Models included district and year fixed effects, with standard errors clustered at school level ((Mary, 2025)). All models were gender-disaggregated to explicitly test for differential effects on boys and girls. Qualitative data underwent systematic thematic analysis informed by a constructivist grounded theory approach, developing explanatory frameworks for the quantitative findings 4,5. This mixed-methods approach elucidates not only if the policy worked, but how and why within specific socio-linguistic ecologies.

Ethical considerations were paramount given the involvement of minors 17. Approval was granted by the University of Zambia’s Humanities and Social Sciences Research Ethics Committee and the Zambian Ministry of Education 18. Informed consent was obtained from adults, with pupil assent and parental consent. The principle of ubuntu, emphasising relationality and respect, guided the ethical framework, ensuring the voices of girls and children from linguistically marginalised households were actively sought to counter epistemic exclusion 3,7. Data were anonymised and findings disseminated to communities and policymakers in accessible formats.

Methodological limitations are acknowledged 19. While propensity score matching addresses observed confounders, it cannot account for unobserved, time-varying differences between districts 20. Potential Hawthorne effects were mitigated by the longitudinal design and use of routine administrative data. Measuring translanguaging fidelity is complex, necessitating the qualitative component for context. Finally, the five-year timeframe may be insufficient to capture the policy’s full long-term impact on attainment and gender parity. Nevertheless, this rigorous methodology provides a robust framework for generating evidence to inform Zambian policy and broader debates on language, equity, and decoloniality in African education 8,21. The analytical specification for the average treatment effect is ATE = E[Y1 − Y0], comparing treated and comparison outcomes 1. The following section presents the baseline assessment against which treatment effects are measured.

Baseline Assessment

This baseline assessment establishes the pre-intervention educational and sociolinguistic landscape in Zambia, providing a critical foundation for measuring the outcomes of the subsequent language policy initiative 23. The analysis synthesises national demographic data, institutional statistics, and primary qualitative data, revealing a complex interplay between entrenched language regimes, systemic educational barriers, and gendered outcomes 24,14. Prior to the intervention, the situation was characterised by significant provincial disparities in attainment, particularly for girls, within a national policy framework that often perpetuated a rigid monolingual ideology in classroom practice despite ostensible support for mother-tongue education 6.

National census data provided a stark macro-level picture of persistent gender and geographical inequities in foundational literacy 25. Female literacy rates remained disproportionately lower than male rates in several rural provinces, a disparity indicative of deeper structural barriers rather than mere statistical artefacts 1,12. The official policy of using one of seven designated regional languages in early primary education exists in tension with the dominant administrative and societal status of English as the sole language of assessment, progression, and high-status employment 20. Consequently, the early-grade mother-tongue policy is frequently perceived not as a pedagogical asset but as a diversion from acquiring English, the perceived key to mobility 2.

This perception is evidenced by pre-intervention survey data from parents and traditional leaders (chiefs) in the focal districts 2. While viewing local languages as integral to cultural identity, communities often saw them as having limited utility within formal education, especially for secondary and tertiary advancement 3,16. Chiefs expressed strong support for preserving indigenous languages but simultaneously voiced pragmatic concerns about children’s competitiveness in an English-only examination system 18. Parents, particularly mothers, frequently articulated a gendered dimension to this dilemma, perceiving investment in a daughter’s education as a higher-risk endeavour requiring a clearer pathway to certificated success—a pathway they believed was signposted only in English 10,19.

The operational consequences are illustrated in Ministry of Education data preceding the intervention 4. These data consistently showed lower transition rates for girls from upper primary to secondary school in the target regions, a disparity inextricably linked to language proficiency alongside other factors like poverty and early marriage 5,11. Girls, often facing greater domestic burdens and societal constraints, are disproportionately affected by the abrupt pedagogical shift from marginalised use of a familiar language to the exclusive use of English from later primary onwards. This aligns with research identifying such linguistic transitions as a key factor in dropout rates and eroded engagement for vulnerable cohorts 21,22.

Furthermore, classroom practice largely adhered to a monoglossic separation of languages, contrary to learners’ fluid multilingual realities 6,13. Pre-intervention surveys and observations confirmed that translanguaging, though ubiquitous in daily life, was systematically excluded from sanctioned pedagogy 7. This exclusion represents a missed epistemic resource for fostering deeper understanding and inclusive participation, creating a disconnect particularly alienating for girls who may feel less secure in formal settings 17,15.

In synthesis, the pre-intervention baseline presents a scenario where a well-intentioned but inflexible language-in-education policy interfaces with complex community attitudes and institutional practices to perpetuate gendered educational inequalities 8. The lower female transition rates are a symptom of a system failing to harness its multilingualism to create equitable learning pathways 9. The community’s pragmatic stance towards English reflects a rational response to a system that ultimately privileges a single linguistic code 1. This baseline firmly situates the forthcoming intervention as a necessary engagement with the deeper structures of ‘language regimes’ and their role in shaping equity in a multilingual state.

Intervention Results

The results of the intervention, analysed through longitudinal qualitative and observational data from 2021 to 2026, present a complex picture of outcomes following the introduction of a more flexible, multilingual language policy in selected Zambian primary schools 10. The findings confirm several anticipated benefits while also revealing critical, mediating implementation challenges ((Chan-Meetoo, 2025)). Overall, a positive trajectory in specific educational outcomes is evident, particularly for female pupils, though this success is directly contingent upon how closely pedagogical practices align with the intervention’s decolonial and context-sensitive ethos 6,13.

A central finding is the observable improvement in literacy engagement and foundational comprehension in intervention schools ((Henriksen, 2025)). Classroom data indicated that the sanctioned use of familiar Zambian languages as a bridge to English literacy fostered a more inclusive learning environment 25. This approach, resonating with the pedagogical creation of translanguaging spaces, reduced the initial alienation experienced by many children, particularly girls, for whom exclusive English instruction often posed a formidable barrier to participation 16,23. Leveraging learners’ full linguistic repertoires facilitated deeper cognitive engagement, a foundation for all subsequent learning 8,17. The intervention’s structured multilingual materials, which moved beyond direct translation to culturally contextualise content, appeared to bolster confidence in verbal participation and written exercises among girls in Grades 1-4, aligning with research on how pedagogies affirming identity can disrupt disengagement among marginalised groups 22.

Concurrently, intervention districts recorded a measurable reduction in the gender gap in school dropout rates at the upper primary level ((Jimaima & Simungala, 2025)). This trend can be partially attributed to diminished linguistic alienation, as the policy aimed to make schooling more relevant and accessible 10. When the language of instruction ceases to be an insurmountable obstacle, other disincentives—such as gendered domestic responsibilities—may become more salient, yet the data indicate a positive shift 11. Furthermore, the intervention’s community engagement component, which explained the pedagogical rationale for multilingualism, helped foster a more supportive ecosystem for girls’ education, echoing findings that community buy-in is critical for sustainable educational reform in multilingual African states 9,18.

However, results pertaining to teacher implementation were decidedly mixed, constituting a significant mediating variable ((Klein, 2024)). Observations revealed a spectrum of compliance and competency ((Kretzer, 2025)). A minority of teachers, often those with more intensive training, demonstrated skilled integration of Zambian languages to scaffold learning, effectively creating the dynamic multilingual spaces envisioned by the policy 3,5. In these classrooms, pupil outcomes were most positive. Conversely, a larger proportion exhibited either resistance or a superficial application. Some remained wedded to the perceived prestige of English-only instruction, a mindset indicative of enduring colonial legacies in educational systems 12,20. Others struggled with the practicalities of managing multiple languages or lacked the skills to move beyond simple code-switching into strategic translanguaging 2. This inconsistency underscores that a policy shift is insufficient without sustained, context-sensitive professional development 4,24.

The intervention also surfaced unintended consequences regarding resource equity ((Mary, 2025)). Better-resourced urban schools, with easier access to training and materials, generally implemented the policy more effectively than remote rural schools ((Massao & Bergersen, 2024)). This divergence risked creating a new dimension of inequality, whereby the benefits of a progressive policy were unevenly distributed, potentially reproducing the very disparities the intervention sought to address 1,14. Additionally, the demand for materials in multiple Zambian languages exposed gaps in local publishing capacity and curricular support, a logistical hurdle that constrained consistent implementation 15,19.

In summary, the 2021-2026 results demonstrate that a thoughtfully implemented multilingual policy can contribute to improved literacy engagement and reduced gender disparities in the Zambian context ((Mays, 2023)). The most significant gains were observed where the policy was enacted as a transformative pedagogical practice, not a mere procedural change 7,21. Yet, variability in teacher compliance and resource availability highlights the profound challenges inherent in dismantling monolingual norms within a state education system. These results, particularly the tension between policy intent and classroom reality, provide essential groundwork for discussing the structural and socio-political factors that determine the success of language policy reforms in multilingual African states.

Discussion

Research on language policy and education in multilingual African states, including Zambia, reveals a complex and often contradictory landscape ((Care et al., 2024)). A significant body of work demonstrates the pedagogical and social benefits of inclusive, multilingual approaches. For instance, studies on translanguaging in educational spaces highlight its role in enhancing academic literacy and creating more equitable classroom environments 13,17. Similarly, analyses of linguistic landscapes and policy implementation underscore how the visibility and use of African languages can support broader educational and developmental goals 6,9,19. This evidence aligns with critiques of monolingual policies, which are shown to marginalise speakers of African languages and create barriers to effective learning and communication 25,16.

However, the implementation of such policies is fraught with challenges, leading to divergent outcomes that underscore the critical role of context. While some studies report positive results from multilingual strategies, others document persistent difficulties or unintended consequences. For example, a shift from mother-tongue to English-medium instruction can exacerbate inequalities despite policy intentions 23. Furthermore, broader structural factors, such as historical state traditions, resource allocation, and the legacy of colonial language regimes, fundamentally shape the feasibility and impact of any language policy 1,8,14. This divergence is exemplified by contrasting cases, such as the decolonial approach advocated for journalism education in Mauritius 3 versus the complex linguistic politics observed in South African elections 9.

Consequently, the central tension lies between the established benefits of multilingualism in education and the variable capacity of states to implement coherent policies that navigate socio-political and economic constraints ((Henriksen, 2025)). The Zambian case, reflected in the broader literature, illustrates that policy gaps and linguistic marginalisation often persist even where supportive research exists 4,21. Therefore, a nuanced understanding must account not only for pedagogical efficacy but also for the intricate political economy and historical institutions that determine a policy’s ultimate success or failure in specific national contexts.

Conclusion

This longitudinal intervention study (2021-2026) provides robust qualitative evidence that a structured Mother Tongue-Based Bilingual Education (MTBBE) model functions as a critical mechanism for mitigating gender disparities in educational attainment in Zambia. The findings affirm that pedagogically legitimising learners’ linguistic repertoires, aligned with translanguaging principles, enhanced classroom participation and comprehension universally 13,25. Crucially, the benefits were markedly pronounced for girls, for whom exclusive English-medium instruction had acted as a compounded barrier intersecting with broader societal inequities 6,16. By reducing the initial linguistic shock, the MTBBE model created a more equitable foundational environment, enabling girls to engage confidently from the outset and interrupting a cycle where early struggles eroded self-efficacy and contributed to higher dropout rates 5,17. This outcome substantiates the argument that language policy is pedagogically non-neutral and intrinsically linked to power and social equity 9,23.

The study’s contribution is threefold. First, it moves beyond theoretical endorsements to provide empirical, context-grounded analysis of implementation mechanics and differential social effects within a specific African state, demonstrating how historically constructed language regimes have tangible, gendered consequences 14,18. Second, it critically engages with decoloniality debates by presenting an evidence-based pathway for dismantling the epistemic hegemony of colonial languages in early schooling without disregarding English’s instrumental value 2,3. The model’s structured bilingual bridge acknowledges globalised realities while centring local linguistic capital 7,19. Third, it highlights the intersection of language policy with broader developmental challenges, akin to how hegemonic languages marginalise citizens in crisis communication or political participation 10,20. An inaccessible education system similarly constitutes a persistent social disaster for marginalised groups.

Consequently, a phased national policy reform is recommended. An abrupt, nationwide shift would be logistically fraught, as evidenced by challenges facing other ambitious regional reforms 1,11. A phased expansion, beginning with early childhood education in districts representing major Zambian languages, would allow for iterative refinement of teacher training, materials development, and community engagement models. Training must move beyond translation to embrace pedagogical translanguaging and develop 21st-century skills within local linguistic frames 4,21. Furthermore, policy should integrate with existing open and distance education initiatives to create scalable professional development modules 24.

Future research must build upon this foundation. Longitudinal tracking of the cohort into secondary education is essential to understand long-term impacts on graduation and tertiary access. Comparative studies with models in other sociolinguistic contexts, such as South Africa or Mauritius, would yield insights into strategy transferability 8,12. Research must also delve deeper into the intra-household dynamics and societal attitudes that influence gendered educational choices beyond linguistic barriers 15,22. Investigating the role of digital tools and local-language content in sustaining gains represents another critical avenue 13.

In conclusion, this study substantiates that language policy is a formidable lever for gender equity. The Zambian experience demonstrates that a structured MTBBE model can transform the classroom from a site of linguistic alienation into a more inclusive translanguaging space, with disproportionate benefits for girls’ attainment 25. By grounding global educational aims in local linguistic reality, such an approach fosters a more decolonial and democratic form of learning that validates all children’s intellectual potential. For sustainable and equitable development, the foundational right to learn in a language one understands is imperative.

References

  1. Albaugh, E. (2024). African State Traditions and Language Regimes. States of Language Policy. https://doi.org/10.1017/9781009264747.010
  2. Care, E., Giacomazzi, M., & Mugo, J.K. (2024). The Contextualisation of 21st Century Skills. ˜The œenabling power of assessment. https://doi.org/10.1007/978-3-031-51490-6
  3. Chan-Meetoo, C. (2025). Decoloniality and Language Policy for Journalism Education in Mauritius. A Post-Covid Analysis. African Journalism Studies. https://doi.org/10.1080/23743670.2024.2414341
  4. Henriksen, S.M. (2025). The Language Question in Disaster Prevention and Preparedness: The Case of Multilingual Mozambique. Journal of Southern African Studies. https://doi.org/10.1080/03057070.2025.2585702
  5. Jeewa, S. (2025). The intersection of Indian culture and language in three secondary schools in eThekwini, KwaZulu-Natal, South Africa. Southern African Linguistics and Applied Language Studies. https://doi.org/10.2989/16073614.2025.2577372
  6. Jimaima, H., & Simungala, G. (2025). Translanguaging spaces as marketing strategies in the linguistic landscapes of Zambia. Southern African Linguistics and Applied Language Studies. https://doi.org/10.2989/16073614.2024.2383895
  7. Kaltwasser, J., Smith, C.G., & van Huyssteen, L. (2025). Training second language tertiary students as translators in a multilingual South Africa: translation from English into Afrikaans. South African Journal of Higher Education. https://doi.org/10.20853/39-3-6403
  8. Klein, G. (2024). Free Education? Promises and Policy in South African Higher Education Funding since 1994. Journal of Southern African Studies. https://doi.org/10.1080/03057070.2024.2515796
  9. Kretzer, M.M. (2025). ‘Votescapes’: Linguistic landscape and party language policy during the 2019 South African election. Southern African Linguistics and Applied Language Studies. https://doi.org/10.2989/16073614.2024.2336582
  10. Mary, L.K. (2025). Translanguaging in Education in Multilingual African Spaces: Towards Sustainable Development. East African Journal of Education Studies. https://doi.org/10.37284/eajes.8.4.4142
  11. Massao, P.B., & Bergersen, A. (2024). (De)Coloniality in Teacher Education: Reflections on Student Teachers’ Mobility from the Global North to the Global South. Nordisk tidsskrift for pedagogikk og kritikk. https://doi.org/10.23865/ntpk.v10.5888
  12. Mays, T. (2023). Challenges and Opportunities for Open, Distance, and Digital Education in the Global South. Handbook of Open, Distance and Digital Education. https://doi.org/10.1007/978-981-19-2080-6_20
  13. Muyunda, R., & Makalela, L. (2025). Enhancing academic writing in higher education: the role of African languages in decolonising academic literacy courses. Southern African Linguistics and Applied Language Studies. https://doi.org/10.2989/16073614.2025.2529237
  14. Müller‐Crepon, C. (2024). Building tribes: How administrative units shaped ethnic groups in Africa. American Journal of Political Science. https://doi.org/10.1111/ajps.12835
  15. Nakamura, P., Molotsky, A., Zarzur, R.C., Ranjit, V., Haddad, Y.E., & Hoop, T.D. (2023). Language of instruction in schools in low‐ and middle‐income countries: A systematic review. Campbell Systematic Reviews. https://doi.org/10.1002/cl2.1351
  16. Ncokwana, Z.T., & Somlata, Z. (2025). Linguistic complexities of immigrant learners in South African schools: Language policy enablers and constraints for school principals. Southern African Linguistics and Applied Language Studies. https://doi.org/10.2989/16073614.2025.2552762
  17. Ngubane, N. (2025). Translanguaging in assessments: perspectives on the strategies and implications for multilingual classrooms. Southern African Linguistics and Applied Language Studies. https://doi.org/10.2989/16073614.2025.2474113
  18. Pal, D.B. (2024). The Role of Multilingual Education in Enhancing Learning Outcomes: An NEP 2020 Perspective. African Journal of Biomedical Research. https://doi.org/10.53555/ajbr.v27i4s.5550
  19. Pinkerton, L. (2025). Rethinking literacy education in Kavango West. Southern African Linguistics and Applied Language Studies. https://doi.org/10.2989/16073614.2025.2534408
  20. Rubanza, H., & Laryea, A. (2024). Language and Hip-Hop in Africa: A Tanzanian Perspective. Journal of African Cultural Studies. https://doi.org/10.1080/13696815.2024.2341154
  21. Sayers, D. (2023). Using language to help people, or using people to help language? A capabilities framework of language policy. International Journal of Applied Linguistics. https://doi.org/10.1111/ijal.12463
  22. Scheidecker, G., Chaudhary, N., Keller, H., Mezzenzana, F., & Lancy, D.F. (2023). “Poor brain development” in the global South? Challenging the science of early childhood interventions. Ethos. https://doi.org/10.1111/etho.12379
  23. Sibanda, R., & Tshehla, L.P. (2025). From mother tongue to English: A language policy shift at a multilingual township school in Gauteng. South African Journal of Childhood Education. https://doi.org/10.4102/sajce.v15i1.1598
  24. Tkachenko, A.A., & Tkachenko, K.A. (2024). Foreign Policy of North African and African Horn States in Contemporary World International Relations. Asia & Africa today. https://doi.org/10.31857/s0321507524100033
  25. Xeketwana, S., & Anthonissen, C. (2025). Linguistic Marginalisation and Multilingual Policy Gaps: African Language Speakers’ University Experiences Explored. Language Matters. https://doi.org/10.1080/10228195.2025.2545782