Impacts of Inequities in Access to Language of Education on Educational Achievement: Evidence from India

Children learning India

The World Bank recently estimated that 70% of students across low- and middle-income countries (LMICs) have been impacted by extended school closures and learning loss due to the COVID-19 pandemic, potentially costing theses students $17 trillion in lifetime earnings in present value.

This learning crisis is exacerbated by language of instruction issues. An estimated 37% of students in LMICs are not taught in a language they best understand. An even larger share of students in LMIC is unprepared to transition to a second language during their primary- or middle-school years, leading to deep and broad structural inequities in learning.

India is a multilingual country and has made progress in developing strong mother tongue (MT) education programming. Research shows that MT-initial instruction increases attendance; improves cost-effectiveness; increases the likelihood of girls and minorities staying in schools; and leads to more effective learning, at least through middle school. Despite this evidence and the leaps in advancing MT education in India, societal pressure to transition to English instruction in earlier and earlier grades is strong, especially given the well-documented link between English and upward socioeconomic mobility.

Our project aims to study these language-driven inequities in India’s education system by examining whether—and to what extent—inequitable access to education in languages that primary grade children use impacts their academic performance.

While much is known about the importance of learning in a child’s own language, much less information is available about the extent to which structural inequities exist in access to education in a child’s own language in India, and how these inequities impact learning.

The main goal of our research is to estimate the extent to which access to education in the ‘right’ language predicts educational achievement. Our secondary hypothesis is that a match between medium of instruction and MT in lower grades (i.e., Grade 3), leads to higher achievement scores in higher grades (i.e., Grades 5 and 8). We will use quantitative and qualitative data to answer these questions. The quantitative approach will primarily focus on analysis using secondary data from the India Human Development Survey (IHDS) conducted in 2005 and 2012, and the Annual State of Education Report (ASER) conducted in 2018 and 2022. IHDS is a panel survey, covering over 42,152 households in 1,420 villages and 1,042 urban neighborhoods across India. IHDS administers the assessment developed by ASER to children that are part of the survey. ASER is an annual survey that aims to provide reliable annual estimates of children’s schooling status and basic learning levels for each state and rural district in India. ASER has been conducted every year since 2005 in almost all rural districts of India. ASER tests children in the age group 5-16 in basic reading and basic arithmetic. ASER 2018 and 2022 surveyed close to 550,000 children in each wave.

We aim to (a) generate evidence on the extent to which structural inequities in language access influences educational outcomes in India; (b) strengthen key stakeholders’ capacity to apply evidence to intentionally plan how to overcome language-driven inequities in education; and (c) connect the findings (and their implications) to other similar bilingual and multilingual education contexts.

Uttara Headshot
Senior Research Economist
Mauricio Estrada Matute headshot
Senior Researcher