Fourteen-year-old Bhavani (name changed) had already learned to survive in ways no child should. Living in Chennai’s Perumbakkam, she had dropped out of school and spent her days in an environment shaped by neglect and abuse. For her, begging was not a choice but an expectation, her routine, and the streets became a place where she was exposed to substances she was allegedly forced to peddle by her own family.
(Sign up for THEdge, The Hindu’s weekly education newsletter.)
When she was finally identified and brought before the Child Welfare Committee, Tamil Nadu’s Department of Children Welfare and Special Services, it appeared to be the beginning of a different life.
Bhavani was sent to a Child Care Institution, where she was expected to receive protection and rehabilitation, and the support needed to rebuild her childhood. However, citing behavioural concerns, including incidents of self-harm, she was restored to the very home from which she had been rescued.
The Information and Resource Centre for the Deprived Urban Communities (IRCDUC), which had referred Bhavani’s case to the concerned authorities, learnt of her return to the same environment only during a subsequent field visit and not a follow up. By then, there had been little communication with the organisation that first identified her. For Bhavani, the rescue did not translate into long-term protection or a pathway back to education.
Her story raises a question confronting India’s child protection as well as our education systems: how many children like Bhavani remain invisible? Children who are out of schools living on the streets, migrating with their families, or trapped in abusive environments.
Without updated demographic data, governments lack a comprehensive picture of children who are outside the education system. They cannot accurately determine how many are out of school, where they live, which communities they belong to, whether they have never enrolled or have dropped out, the reasons behind their exclusion, or whether they have eventually returned to school. This makes it more difficult to identify vulnerable children who have slipped through the cracks, like Bhavani.
The cost of missing data
K. Narayanan Unni, a retired officer of the Indian Statistical Service and former Deputy Registrar General of India who served on the advisory committees for the 2011 Census and subsequent census cycles, explained that while the Census cannot directly determine number of children who are out of school due to confidentiality issues, it remains one of the most important tools for identifying children who are “likely” to be outside the education system or number schools needed in an area.
Without an updated enumeration of the school-age population across local administrative units, governments face constraints in estimating the needs for educational infrastructure, teacher allocation, transport, hostels, scholarships and other targeted interventions/policies.
The Census data provides a demographic baseline with the information on a children, including attendance at an educational institution and the highest level of education attained. When analysed with age and geographic location it identifies the number of school-age children who are currently outside the education system, and supports evidence-based planning at the village, ward, district and state levels – an analytical capability that sample-based education surveys are often unable to provide.
Mr. Unni explains that the Census also provides the population denominator used to calculate key education indicators such as the Gross Enrolment Ratio (GER) and Net Enrolment Ratio (NER). Without an updated population baseline, these indicators become less reliable, making it difficult to accurately assess enrolment, monitor progress towards universal education, or evaluate whether government policies are reaching all eligible children.
The last full Census was conducted in 2011. Fifteen years later, India’s education system is still planning for lakhs of children; In the absence of current demographic data, governments have increasingly relied on population projections rather than actual counts to design and implement education policies.
Rajendran Narayanan, Associate Professor at Azim Premji University, Bengaluru, whose work primarily focuses on bridging the gap between statistical data and social justice, particularly regarding welfare policies in India, explained that the sampling frame or the basis on which any national survey collecting information on education is the Census. Owing to delays in Census, our national survey estimates would also be less reliable. Moreover, while surveys provide an aggregate picture of education related metrics at district or state level, for small area estimations like educational metrics at a village or ward level, only Census will give an accurate picture. Timely census is needed to plan, allocate and arrive at better facilities for all, but more importantly for children of historically marginalised sections like Dalits and Adivasis. Scholarships, Direct Benefit Transfer (DBT) schemes, hostel facilities, residential schools and other targeted interventions are designed around estimates of the eligible population which the Census can give.
However, over the past 15 years, migration, urbanisation and changing settlement patterns have significantly altered the distribution of these communities. Without updated Census data, governments risk planning and allocating resources based on outdated population estimates.
Mr. Unni informed that constitutional guarantees for linguistic minorities can only be meaningfully implemented when backed by reliable census data. While Article 30 gives linguistic and religious minorities the right to establish and administer educational institutions of their choice, and Article 350A places a duty on states to provide adequate facilities for primary education in the mother tongue of children belonging to linguistic minority groups, ensuring these rights are upheld requires accurate information on the distribution of mother tongues. He pointed out that over the past two decades, large-scale migration from northern and eastern India to southern States has significantly altered the linguistic composition of many regions. Whether the number of children belonging to linguistic minorities is sufficient to justify the establishment of mother tongue schools or the provision of such educational facilities can only be determined through up-to-date census data.
The data gap also undermines efforts to improve girls’ education across every stage of schooling. Policies aimed at ensuring girls’ enrolment, regular attendance, retention, access to healthcare and nutrition, reducing dropout rates, preventing child marriage, and supporting their transition from primary to upper primary and secondary education all depend on accurate demographic information. Governments need reliable data on how many girls live in a particular area, their age distribution, from pre-school children requiring Anganwadi services to those in primary and secondary school, and whether they are enrolled and attending school regularly.
Without updated population data, it becomes difficult to identify girls who have never entered the education system, have dropped out, or are at risk of being left behind, limiting the effectiveness of targeted interventions.
Updated demographic data is also critical for planning gender-responsive infrastructure such as hostels, separate sanitation facilities, transport support and menstrual hygiene initiatives. In the absence of a current population baseline, governments may find it difficult to assess whether these interventions are reaching all eligible students, particularly those living in remote, migrant or socio-economically vulnerable communities.
Infrastructure planning has also become increasingly difficult. Decisions on where to build new schools, or shut down, expand classrooms, appoint teachers, provide transport, or allocate funds for midday meals and other support services are being made without an accurate understanding of the current school-age population. As a result, some schools may remain understaffed and overcrowded, while others continue to receive resources disproportionate to their actual enrolment, particularly in regions transformed by rapid migration and urban expansion.
The challenge is equally significant for policy implementation. Ambitious reforms under the National Education Policy (NEP) 2020 require precise demographic data to set realis evaluate outcomes. Without a current Census, governments face limitations in measuring whether policies are reaching intended beneficiaries or responding to changing educational needs.
Alternative data
National Sample Survey (NSS), the Annual Status of Education Report (ASER), and the Unified District Information System for Education Plus (UDISE+) offer valuable snapshots of enrolment, dropout, retention, learning outcomes and school infrastructure, helping policymakers fill the information gap.
But these datasets are no substitute for a Census.
Most are sample surveys that cannot provide a complete count of the population or capture demographic changes at the village, ward or neighbourhood level. These reports and surveys are limited in their ability to track migration, reasons of migration, identify every out-of-school child, estimate future demand for schools, teachers and Anganwadi staff, centres, or provide detailed cross-sectional information on age, gender, caste, disability and socio-economic status simultaneously.
While alternative datasets remain indispensable, they must be complemented by stronger local data systems. Suggested measures can include continuous tracking of school-age children through technology-enabled databases, integration of education records with local administrative and Anganwadi data, periodic household-level verification, and regular updates by local bodies.
Such systems can improve monitoring between Censuses, but they cannot fully replace the comprehensive demographic picture that only a national Census provides. Without that foundation, education planning risks remaining reactive rather than evidence-based, leaving the most vulnerable children least visible in the system.
Policies in place
Five years after the National Education Policy (NEP) 2020 was unveiled, governments have rolled out a range of reforms—from introducing the 5+3+3+4 curricular structure and strengthening early childhood education to expanding foundational literacy initiatives under NIPUN Bharat and implementing the Samagra Shiksha scheme. Progress is monitored through administrative datasets such as UDISE+, the PRABANDH portal, National Achievement Survey (NAS), PARAKH, and periodic reviews conducted by the Ministry of Education. These systems provide valuable information on enrolment, infrastructure, teacher deployment, learning outcomes and financial utilisation. However, they are ultimately built upon population estimates that continue to rely on the 2011 Census.
This creates a fundamental challenge in assessing whether the policy is actually meeting its objectives. Many of NEP 2020’s goals, including universal access to education, reducing dropout rates, improving Gross Enrolment Ratio (GER), expanding Early Childhood Care and Education (ECCE), and ensuring equitable access for disadvantaged groups, are measured against the size and characteristics of the school-age population.
Without updated demographic data, the denominator itself becomes uncertain. While enrolment numbers may be available through administrative records, it is increasingly difficult to determine whether they represent all eligible children or only those already within the education system.
Administrative databases can record the children who are enrolled in schools, but they are less equipped to identify children who have never entered, have dropped out, or have migrated with their families. As a result, governments may continue planning schools, recruiting teachers and allocating funds using population projections rather than an accurate, current count of children across age groups and locations.
The same challenge extends to Samagra Shiksha, India’s flagship school education programme. Although the scheme tracks expenditure, infrastructure and service delivery, the absence of an updated demographic baseline limits its ability to align resource allocation with actual need. Consequently, decisions on teacher deployment, school infrastructure, midday meals and other interventions continue to be guided by population projections rather than current demographic realities.
Children like Bhavani, who exist on the margins of official records as much as they do on the margins of society, expose the cost of data invisibility. Had governments possessed updated demographic information, the gaps between population estimates and actual beneficiaries would have been far more visible.
Such data cannot prevent abuse or exploitation on its own, but it can serve as an early warning system, enabling authorities to identify children who have dropped out from classrooms, strengthen enrollment/retention outreach, and intervene before they slip further through the cracks like Bhavani.
The Hindu attempted to contact Bhavani’s mother through IRCDUC, a Tamil Nadu-based non-governmental organisation working for children, women and marginalised families in need of protection, education, housing, and social inclusion while facilitating access to essential services and government entitlements. However, the mother declined to speak, fearing her previous interactions with and in connection with child protection interventions. IRCDUC has documented several similar cases across Chennai and its surrounding areas involving children who are out of school.
(People seeking assistance for a child in need of care and protection can call the toll-free Childline helpline, 1098.)
Remove