3 min readJul 14, 2026 07:30 PM IST
First published on: Jul 14, 2026 at 06:27 PM IST
Written by Gowd Kiran Kumar
Recent news reports indicate that the rehearsal for the second phase of Census 2027 includes an “open column” for recording caste. While the government has clarified that the final methodology will be decided after evaluating feedback from the test exercise, experience from previous surveys suggests that relying solely on an open-text response without a standardised coding framework can undermine the objective of caste enumeration.
The experience of the Socio-Economic and Caste Census (SECC), 2011 is a cautionary example. The use of an open column resulted in over 46 lakh different caste entries, largely due to spelling variations, regional names, sub-castes, synonyms, and inconsistent reporting. As a result, the data could not be effectively classified or officially released for policy purposes.
This stands in contrast to earlier and more structured exercises. The 1931 Census recorded approximately 4,147 castes, while the Anthropological Survey of India identified around 6,325 castes in 1951. More recently, states that adopted a standardised coding methodology have shown that accurate enumeration is feasible and reliable.
The Bihar Caste Survey (2023) identified 209 castes, including 112 castes in the recognised State list of Extremely Backward Classes (EBCs), 30 castes under Other Backward Classes (OBCs), 22 Scheduled Castes (SCs), 32 Scheduled Tribes (STs), 7 Upper Castes, and 2 Other Related Castes (ORCs). The Telangana Socio-Economic, Educational, Employment, Political and Caste Survey, 2024 identified 242 castes, comprising 133 Backward Classes, 59 Scheduled Castes, 32 Scheduled Tribes, and 18 Other Castes. Both relied on officially recognised caste lists and assigned standardised codes to each caste, ensuring consistency and minimising duplication.
As the Union government finalises the methodology for Census 2027, it must constitute an expert committee to determine the authoritative source for caste coding. The committee should examine what the coding framework should be based on: State-wise lists of SCs, STs, and OBCs;
the central list of OBCs maintained by the National Commission for Backward Classes; the caste database maintained by the Anthropological Survey of India; or a harmonised national master list integrating these while accommodating state-specific variations.
The identification of Upper Castes also requires a clear framework. Since EWS certificates are issued to persons belonging to communities not included in the SC, ST, or OBC categories, the government already has an administrative basis for identifying such communities. This can be refined and incorporated into the Census coding methodology.
An open column should remain available only to capture genuine omissions, new caste names, or regional variations. However, every enumerator should also have access to a standardised coded list so that recognised castes are recorded uniformly.
The Modi government has a historic opportunity to conduct India’s first comprehensive caste enumeration after Independence as part of the regular Census. This should not be compromised by repeating the methodological shortcomings of SECC-2011.
A well-designed caste census is not merely about counting communities; it is about generating accurate data to identify deprivation, measure inequality, and formulate targeted public policies for communities that continue to remain socially and educationally backward. The success of Census 2027 will depend on the government’s ability to classify every recorded caste through a transparent, standardised, and nationally accepted coding system.
The writer is national president, All India OBC Students Association
