Former Civil Servants Urge Transparency With Census 2027

New Delhi: The Constitutional Conduct Group which is made up for former civil servants has written to the Registrar General and Census Commissioner voicing concerns about the delayed 2027 Census.

While acknowledging the COVID-19 disruption, the signatories have sought transparency, warning against apprehension that it aligns with constituency delimitation before the 2029 elections.

The former IAS and other services' officers have urged adherence to UN guidelines, stressed data quality over technological fixes, and recommended streamlining questions. The group has called for clarity on caste enumeration, proper classification of OBCs and tribes, and careful handling of sensitive religious data.

The full letter is below.

§

CCG LETTER TO THE REGISTRAR GENERAL

AND CENSUS COMMISSIONER OF INDIA

23 February 2026

To
Shri Mritunjay Kumar Narayan
Registrar General and Census Commissioner of India
New Delhi

Dear Shri Narayan,

We are members of the Constitutional Conduct Group, a collective of former civil servants belonging to the All-India Services and the Central Services. Our group, which has no political affiliation, is committed to the promotion of the foundational values of our Republic and the observance of norms of Constitutional conduct.

We wish to bring to your attention some aspects of the 2027 Census currently under way.   The Decennial Census exercise was carried out in independent India every ten years from 1951 to 2011. While we can understand that the Census could not be carried out in 2021 because of the COVID pandemic, we fail to comprehend why the Census could not have been carried out by 2023, as was done in 143 other countries. The reasons for delaying the Census by six years instead of two to three years have not been made public. This lack of transparency gives rise to unnecessary apprehensions in the public mind that the Census is being conducted at this juncture to enable the completion of the exercise of delimitation of constituencies in 2027-28, in time for the 2029 Lok Sabha elections. We would certainly hope that no such extraneous considerations have influenced the timing of the 2027 Census.

We sincerely expect that the Census exercise will be unexceptionable and in conformity with the United Nations guidelines laid down in the Principles and Recommendations for Population and Housing Censuses (Revision 4 March 2025), to which India is a signatory. We understand that the main reasons for the delay in the processing and release of the data of past Censuses were:

(a) the need for coding of descriptive answers to several questions; and

(b) the lack of sufficient expertise within the Census Commissioner’s office to check the quality of data.

Providing mobile phones to code everything at field level, where the enumerator is required to select the correct option from a dropdown menu, does not allow for correction of errors in the recorded code. Past experience, especially in the 2001 and 2011 Censuses, has shown that mere technological advance in computing facilities does not necessarily speed up release of data. There is need to be open to the possibilities of errors, with effective measures being put in place to ensure data quality. 

Dropping questions on data items that are not required, cannot be collected or where alternate sources of data are available would help in streamlining the data collection process, reducing respondent fatigue and resulting in better quality data. For example, the questions on children born/surviving are better collected in the National Family Health Surveys.

Other Backward Classes (OBC) have not been specifically classified in the Census. The methodology for caste enumeration is yet to be announced. While one option could be to compile a list of castes for people to select from (as was done in the Bihar caste survey), we feel the better option is to leave the field open in the Census form, as was done in the 2011 Socio Economic and Caste Census (SECC). The methodology of surveying and enumerating languages could be used for condensing the Census data. However, this would require the government to keep the data open for scrutiny by scholars and involve institutions like the Anthropological Survey of India. The process can begin with collecting information on the 1369 mother tongue languages listed in the 2011 Census. An institution like the ASI could then certify the caste based on markers of common language, ancestry, lifestyle, relatives, marriages and kinship bonds. 

Data on tribes were being collected in past Censuses only from the Scheduled Tribe (ST) population. If all tribes, other than those in the ST list, are classified and recorded, a long existing injustice to the Denotified Tribe communities, which account for more than 100 million people, would be rectified.  

The issue of religion is, and has been in the past, a sensitive area for the Census. At a time when political leaders openly express their opposition to the inclusion of so-called “Bangladeshi Muslims” in the electoral rolls, care must be taken to ensure that the Census fully records the population of various minority groups in the country, covering religion, caste and tribe. 

 As former civil servants, many of us have been, during our careers, involved in the Census exercises at district, state and national levels. We are sure that you will exercise the highest level of professional competence in ensuring that the upcoming Census meets the threefold goals of accuracy, transparency and accessibility.

We wish the Census exercise all success. 

SATYAMEVA JAYATE

Yours sincerely,

Constitutional Conduct Group

The Wire is now on WhatsApp. Follow our channel for sharp analysis and opinions on the latest developments.