Infirmities to caste-tech: How Census questions have changed since 1872

The evolution of India's Census questions over 150 years reflects the state's shifting priorities, from caste and infirmities to migration, fertility and tech-driven caste data

The 1872 House Register, the precursor to the first full-fledged Census in 1881, had 17 questions to which responses were sought. These were whittled down to 13 for the subsequent Census.

 

For the 1881 Census, data were collected for questions on religion, caste, mother tongue, “occupation of men, also of boys and females who may do work”, and whether the respondent received education “under instruction”, or was “not under instruction but able to read and write”, or was illiterate. It also tabulated “infirmities”, including whether someone was of “unsound mind” or was a “leper”.

 

Apart from asking the religion of a respondent, the 1891 Census also sought to know the sect of religion, “main caste” and “sub-division of caste”, and “foreign languages known, if any”.

 

The 1901 Census had a more detailed questionnaire on occupation, including principal and subsidiary occupation or means of subsistence of “actual workers” and “means of subsistence of dependents of actual workers”. It also, for the first time, sought to know if respondents knew English. 

 

The 1911 Census asked officials to fill in the census number painted on the house, which was not the case in previous censuses, and specifically sought to know sects of Christians. The 1931 Census was the last to include a question on “race, tribe or caste” of the respondent. The 1941 Census, which could not be conducted fully because of the Second World War, included questions on the number of children born to a married woman and the number of surviving children, as well as whether a respondent employed paid assistants.

 

The 1951 Census, the first conducted after Partition and Independence, included questions on nationality, religion and special groups, and had a detailed section to collect data on “indigenous persons of Assam”.

 

The 1961 Census was the first to introduce a question on whether a respondent belonged to either a Scheduled Caste or Scheduled Tribe. It also collated data on cultivators, agricultural labourers, those “working at household industry”, and the nature of that work.

 

The 1971 Census had 17 questions and asked married women their age at marriage, whether “any child [was] born in the last one year”, and questions on migration. The 1981 Census had a specific question on the “reason for migration from the place of last residence”. For married women, it asked the number of children surviving and the number of children ever born alive. The number of questions increased to 23 in the 1991 Census. 

 

The 2001 Census fine-tuned data collection on SCs and STs, asking officials to write the name of the caste in the case of SCs and tribe in the case of STs from the list supplied. It also asked whether a respondent had worked at any time in the last 12 months, whether they were a main or marginal worker or a non-worker, the distance from residence to place of work, mode of travel to the workplace, migration, and fertility details, including specific questions on the number of son(s) and daughter(s) surviving at present and ever born alive.

 

The 2011 Census, the last conducted so far, had 29 questions covering economic and non-economic activity, whether seeking or available for work, and migration and fertility characteristics.

 

The 2027 Census is set to use technology extensively. It will include questions to enumerate the caste identity of respondents, and enumerators are likely to be asked to punch the caste of a respondent into their tablets from a drop-down menu.