"India has begun counting its population in the world's largest census, which will include caste enumeration for the first time in nearly a century. This year's census is a $1.24bn exercise during which more than three million Indian officials will spend a year surveying about 1.4 billion Indians about their household composition, living conditions and access to basic amenities. The last census was conducted in 2011. Another one was due in 2021, but it was delayed due to the COVID-19 pandemic, leaving India's data on such things as its demographics, housing conditions and welfare amenities outdated."

"The two-phase exercise, billed as the world's most ambitious of its kind, will see more than three million officials spend a year counting every person in India. India's 16th census — the eighth since independence in 1947 — will also include caste data and is seen as crucial for policy, welfare delivery and political representation in the world's most populous country. With more than 1.4 billion people, India overtook China in 2023, according to the United Nations Population Fund. The exercise will span 36 states and federally-administered territories, more than 7,000 sub-districts, over 9,700 towns and nearly 640,000 villages, with fieldwork carried out by enumerators and supervisors — typically schoolteachers, government staff and local officials. For the first time, the census will be conducted digitally, with enumerators using mobile apps to collect and upload data."

"given that the population has grown at very different places in different parts of the country, the information from this census could become highly politically relevant"

"Since it is a full enumeration of the entire population, all large-scale surveys — which, by design, capture only a subset of households — rely on the census as their sampling frame. With India's last census now well over a decade old, every major survey conducted in this period is working off a frame that no longer reflects the population it is meant to represent. That is not a minor technical inconvenience. It introduces systematic errors into the data that policymakers, researchers and planners depend on"