Recent attacks on field enumerators in some parts of the country have put the upcoming Census procedure in “deep trouble” and may risk contaminating data collected in household surveys, former Chief Statistician of India Pronab Sen has said.

In recent days, some health researchers and government survey officials have been attacked by those who mistook them for people collecting data for the controversial National Population Register. As first reported by Scroll.in, the NPR is the first step to creating an all-Indian National Register of Citizens, which would identify undocumented migrants residing in India. The government has denied any link between the NPR and the NRC, but since the Census website itself made a connection, the population register exercise has also been a target of anti-NRC protestors.

Field investigators of the National Sample Survey were attacked in the past, but by now citizens had got comfortable with them as they got familiar with the process, Sen told The Indian Express in an interview. What is new this time is that the house-listing operations, a key step in the Census, will be conducted along with the population register exercise, said Sen, who heads the standing committee on economic statistics.

The population register will be a list of “usual residents” – those who have stayed at a place for six months or intend to stay there for the next six months. The exercise to update the register will be carried out across the country along with the house-listing phase of the decennial Census of 2021 from April 1 to September 30, 2020.

“The house-listing operation is a key step in our critical analysis because that’s the basis on which the entire country is divided into enumeration blocks and then, the Census enumerators are given specific enumerator blocks,” he said. “So, the enumeration blocks are at the heart of the Census operation. If you have problems in getting that done and if you get resistance to that, the Census is in deep trouble.”

He added: “So, you may well have a situation where you are unable to do the Census properly and if the Census is not done properly, then for the next 10 years, no household survey would be reliable because all household surveys rely on the Census as the frame. If this runs into problems, and there’s a danger that it might, then for the next 11 years, you are in trouble.”

The phenomenon of such attacks may have an effect not just on the Census but other surveys too, Sen said. “There’s a second problem,” he said. “Whatever you do, you are going to have people who will be suspicious. They may answer the expenditure part of it accurately, but the front part, which is household characteristics, they may resist that. All household details begins with asking household members, names, ages, educational details, so all that is in the first sheet. And what may end up happening is that the data might get contaminated pretty badly…it’s a huge problem. We may still be able to get enterprise surveys done but we are in deep trouble for household surveys.”

Sen said a possible solution would be to cut down on the questions in the Census, and ask only the bare minimum. “The only hope that we have is that for the house-listing operation, cut that form down,” he said. “A lot of stuff is collected in the form, just cut it down to the bare minimum that you need for the Census. And all you actually need for the Census is how many people, that’s all you need. So in each house or each location, you need to know how many people are there. That’s the basic knowledge.”