Rediff.com's Priyanka meets Dr C Chandramouli, the Census Commissioner of India, where he explains what it takes to organise the mammoth exercise.
Counting a population of 1.21 billion Indians, comprising of about 623 million males and 586 million females, is no easy task.
The Census of India undertakes a mammoth mission after every ten years. It does an official count of each individual from all corners of the country and tallies the overall numbers. The man-in-charge, Dr C Chandramouli, Census Commissioner of India, had presided over the census count of 2001, and 2011. A few of the results of Census 2011 are out, and the rest like those of the caste-based census would come out by the end of March this year.
Mapping his journey of the two censuses that he has overseen, Dr C Chandramouli, in a public meeting held in the capital recently, talked about his experiences, and answered a few questions on the big count.
"As long as one is a citizen of the country, he or she had to be covered under the census," said Dr Chandramouli, asserting there were no two doubts about it. "But from the difficult terrain to dealing with people, we faced a whole lot of problems," he also added.
Dr Chandramouli assured that for the first time that the entire country was covered and it was done with the help of district collectors, local level administrators.
"It was really an achievement that even the left-wing extremist affected areas of our country were covered," he said.
"The Census faced difficulties in getting data from places in the north east, as a big propaganda saying that the census is the coming of anti-Christ and hence it must be boycotted was built up," he said to an amused audience.
"It required a lot of persuasion at the ground level to convince the people that the government did not have any such sinister designs," Dr Chandramouli added.
He clearly lays out his objectives which are covering the entire population, ensuring quality of the data, making the exercise cost effective and delivering the results in time.
"We used all the available technology to improve the quality, primarily involved a lot of training," he said. Manual instructions were prepared in 18 languages.
"The problem has always been that the data reaches the users far beyond its time," Dr Chandramouli said. "And hence technology was employed to ensure that the numbers are out within the shortest period of time. Our attempt to bring the entire findings within two years, the first year of which is already over and we hope to bring out the first set of results by the end of January 2012 and after that a series of results over the current year."
An allocated budget of Rs 2,200 crores was spent on the Census 2011, with a per capita cost to about 18 rupees. About 27 lakh functionaries were assigned the task of going to places and taking data. About 3.4 crore forms were printed.
"We used satellite imagery for generating maps of cities which shows houses to where functionaries then went and collected data," Dr Chandramouli said.
The printing of forms, tutorials and the training manuals had to be completed within three months so that every tehsil and district gets the right number of manuals in the right language. A standard mode of training was adopted where trainers at the national, state and district level were equipped.
"A little known fact is that the Census of India has always been at the forefront of cutting edge computing technology," Dr Chandramouli said. "In the Census of 1961, unit record machines were used, whereas key punching for data entry was used in the 1971 Census."
"In 2001 we were the first country in the world to switch over to image based recognition of data; a further improved version of this technology was used in 2011," he added. And hence, unlike the early days when 13,000 people worked for 5-6 years to produce the same data which is now being generated by 2,000 people in just about 2 years.
Among the key findings of the Census of India 2011, Dr Chandramouli said that for the first time there has been a significant dip of the rate at which the population is growing in states of UP, Bihar, Uttarakhand, Jharakhand, Madhya Pradesh, Chhattisgarh and Odisha, which were also considered as the main contributor to the high population growth.
"After three consecutive censuses, and after having showing a growth rate of 25 percent, these states are now showing a dip," he said.
The overall growth rate of population of the country is also declining from an earlier 21 percent, and is now hovering about 17 percent. The ratio of females per 1000 males in the country is now 940, and the literacy rate is up to about 74 percent now. However, the child sex ration, that is the number of girls per 100 boys, in the minors category is lowest ever this year.
Dr Chandramouli also answered to a few questions put forward by the audience:
What were the learnings from the 2001 Census, of which also you were a part, which were used in the 2011 Census?
There were two major learnings; the first was to reach out to those sections that are usually left out. It comprises of the poor, the marginalised and women and little children. It is a general practice to miss out the very young children and the elderly, but this time we made sure we do not leave out the too young and the elderly.
Second, this time I feel we were better equipped to negotiate the technology we wanted to use, which we had learnt the hard way after our experiences in the last Census.
It was very interesting to see that the Census made a detailed note of the occupation of women and accounted for it, making it clear whether it was paid or non-paid work.
This was interesting because the Supreme Court made scathing remarks about the classification of the work by women and categorising in into paid and non-paid work. Now, all we can offer to say is that Census is not the author of defining what is work and what is not, the National System of Accounts does that. We merely use the prescribed definition and use it to collect data.
We realised that the work done by women was not being captured to its best and there were hundreds of villages which reported that not a single woman was working. So we went up to those villages and enquired if there was a teacher or an anganwadi worker in the village. Such campaigns were carried out in gender sensitive places and the emphasis was to capture the overall picture of the status of the contribution of women to the work done more accurately.
To each of the sates that you went to, how was their response in conducting the Census?
I had personally been to most of the states and individually met the chief ministers of the states. I can say all were very aware and willing to share their services, but general apathy was at the collectorate and the sub-district level.
Census is a long process. We have just completed a mammoth of an exercise, now the task of taking data for the National Population register is on alongside of caste-based census. Hence, we really need to ensure the participation of administrative officers at the junior levels.
The Chinese will apparently reach their peak in population in 2030. When are we reaching the peak, before the population begins to decline? China is also the only other country that has to do an equally big task in conducting the Census. Is there any exchange of the technology or ideas, so that the two countries can learn from each other's methods?
We did visit and we did talk to the Chinese, but we didn't get much of a response. All we got to know was that they employ about 6 million people. But I feel their process is not very transparent, so we don't know what their actual numbers are.
The second difference is they do a lot of cross-checking and validation of records. So, they really don't have to do a Census to get their numbers. There is a lot of regimentation and thus there are many records from where they get their numbers.
However, there is no such pool in India. So, we literally go blindfolded with every Census, not knowing what to expect.
With the introduction of the UID, how is the Census going forward?
The UID is a completely different ball game. The Census information is strictly geographical and we do not reveal the individual characteristics of a person; we are very strict about maintaining privacy. Even the courts cannot direct us to reveal data that we have collected. There have been instances when we have been asked to give out details of public personalities and about what religions they listed or other personal information. We have never revealed.
However, there is another exercise going on where data is being compiled under the National Population Register. In this case, we collect some generic information like name, fathers' name, mother's name and so on. Now this information would be given to the UID, which, in turn, would undertake a de-duplication based on the biometric that we have collected and generate a number that would be unique to everyone.
Hence what the UID is generating a unique number based on the de -duplication of biometrics. These are two very separate schemes.
Now the on-going buzz is not about whether the UID should continue or not, but the question is who will collect the biometric data. The home ministry in charge of creating the NPR has been asked to collect biometric data from the entire country.
In the meantime, several other registrars of the UID ecosystem have also been doing the same job. So, the concern from our ministry is that there should not be any duplication of work and there should be only one agency assigned to collect biometric. Ultimately, it will be all thrown into the UID and unique numbers would be generated.
The only question is who would be collecting the biometric data. We don't want multiple agencies doing the same job. So, we want the government to decide whether there should be multiple agencies doing the same job or it should be assigned to one.
Is it possible for an individual to register as a non-believer in any religion in the Census report?
Now, we understand there is a major tendency among the enumerators who are assigned the task to collect data to report only the major religion. In some other cases, a person may not either reveal his or her religion or the enumerator assumes that a person belongs to the majority religion. We have made an attempt to make sure that the enumerators ask each person what their religion is and note down each response.
Caste is a very important aspect of the Indian societal fabric. How do you plan to incorporate it in the Census?
It is common understanding that while religion has been widely accepted, caste does have different political connotations attached to it. There is a tendency to boost the numbers of a particular caste so that they become politically more relevant. Our fear was that if every small caste tries to bulge its numbers, then this would eventually swell their overall numbers to such an extent that the entire Census would go haywire. After all, there is no verification done in Census.
And we had no numbers to verify if there was a boosting of numbers of a particular caste. Thus we requested the government to conduct the caste census as a separate exercise, and only after the main Census is complete. Today, the caste-based Census is on, and we have the actual numbers to check any improper reporting of caste numbers in any village.