The Economist in a recent article gave some interesting facts about how Population Census is carried out in different parts of the world. I learnt some surprising aspects:
1. I always assumed that Population Census has to be done physically i.e. someone has to go and physically visit each house and assess the number of people. This was actually so in all countries until recently. However, increasing privacy concerns - and rising costs - is changing this. In countries like Germany, Census is carried out almost entirely virtually - through accessing government databases such as National Employment records, birth and death registers etc. Subsequently, only 10% of the sample is randomly selected for a physical old style face to face interviewing for the purpose of verification.
2. Interestingly - for this 10% of the selected sample - there is a penalty if they refused to co-operate with the enumerators. This is unthinkable in traditional Market Research.
3. This pattern of using secondary sources to carry out the population census is becoming more and more popular - 17 countries in Europe do it in this manner. And even the US is planning to shift more and more of its census data collection to online.
4. The US currently carries out its Population Census still in the old fashioned manner (as in India). It cost them USD 43 (i.e. about Rupees 2000) to count each person. The equivalent cost for counting each person in India was only 40 cents (i.e. about Rupees 17).
For me there were 2 important take-outs from this article:
- Data collection, even for something like the Population Census, will increasingly move to a secondary data collection methodology using online databases. Hence with the national biometric ID being issued to every Indian, it is possible that the Census methodology as we know it today will vanish within the next couple of decades even in India
- The other take-out is the cost. If data collection continues to be so cheap in India then there will be no immediate cost incentive to move to a Secondary data collection system as in other countries. The only reason why India will abandon the face-to-face primary methodology and move to a secondary system would be in the interests of time and accuracy.
1. I always assumed that Population Census has to be done physically i.e. someone has to go and physically visit each house and assess the number of people. This was actually so in all countries until recently. However, increasing privacy concerns - and rising costs - is changing this. In countries like Germany, Census is carried out almost entirely virtually - through accessing government databases such as National Employment records, birth and death registers etc. Subsequently, only 10% of the sample is randomly selected for a physical old style face to face interviewing for the purpose of verification.
2. Interestingly - for this 10% of the selected sample - there is a penalty if they refused to co-operate with the enumerators. This is unthinkable in traditional Market Research.
3. This pattern of using secondary sources to carry out the population census is becoming more and more popular - 17 countries in Europe do it in this manner. And even the US is planning to shift more and more of its census data collection to online.
4. The US currently carries out its Population Census still in the old fashioned manner (as in India). It cost them USD 43 (i.e. about Rupees 2000) to count each person. The equivalent cost for counting each person in India was only 40 cents (i.e. about Rupees 17).
For me there were 2 important take-outs from this article:
- Data collection, even for something like the Population Census, will increasingly move to a secondary data collection methodology using online databases. Hence with the national biometric ID being issued to every Indian, it is possible that the Census methodology as we know it today will vanish within the next couple of decades even in India
- The other take-out is the cost. If data collection continues to be so cheap in India then there will be no immediate cost incentive to move to a Secondary data collection system as in other countries. The only reason why India will abandon the face-to-face primary methodology and move to a secondary system would be in the interests of time and accuracy.
1 comment:
Interesting topic. Only recently we had a census survey and the person who comes to survey, perhaps to cut short her time, would go ahead and assume so many things. Instead of filling out box by box she would directly go to certain particular boxes jumping and avoiding the rest of them. When enquired she would reply that it was not so important or that she would fill that up later. In Kerala most of the Christians have Hindu sounding first names and if surveys are done in this assumption method then Hindus would certainly be a majority still, thus preserving the benefits of minority status of the Christians here. There is a talk here that this a ploy played by the church in collusion with the corrupt ministers here.
Post a Comment