Sri Lanka’s Census 2012: What should have been asked? What could have been done better?

Conducting a census is an important activity for any country as the data gathered from it would serve as the foundation for policies, development related activities and future planning of not only government institutions but also non state actors such as academics, development and aid agencies. The idea behind collecting feedback on the 2012 Census in Sri Lanka is to identify the positive and negative aspects of the census, and to encourage discussion on how it can be improved without merely identifying the faults. This year’s census was held after 30 years and covered the entire island. The importance of this census and the data it gathered is obvious to us all. Feedback on Census 2012 was launched in late March. Some initial feedback from people who shared their comments via the site and also via email follow.

Enumeration stageEnumerators for the Census 2012 underwent a training whereby they were briefed on the questions in the data collection tool and the methodology. Each enumerator was also given a handbook which they could refer to should they have any doubt during the enumeration stage. While there was some positive feedback with regard to the professional manner in which data collectors gathered information from each household, most of the feedback received indicates several flaws in the enumeration stage.

One key issue that was evident from the comments was that it appears that many enumerators have taken the liberty of filling out some of the answers themselves. The census questionnaire is a lengthy one and in a household where there are a lot of people, it would take a fair amount of time. However that is not a valid reason to assume the answers and select an answer on behalf of the people in the household. Not only does it create a bias based on the enumerators own opinion but they could also be filling out the wrong answers.

An example for bias was flagged by an individual whose parents were of two different ethnicities and religions. The enumerator had not asked her what her religion and ethnic group was and instead filled in the same answers as her father’s, when in fact the religion she follows is that of her mother’s. The enumerator assumed that the children’s religion and ethnic group is the same as the father’s and it was only because she was observant that she noticed the enumerator had filled out these two questions on his own that she was able to point out the error.

Another question posed on the Feedback on Census 2012 website was on what questions should have been asked by the census, but were not. One person commented saying that they should have asked about Internet usage. When asked to clarify her point as internet usage was a question that was asked in the census, the answer given by the individual – “I’m sorry, it’s just that our enumerator didn’t ask us where we access the internet from, so I was under the impression that there wasn’t a single question with regard to internet usage. He did ask us about our computer literacy though. He might’ve made an assumption based on that.” – was an interesting one as it highlighted the same issue of the enumerator assuming the answer and not asking the question at all.

There were several people who said that their households were not enumerated at all. This included people who lived in apartment complexes as well. Some households had not been visited at all whereas some had been visited when no one except the domestic worker had been home and afterwards never revisited at a time when everyone was home. While there were articles that ran in the print media that stated that those who have not been visited by a census enumerator should contact their relevant Grama Niladhari officer, it is not a practical option as a) not everyone reads newspapers b) some people do not even know who their GN officer is and therefore would not take the time to personally go meet him or her to provide information. A more realistic approach to ensure that every household is counted would have been to provide enumerators with official printed notes which they could put in a household’s post box or leave with whoever that was at home stating that that household had been visited by a census enumerator and to contact the enumerator using the provided contact details so that a revisit can be made at a time when everyone in the household was at home.

Enumerators did not have to visit the households of people who had opted to fill out the census form online. However a comment made by an individual on the 6th of April 2012 – Information was submitted on-line. Hope they received it since no acknowledgement. When I phone them on 24 hr line, the officer answered said it should have been OK. But up to now, no response from the Dept.” – indicates that better monitoring of online submissions should have been in place to ensure that every single household was counted in the census.

Questions that should have been asked – The 2012 census asked a wide range of questions that have not been asked before in Sri Lanka’s census history. From educational qualifications to detailed questions about occupation, disabilities, literacy and household information, the data gathered would give valuable information about Sri Lankans. We asked the public whether they thought there were any questions or sections that should have been included in the 2012 census in order to identify missed opportunities which can hopefully be included in the next census.

One question that several people felt should have been included was a question about number of vehicles and type of vehicles owned by each household. The number of vehicles on the roads increases significantly every year and this has an impact on the traffic, road infrastructure, parking facilities of each city and therefore it would have been useful at a district level or even a DS level to know the number of vehicles to identify what infrastructural gaps would need to be filled in the years to come.

The census 2012 did not include any questions with regard to documentation – whether each individual had a birth certificate and where relevant, a National Identity Card, deeds for land owned by them, marriage certificates and so on. Birth certificates and NICs are the most important documents that an individual should possess and including a section on documentation would have proved to be invaluable to ascertain whether there are particular districts or regions that the state should focus on with regard to setting up or improving existing institutions to ensure a better documentation system of the country’s citizens.

In addition to any comments you may want to leave on Groundviews, please leave your feedback on the Feedback on 2012 census site as well.