Over the last few days it has become evident that the British government is questioning the worth of its financial and administrative commitment to conducting a compulsory decadal census of the whole population. It was announced that the Office of National Statistics has initiated a government consultation about the future of the census, comparing it to alternatives such as using an online survey or efforts to collate data already held by government. In the era of ‘big data’ and increasingly sophisticated databases and forms of data collection is there still a role for this 212 year old institution? What would the loss of the census mean for research in areas such as epidemiology and human geography, or for future historians?
One of the main concerns with the UK census seems to be its cost, with the 2011 census costing the Government a reported £480 million. Furthermore, concerns have been raised about the quality and usefulness of census data, as compared to data compiled more regularly for administrative purposes. Another common argument against the census is that it requires citizens to answer unreasonably intimate and personal questions, for example concerning religion and marital status, which might even count as an infringement of human rights.
Image credit: Johnny Cyprus
Some researchers as well as bodies like the conservative think tank Policy Exchange have claimed that the use of currently collected administrative data as an alternative to the decadal census would provide the same quality of information at the fraction of the cost. However others, such as the economic geographer Danny Dorling, have argued strongly that the census is the only reliable source of important statistics such as mortality rates and population increase or decrease. For example, the 2011 census revealed that there were half a million more people living in England and Wales than had been assumed in official estimates – this is not an insubstantial figure, especially considering the importance of some of the policy decisions made on the basis of these statistics.
And this is not the first intervention that Danny Dorling has made in the debate about the merits of the UK census. Almost a decade ago Dorling and fellow economic geographer Paul Boyle wrote a guest editorial in the journal Area introducing a series of papers which used the 2001 census data. Boyle and Dorling ask whether the census is “a remarkable resource or bygone legacy of the ‘pencil and paper’ era?” and end up arguing emphatically for the continuing importance and usefulness of the census. They point out that freely available census data is utilised (though often without acknowledgement) for a wide-range of academic and non-academic uses, from epidemiological studies to historical accounts of important figures, to informing market research and infrastructure planning.
As Boyle and Dorling point out, one of the things that makes census data unique is its longitudinal range. A census has been conducted in the UK every decade since 1801 (with the exception of 1941 during World War Two), with some of the questions unchanged. This gives us the opportunity to follow the major changes in UK population and society. The census is also the only time when a data set is collected nationally but also provides a detailed picture of the local level. This kind of information is invaluable for resource allocation by highlighting the different needs of localities and giving information about local migration patterns.
Historically the creation of the census was tightly bound up with state acknowledging its responsibility to know about (and implicitly, to act on) the economic and cultural situation of its people. In several cases the census has even been involved in revealing the extent of major social problems and ringing the alarm bells. For example, Dorling has highlighted how the 1971 census prompted government action by showing for the first time how many families were living without hot water. Consequently, freely available statistics from the census have often been an important resource for campaigners trying to improve living conditions or deal with other social problems.
Some of the media coverage of the census debate has also highlighted the needs and interests of imagined future researchers, or even perhaps historical geographers. Following census data helps to identify major societal landmarks, such as the census in 1851 which was the first to record more of the population living in urban areas that rural ones. The census also makes it possible to chart the changing occupations and living standards and individuals, whether they are important figures from history or a long lost member of your family. Even changes in the way censuses have been conducted and the questions which have been asked over the years are very revealing of key changes in society over time and give us an indication of which issues were of concern and interest to administrators at the time.
If the 2021 census goes ahead what will it tell us about changes in our society? And what invaluable information about our living conditions, population and eccentricities will it provide to future researchers?
Paul Boyle and Danny Dorling, 2004 Guest editorial: the 2001 UK census: a remarkable resource or bygone legacy of the ‘pencil and paper’ era? Area 36 101-110
Ending the national census would make us blind to our society The Guardian, 2 September 2013
Loss of census seen as threat to UK historical insight The Financial Times, 1 September 2013
Researchers in UK count the cost of plan to scrap census The Financial Times, 1 September 2013
Census consultation has option to replace 200-year-old survey BBC News, 3 September 2013
Geography Directions 
A short overview of the UK census history: http://demotrends.wordpress.com/2013/03/03/the-origins-of-population-statistics-a-glimpse-at-the-uk/
I’m watching the debate with interest, as I make much of the 2011 census available on a website that I run ( http://www.streetcheck.co.uk ).
Another point worth mentioning is that the census data is now freely available under the government’s open data initiatives to anyone, thus providing a single public dataset that contains all of the various data points. If would be a great loss if this data source was lost and replaced with dozens of smaller niche datasets from across government (many of which are not currently open to the public).