Counting without the census

Crowd of people13 December 2011

It has had a run of more than 200 years, but the England and Wales census of 2011 may have been the last of its kind.

A 'one size fits all' count of the population every ten years fitted the purpose as long as society was homogenous and households were stable. But with a more mobile population and more complex households and ways people live, it becomes more difficult to capture changes in society - not least with a snapshot only once in a decade. Add to this the rising costs of achieving a high response rate, and the status quo becomes unsustainable.

"To date the decennial census has been at the heart of the population statistics system and is used for a wide variety of purposes, including the allocation of public resources of more than £100bn each year. However, the task of census enumeration is becoming increasingly difficult and expensive," says Professor David Martin, Director of the ESRC Census Programme.

But in order to make informed policy decisions on national and local issues, for instance the planning of key local services in areas like health and crime, it is still crucial to collect information for population and socio-demographic statistics.

The Office for National Statistics has set up the Beyond 2011 project to explore future alternatives to the traditional census, such as:

  • a short form census
  • a rolling programme of census data collection
  • an enhanced programme of government surveys
  • linking existing administrative registers
  • continuing the conventional census with increased emphasis on internet-based data collection.

"The one thing we can be sure about is that no one of these options will meet all our data requirements on its own," says Professor Martin. "The solution will almost certainly be a hybrid system, drawing data from multiple sources to meet the wide variety of user needs that were met by the traditional census."

Most of the Beyond 2011 Programme will focus on understanding how existing data sources could be combined with surveys to provide the necessary statistics, reporting on findings in 2014.

The ESRC is working with ONS on three projects to complement the Beyond 2011 Programme:

  • A review of the rolling census approach, carried out by Dr James Brown at the University of Southampton. The review will be looking at the methodological and policy issues surrounding a rolling census, assessing whether this is a usable option.
  • A series of seminars, led by Professor Martin, is bringing together methodological researchers in the academic community and government statistical agencies to address key debates relating to census replacement options - including administrative data linkage, small area estimation, address listing and lessons from the census longitudinal studies.
  • Four jointly funded ESRC/ONS studentships focusing on three areas: quality measures for population and demographic statistics; small area estimation; and statistical disclosure control for derived and administrative data.

"It is clear from international experience that any decision to replace a traditional census with alternative data sources can require a long development period, so there is an urgent need for research on the options now if we are to adequately meet our data requirements at the end of the decade," adds Professor Martin.