Census Methodology

This section provides additional information on the processes applied to the 2011 Scottish census data, arranged by process type. The Office for National Statistics (ONS) developed the key elements of statistical processing which we adapted for use in Scotland. Click here to find out more about the detailed methodologies that ONS use to produce census estimates for England and Wales.

Coverage Estimate and Adjustment

The main purpose of the census is to provide an accurate population count. Although every effort is made to ensure everyone is included in the census, inevitably some individuals are missed. This under-counting does not usually occur uniformly across all geographical areas or across other sub-groups (for example, by age and sex) of the population.

To fill the gap, NRS implemented a coverage assessment and adjustment process to estimate the population that was missed. This process identified and adjusted for the people who were counted more than once or who were counted in the wrong place. More details on how this process  can be found in the Estimation and Adjustment Strategy for Scotland’s Census. Carrying out this work allowed a census estimate of the entire population to be obtained. A summary of the statistical methodology used to derive the 2011 Census population and household estimates for Scotland for Release 1C is detailed in this paper:-

Release 1C - How the 2011 Census population estimates were obtained.
(Adobe Acrobat Portable Document Format) (134 Kb)

Link to supporting tables

You can see the summary produced to accompany previous Releases:-

The Scottish strategy was built upon the methodology developed by the Office for National Statistics (ONS). For more details of the ONS coverage methodology please see the ONS website.

For an accessible explanation of Dual System Estimation methodology which forms a core part of the estimation process please see the ONS website.

A report on the data quality issues which originate from the estimation and adjustment process can be found in the report Scotland’s Census 2011 Data quality for coverage records

Edit and Imputation

Some of the returns from the 2011 Census were incomplete - that is, respondents did not answer every question - and sometimes inconsistent (for example, in the 2001 Census, some respondents recorded parent-child relationships the 'wrong-way-round', recording the child as the parent).

The edit and imputation process is designed to produce a dataset that is internally consistent, and has no relevant missing values for any returned questionnaire.

The development of the ONS Edit and Imputation Methodology, which was adapted for Scotland, is detailed on the ONS website

Statistical Disclosure Control

While most census outputs take the form of statistical counts, there is a risk that information about an individual person could be deduced from census outputs. For example, if everybody in a particular geographic area was aged under 50 apart from one old-age pensioner living in a single person household, a cross-tabulation of age and general health would reveal the response of that pensioner to the census question on general health. The census form gives respondents an assurance that their information will be treated as confidential, and statistical disclosure techniques are employed to ensure that the risk of inadvertent disclosure in statistical outputs is minimised.

For specific detail on Statistical Disclosure Control see the ONS website. This describes the principles being applied throughout the UK. In order to ensure that census outputs across the UK are comparable, similar processes are being applied to each of the Censuses in the UK.

Output Geography

A full Geography Prospectus is currently under development. Guided by consultation feedback from users, NRS will take the same general approach to output geographies for Scotland’s 2011 Census as was adopted for the previous census in 2001. This consultation feedback was published in the Spring 2010 Consultation on Statistical Outputs: Analysis of Responses document (see consultation points 5-8). Census output areas (averaging around 50 households in size, with minimum thresholds for confidentiality set at 20 households and 50 individuals) will continue to be the smallest geography for which census results will be produced and will form the building bricks for census outputs for all higher geographies. With the exception of local authority areas, which will be produced on an ‘exact fit’ basis, all higher geographies will be generated as aggregations of output areas on a ‘best fit’ basis. Further background information on census output geography can be found in section 4 of the February 2011 Census Outputs Consultation document on the NRS website. The matrix showing which set of pre-defined statistical tables will be produced for each output geography is provided at Annex A of the Scotland’s Census 2011 Outputs Prospectus. Further information on UK Census Geography can be found on the ONS website.