Data Quality Issues

On this page you can find out about any known data quality issues. Information on known quality issues with specific census variables can be found in the Metadata - Variables and Classifications section.

Below is a summary of the issues which were identified in data related to the relationships of residents within households, to accompany Release 2C:

Households Relationships Data Quality Issues (Adobe Acrobat Portable Document Format) (28 Kb)

Impact of adjustments to published census estimates to avoid disclosure through differencing between geographical areas.

Some manual adjustment was required to avoid disclosure within the Census dataset.  This has resulted in a small number of cases where published tables will not match tables that could be created by combining the figures of individual output areas for certain geographies using GIS packages.  In order to help users who have an interest at this detailed level, we have documented the areas affected by this process.  Further information is available in the following Excel document.

Further documents on known data quality issues will be published in due course.

Impact of building blocks on 2011 Census population and household estimates for 2001 datazones.

2011 Census population and household estimates for 2001 datazones can be produced using two different methods. One method aggregates Census 2011 household and population postcodes estimates whereas the other aggregates Census 2011 household and population output area estimates.

This document provides an example of how the two methods can produce different estimates, compares the 2011 census population and household estimates for 2001 datazones produced using the two different methods and it also gives the size of the population/number of households which are common to both definitions or are in one but not the other.

It is important that users do not interpret the two estimates as a lower and upper bound as the true estimate may not lie between the two.

The underlying assumption is that the estimates produced by aggregating 2011 Census postcode estimates are more accurate but there may be exceptions.

Further information is available in the following Excel document.