-
Information on the amalgamation of areas that fell below the threshold values set to prevent accidental disclosure of personal information.
A number of measures are applied to all 2001 Census output for England and Wales to prevent the inadvertent disclosure of information about identifiable individuals. These include:
factors involved in the design of output tables
record swapping occurring in the output database
application of thresholds to determine which areas can have certain results produced for them
application of a small cell adjustment procedure to final tables to modify the values contained in small count cells
conditions of use applied to all output products
These methods are described in more detail below.
Design of tables
A general principle of making the average cell count in a table greater than or equal to one is applied to the design of all 2001 Census output.
Record swapping
The individual records on the output database are slightly modified by record swapping. This means that a sample of records were 'swapped' with similar records in other geographical areas. Information about the proportion of records swapped cannot be provided as this may compromise confidentiality protection.
Thresholds
Minimum resident population and household threshold values are applied to sets of 2001 Census data. Areas with counts which fall below these threshold values may either have some less detailed summary statistics produced, or be amalgamated with other areas to allow the release of more detailed sets of results for the combined area. All threshold values apply to population and household values on Census day, April 2001.
These measures are applied to predefined areas, such as parishes, wards or postal sectors, for which a particular set of results are generally available, but which may sometimes have populations below the thresholds. All Output Areas are above the threshold for Census Area Statistics (CAS) and Key Statistics.
There are three sets of threshold values which are applied to the different series of data produced. The thresholds are:
1,000 residents and 400 households for Standard Tables
100 residents and 40 households for Census Area Statistics (CAS) and Key Statistics
50 residents and 20 households for parish/community profiles
Where civil parishes in England, communities in Wales, or wards fall below either of the thresholds for CAS, but contain more than 50 people and 20 households, summary statistics called profiles are released. The 1991 Census equivalent for this threshold were 50 people and 16 households, and the increase in the threshold of households to 20 reflects the change in average household size between 1991 and 2001.
Where civil parishes or communities have less than 50 people and less than 20 households, counts of the total numbers of residents, males, females and resident households are released. These are known as 'headcounts' in Census products. There are no thresholds for headcounts, but small counts are still adjusted as part of disclosure protection.
When areas are amalgamated, they are done so with contiguous areas, and areas of the same type wherever possible. This was done in consultation with the local authorities of the areas concerned. The few postal sectors which fell below thresholds were not amalgamated with other sectors as there was no definitive information about contiguity of sectors. More information can be found using the 'Amalgamation of geographic areas' link above.
Small cell adjustment
As a final pre-release means of ensuring confidentiality of results, small counts in tables are subject to a small cell count adjustment routine. During this process:
a small count appearing in a table cell is adjusted (information on what constitutes a small cell count cannot be provided as this may compromise confidentiality protection)
totals and sub totals in tables are calculated as the sum of the adjusted counts so that all tables are internally additive (within tables, totals and sub totals are the sum of the adjusted constituent counts)
tables are independently adjusted (this means that counts of the same population in two different tables may not necessarily be the same)
tables for higher geographical levels are independently adjusted, and, therefore, will not necessarily be the sum of the lower component geographical units
output is produced from one database, adjusted for estimated undercount, and the tables from this one database provide a consistent pictures of this one population
Conditions of use
Condition of use are included in all end user licences. These state that Census material shall not be used to attempt to derive information relating to an identified person or household nor shall a claim be made that such information has been obtained or derived.