Coding and naming policy for statistical geographies in Scotland

This section outlines the format and practices for creating and maintaining Codes for statistical geographies in Scotland.

Format

Geogrpahy codes, or GSS codes, consist of a nine-digit alpha-numeric (ANNNNNNNN) consisting of two parts. These are the entity and the instance.

​​​​​​​The first part (ANN) identifies the geographical entity (area type). The first alpha character of the entity code indicates the country (or 'cross-border' status) within which the entity is bounded. The 2nd and 3rd numeric characters will represent a specific geographic entity, but with no other intelligence in the code.

Scotland has been assigned the country designation ‘S’, and as such GSS codes for Scottish geographies will have the format 'SNNNNNNNN'. The characters T, U, and V have also been reserved for Scotland.

The final 6 digits of the code represent the area instance. This instance will be uniquely identified within the UK by its combination with the entity code. Instances should not normally be based on any inbuilt intelligence. For example, codes should not be assigned based on alphabetical or hierarchical properties. This is because any later changes to the geography such as renaming will likely disrupt this ordering. The nature of hierarchical relationships and any other classification-type of intelligence should instead be recorded and managed separately to the coding.

An exception to this rule is the coding of geographies such as Census Output Areas. These geographies are generally coded in groups by Local Authority to help users with ranges of areas. As Output Areas do not change within the inter-census period this is deemed acceptable. Upon Local Government boundary reviews, wards will also normally be coded within groups of Local Authorities, however, as council and ward boundaries can change in between review periods via a statutory instrument, this ordering can become eroded over time.

Note that cross-border geographies will have instances that use the designation ‘K’. An example of this would be the Travel to Work Areas geography managed by ONS. Particular instances in this geography do not respect the border between Scotland and England and are therefore coded with a ‘K’ prefix rather than ‘S’ (i.e. Berwick and Carlisle).

​​​​​​​Policy

The key principles of the coding policy are as follows:

• The coding system is not hierarchical and does not contain embedded intelligence (other than the first country character).
• Codes cannot be re-used. Re-use of codes would be bad practice and presents risks to the analyses of time-series geographic data.
• Except for a few special circumstances detailed below, a new instance of the area will be created with a new code only when a change occurs to the boundary of an area instance. For example, name changes alone do not initiate a code change for the area instance. Note that name changes are recorded elsewhere (in the Scottish Register of Geographic Codes).

Generally, GSS codes are only assigned to geographies that are required for government administrative functions, legislation or publication of official statistics. This definition can be interpreted relatively easily and consistently by the consideration that “if the geography did not exist, it would have to be created to support a specific government business area”.

Geography codes are normally recommended where one of the following criteria applies:

• The boundaries are defined by legislation or some other formal process.
• The boundaries are used for Government administrative functions.
• The boundaries are expected to be relatively stable.