Scottish Neighbourhood Statistics Data Zones Background Information
1 Frequently Asked Questions
How will the Scottish Executive use data zones?
Data zones will play a key role in Scottish Neighbourhood Statistics and the Scottish Multiple Index of Deprivation to help develop and monitor Scottish Executive and its partners' policies. particularly in terms of area regeneration. Data zones will increasingly be the core geography for making available small area statistics across most social topics including information about benefits, education, health and the labour market. For the first time, this will allow users to readily (and regularly) bring together information from various sources on a common small area geography.
Do data zones represent communities on the ground?
It was always recognised that data zones are statistical areas and do not necessarily delineate communities. The key feature of data zones is that they are significantly smaller than previous geographies for which statistics have been available (postcode sector or ward) and are much more effective in identifying small areas with particular social characteristics, and are also more flexible in aggregating to specific areas of user interest. While at the same time data zones are large enough to protect confidentiality and to allow regular updates to be made available.
How were data zones produced?
Data zones meet tight constraints on population thresholds (500 - 1000 household residents), they all nest in to local authorities and are built up from 2001 Census output areas. The aim was also to build data zones by grouping together output areas with similar social characteristics, for data zones to have a fairly compact shape, and to take account of physical boundaries. To do this (in most areas) we used primary school catchment areas to get a first cut of data zones and refined these using a measure of local homogeneity (based on employment levels, over-crowding and car-ownership) and underlying geographical features. A report from St Andrews describing the process for producing data zones is included at Annex A.
How were users consulted?
Over the last couple of years the Scottish Executive consulted widely on the development of data zones. A Scottish Neighbourhood Statistics Neighbourhood Definitions Group was established in 2001 to oversee the development of data zones which included representatives from local authorities. The project was split into five phases.
The first phase invited interested parties to develop strategies for the production of data zones and this was followed by a review of the various strategies by St Andrews University. The next two stages involved St Andrews producing a prototype set of data zones for the whole of Scotland based on 1991 Census information and users throughout Scotland commenting on the prototype set. The prototype set of data zones were generally well received and a review of the first four phases was presented at a seminar in Glasgow at the end of 2002.
The main findings from the first four phases were that the methodology devised by St Andrews could be used to produce data zones and that a final set of data zones should be created using 2001 Census information. After the 2001 Census information became available in 2003, St Andrews produced a final set of data zones and invited local authorities to comment on the data zones and provide suggested amendments based on local knowledge. At this stage we were expecting minimal change, however a small number of local authorities suggested significant changes to the data zones in their areas.
Were all suggested changes to data zones accepted?
All suggested changes that were consistent with the tight constraints on population were accepted. Some local authorities suggested changes to data zones so that they better respected settlement boundaries or some administrative geographies.
What effect have the accepted changes based on local knowledge had?
Because there were a number of criteria to be met, there was no one optimal solution and there would always be some trade-offs between the various criteria. However, it is clear that constraining the data zone populations tightly has had the largest effect on the overall set of data zones. Any changes resulting from local knowledge is likely to have resulted in only a very marginal improvement in the set of data zones in terms of local homogeneity and/or respecting various boundaries.
Was enough time made available for the final phase of consultation?
St Andrews provided the data zones to local authorities as they became available and in the final phase local authorities were given 4 to 6 weeks to comment on the data zones. This final phase of consultation coincided with holiday period and some local authorities asked for the consultation to be extended. The consultation was further extended from mid October to mid-November.
Which boundaries do data zones respect?
Data zones have been designed to respect local authority boundaries as at 2001 Census.
During the consultation some users asked for data zones to take account of ward, old district, or settlement boundaries. There was no Scotland wide consensus as to which boundaries to take account of. A practical Scotland wide solution was needed and increasing the number of boundaries to be respected would have made it increasingly difficult to produce data zones (increasing the number of different boundaries to be respected would have an impact on the meeting other constraints).
During the final phase of consultation some local authorities suggested specific amendments to data zones so that they respected better some of these listed boundaries. These changes were accepted where they met the tight constraints, however, future changes to those listed boundaries will not in itself direct change to data zone boundaries.
Are 'doughnut' shaped data zones acceptable?
We have tried to avoid the creation of 'doughnut' shaped data zones. However, adjustments to data zones which result in doughnut data zones have been accepted where they are based on local knowledge.
Will data zones be a stable geography?
The intention is for data zones to be a stable geography. The Scottish Executive are currently developing the process for maintaining data zones. With each yearly update of Scottish Neighbourhood Statistics we will need to ensure that data zones meet the minimum population thresholds. Once data zones have bedded in to the Scottish Neighbourhood Statistics we intend to undertake an evaluation which will feed into the longer term process for maintaining data zones.
How will data zones take account of population change?
As part of Scottish Neighbourhood Statistics the General Register for Scotland (GROS) is developing a methodology to produce population estimates at the data zone level. 2001 Census population estimates were used in the creation of data zones.
Will data zones take account of current and future planning applications?
Information on population change as a result of development and regeneration will come from the General Register for Scotland small area population estimates.
Will care be needed when analysing trends in data zone that have had relatively large changes in population?
Yes. The Scottish Executive are considering developing an indicator to flag up data zones which have had significant population change.
Will data zones be named?
There are 6,505 data zones in Scotland and it is not practicable to name each of the zones. Each data zone will be assigned an individual code; and each code will have attributes including relationships to some other higher area geographies. Data zones are constructed from census output areas (whose boundaries are fixed) and the historic link between census output areas and data zones will be maintained.
Why did you use 2001 Census output area as the geography to build data zones?
There are various operational and practical reasons why the 2001 Census output area geography was used to build the data zones. these include Census output areas are readily available and cover the whole of Scotland; Census output area boundaries are fixed; a great deal of information is now postcode referenced and the links between postcodes and census outputs areas are maintained. This allows ready aggregation of postcoded information to data zones via the Census output area.
Will the links between postcodes and data zones be maintained?
The links between postcodes and data zones are needed to allow statistics to be produced for SNS. The intention is to continue to make the links available to SNS users. GROS will also maintain the relationships on the GROS Postcode Index.
What data will be made available at the data zone level?
Given confidentiality constraints are overcome, the intention is to provide data across all Scottish Neighbourhood Statistics topics at the data zone level. See www.sns.gov.uk for the current topics included. The Scottish Executive have recognised that not all data can be made available at the data zone level and have plans to develop an intermediate geography based on aggregations of data zones during 2004.
Are there any confidentiality issues with using data zones?
The Scottish Executive statisticians work to a Code of Practice which prevents us from releasing analyses which could identify specific individuals. We use several methods to prevent this, including rounding the results.
What other geographies will be supported in Scottish Neighbourhood Statistics?
The key geography for disseminating results will be the data zone, results will also be available at the Local Authority, Health Board and Scottish Parliamentary Constituency level. During 2004 the Scottish Executive will undertake a consultation on an intermediate geography that will be aggregated from data zones.
How do I get a copy of the data zone boundaries and look-up tables?
This information is now available from the Data Download section of the Scottish Neighbourhood Statistics website ( http://www.sns.gov.uk/Downloads/DownloadHome.aspx). Please choose the 'Download Geography' option, read and accept the license, then click the 'Get Geography Download' button. This will then present you with the data zone boundaries and look up tables which you require.