SUBJECTS ON THIS PAGE:
Government data sources do not use a consistent rural/area classification system. Broadly speaking, the sources used by this website can be divided into four groups:
- Classification system is available, whereby all the data is classified according to the Government’s 2004 classification system for small areas (see below). However, only two sources provide this, namely the British Crime Survey and the English Housing Survey.
- Classification system is available in principle but not in practice, either because the relevant data in the publicly available dataset is suppressed for supposed data confidentiality reasons (e.g. the Labour Force Survey and associated Annual Population Survey) or because the method of classification is neither clear nor documented (e.g. the Health Survey for England).
- A less than optimal classification system is derivable: many sources provide data by local authority and this can be used to analyse rural/urban differences using the DEFRA 2009 classification system (see below).
- No classification system is possible: where data is only obtainable via Government publications rather than from the underlying detailed data, it is usually not possible to undertake any rural/urban analyses.
In this context, the indicators in this website define ‘rural’ areas in a variety of different ways, with the selection determined by data availability.
The government’s preferred level of rural/urban classification is at a small area level. Its 2004 classification system for small areasclassifies a small area as ‘urban’ if the majority of its population live within settlements of more than 10,000 people and as ‘rural’ if this is not the case. Within the rural category, there is a further subdivision into ‘town and fringe’, village’ and ‘hamlets and isolated dwellings’, giving an overall 4-way classification system. 1
In principle, all the national survey datasets could include the small area rural/urban classification system, based on each household’s address. In practice, however, this is not typically done, with the only overt exceptions being the British Crime Survey and the English Housing Survey 2.
Rather, what many data sources record is the lower tier local authority (district councils, unitary authorities, metropolitan boroughs and London boroughs), often referred to as local authority district. Using the DEFRA 2009 classification system 3, each lower tier authority is classified as one of:
- ‘Very rural”: if 80% or more of their population live in either rural settlements or market towns, where a ‘rural settlement’ is any settlement of less than 10,000 people and a ‘market town’ is a settlement of between 10,000 and 30,000 people which provides certain functions and services to its wider rural hinterland. 4
- ‘Mostly rural”: if between 50% and 80% of their population live in rural settlements or market towns. 5
- ‘Part rural”: if between 26% and 50% of their population live in rural settlements or market towns. 6
- ‘Major urban’: if not any of the above but either at least 50% or at least 100,000 of their population live in an urban area with a total population of 750,000 or more.
- ‘Large urban’: if not any of the above but either at least 50% or at least 50,000 of their population live in an urban area with a total population of 250,000 or more.
- ‘Other urban’: if not any of the above.
In most of the graphs on this website, the ‘major urban’, ‘large urban’ and ‘other urban’ districts are then amalgamated into a single ‘urban’ category.
Where it is not possible to use lower tier local authority designations, upper tier local authority classifications have sometimes had to be used. This is, for example, the case for social service statistics given that social services is an upper tier (e.g. county) rather than lower tier (e.g. district) responsibility. Each upper tier authority is classified as: ‘very rural’, ‘mostly rural’, ‘part rural’ or urban using the same definitions as above (but with the three urban categories all grouped together).
These differing classification systems illustrate that there is no single ‘correct’ rural-urban classification. In fact, there are at least three types of decision involved in any classification: 7
- ‘Level of magnification‘: as discussed above, the ideal would be small area classifications but most data is only classified at the lower tier local authority level.
- ‘Method of identification‘ – i.e. how to decide whether an area should be classified as rural or urban. The size of the settlement is the main starting point, but, as per the discussion above, the lower tier local authority classifications involve a variety of other rules.
- ‘Strictness of definition‘ – i.e. where to set the cut-off point between rural and urban. The lower tier local authority classifications involve a cut-off point which results in two-thirds of the population being in the urban categories and one-third being in the various rural categories.
Another implication of the discussion above is that urban/rural classifications are determined by the classification of the majority of the population in the area and there may well be a significant proportion of the population in the area who are in the opposite classification. So, for example, a ‘rural area’ may well contain many ‘urban people’ and vice versa. This inevitably tends to lessen any observed differences in statistics between types of area.
Lower tier local authority classifications (districts)
As listed in the DEFRA 2009 classification system.
|'Very rural'||'Mostly rural'||'Part rural'||Urban|
Forest of Dean
Isle of Wight
Isles of Scilly
Central Bedfordshire 9
Cheshire East 10
East Riding of Yorkshire
King's Lynn and West Norfolk
Newark and Sherwood
North East Derbyshire
North West Leicestershire
Tonbridge and Malling
Vale of White Horse
Basingstoke and Deane
Bath and NE Somerset
Cheshire West and Chester 15
Epping Forest 19
Hinckley and Bosworth
Mole Valley 21
Redcar and Cleveland 22
'Major urban'Barking and Dagenham
City of London
Epsom and Ewell
Hammersmith and Fulham
Kensington and Chelsea
Newcastle upon Tyne
Brighton and Hove
Oadby and Wigston
Blackburn with Darwen
North East Lincolnshire
Nuneaton and Bedworth
Reigate and Banstead
Telford and Wrekin
Weymouth and Portland
Windsor and Maidenhead
Upper tier local authority classifications
Where the upper tier authority is also a lower tier authority (i.e. unitary authorities, London and metropolitan boroughs)
The allocation to ‘very rural’, ‘mostly rural’, part rural’ and ‘urban’ is as per that for the lower tier classifications above.
Where the upper tier authority is a grouping of lower tier authorities (i.e. a county)
|‘Very rural’||‘Mostly rural’||‘Part rural’||Urban|
|Cornwall and The Isles of Scilly||Cambridgeshire
In practice, these allocations mean that the only upper tier authority that is classified as ‘very rural’ is Cornwall and The Isles of Scilly because all of the ‘very rural’ lower tier authorities are combined with other authorities at the upper tier level. For example, the ‘very rural’ lower tier authorities of Craven, Hambleton, Richmondshire, Ryedale and Selby are combined with the ‘part rural’ Harrogate and Scarborough to form the ‘mostly rural’ North Yorkshire upper tier authority. Because of its consequent very small population, ‘very rural’ is therefore not shown on the any of the upper tier graphs.
Graph 3: Retirement-age over time
Rural/urban analyses can be undertaken using most of the major surveys, the main limitation being that the Labour Force Survey does not have any rural/urban codes and its substitute for work-related statistics – the Annual Population Survey – only goes back to 2004. The table below provides some summary information about some of these sources.
|Subject area||Data source||Rural/urban classification||Earliest data|
|Income||Households Below Average Income dataset||Districts||1994/95|
|Work||Annual Population Survey||Districts||2004|
|Work and Pensions Longitudinal Study, DWP||Districts||2000|
|Annual Survey of Hours and Earnings||Districts||2002|
|Geographical Analyses, HM Revenue & Customs (tax credits)||Districts||2003|
|Education||English National Pupil Database||Districts||2002|
|Department for Education publications||Upper tier local authorities||Varies by subject|
|Health||Key population and vital statistics, ONS||Districts||1998|
|Conception statistics, ONS||Districts||2001|
|Mortality Statistics Division, ONS||Districts||1991|
|Housing||Stock data from the English Housing Survey||Small area urban/rural classifications||2006 29|
|Household data from the English Housing Survey||Small area urban/rural classifications||2008/09 30|
|DCLG Statutory Homelessness Statistical Releases||Districts||2003|
|Services||Family Resources Survey||Districts||1994/95|
|Commission for Social Care Inspection performance assessment framework (PAF) reports||Upper tier local authorities||2004/05|
|Social cohesion||British Crime Survey||Small area urban/rural classifications||2000|