Guidance

Road lengths in Great Britain statistics: Methodology and quality note

Updated 21 March 2024

Applies to England, Scotland and Wales

This methodology and quality note relates to the road lengths statistics published on the Department for Transport website.

For further information on these statistics please email road length statistics.

Section 1: Background to the statistics

Overview of road length statistics

The Department for Transport (DfT) publishes road length statistics in Great Britain on an annual basis. The data is based on a snapshot of the road network as of April each year.

These official statistics were independently reviewed by the Office for Statistics Regulation in February 2013. They comply with the standards of trustworthiness, quality and value in the Code of Practice for Statistics and are labelled ‘accredited official statistics’.

The statistics are based on information from a range of sources including local authorities, the Scottish Government, the Welsh Government, National Highways (formerly Highways England) and Ordnance Survey (OS).

Road lengths are categorised by road types, incorporating major and minor roads and also whether the roads are in an urban or rural location. Major roads are also split into trunk roads which are centrally managed, and principal roads, which are managed by local authorities. A full list of road categories is available in the associated notes and definitions. Road lengths are presented for Great Britain as a whole, and are also available by country, region and local authority.

Data sources

The OS MasterMap Highways Network (Highways) provides a central database containing a variety of transport data, including road lengths data and is published monthly. Highways contains data from the National Street Gazetteer (NSG) and utilises OS geography data.

The NSG contains road lengths data on all roads in England and Wales and is maintained by GeoPlace. This data is obtained directly from local authorities who maintain their own Local Street Gazetteers (LSG), and from national and regional highways authorities.

Road length data is first recorded by local authorities in their LSG and then submitted to GeoPlace. GeoPlace then publish the NSG at the start of the next month and send their streets data to OS. OS then incorporate this data into their Highways product which is released in the following month. This process is represented visually in the diagram below (see Diagram 1).

Diagram 1: A flow chart showing the process of data travelling from local authorities to DfT via the NSG and Highways

DfT Road length estimates are generated so that the data is accurate as of April each year. As there is a delay between the data being reported by local authorities and becoming part of Highways, the May release of Highways is used.

Since 2018, OS’s MasterMap Highways Network dataset has been used in place of OS’s Integrated Transport Network (ITN). This coincides with the withdrawal of OS’s ITN product and migration to Highways that took place on 31 March 2019. However, the data used from OS Highways has historically been based on the route hierarchy field which is equivalent to the classification used in the ITN. This field does not allow differentiation between ‘C’ and ‘U’ Roads, therefore these figures have not been presented separately unless this information is collected from local authorities in an R199b consultation.

References to ‘Ordnance Survey data’ in the rest of this document relate to this data in Highways (for 2018 onwards) or ITN (for previous years).

R199b consultation  

The department undertook an R199b consultation in 2022 in which road length information for all local authorities in England was collected. The data from the exercise may be used by the Department for Levelling Up, Housing and Communities (DLUHC) (formerly MHCLG) in calculating future Local Government Finance Settlements and in devising government policy over the funding of local government. The data from the exercise was also used most recently by DfT to assist with the production of the 2022 Road Lengths in Great Britain statistical release. In years in which the R199b consultation is not run, data from the most recent R199b is updated with data from Ordnance Survey to calculate the length of minor roads, as explained in the minor roads section below.

Previously, road length information has been used for funding allocations made by DLUHC and DfT. Due to changes to the system of Local Government Funding introduced by DLUHC (then DCLG) in the financial year ending 2014 and the stable nature of the statistical series, it was decided that the length estimates of roads maintainable at public expensive used in this process did not need to be updated each year. Due to the limitations in the attributes of the OS data, ‘C’ and ‘U’ road length can only be estimated when an R199b consultation is carried out.

The length of ‘C’ and ‘U’ roads is combined into a single figure in 2013, 2014, 2016, 2017, 2019, 2020, 2021 and 2023, when an R199b was not completed. The frequency and nature of future consultations has depended on the needs of DfT and DLUHC, but improvements have been made to OS data in recent years which will allow the department to use the road classifications recorded. Therefore, the department aims to publish road length estimates for each road classification every year when these improvements have sufficiently increased the accuracy of the data.

Major roads

The lengths of major roads (motorways and ‘A’ roads) are obtained from the Major Roads Database (MRDB), maintained by DfT, using information from Ordnance Survey data, local authorities, the Scottish Government, the Welsh Government and National Highways (formerly Highways England). The MRDB shapefiles are available to download.

The MRDB contains a series of links which are unique sections of road and together make up all of the major roads in Great Britain. This database contains various fields of information about these links, including link length. As this database contains links which cover all major roads in Great Britain, the length of major roads can be estimated by summing all of the relevant links to give a total road length for a local authority area, a particular road type, or the whole country.

The road length estimates are fixed as of April each year. Only those links in the database which were recorded as active/open on 1 April are included when calculating our estimates each year. This prevents any links which have been closed over time, or opened since this date, from being included in these estimates.

Major road estimates produced from the MRDB are split by road classification and management. Minor roads are split only by classification as they are all principally managed (see Diagram 2).

The diagram below shows the road types that make up trunk and principal, and major and minor roads. Trunk roads are always major roads and are referred to in these statistics as trunk motorways and trunk ‘A’ roads. Principal roads can be major roads, referred to as principal motorways and principal ‘A’ roads, or they can be minor roads, referred to as ‘B’ roads, ‘C’ roads and ‘U’ roads.

Diagram 2: A diagram presenting the road types in Great Britain and their management

Minor roads

Road length information for minor roads, further classified as ‘B’, ‘C’ and ‘U’ roads, are typically obtained from Ordnance Survey data, local authorities, the Scottish Government and the Welsh Government. Generally, the total minor road length is calculated by summing the full length of roundabouts and single carriageways, and then half the length of slip roads, dual carriageways, traffic island links, traffic island links at junctions and enclosed traffic area links. These calculations are undertaken so that road length statistics can be combined with figures from traffic count points to produce traffic statistics. This process is shown in the formula below.

Formula 1: The formula used by DfT to calculate minor road length estimates from the Ordnance Survey data

The image above shows the formula used to estimate the road length held within OS data. Half of the length of slip roads, dual carriageways, traffic island links, traffic island links at junctions and enclosed traffic area links is totalled and added to the full length of roundabouts and single carriageways.

In years where an R199b consultation does not take place (including 2023), for English local highways authorities, minor road estimates are produced taking data from the previous year and multiplying by a growth factor derived from changes in road length in OS data over the previous year. For Scottish and Welsh authorities, estimates are based directly on OS data.

In years where an R199b consultation takes place, minor road lengths for each authority in England are initially calculated using the information held by DfT and OS data. These are then sent to local authorities and may be changed where the local authority can provide evidence that these lengths are incorrect.

‘C’ and ‘U’ roads

In years where an R199b consultation does not take place, the total length of ‘C’ and ‘U’ roads for each local authority in Great Britain is calculated by summing the ‘minor roads’ and ‘local roads’ categories from the OS data. These categories do not allow ‘C’ and ‘U’ roads to be separated, hence in these years the figures have been combined for ‘C’ and ‘U’ roads.

In the 2022 R199b, the road classification field was used to produce initial ‘C’ road estimates which were sent to local authorities in England to validate. Previously, the ‘C’ road figures from the last R199b would be used as the OS road classification field for ‘C’ roads and ‘U’ roads was incomplete.

Minor roads figures in 2013, 2014, 2016, 2019, 2020, 2021 and 2023 have been derived with ‘C’ and ‘U’ roads combined, as no R199b road length consultation with local authorities took place.

Urban or rural breakdown

Urban areas for England and Wales are defined as built-up areas with a population of at least 10,000, and for Scotland, a population of at least 3,000. All other locations are rural areas. The urban areas definitions used in the production of these statistics are set by the Office for National Statistics (ONS) for England and Wales, and by the Scottish Government for Scotland.

Since 2017, the department has used an improved methodology to calculate urban and rural road lengths so that urban and rural road length estimates are calculated consistently for both major and minor roads. The urban and rural definition for England and Wales used urban polygon data from ONS based on the 2011 census. The Scottish government provides the urban and rural definition for Scotland and this is now based on population data from 2020. The process for producing urban and rural road estimates involves matching data from OS to the urban polygon data with roads in urban areas being classed as urban roads and all other roads classed as rural roads. Prior to 2017, this approach was only used for minor roads, whereas for major roads (sourced from the department’s MRDB) each road section would either be assigned as ‘urban’ or ‘rural’ depending on where the majority of its length fell. This resulted in less precise estimates, as a large number of major road sections would span both urban and rural areas.

The 2023 statistics for urban and rural roads in Scotland used the 2020 urban and rural definition for Scotland (the latest version available). The 2022 statistics were produced using the 2020 urban and rural definition for major roads whilst the minor roads figures continued using the 2016 definition. The 2020 and 2021 road length statistics for Scotland were calculated using the 2016 urban and rural classification.

As the population data is not updated annually, accurate population information may not be available for the areas where newly built or adopted roads are located, limiting the ability to accurately classify these roads as urban or rural. When urban rural definitions are updated, a step change may be seen in the amount of road classified as urban or rural. Due to this, year-on-year comparisons of urban and rural classifications should be looked at with caution and instead longer term trends should be observed.

Local highway boundary changes

From April 2023, the former county of Cumbria was separated into 2 unitary authorities. The first unitary authority known as Cumberland consists of the districts of Allerdale, Carlisle, and Copeland. The second unitary authority known as Westmorland and Furness consists of the districts of Barrow-in-Furness, Eden, and South Lakeland.

When a change occurs, the existing road length data for all LAs affected are firstly aggregated. The proportion of road length within the new boundaries is then estimated and adjustments are applied to existing aggregated estimates to produce road length estimates for the LAs within the newly defined boundaries. The process used to separate Cumbria into the 2 new authorities is the same process that was used in 2021 to separate Northamptonshire into West Northamptonshire and North Northamptonshire.

There will be a step change in the reported figures for the affected LAs, but it is important to note that the aggregated data of the LAs affected can still be compared historically as the geographies will be identical at this level. The local authority boundary changes do not impact the regional or national totals as the old and new authorities are still part of the same region.

The map below shows the boundary of the former local authority Cumbria on the left and the boundaries of the newly formed Cumberland and, Westmorland and Furness on the right.

Map 1: A map showing the former local authority Cumbria and the local authorities Cumberland and, Westmorland and Furness which replaced it

Prior to the R199b in 2023, there were other changes made to local authority boundaries. The paragraphs below will detail changes made since 2019. As these changes were made prior to the R199b in 2022, these local authorities have had the opportunity to provide the department with information evidencing the road length owned by their authority. As a result, the boundary changes outlined below should have minimal impact on the published statistics from 2022 onwards.

From April 2019, Poole Unitary Authority (UA), Bournemouth UA and Christchurch District Council (previously part of Dorset County Council) have joined together to form a new Unitary Authority known as Bournemouth, Christchurch and Poole (BCP) UA. Due to this, Dorset has also ceased to exist as a County Council and formed a new Unitary Authority by combining its remaining district councils, Weymouth and Portland, West Dorset, North Dorset, Purbeck, and East Dorset.

From April 2021, the former county of Northamptonshire was separated into 2 unitary authorities. West Northamptonshire UA consists of the districts of Daventry, Northampton Borough, and South Northamptonshire. North Northamptonshire UA consists of the districts of Corby, East Northamptonshire, Kettering and Wellingborough.

Data and methodological improvements

A small update to the methodology was announced in the ‘Road Lengths in Great Britain 2017’ statistical release, published in July 2018, which changed how roundabouts are treated from 2018 onwards. This outlined that the full length of the road link for roundabouts are included in the road length formula for the published statistics. This will have a relatively minor impact overall on the road length estimates. In 2017 there were some small improvements to the methodology used to estimate road lengths, particularly for major roads. These generally meant that the 2017 figure is slightly more accurate than those produced in previous years.

There are step changes in the minor road length figures in 2004 and 2006 due to changes in the base data. In 2004, amendments were made to the data for private roads in Scotland which had been incorrectly recorded as public roads. From 2006, minor road length estimates were made using the Integrated Transport Network (ITN) dataset rather than the Ordnance Survey Centre Alignment of Roads (OSCAR) dataset. This change in methodology led to an increase in minor road lengths due to the greater accuracy of ITN. While road length estimates used the OS MasterMap Highways Network product from 2018 onwards rather than the ITN, the route hierarchy field in Highways was used which is equivalent to the classification in ITN.

Since transitioning to using the OS MasterMap Highways Network product in 2018, improvement works to the quality of this data have been ongoing. An anticipated methodology change in the future is that the data produced by OS will fully replace the R199b consultation.

Section 2: Quality assessment

Relevance

Uses of data

Road length statistics underpin many statistics produced by DfT. These include road accident and safety statistics, road traffic statistics and road condition statistics. This data is also used by the department and DLUHC in local authority funding calculations, including road maintenance grants.

Meeting users’ needs

Road length statistics provide measurements of the physical length of roads maintained at public expense in Great Britain. This data is clearly defined and the definitions for classifications within these measurements are set out in the accompanying information documents published alongside the statistics.

This publication provides total road length estimates at a national level dating back to 1951. It also provides estimates for a number of years between 1914 and 1951. Local authority and regional breakdowns of the estimates are available from 2005 onwards. It is intended that this time series will be continually reviewed to consider user needs. Estimates are broken down by multiple classifications to allow a wide range of user needs.

Accuracy and reliability

Data process

The road length statistics produced by DfT are a combination of data from multiple sources. Local authorities provide GeoPlace information on the roads that they own or manage via the NSG. The NSG undergoes set data validation processes before this information is sent to OS. OS will then use this information alongside other data sources to match the geographies provided by GeoPlace. Where there are any mismatches in the data, these are flagged and investigated further. This means that the department are using information that has been verified multiple times before it reaches the department. OS are a long-standing and well-respected national mapping agency, and their data is widely used and trusted. There are annual improvement plans and the department works closely with both bodies to ensure continuous improvement and maintenance of the information.

In years where an R199b consultation is performed, the data is further validated directly by local authorities. This gives them the opportunity to provide evidence for any data they suspect to be missing from the estimates we hold which further increases the accuracy of our statistics. This information is also passed onto GeoPlace and OS to ensure accuracy of all networks used in the process.

Road length statistics are based on estimates and so there is some uncertainty associated with this release. The geographic surveying tools used by OS are highly accurate but will have a very small amount of measurement errors. The road classification information in Ordnance Survey data is not always up-to-date, and in particular, does not distinguish robustly between ‘C’ and ‘U’ roads. As explained above, the department mitigates this by collecting data from local authorities as part of the R199b road length consultation. However, local authorities use different computer systems to map their local highways, and so data from different local authorities will contain some relatively small inconsistencies.

Major road lengths are provided from the MRDB, which is managed internally, with the length of road links summed to calculate estimates. The length of each link is calculated by summing the length of the relevant ‘TOIDs’ from OS data. The length of each link is rounded to the nearest hundred metres. ‘TOIDs’ are provided with lengths rounded to the nearest centimetre. These TOIDs generally represent smaller sections of road than MRDB links. Minor road lengths are provided from OS data with the length of each ‘TOID’ is summed to calculate estimates. As each ‘TOID’ and each MRDB link are rounded, the estimates calculated from these will have a level of rounding error.

Urban or rural definitions

Urban or rural definitions can vary depending on devolved policies. These classifications are set by ONS for England and Wales and by the Scottish Government for Scotland. Urban areas for England and Wales are defined as built-up areas with a population of at least 10,000, but for Scotland, an urban area requires a population of at least 3,000.  This difference in what is considered urban or rural makes national level figures less comparable, but since 2017, the Scottish definition has been updated to make it more in line with the rest of Great Britain.

The current urban or rural definitions are based on populations data from 2020 for Scotland and 2011 for England and Wales.

Timeliness and punctuality

Road length statistics are usually published within a year of the time period they refer to. This allows improvements to be implemented in the MRDB. The MRDB is managed by DfT and each year improvements are made as a result of new information from traffic counts performed on behalf of the department. Road lengths tend to undergo a steady increase year-on-year with a consistent low rate of new roads built or adopted by local authorities. Therefore, despite the time delay between the data cut and publication, the rate of change of length estimates means that this has a small impact.

Regarding punctuality, we adhere to the requirements of pre-announcement of publication.

Comparability

Standardised data

The data in this publication has been published in the same format since 2005. Data has been published annually since. The definitions of the 5 road classifications are consistent over time but there are small updates to urban and rural classifications. New local authorities are created periodically, generally as a joining or separating of existing lower tier authorities. This results in breaks in the time series for some authorities. ONS local authority codes are used as standard throughout, which allows this information to be linked to other local authority data.

In most years, ‘C’ and ‘U’ road lengths are presented as a total. In years where the R199b consultation is run, data for these road types are reported separately. This means that there are 2 time series; combined ‘C’ and ‘U’ road data, and separate ‘C’ and ‘U’ road data, with the combined data having a more consistent time series.

Difference to other publications

This publication refers to roads that are managed by regional or national authorities as trunk roads and roads that are managed by local authorities as principal roads. This differs to road traffic publications which refer to the Strategic Road Network (SRN) instead of trunk roads. The SRN is a definition by National Highways for England, but since we report data for all of Great Britain we use the more universal terms of trunk and principal.

The Welsh and Scottish governments also publish their own road length statistics. The figures included will differ from these statistics as different methodologies are used to produce the estimates.

Coherence

The information within this report has consistent definitions throughout Great Britain, with the exception of urban and rural roads in Scotland. The difference in this definition is clearly stated and the year it was updated has been given.

Accessibility

This publication complies with GOV.UK and Government Statistical Service requirements for accessibility. Statistical releases are published in HTML format and data tables are available in ODS.

Clarity

The published tables maintain consistent terminology and the format varies minimally between publications. Language in the statistical release is chosen to be plain and clear whilst retaining technical terms.

Trade-offs

Road condition statistics are also produced by the team. These statistics have a wider audience and so there is a greater need for timely release. As a consequence, road length statistics are published later in the year. In addition, road length statistics are used primarily to feed into statistical calculations and budget allocations, rather than to make timely decisions, and so accuracy is prioritised over timeliness. The time between data collection and publication is used to validate the data, with thorough checks and improvements carried out to result in robust and trustworthy statistics.

Future road length statistics publications

The department hopes that the improvements made to OS data in recent years will remove the need for an R199b road length consultation. The department intends to use the OS MasterMap Highways Network as the single source of road length information. This will ensure efficient and timely delivery of statistics, alongside more comparable data across Great Britain.

If you have any questions, please email road length statistics.

Instructions for printing and saving

Depending on which browser you use and the type of device you use (such as a mobile or laptop) these instructions may vary.

You will find your print and save options in your browser’s menu. You may also have other options available on your device. Tablets and mobile device instructions will be specific to the make and model of the device.

How to search

Select Ctrl and F on a Windows laptop or Command and F on a Mac.

This will open a search box in the top right-hand corner of the page. Type the word you are looking for in the search bar and press enter.

Your browser will highlight the word, usually in yellow, wherever it appears on the page. Press enter to move to the next place it appears.