New analysis of sports facilities across English Local Authority Districts (LADs) shows that where there are more sports facilities per 10,000 people, a greater proportion of adults are active for over two and a half hours per week.

We looked at the relationship between the number of sports facilities per 10,000 people and the proportion of adults who exercised for over the government-recommended amount of exercise, using a method called the Pearson's Correlation Coefficient. The coefficient was 0.49, suggesting the two were correlated.

When grass pitches are excluded from the analysis, the relationship of sports facilities to activity is stronger still (Pearson's Correlation Coefficient 0.63). This suggests that the availability of grass pitches may have less impact on levels of physical activity in a local area than the availability of other types of sport facility.

It should be noted that this could relate to demographic and socio-economic differences between rural, suburban and densely populated areas.

Explore the interactive map to discover how sports facilities are distributed in your local area

Embed code

Download this chart


ONS analysis of sports facilities forms part of the Levelling Up Subnational Data project, our collaboration with the Department for Levelling Up, Housing and Communities (DLUHC). It is intended to support local policy making by providing more information on place-based disparities.

This analysis looked at the 15 local authorities with the highest and lowest number of publicly accessible sports facilities per 10,000 people.

It found that 14 of the 15 local authorities with the highest number of these facilities are predominantly rural.

Meanwhile, all 15 of the local authorities with the lowest number of publicly accessible sports facilities per 10,000 people are predominantly urban areas, including 12 London boroughs. The 12 local authorities with the highest percentage of sports facilities that are not grass pitches are all London boroughs.

Cities in England tend to have fewer sports facilities per 10,000 people than the surrounding areas. This is apparent in Birmingham, Bristol, Leicester, Lincoln, Norwich, Plymouth, Southampton, and Stoke-on-Trent.

Neighbouring disparities in access to sports facilities

The local authorities of Slough and neighbouring Windsor and Maidenhead are at opposite ends of the scale when it comes to sports facilities per 10,000 people excluding grass pitches. While Windsor and Maidenhead is ranked fifth (15.77 facilities per 10,000), Slough is positioned sixth from bottom of the list (5.62 per 10,000).

Concentration of sports facilities, excluding grass pitches, by region

Embed code

Download this chart


1. City of London and Hackney Local Authority Districts have been combined due to low population counts.
2. Isles of Scilly and Cornwall Local Authority Districts have been combined due to low population counts.

Supermarket locations are unevenly spread across the country

There is a difference in availability of supermarkets at UK, regional, and local authority levels.

The UK has an average of 2.70 chain supermarkets and 1.11 large chain supermarkets per 10,000 people. For UK countries, Scotland has the highest number of large chain supermarkets per 10,000 people at 1.44. England had the lowest number of large chain supermarkets per 10,000 people at 1.07.

Of the 15 local authorities with the fewest chain supermarkets per 10,000 people, more than half were in London. The fewest being in the London borough of Newham. Newham was also among the top 3% most densely populated English local areas at the last census in 2021.

London areas such as Enfield, Ealing and Barking and Dagenham also featured in the list of areas with the fewest chain supermarkets per 10,000 people. Most of these boroughs have relatively high levels of deprivation, as scored in the Indices of Multiple Deprivation.

The East of England has the largest range between local authorities, of large supermarkets per 10,000 people; values ranging from 0.37 in South Cambridgeshire, to 1.75 in Tendring. The region with the smallest range between local authorities is the North East of England.

Concentration of large food retail points by country and English regions

Embed code

Download this chart


1. Large supermarkets are classed as those over 280 square metres.
2. City of London and Hackney Local Authority Districts have been combined due to low population counts.
3. Isles of Scilly and Cornwall Local Authority Districts have been combined due to low population counts.

Explore how sports facilities, chain supermarkets, and museums are distributed in your area compared with others

Embed code

Download this chart


1. Channel Islands and Isle of Man are available in the data though not included in the map.
2. City of London and Hackney Local Authority Districts have been combined due to low population counts.
3. Isles of Scilly and Cornwall Local Authority Districts have been combined due to low population counts.

Where is there better access to museums?

Urban areas, in particular large cities, have fewer museums relative to their population than rural areas. Seven of the ten local authorities with the fewest museums per 100,000 people in the UK are London boroughs. Those local authorities with the higher proportion of museums tend to have small not-for-profit local history museums.

This pattern is also apparent in Scotland and Wales. In Scotland, the more densely populated central belt has fewer museums per 100,000 people than the surrounding Southern Uplands and the Highlands. In more densely populated South Wales, there are fewer museums relative to the population than Mid and West Wales.

Scotland has far more museums per 100,000 people than the rest of the United Kingdom at an average of 9.14 compared to the whole of the UK’s 4.94.

The Shetland Islands have the most museums per 100,000 people across the UK at 74.2. This is more than 14 times higher than the UK average and eight times higher than the Scottish average.

Of the 12 museums in the UK classified as ‘huge’, because they receive a million visitors or more annually, 10 are located in London boroughs, the remainder in Edinburgh, and Glasgow.

Most large and medium museums tend to be in cities, and in busy tourist hotspots with better public transport provision.

Future developments

The Office for National Statistics (ONS) continue to develop our subnational statistics. We plan to extend this analysis to a wider range of amenities within local communities at small geographical areas, including exploring methodology to produce distance and travel times to amenities.

The ONS encourage feedback from users on whether these statistics meet their needs. You can contact us at

Measuring the Data

Presenting Small Area Counts

In addition to Local Authority Districts (LAD), amenity counts for small area statistical geographies for supermarkets (UK) and sport facilities (England only) are presented. These counts are not comparable between different countries because of their differing population sizes.

England and Wales – Middle layer Super Output Areas (MSOA) from Census 2021, have a usual resident population between 5,000 and 15,000 persons.

Scotland – Intermediate Zones (IZs) from Census 2011, have a household population between 2,500 and 6,000 residents.

Northern Ireland – Super Data Zones (SDZs) from Census 2021, have a usual resident population between 1096 and 4077 persons.

Calculating Per Capita estimates

Per capita estimates for local authorities are calculated using the usual resident population estimates from the most recent census. These per capita estimates are not directly comparable between the different countries because of the non-consistent census dates that the population estimates were taken on.

England and Wales: 21 March 2021 – Lower Tier Local Authorities

Scotland: 20 March 2022 – Council Area

Northern Ireland: 21 March 2021 – Local Government District

Isle of Man: 30 to 31 May 2021

For the Channel Islands, we combine the resident populations of Guernsey: 31 March 2021 and Jersey: 21 March 2021.

Because of low populations estimates we have combined the Isles of Scilly local authority with Cornwall, and the City of London local authority with Hackney when presenting per capita estimates in the article. Individual estimates are presented separately in the accompanying data tables.

In addition, per capita estimates at MSOA for sport facilities in England are calculated using population estimates from the 2021 Census for England and Wales. The ONS plan to include per capita estimates for supermarkets at smaller geographical areas following the release of Intermediate Zone population estimates from the Scottish Census.

Sports Facilities

Data about Sports facilities in England are from the Sport England Active Places data on 5 October 2023. Only facilities listed as operational or temporarily closed are included.

Sport England manage and maintain the sports facility database Active Places. The database records 15 facility types with over 200 attributes used to describe a facility and its associated site. At present the database holds information on approximately 115,000 sports facilities, located at approximately 41,000 sites.

The database is updated daily with each facility being audited on a rolling annual basis.

The analysis refers to whether a facility is publicly or privately accessible. Whether a facility is classified as publicly or privately accessible depends on which access type is the most frequent access type by hours. The publicly accessible category includes facilities that are free for public access, pay and play, Sports Club or Community Association, and Registered Membership use.

Facility counts do not consider the properties such as size, capacity, attractiveness, accessibility, or cost.

Engagement in sport

Data about engagement in sport are from the Sport England Adult Active Lives Survey. Data covers the November 2019 to November 2021 period, and 122,403 adults were surveyed. For further information on the survey, please see the Sport England methodology.

Estimates of weekly duration of physical activity, club membership, inclusive and exclusive of fitness clubs, attendance of sports events and whether people feel they have opportunities to participate in physical activity are presented.

We examined these variables for males and females. The data from Sport England also included ‘other’ as a sex category but due to statistical disclosure rules are not reported.

The Adult Active Lives Survey data required two separate weightings. The ‘wt_final’ was used when the question was asked to everyone who had taken part in the survey. This applies to variables relating to duration, opportunity, and sport event attendance. The ‘wt_final_B’ weighting was used for questions asked to the online group B participants and applies to the club membership variables.

The differences in sports facility engagement could be due to preferences regards the type of facility itself or other factors pertaining social, economic and demographic differences between densely populated areas and suburban or rural areas where grass pitches are more prevalent. Due to statistical disclosure rules, counts below 10 are suppressed. To minimise these instances the Likert scale used to report opportunity to be active has been aggregated from a 5-point Likert scale (Strongly Agree – Strongly Disagree) to a 3-point Likert scale (Agree – Disagree).


Data about chain supermarkets in the UK are from the Geolytix Supermarket Retail Points project. For this analysis we have used the August 2023 dataset of 18,108 supermarkets. The complete list of included supermarket retailers and their categorisation can be accessed via the user guide. This data is complemented by 52 stores from selected food retailers not in the Geolytix dataset.

The analysis refers to supermarket size. Large supermarkets are those over 280 square metres, for which Sunday trading hours legislation apply in England and Wales and Northern Ireland. There are no trading hours restrictions in Scotland.

We recognise that not all food retailers are included in this dataset. Notable exceptions are most independent food retailers, including those part of a symbol group. Whilst acknowledging the limitation this has on the dataset, we emphasise that all major chain retailers are included.


Data about museums in the UK are from the Mapping Museums project produced by Birkbeck, University of London. The database was released on the 30 September 2021, and we only include the 3,349 museums listed as open as of this date. For the purposes of this study, a museum is classed as closed when there are no regular opening days or hours.

For conciseness, data has been aggregated by governance type and subject matter.

The analysis refers to a museum’s accreditation status and size. We apply the accreditation and size classifications used by the Mapping Museums project.

View all data used in this article


Subnational Statistics and Analysis Division