EPRSC Successful Suburban Town Centres Project
 

Data Description
Description of data used in the suburban profiler

This page gives details of the classifications and data sources used to create the suburban town centres profiles

The data presented on the website have been categorsied into two distinct groups. The first been a random sample of 20 town centres located in London's Outer Suburbs. The second a group 6 slightly larger town centres located both within or near to the edge of the outer suburbs, this group is called the control group.


Socio-economic Data

  1. Infrastructure maps
  2. Car ownership
  3. Commuting patterns and method of travel to work
  4. Functional activity maps
  5. Social economic classification of occupation maps

Space Syntax Data

  1. Axial Integration
  2. Segment angular choice
  3. Segment angular integration
  4. Combined integration-choice

| Back | Top of page | Print this page |

Social Economic Data
Infrastructure maps

These maps illustrate the location of the transport infrastructure in each town centre, The information used to create the infrastructure maps is derived from 2 Ordnance Survey products. Meridian was used to classify the road, rail lines and tube and rail stations. Address Layer 2 was used to extract the locations of car parks. Data used was from 2007.

| Back | Top of page | Print this page |

Car ownership

These maps show the percentage of households within each output area that have no car or van, one or more cars or vans, two or more cars of vans. The car ownership data was derived from the 2001 census. The census collated information on car or van ownership per household. The results of which were aggregated to the output area which is the smallest level of geography that data is released to the public and researcher. The percentage of the various car ownership variables were calculated for each one within the 15 minutes walking neighbourhood of our suburbs.

The resultant maps are created using a diverging colour scale, where the scale has been classified according to various other areas such as; London, London's outer suburbs and the south east. A diverging scale is one where both ends of a data distribution are potentially of interest. The break point values show clusters in the data that are above or below such points. The blue colours represent low percentages and the reds high percentage values. The premise behind classifying the maps like this is to enable the user to compare the results for a local area to other larger regional scales.

| Back | Top of page | Print this page |

Commuting patterns and method of travel to work

The commuting patters were derived from 2001 Census data and provided to the project via the Census Interaction Data Service (CIDS).The analysis uses the journey-to-work flows within and between output areas in England and Wales, for total number of commuters per output area.(Source: 2001 Census: Special Workplace Statistics (Level 3)).

It is worth noting that the data has been adjusted in accordance with the ONS disclosure control mechanism. Output areas with very low numbers (between 0 to 3) are adjusted. Output areas with an initial value of 1 have been rounded to either 0 or 3, with 0 being the more likely result. Cells with an initial value of 3 have also been rounded to either 0 or 3, but with 3 being the more likely result. Cells with initial values of either 0 or 3 have retained these values although in each case it is impossible to distinguish between rounded values and 'genuine' 0s or 3s. All sub-totals and totals have been re-calculated from the rounded cells. Further information is available from the following reference: Office for National Statistics, 2001 Census: Special Workplace Statistics (Level 3) [computer file], ESRC/JISC Census Programme, Census Interaction Data Service, University of Leeds and University of St. Andrews

Commuting patterns- inflow maps represent the number of people per output area who are commuting into output areas within the sample town centres and its surrounding walkable neighbourhood. It highlights the town centres as places of work.

Commuting patterns - outflow maps represent the number of people per output area who are commuting out of output areas within the sample town centres and its surrounding walkable neighbourhood. It shows the places of work for the residence of the sample town centre.

The walkable neighbourhood was approximated by drawing an 800m buffer around the official Department of Communities and Local Government (DCLG) town centre boundaries and finding output areas that intersect with that buffer.

Method of travel to work - outflows . These maps were derived from the 2001 Census table UV39. The information shows the usual resident population aged 16 to 74 by their method of travel to work. The method of travel to work is for the longest part, by distance, of the usual journey to work. This information enables us to determine which method of transport (if any) is used by the residents of the sample centres to get to their place of work. Two sets of maps have been created: percentages and counts.

Method of travel to work - inflows These maps were derived from the 2001 Census table UV37. The table shows the daytime population aged 16 to 74 by the method of travel to work. The day-time population is defined for people aged 16 to 74, as those people who live and work in the area (or do not work) and those people who live outside the area and work inside the area. 'No fixed place of work' is counted as if working in the area. The method of travel to work is for the longest part, by distance, of the usual journey to work. Two sets of maps have been created: percentages and counts.

| Back | Top of page | Print this page |

Functional activity maps

The information used to create the functional activity maps is derived from an Ordnance Survey product: Address Layer 2 (AL2) 2007. AL2, an Ordnance Survey product, was used to identify 4 major groups of activity generating land uses occurring in and around the town centres: retail, industry, community services and offices and general commercial.

Address Layer 2 is a product derived primarily from the royal mail postal address file (PAF), whereby each postal address (delivery point) has been allocated a unique reference and national grid reference. Supplementary to these data Address Layer 2 incorporates information about geographically derived address locations that do not have specific postal addresses. The resultant dataset contains address locations (of which there are over 29 million in Great Britain (England, Scotland and Wales) which have a corresponding classification; residential or commercial. This classification is further broken down into more detailed classifications which were sourced from 3 different data records.

  • 1. Allocation by OS field surveyor (OS base function), of which there are 1500 categories;
  • 2. Special category code by valuation office agency (SCAT), of which there are 400;
  • 3. National Land Use database of which there are 13 Land Use order descriptions which are further categorised into 41 more detailed Land Use Group descriptions.

| Back | Top of page | Print this page |

Functional activity surface maps

The functional activity surface maps have been created using the landuse activity that was assigned to postcode point location, derived from Address Layer 2 data. A technique called kernel density estimation (KDE) was then used to create a density surface of the point locations for the different types of land use. KDE is most commonly used to estimate population density and diseases or the incidence of crime. In this study the process was used to create a probabilistic density surface of different types of land uses, based upon the point locations of postcodes and their corresponding landuse classification. The reason for doing this was to explore the presence of any inherent spatial pattern, and to produce a generalised surface that could be used for visualisations with the Space Syntax maps.

One limitation of this method which must be noted, is that the surfaces do not predict sizes or concentrations of employment, because that attribute data was not available. The surfaces merely present a generalised view of the types of landuses that are occurring and the concentrations of postcode locations to consider the question of where they are situated.

Social economic classification of occupation maps

The National Statistics Socio-economic Classification (NS-SeC) was introduced by the Government to replace Social Class based on Occupation (also known as the Registrar General’s Social Class) and Socio-Economic Groups (SEG)

Social economic classification of occupation - residents (outflows) The table shows the usual resident population aged 16 to 74 by their socio-economic classification. It represents the number of resident people per occupation classification for output areas within the sample town centres and its surrounding walkable neighbourhood. The data is derived from the 2001 Census tables UV31.

Social economic classification of occupation - workplace (inflows) The table shows the workplace population by the National Statistics Socio-economic classification (NS-SeC).The workplace population is defined as the people aged 16 to 74 who are in employment and whose usual place of work is in the area. People with no fixed place of work are treated the same as people who work mainly at or from home and are counted as working in their area of residence. The maps represent the number of people per output area who are commuting into the output areas within the sample town centres and its surrounding walkable neighbourhood.

| Back | Top of page | Print this page |

Space Syntax Data Description
Axial Integration

The axial map is used in space syntax analysis to represent and analyse all open public space as a continuous spatial network in order to measure how well connected each street space is to its surroundings. This is done by taking an accurate plan of a built up area and drawing the set of least and fewest lines that cover all the open space ensuring that lines intersect where adjacent spaces are contiguous.

Space syntax analysis computes all the lines in the network according to their relative depth from each other. Depth increases with the number of changes of direction between lines. The terminology used to describe the shallowness or depth of a space in relation to other spaces refers to spatial integration or segregation. The resulting numbers form the basis for coloured-up visualisations which represent the distribution of spatial accessibility. The scale is coloured from red for the most accessible (integrated) lines, through the colour spectrum to blue for the least accessible (segregated) lines. See Hillier, B. (2007) Space is the machine: a configurational theory of architecture. Space Syntax, London, UK. ISBN 9780955622403

Radius -2: The maps are sections extracted from the axial map of the Greater London area bounded by the M2. They visualise local integration, or the relative depth of each axial line from all other lines within two changes of direction. This measure identifies the localised accessibility of a space within a limited area which may be considered as an area approximately equivalent to a walkable neighbourhood

Radius -3: The maps are sections extracted from the axial map of the Greater London area bounded by the M25. They visualise integration at radius-3, or the relative depth of each axial line from all other lines within three changes of direction.

Radius -6: The maps are sections extracted from the axial map of the Greater London area bounded by the M25. They visualise integration at radius-6, or the relative depth of each axial line from all other lines within six changes of direction.

Radius -8: The maps are sections extracted from the axial map of the Greater London area bounded by the M25. They visualise integration at radius-8, or the relative depth of each axial line from all other lines within eight changes of direction.

Radius -10: The maps are sections extracted from the axial map of the Greater London area bounded by the M25. They visualise integration at radius-8, or the relative depth of each axial line from all other lines within ten changes of direction.

Radius -27: The maps are sections extracted from the axial map of the Greater London area bounded by the M25. They visualise integration at radius-27, or the relative depth of each axial line from all other lines within twenty-seven changes of direction.

This radius is known as radius-radius calculated on the basis of the average depth of each line from all other lines in the system. It helps to eliminate problems of the edge effect in radius-n maps. The edge effect describes the fact that the edge of axial models appears disproportionately segregated due to the fact that streets on the edge of the map are not connected onwards.

Radius-n: These maps are sections extracted from the axial map of the Greater London area bounded by the M25. They visualise global integration or the relative depth of each axial line from all other lines in the system.

| Back | Top of page | Print this page |

Segment angular choice

Segment analysis takes each axial line and breaks it into segments at the intersections between axial lines (street junctions). This representation is referred to as a segment map. Segment analysis is concerned with the angular properties of graphs which involves calculating the relative straightness - least angular deviation or angular depth - of each segment from all other segments in the system.

Choice is calculated by counting the number of times each segment falls on the shortest path between all pairs of segments within a selected distance-radius where shortest path refers to the path of least angular deviation or straightest route through the system. More information can be found by reading Hillier, B. and Iida, S. (2005) Network Effects and Psychological Effects: a Theory of Urban Movement, 5th International Space Syntax Symposium, i. Delft, TU Delft, Faculty of Architecture: 553-564.

A high choice value means that a segment is frequently on the shortest path, a low choice value that it is less frequently or never on the shortest path. Choice values have been logged and represented using an equal ranges scale along a colour spectrum of warm colours (high choice) to cold colours (low choice). The range of choice values is restricted to those 20 centres selected as cases studies by the SSTC project and those additional 6 centres which act as our control cases. This is to emphasise visually the relatively high choice routes in the area of the case studies which would appear less differentiated in comparison with London as a whole.

Raduis-1250m: The maps are sections extracted from the segment map of the Greater London area bounded by the M25. They visualise Global Choice at radius-1250, which refers to the shortest angular path between all pairs of segments in the system within 1250 metres of each segment.

Raduis-1600m: The maps are sections extracted from the segment map of the Greater London area bounded by the M25. They visualise Global Choice at radius-1600, which refers to the shortest angular path between all pairs of segments in the system within 1600 metres of each segment.

Raduis-2000m: The maps are sections extracted from the segment map of the Greater London area bounded by the M25. They visualise Global Choice at radius-2000, which refers to the shortest angular path between all pairs of segments in the system within 2000 metres of each segment.

Radius -n: The maps are sections extracted from the segment map of the Greater London area bounded by the M25. Radius-n visualises Global Choice which refers to the shortest angular path between all pairs of segments in the system.

| Back | Top of page | Print this page |

Segment angular integration

Segment analysis takes each axial line and breaks it into segments at the intersections between axial lines (street junctions). This representation is referred to as a segment map.Segment analysis is concerned with the angular properties of graphs. This involves calculating the relative straightness - least angular deviation or angular depth - of each segment from all other segments in the system. Using this angular metric, lines that are said to be relatively deep from each other are said to be segregated, lines that are said to be relatively shallow are said to be integrated. More information can be found by reading Hillier, B. and Iida, S. (2005) Network Effects and Psychological Effects: a Theory of Urban Movement, 5th International Space Syntax Symposium, i. Delft, TU Delft, Faculty of Architecture: 553-564.

Radius -200m: The maps are sections extracted from the segment map of the Greater London area bounded by the M25. They visualise angular integration at radius-200, which refers to the linearity of each segment in relation to all other segments within 200 metres.

Radius -400m: The maps are sections extracted from the segment map of the Greater London area bounded by the M25. They visualise local angular integration at radius-400, which refers to the linearity of each segment in relation to all other segments within 400 metres. This measure identifies the localised accessibility of a space within a limited area which may be considered as an area approximately equivalent to a neighbourhood defined by a ten minute walk.

Radius -800m: The maps are sections extracted from the segment map of the Greater London area bounded by the M25. They visualise angular integration at radius-800, which refers to the linearity of each segment in relation to all other segments within 800 metres.

Radius -n: The maps are sections extracted from the segment map of the Greater London area bounded by the M25. They visualise global angular integration, which refers to the linearity of each segment in relation to all other segments in the system.

| Back | Top of page | Print this page |

Combined integration-choice

Segment analysis takes each axial line and breaks it into segments at the intersections between axial lines (street junctions). This representation is referred to as a segment map.Segment analysis is concerned with the angular properties of graphs. This involves calculating the relative straightness - least angular deviation or angular depth - of each segment from all other segments in the system. Using this angular metric, lines that are said to be relatively deep from each other are said to be segregated, lines that are said to be relatively shallow are said to be integrated

Choice is calculated by counting the number of times each segment falls on the shortest path between all pairs of segments within a selected distance-radius where shortest path refers to the path of least angular deviation or straightest route through the system.

Raduis-n: The maps are sections extracted from the segment map of the Greater London area bounded by the M25. They visualise a Global Combined Choice-Angular Integration measure at radius-n. Each segment is assigned a value of which radius-n and Angular Integration radius-n are the factors

Disclaimer and Copyrights

Every effort has been made to ensure that the data contained on this website are accurate. Nevertheless, there will be some inaccuracies in the aggregations presented and the project will not be held liable for any losses incurred from their use.

The 10K raster map backgrounds are based on the OS maps with permission of Ordnance Survey on behalf of The Controller of Her Majesty's Stationery Office, © Crown Copyright. All rights reserved.

The functional activity maps (points and surfaces) are based on the OS Address Layer 2 with permission of Ordnance Survey on behalf of The Controller of Her Majesty's Stationery Office, © Crown Copyright. All rights reserved.

Census output is Crown copyright and is reproduced with the permission of the Controller of Her Majesty's Stationery Office, © Crown Copyright. All rights reserved.

The initial Space Syntax graphs were kindly supplied to the project by Space Syntax Limited

| Back | Top of page | Print this page |

Acknowledgements

The profiler was created to assist the SSTC project team in developing their project hypotheses, and therefore would not have come to fruition without:
Dr Laura Vaughan, Dr Muki Haklay, Sam Griffiths and Catherine Jones



Together with the project team a number of other people and organisations have been very helpful to the project:

Ordnance Survey - who kindly provided us with access to the address layer 2 dataset and granted us permission to make the profiler and its maps publicly available.

Noémie Grémaux - our project intern of summer 2007 - whose dedication and hard worked assisted in the creation of many of the maps used in the website

URBED - who assisted us in developing the control case studies

Space Syntax Limited - who gave us access to the M25 Syntax map

Alejandra Celedon - MSc student from 2007 whose research has contributed to the project

Ozlem Shabaz - who created the historical doughnut - which enabled us to identify London's outer suburbs

| Back | Top of page | Print this page |