Classifying and mapping residential structure through the London Output Area Classification
Alex D Singleton; Paul A Longley (2024). Environment and Planning B: Urban Analytics and City Science, 51(5), 1153-1164. DOI: 10.1177/23998083241242913
Abstract
This paper outlines the creation of the London Output Area Classification (LOAC) from the 2021 Census, set within the broader context of geodemographic classification systems in the United Kingdom. The LOAC 2021 was developed in collaboration with the Greater London Authority (GLA) and offers an enhanced, statistically robust typology adept at capturing the unique spatial, socio-economic and built characteristics of London’s residential neighbourhoods. The paper asserts the critical importance of nuanced, area-specific geodemographic classifications for urban areas with unique geography relative to the national extent.
Extended Summary
This research develops a specialised neighbourhood classification system to better represent London’s unique residential geography compared to national demographic models. The study addresses how London’s distinctive characteristics are poorly captured by England and Wales national classifications, which represent the capital using only a narrow range of cluster types. Working in partnership with the Greater London Authority, the research creates the London Output Area Classification (LOAC) 2021 using data from the 2021 Census covering all small geographical areas within Greater London. The methodology mirrors established approaches used for national classifications but applies clustering algorithms specifically to London’s 25,053 Output Areas. The study uses 68 variables spanning demographics, ethnicity, living arrangements, health, education, and employment, with additional industry variables incorporated after consultation with stakeholders to reflect London’s distinctive employment structure. Data transformation techniques including inverse hyperbolic sine transformation and range standardisation ensure comparability across all variables. The research employs k-means clustering analysis with 10,000 iterations to ensure stability, creating a two-tier nested classification system. Clustergram visualisation techniques help determine optimal cluster numbers, resulting in seven Supergroups subdivided into 16 detailed Groups. Extensive consultation with an Advisory Group comprising representatives from London boroughs, Metropolitan Police, and Transport for London ensures the classification meets practical policy needs. The final classification reveals London’s residential complexity through categories such as ‘Professional Employment and Family Lifecycles’, ‘Central Connected Professionals and Managers’, ‘Suburban Asian Communities’, and ‘Social Rented Sector Families with Children’. Each category receives detailed descriptions using ‘grand index’ scores showing how variables differ from London averages. The research demonstrates significant improvements over national classifications through better representation of London’s diversity. For example, Camden shows diverse neighbourhoods spanning professional areas and social housing communities, while Hounslow predominantly features suburban Asian communities - patterns obscured in national models. The classification provides enhanced utility for policy applications including population estimation, school planning, transport strategy development, and public service resource allocation. Previous versions supported Transport for London’s travel behaviour segmentation and Greater London Authority’s school roll forecasting, demonstrating practical value for urban governance. This work emphasises the importance of area-specific geodemographic classifications for cities with unique characteristics, offering a methodological framework applicable to other distinctive urban areas requiring nuanced residential analysis beyond national typologies.
Key Findings
- LOAC 2021 creates a seven-Supergroup, 16-Group classification system specifically designed to capture London’s unique residential geography
- National classifications inadequately represent London’s diversity, using only narrow cluster ranges that obscure local variation patterns
- The classification incorporates 68 variables including London-specific industry data, analysed through k-means clustering with 10,000 iterations
- Extensive stakeholder consultation with London boroughs and agencies ensures practical utility for policy applications and service planning
- The methodology demonstrates how area-specific geodemographic systems provide superior representation compared to national models for distinctive cities
Citation
@article{singleton2024classifying,
author = {Alex D Singleton; Paul A Longley},
title = {Classifying and mapping residential structure through the London Output Area Classification},
journal = {Environment and Planning B: Urban Analytics and City Science},
year = {2024},
volume = {51(5)},
pages = {1153-1164},
doi = {10.1177/23998083241242913}
}