Output Area Classification Research Articles

Classifying and mapping residential structure through the London Output Area Classification

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.

A neighbourhood Output Area Classification from the 2021 and 2022 UK censuses

AbstractUK‐wide multivariate neighbourhood classifications have been built using small area population data following every census since 1971, and have been built using Output Area geographies since 2001. Policy makers in both the public and private sectors find such taxonomies, typically arranged into hierarchies of Supergroups, Groups and Subgroups, useful across a wide range of applications in business and service planning. Recent and forthcoming releases of small area census statistics pose new methodological challenges. For example, the 2022 Scottish Census was carried out a year after those in other UK nations, and some of the variables now collected across different jurisdictions do not bear direct comparison with one another. Here we develop a methodology to accommodate these issues alongside the more established procedures of variable selection, standardisation, transformation, class definition and labelling.

A graph neural network framework for spatial geodemographic classification

Geodemographic classifications are exceptional tools for geographic analysis, business and policy-making, providing an overview of the socio-demographic structure of a region by creating an unsupervised, bottom-up classification of its areas based on a large set of variables. Classic approaches can require time-consuming preprocessing of input variables and are frequently a-spatial processes. In this study, we present a groundbreaking, systematic investigation of the use of graph neural networks for spatial geodemographic classification. Using Greater London as a case study, we compare a range of graph autoencoder designs with the official London Output Area Classification and baseline classifications developed using spatial fuzzy c-means. The results show that our framework based on a Node Attributes-focused Graph AutoEncoder (NAGAE) can perform similarly to classic approaches on class homogeneity metrics while providing higher spatial clustering. We conclude by discussing the current limitations of the proposed framework and its potential to develop into a new paradigm for creating a range of geodemographic classifications, from simple, local ones to complex classifications able to incorporate a range of spatial relationships into the process.

Microdata selection for estimating household consumption-based emissions

To estimate household emissions from a consumption-perspective, national accounts are typically disaggregated to a sub-national level using household expenditure data. While limitations around using expenditure data are frequently discussed, differences in emission estimates generated from seemingly comparable expenditure microdata are not well-known. We compare UK neighbourhood greenhouse gas emission estimates derived from three such microdatasets: the Output Area Classification, the Living Costs and Food Survey, and a dataset produced by the credit reference agency TransUnion. Findings indicate moderate similarity between emission estimates from all datasets, even at detailed product and spatial levels; importantly, similarity increases for higher-emission products. Nevertheless, levels of similarity vary by products and geographies, highlighting the impact microdata selection can have on emission estimates. We focus our discussion on how uncertainty from microdata selection can be reduced in other UK and international contexts by selecting data based on the data generation process, the level of disaggregation needed, physical unit availability and research implications.

A Principal Component Analysis (PCA)-based framework for automated variable selection in geodemographic classification

ABSTRACT A geodemographic classification aims to describe the most salient characteristics of a small area zonal geography. However, such representations are influenced by the methodological choices made during their construction. Of particular debate are the choice and specification of input variables, with the objective of identifying inputs that add value but also aim for model parsimony. Within this context, our paper introduces a principal component analysis (PCA)-based automated variable selection methodology that has the objective of identifying candidate inputs to a geodemographic classification from a collection of variables. The proposed methodology is exemplified in the context of variables from the UK 2011 Census, and its output compared to the Office for National Statistics 2011 Output Area Classification (2011 OAC). Through the implementation of the proposed methodology, the quality of the cluster assignment was improved relative to 2011 OAC, manifested by a lower total within-cluster sum of square score. Across the UK, more than 70.2% of the Output Areas (OAs) occupied by the newly created classification (i.e. AVS-OAC) outperform the 2011 OAC, with particularly strong performance within Scotland and Wales.

A spatial analysis of health status in Britain, 1991–2011

Using Census-derived data for consistent spatial units, this paper explores how the population of Britain in 1991, 2001 and 2011 was spatially structured by self-reported health including exploring the trajectories of change. This paper uses consistent small area units to examine the changing spatial structure of census-derived Limiting, Long-Term Illness (LLTI) in Britain over the twenty year period and utilises the 2011 Office for National Statistics Output Area Classification (OAC) as a geodemographic indicator. The results allow the geography of change to be captured, highlighting how health is inextricably linked to geography, demonstrating quantitatively a complex, yet distinctive, spatial organisation of health inequalities within Britain. Overall decreasing unevenness values, coupled with increased positive spatial association suggests that neighbouring areas have become more similar over time – the distinction between areas characterised by poor health or by good health is decreasing.

Locating areas with high use of critical care

Admissions over a three-year period to the general surgical and medical critical care unit at King's College University Hospital were allocated to the output area of residence for each patient. Each output area contains around 300 people of roughly similar social and demographic characteristics. Output area social grouping (output area classification) and deprivation (index of multiple deprivation) are shown to effect critical care admission rates. The admission rate doubles between the least and most deprived areas (excluding the effect of social group), however there is in excess of a 10-fold variation between the various deprivation-banded output area classification social groups. Particular social groups also show differing proportions of patients admitted to the critical care unit for different conditions such as sepsis or drug overdose. Having identified small areas with high risk of critical care admission should enable primary care organisations to target admission avoidance and/or appropriate end-of-life care.

An Open Source Geodemographic Classification of Small Areas in the Republic of Ireland

Geodemographic classifications have progressed from manual classifications of areas through to complex, highly marketable products used in both the public and private sectors. As their production became commercialized, input variables moved beyond census variables to include other, often not publicly available datasets, and hence the resultant black-box approach increased in sophistication, but was less open to scrutiny. In the UK this was somewhat reversed with the production of the Output Area Classification (OAC) from the 2001 census. As an alternative approach, in this paper we demonstrate the production of a geodemographic classification for the Republic of Ireland, using a different approach to OAC, and extending the ethos of transparency and reproducibility.

Geography of Diet in the UK Women’s Cohort Study: A Cross-Sectional Analysis

Diet can influence health outcomes and chronic disease risk, therefore a better understanding of factors influencing diet is important in promotion of healthier dietary choices. Many factors influence food choice, including the environment in which we live. This study aims to explore differences in dietary pattern consumption by two spatial measures: Government Office Region (a large regional unit of geography) and Output Area Classification (a small area geography combined with demographic characteristics). A cross-sectional analysis using data from the UK Women’s Cohort Study was carried out. This cohort included ~35000 middle aged women recruited between 1995 and 1999. Dietary patterns were derived using a k-means cluster analysis from diet data collected using a validated 217 item Food Frequency Questionnaire. Multinomial logit regression was used to test whether the area in which the women live, predicts their dietary pattern consumption. Results show that dietary patterns vary significantly by both spatial measures. The Government Office Region, the North West of England has the highest proportion of individuals consuming the least healthy, monotonous diets, while Greater London has the highest proportion of vegetarian diets. Individuals living in Supergroups ‘Countryside’ and ‘Prospering Suburbs’ consume healthier, more diverse diets. Those in ‘Constrained by Circumstance’ and ‘Blue Collar Communities’ consume monotonous, less healthy diets. Using a combination of spatial scales such as Government Office Region and Output Area Classification Supergroup could have a beneficial impact on targeting of public health dietary interventions and subsequent health.

Putting the geography into geodemographics: Using multilevel modelling to improve neighbourhood targeting – a case study of Asian pupils in London

This paper explores the use of multilevel modelling to provide a statistical framework for geodemographic analysis. It argues that combining a neighbourhood classification with a modelling approach allows the levels of the geodemographic hierarchy to be considered simultaneously, identifying those that are most appropriate to the analysis and allowing the apparent differences between neighbourhood types to be considered in regard to their statistical significance, and to the uncertainty of the estimates. The paper shows how the model can be extended to create a cross-classified multiscale model that makes better use of the locational information available to improve the efficiency of the neighbourhood targeting. The ideas are illustrated with a case study using a sample of data and the freely available London Output Area Classification to predict which neighbourhoods in London have the highest percentages of Asian school pupils. The multiscale model is shown to outperform the predictions made using geodemographics alone.

The internal structure of Greater London: a comparison of national and regional geodemographic models

Geodemographic classifications are categorical measures representing salient multidimensional population and built environment attributes of small areas. The UK Output Area Classification (OAC) is one such classification, created on behalf of the Office for National Statistics, and was built with an open methodology and entirely from 2011 Census variables. However, one criticism of national classifications such as OAC is that they do not adequately accommodate local or regional structures that diverge from national patterns. In this paper we explore this issue with respect to Greater London. We develop a London classification based upon the OAC methodology, and explore the extent to which these patterns are divergent from the national classification.

Modelling Class Uncertainty in the Geodemographic Output Area Classification

Geodemographics is the study of people by where they live, and historically has relied in part on census information to provide essential variables. In 2005 the Office for National Statistics for England and Wales published a geodemographic classification of output areas (OAs; the smallest published areas of enumeration for the UK) of the 2001 UK Census. Uniquely among UK examples of such classifications not only are the classes to which each OA is assigned published for free with no restriction on use, but the difference between each OA and all class centroids are also made available. In this paper these differences (being Euclidian distances in the classification feature space) are used to introduce uncertainty by softening the classification using the methods derived from fuzzy c-means and the possibilistic c-means classifications. The former applies a probabilistic assumption that all fuzzy memberships for a particular OA sum to unity while the latter does not. This assumption (which is the same as that behind the published OA classification) may generate results which are potentially misleading because the fuzzy c-means method forces OAs into classes to which they may have relatively little similarity while minimising the importance of classes to which they may have a larger affinity. If the fuzzy memberships or possibilities were to be integrated with other socioeconomic variables which are also conceived as fuzzy sets for geodemographic applications (decision making for government resource allocation, site location modelling or marketing, for example), then the two results would produce different outcomes. It is suggested that the possibilistic c-means are the more useful although they do lead to the possibly confusing but analytically rich situation where one OA may have minimal affinity with any class, at the same time as others may have strong affinity with many classes.

OACoder: Postcode Coding Tool

Geodemographic classifications are small area classifications of social, economic and demographic characteristics. The Output Area Classification (OAC) is a free geodemographic classification. It is an Office of National Statistics validated measure that summarises neighbourhood conditions at the Output Area Level across the United Kingdom. Linkage of these valuable statistics has been problematic for users more used to address records that are georeferenced using unit postcodes. OACoder resolves this problem by allowing users to link corresponding OAC codes to each of the postcode addresses. OACoder is an open source software, and it is developed and tested to work on different versions of windows operating systems. It is stored in Figshare. The source code of the OACoder is stored in SourceForge. As open source software, OACoder has reuse potential across a range of applications. The functionality of OACoder can be extended to work with the new version of OAC (2011 OAC). It is also possible to reuse the source code and extend the functionality to work on different operating systems other than Windows. Different components of the software can be reused for the purpose of reading/writing CSV files and handling large data sets.

Temporal Uncertainty in a Small Area Open Geodemographic Classification

AbstractThe 2001 Output Area Classification (2001 OAC) is an open source geodemographic classification of the UK built exclusively from 2001 UK Census data. There has been considerable user interest in its applicability to subsequent time periods, particularly given the potential propensity of characteristics and attributes in some areas to change during inter‐censual periods. Users often purchase commercial geodemographic classification products in the belief that purely census‐based classifications such as the 2001 OAC are uniformly unreliable because there is no temporal updating of input data. Yet there is evidence to suggest that whilst some UK neighborhoods are prone to sudden changes, many others change very little over protracted time periods. Using measures that are available at the small area level, temporal uncertainty indicators can be constructed to identify those areas that are less stable. Using mid‐year population estimates and dwelling stock data, this article develops three temporal uncertainty indicators. These provide a reliable means of gauging the stability or otherwise of neighborhood conditions. The conclusion from this is that while a large number of small areas in the UK do experience change over time, this change is not uniform in either degree or distribution, or by geodemographic type.

Exploring the geographical distribution of wealth using the output area classification

The results of the first wave of the Wealth and Assets Survey were published in the report, ‘Wealth in Great Britain’ in December 2009 with datasets released in March 2010. However, apart from the presentation of the aggregate results on a Government Office Region basis, no more disaggregated geographical analysis of the data has so far been carried out, not least because of the potential disclosivity of the data when presented for low level geographies. This paper aims to provide an understanding of the geographical distribution of wealth and its components by areas with common socio-economic and demographic characteristics from relatively high to low levels of aggregation while overcoming the disclosivity problem. This is achieved by using the Output Area Classification (OAC) groupings, based on the 2001 Census.

Household expenditure by Output Area Classification

Household expenditure by Output Area Classification

Geodemographics as a tool for targeting neighbourhoods in public health campaigns

Geodemographics offers the prospects of integrating, modelling and mapping health care needs and other health indicators that are useful for targeting neighbourhoods in public health campaigns. Yet reports about this application domain has to date been sporadic. The purpose of this paper is to examine the potential of a bespoke geodemographic system for neighbourhood targeting in an inner city public health authority, Southwark Primary Care Trust, London. This system, the London Output Area Classification (LOAC), is compared to six other geodemographic systems from both governmental and commercial sources. The paper proposes two new indicators for assessing the performance of geodemographic systems for neighbourhood targeting based on local hospital demand data. The paper also analyses and discusses the utility of age- and sex standardisation of geodemographic profiles of health care demand.

Rectangular Hierarchical Cartograms for Socio-Economic Data

Please click here to download the map associated with this article. We present rectangular hierarchical cartograms for mapping socio-economic data. These density-normalising cartograms size spatial units by population, increasing the ease with which data for densely populated areas can be visually resolved compared to more conventional cartographic projections. Their hierarchical nature enables the study of spatial granularity in spatial hierarchies, hierarchical categorical data and multivariate data through false hierarchies. They are space-filling representations that make efficient use of space and their rectangular nature (which aims to be as square as possible) improves the ability to compare the sizes (therefore population) of geographical units. We demonstrate these cartograms by mapping the Office for National Statistics Output Area Classification (OAC) by unit postcode (1.52 million in Great Britain) through the postcode hierarchy, using these to explore spatial variation. We provide rich and detailed spatial summaries of socio-economic characteristics of population as types of treemap, exploring the effects of reconfiguring them to study spatial and non-spatial aspects of the OAC classification.

Household expenditure by Output Area Classification (OAC): an overview

Household expenditure by Output Area Classification (OAC): an overview

A geographically and socio-economically disaggregated local household consumption model for the UK

A key challenge for local environmental and economic planning is to predict resource flows and material waste arisings at a high level of geographical localisation. This paper presents an approach to household consumption modelling in which the material demands of small socio-economically homogeneous neighbourhoods are estimated on the basis of consumer expenditure data in conjunction with Census data. Household waste arisings are estimated through incorporation of a household metabolism model. Selected material flows of consumer commodities, such as household appliances, carpets and clothing, are examined in case study areas representing extremes of relative deprivation, and neighbourhoods representing typical examples of the UK National Output Area Classification (OAC). The model will be of use in sustainable consumption policy-making and local waste strategy planning.