ESRC NCRMEvaluating and improving small area estimation methods


This network, funded by the ESRC's National Centre for Research Methods (NCRM) Programme, brings together experts in small area estimation (SAE) techniques from the academic and policy (eg Office for National Statistics, Teasgasc) communities in the UK and internationally in order to seek innovative ways to advance knowledge and understanding in SAE methodologies.

Listen to a podcast to hear more about the network's activities

Context to the network

Guanajuato, Mexico © Robert Crum

SAE methodologies have become a widely used method across various disciplines as a result of growing demand by policy makers and researchers for spatially detailed information alongside advances in small area data availability and computing power (Clarke, 1996). Various small area estimation methods have emerged and these can broadly be disaggregated into statistical regression-based approaches and spatial microsimulation approaches. Although less widely used in this context, agent-based modeling may also hold benefits for SAE.

Currently, despite the potential of these approaches and the growing demands placed upon them, there is little agreement within the academic and policy community as to which method(s) work best, whether different approaches are best suited to different local contexts, how best methods can be implemented and how best results can be validated. Experts from across each of these methodological strands and across a range of academic disciplines are included in the network so as to enable not only improvements in each separate approach but also overall methodological progress through the cross-pollination of ideas and skills.


The network has five core aims:

  1. To evaluate the performance of alternative methods across different small area contexts and to define general rules of guidance as to when specific methods seem to work most effectively and why
  2. To enhance the performance of the separate methodologies themselves
  3. To explore whether general rules of validation can be described
  4. To explore methods to ‘roll forwards’ small area covariate (eg census) data in light of anticipated changes to the UK census after 2011
  5. To disseminate findings and methodological skills widely

Network members

Principal Investigator: Dr Adam Whitworth, Dept of Geography, University of Sheffield adam.whitworth@sheffield.ac.uk

Fiona Aitchison, Office for National Statistics Small Area Population Estimation group, Southampton

Dr Ben Anderson, Deputy Director, Centre for Research in Economic Sociology and Innovation, University of Essex

Prof Peter Atkinson, Centre for Geographical Health Research, University of Southampton

Dr Dimitris Ballas, Dept of Geography, University of Sheffield

Prof Nicky Best, Faculty of Medicine, Imperial College London

Prof Graham Clarke, Dept of Geography, University of Leeds

Dr Chris Dibben, School of Geography and Geosciences, University of St Andrews

Dr Cathal O’Donoghue, Head of Rural Economy Research Centre, Teagasc (Agriculture and Food Development Authority, Ireland)

Dr Kimberley Edwards, School of Clinical Sciences, University of Nottingham

Dr Alison Heppenstall, School of Geography, University of Leeds

Dr Dimitris Kavroudakis, Researcher, University of the Aegean, Lesvos, Greece

Dr Nick Malleson, Dept of Geography, University of Leeds

Mr David McLennan, Deputy Director, Social Disadvantage Research Centre, University of Oxford

Prof Graham Moon, Centre for Geographical Health Research, University of Southampton

Dr Karyn Morrissey, Dept of Geography, University of Liverpool

Dr Tomoki Nakaya, Dept of Geography, Ritsumeikan University, Japan

Dr Robert Tanton, NATSEM, University of Canberra, Australia

Dr Nikos Tzavidis, School of Social Sciences, University of Southampton

Dr Paul Williamson, Dept of Geography, University of Liverpool