Poverty on the Cheap Project(PoC)


Contact Info

UCL - London

Project Description

Knowledge of the distribution of wealth and poverty in developing countries country is often acquired by means of manually collected household survey data (e.g., Demographic and Health Surveys). However, the cost associated with this method is such that poorer nations can only run such surveys every 10 years or more, and on sample sizes rarely large enough to provide statistically significant estimates for small geographical units, such as municipalities and villages.

The aim of the project is to develop and validate a practical methodology that governments of developing countries can use to accurately estimate socioeconomic deprivation at a fine level of spatiotemporal granularity and low cost.

We propose to do so using unobtrusively collected mobile call detail records obtained from local network operators, whose penetration in developing countries is so high to offer a relatively unbiased picture in terms of demographics. Through analysis of patterns inherent in mobile phone users’ collective behaviour, aggregated to the cell tower level, we define features that can be extracted and used as proxy indicators of poverty levels. The intended outcomes are interpretable results that local governments and NGOs can use to develop timely and targeted interventions.

Reference publications

  • C. Smith-Clarke, A. Mashhadi and L. Capra. "Poverty on the Cheap: Estimating Poverty Maps Using Aggregated Mobile Communication Networks". In 32nd ACM SIGCHI. Toronto, Canada. April 2014

  • C. Smith-Clarke and L. Capra. "Beyond the baseline: Establishing the value in mobile phone based poverty estimates". In 25th International World Wide Web Conference (WWW 2016). Montreal, Canada. April 2016

  • C. Smith-Clarke and L. Capra. "Information Diffusion and Economic Development". In IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM 2017). Sydney, Australia. August 2017 

Fig. 1 Cote d’Ivoire: Poverty Rates Computed via surveys at Prefecture Level

Fig. 2 Cote d’Ivoire: Poverty Rate Estimated from CDR at Sub-prefecture Level

Research Funded by:


Licia Capra

She is Professor of Pervasive Computing in the Dept. of Computer Science at University College London. She conducts research in the area of computational social science, investigating how new technology is chaning society, and also how society is appropriating and shaping such new technology. Licia has specific expertise and interest in urban informatics, investigating how to extract knowledge from a variety of data sources to better understand the functioning of cities, and to develop technologies and interventions aimed at improving life of citizens.

She has conducted pioneering research in analysing a broad variety of big datasets, including public transport data, telecommunication data, social media data, and crowd-sourced data, in order to model cities’ and citizens’ dynamics. These models have been used, for example, to build personalised travellers’ services, to estimate cities’ economic wellbeing, and to help regulate novel markets.