Estimating Poverty Using Cell Phone Data : Evidence from Guatemala

Hernandez, Marco ; Hong, Lingzi ; Frias-Martinez, Vanessa ; Whitby, Andrew ; Frias-Martinez, Enrique

World Bank, Washington, DC

关键词: cell phone data;    poverty estimation;    big data;    poverty measurement;    call detail records;   

DOI  :  10.1596/1813-9450-7969
RP-ID  :  WPS7969

学科分类：社会科学、人文和艺术（综合）

来源: World Bank Open Knowledge Repository

【 摘 要 】

The dramatic expansion of mobile phoneuse in developing countries has given rise to a rich andlargely untapped source of information about thecharacteristics of communities and regions. Call DetailRecords (CDRs) obtained from cellular phones provide highlygranular real-time data that can be used to assesssocio-economic behavior including consumption, mobility, andsocial patterns. This paper examines the results of a CDRanalysis focused on five administrative departments in thesouth west region of Guatemala, which used mobile phone datato predict observed poverty rates. Its findings indicatethat CDR-based research methods have the potential toreplicate the poverty estimates obtained from traditionalforms of data collection, like household surveys orcensuses, at a fraction of the cost. In particular, CDRswere more helpful in predicting urban and total poverty inGuatemala more accurately than rural poverty. Moreover,although the poverty estimates produced by CDR analysis donot perfectly match those generated by surveys and censuses,the results show that more comprehensive data could greatlyenhance their predictive power. CDR analysis has especiallypromising applications in Guatemala and other developingcountries, which suffer from high rates of poverty andinequality, and where limited fiscal and budgetary resourcescomplicate the task of data collection and underscore theimportance of precisely targeting public expenditures toachieve their maximum antipoverty impact.