Estimation of Poverty in Somalia Using Innovative Methodologies

Pape, Utz ; Wollburg, Philip

World Bank, Washington, DC

关键词: POVERTY MEASUREMENT;    CONSUMPTION;    SURVEY DESIGN;    LIVELIHOODS;    FRAGILE STATES;   

DOI  :  10.1596/1813-9450-8735
RP-ID  :  WPS8735

学科分类：社会科学、人文和艺术（综合）

来源: World Bank Open Knowledge Repository

【 摘 要 】

Somalia is highly data-deprived, leavingpolicy makers to operate in a statistical vacuum. Toovercome this challenge, the World Bank implemented wave 2of the Somali High Frequency Survey to better understandlivelihoods and vulnerabilities and, especially, to estimatenational poverty indicators. The specific context ofinsecurity and lack of statistical infrastructure in Somaliaposed several challenges for implementing a household surveyand measuring poverty. This paper outlines how thesechallenges were overcome in wave 2 of the Somali HighFrequency Survey through methodological and technologicaladaptations in four areas. First, in the absence of a recentcensus, no exhaustive lists of census enumeration areasalong with population estimates existed, creating challengesto derive a probability-based representative sample.Therefore, geospatial techniques and high-resolution imagerywere used to model the spatial population distribution,build a probability-based population sampling frame, andgenerate enumeration areas to overcome the lack of a recentpopulation census. Second, although some areas remainedcompletely inaccessible due to insecurity, even mostaccessible areas held potential risks to the safety of fieldstaff and survey respondents, so that time spent in theseareas had to be minimized. To address security concerns, thesurvey adapted logistical arrangements, sampling strategyusing micro-listing, and questionnaire design to limit timeon the ground based on the Rapid Consumption Methodology.Third, poverty in completely inaccessible areas had to beestimated by other means. Therefore, the Somali HighFrequency Survey relies on correlates derived from satelliteimagery and other geo-spatial data to estimate poverty insuch areas. Finally, the nonstationary nature of the nomadicpopulation required special sampling strategies.