Design of a Multi-Stage Stratified Sample for Poverty and Welfare Monitoring with Multiple Objectives : A Bangladesh Case Study

Ahmed, Faizuddin ; Roy, Dipankar ; Yanez-Pagans, Monica ; Yoshida, Nobuo

World Bank, Washington, DC

关键词: stratified clustered sampling;    household surveys;    poverty measurement;   

DOI  :  10.1596/1813-9450-7989
RP-ID  :  WPS7989

学科分类：社会科学、人文和艺术（综合）

来源: World Bank Open Knowledge Repository

【 摘 要 】

This paper describes the design of amulti-stage stratified sample for the Bangladesh HouseholdIncome and Expenditure Survey 2016/17. This surveyinstrument will be used by the Government of Bangladesh toestimate reliable poverty and welfare statistics at threedifferent levels: (i) annual estimates at the districtlevel, (ii) quarterly estimates at the national level, and(iii) annual estimates at the division level for urban andrural areas. The sample for this survey was designed toachieve these three objectives. The paper explains how thethree objectives are prioritized and how inconsistencies inachieving more than one objective can be reconciled.Further, the paper modifies the standard formulas toestimate the optimal sample size and the allocation of thesample across strata by explicitly taking into considerationthe effect of clustering in the sample.