Correcting for Sampling Bias in the Measurement of Welfare and Poverty in the Cote d'Ivoire Living Standards Survey

The sampling aspects of a household data set are important toanalysts. The early years of the Côte d'Ivoire LivingStandards Survey (CILSS) had a sampling bias, which seriouslyaffected estimates of population statistics such as householdsize. The bias arose from sampling procedures that overrepresentedlarger dwellings. Assuming that samples drawn in later yearswere unbiased, a correction procedure is applied that uses weightsbased on household size. Results from the weighted data arethen compared with the unweighted findings to assess the seriousnessof the bias. Estimates of household expenditure per capita inthe early years of the survey are found to be significantlyunderestimated, resulting in an overestimation of poverty. Thesampling bias also resulted in an underestimation of the upwardtrend in poverty during 1985–88. The CILSS has been apopular and fruitful data set for policy analysis. These findings,however, cast doubt on the robustness of earlier work. Thus,the effort to trace sampling information is particularly worthwhilefor policy-oriented applied research.