Premium
DOES SAMPLE DESIGN MATTER FOR POVERTY RATE COMPARISONS?
Author(s) -
Howes Stephen,
Lanjouw Jean Olson
Publication year - 1998
Publication title -
review of income and wealth
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.024
H-Index - 57
eISSN - 1475-4991
pISSN - 0034-6586
DOI - 10.1111/j.1475-4991.1998.tb00254.x
Subject(s) - econometrics , sampling design , economics , poverty , sample (material) , welfare , sampling (signal processing) , standard error , statistics , point estimation , simple random sample , point (geometry) , mathematics , computer science , population , economic growth , sociology , demography , market economy , chemistry , geometry , filter (signal processing) , chromatography , computer vision
Poverty comparisons—an increasingly important starting‐point for welfare policy analysis‐are almost always based on household surveys. Therefore they require that one be able to distinguish underlying differences in the populations being compared from sampling variation: standard errors must be calculated. This has typically been done assuming that the household surveys are simple random samples. However, household surveys are more complex than this. We show that taking into account sampling design has a major effect on estimated standard errors for well‐known poverty measures. In our samples they increase by around one‐half. We also show that making only a partial correction for sample design (taking into account clustering, but not stratification, whether explicit or implicit) can be as misleading as not taking any account of sampling design at all.