ST/ESA/STAT/POVERTY/WWW
1 February 2004
Handbook on Poverty Statistics:
Concepts, Methods and
Policy Use
Chapter 4. Statistical tools and estimation methods
for poverty measures based on household surveys
To discuss the surveys - income and expenditure surveys, LSMS, time-use
surveys, DHS, labour surveys, appraisal surveys - as sources of data for
poverty assessments based on monetary as well as non-monetary approaches;
To highlight the practical difficulties involved in generating reliable
and comparable estimates - definition of terms, sampling, periodicity,
frequency, regional differences and other sources of nonrandom error,
costliness and other constraints; to offer options to address specific
survey design issues that could potentially affect the interpretation
of - or bias- poverty estimates and changes in the estimates - income
or consumption; the use of a reference person rather than a "household
head" as unit of measurement; imputations and value of non-market
services; To pay due consideration to survey techniques relevant for assessing
the well-being of specific target groups - the poorest, earnings from
informal enterprise, itinerant and refugees populations, social minorities
- and for collecting information of non-economic components of well-being
- to consider characteristics having different unit of analysis - individual,
household, community, regional and national; To address specific statistical
and data issues in longitudinal analyses - attrition of the sample over
time; high mobility among specific groups - and to describe how measurement
errors can particularly bias analyses of transience and vulnerability,
and to provide guidance for the analysis and interpretation of the data;
To address the need for developing gender-specific data collection instruments
to enable poverty analysis from a gender perspective.
|