Egyptian Integrated Household Survey (EIHS)
(1997)
Description
The Egypt Integrated Household Survey (EIHS) is a multi-topic,
nationally representative household survey carried out by the
International Food Policy Research Institute (IFPRI) in
coordination with the Ministry of Agriculture and Land Reclamation
(MALR) of the Government of Egypt (GOE) and the Ministry of Trade
and Supply (MOTS) of the GOE.
Collaborating institutions
International Food Policy Research Institute (IFPRI), the
Ministry of Agriculture and Land Reclamation (MALR) of the
Government of Egypt (GOE), and the Ministry of Trade and Supply (MOTS)
of the GOE.
Sponsor
US Agency for International Development
Related publications
Datt et.al., 1998
Contact person
Manohar Sharma
Access to data
IFPRI website
For more details
For the full text of the discussion paper, see
http://www.cgiar.org/ifpri/divs/fcnd/dp/papers/dp49.pdf
Survey Design and
Implementation
The questionnaire was administered to 2,500
households from 20 governorates using a two-stage, stratified
selection process. The sample frame used for the selection process
was supplied by the Central Agency for Public Mobilization and
Statistics (CAPMAS) and is based on the 1986 Egypt census frame
and a 1993 listing of households in selected primary sampling
units. This sample frame is used by CAPMAS as a master sample for
much of their survey work. It consists of 492 sampling units, 296
of which are urban and 196 are rural. Households were selected
from the master sample in a two-stage process. The first stage
entailed randomly selecting 125 primary sampling units (PSU) with
probability proportional to size from the master sample. The
second stage then randomly selected 20 households from each PSU.
The advantage of a two-stage process relative to a pure random
selection process is that it dramatically reduces the scope of
field work and therefore reduces the cost of the survey. The
disadvantage is that standard errors resulting from two-stage
samples tend to be significantly larger than those resulting from
pure random samples. With a representative sample of 2,500
households it is possible to examine average characteristics of
the poor at the national level, and also into most types of
dichotomous breakdowns, such as: urban-rural or Upper-Lower. In
order to allow analyzing the data by a somewhat more extensive
breakdown, the sample is stratified on five regions of Egypt:
Metropolitan, Lower urban, Lower rural, Upper urban, and Upper
rural. The stratification means that instead of allowing the data
to be self-weighted, the design requires that a fixed number of
households fall in each strata. Without stratification the sample
sizes in the Lower and Upper urban regions could have potentially
been too small for analysis.
Field work began during the first week of March 1997 and
concluded in the third week of May 1997. |