Egypt Multiple Indicator Cluster Survey (1996)

Description
In the 1996 Egypt Multiple Indicator Cluster Survey (EMICS), 8,409 households in one sampling frame and 2,555 households in the second, were interviewed, in urban and rural areas of 21 of the 26 governorates of Egypt. Topics included indicators of child well-being immunization such as breast-feeding and anthropometry, diarrhoea and acute respiratory infections, disability, school enrollment, deprivation from education, child labour, and safe water and sanitation.

• developing sample design and listings that can be used repeatedly, ie each time, with a different set of sample households from the same primary sampling units and using the same listings
• employing a well-tested instrument easily implemented in the field that facilitates information-gathering on a concise yet comprehensive set of indicators
• developing and testing the logistics of field operations necessary to ensure timely data collection of high quality

EMICS sets out to provide estimates of priority indicators at the national level disaggregated by urban and rural residence, and at governorate level - in particular Greater Cairo, Alexandria, Assiut, Sohag, Qena and Aswan.

The survey also aims to provide estimates of these priority indicators at the level of unplanned urban districts (random housing areas) as a separate stratum, in order to study disparities within the urban population and between the growing unplanned communities.

• Percentage of children aged 12-23 months who received triple DPT vaccination before their first birthday
• Percentage of children aged 12-23 months who received triple polio vaccination before their first birthday
• Percentage of children aged 12-23 months who received triple hepatitis B vaccination before their first birthday
• Percentage of children aged 12-23 months who received measles vaccination before their first birthday
• Percentage of children aged 12-23 months who received BCG vaccination before their first birthday
• Percentage of mothers of infants under one who received at least two doses of tetanus toxoid vaccination within three years of the child’s birth

• Percentage of children under one who ever received breast milk
• Percentage of children under four months who are exclusively breastfed
• Percentage of children aged 6-9 months who are receiving breast milk and supplementary feeding
• Percentage of children aged 9-23 months who are still breastfed
• Percentage of under-fives who fall below minus two standard deviation from the median weight-for-age of NCHS/WHO reference population (underweight moderate to severe)
• Percentage of under-fives who fall below minus three standard deviation from the median weight-for-age of NCHS/WHO reference population (underweight severe)
• Percentage of under-fives who fall below minus two standard deviation from the median height-for-age of NCHS/WHO reference population (stunting moderate to severe)
• Percentage of under-fives who fall below minus three standard deviation of the median height-for-age of NCHS/WHO reference population (severe stunting).
• Percentage of under-fives who fall below minus two standard deviation of the median weight-for-height of NCHS/WHO reference population (wasting moderate to severe)
• Percentage of under-fives who fall below minus three standard deviation of the median weight-for-height of NCHS/WHO reference population (severe wasting)

• Percentage of under-fives who had diarrhoea during the two weeks preceding the survey
• Percentage of diarrhoea episodes among under-fives treated by ORS and/or recommended home fluids
• Percentage of diarrhoea episodes among under-fives treated by increased fluids and continued feeding
• Percentage of mothers of under-fives who have ORS at home or who know where to get ORS
• Percentage of mothers of under-fives who are aware of the signs of pneumonia
• Percentage of under-fives who had signs of pneumonia (fever, cough and fast breathing) during the two weeks preceding the survey
• Percentage of under-fives who had signs of pneumonia during the two weeks preceding the survey and were seen by health facilities

• Prevalence of any disability condition per 100,000 children by age
• Prevalence of disability condition per 100,000 children by age and type of disability
• Percentage distribution of disability conditions by type of disability

• Percentage of children under six who go to nursery school
• Percentage of children aged 6-7 years who are enrolled in nursery or primary school, by gender
• Percentage of children enrolled in primary school who belong to the 6-10 year age-group, out of the total number in that age-group (net enrolment ratio)
• Ratio of children enrolled in primary school to the total number of children in the 6-10 year age-group (gross enrolment ratio)
• Percentage of children entering the first grade of primary school who reach grade five, by gender
• Grade-specific repeat rates among children entering school

• Percentage of children aged 6-14 years who never attended school, by gender
• Grade- and gender-specific drop-out rates among children aged 6-14 years entering school

• Percentage of children aged 6-11 years who are regularly engaged in unpaid family work (in business or on the farm)
• Percentage of children aged 6-11 years who have ever worked for cash
• Percentage of children aged 6-11 years who are currently in paid employment
• Percentage of children aged 12-14 years who are regularly engaged in unpaid family work (in business or on the farm)
• Percentage of children aged 12-14 years who have ever worked for cash
• Percentage of children aged 12-14 years who are currently in paid employment

• Percentage of households having access to safe water
• Percentage of households having access to sanitary means of excreta disposal

Sample design and implementation
The EMICS sample is a multi-stage, stratified probability cluster sample representative of Egypt’s population in 21 governorates. The five frontier governorates (North and South Sinai, Red Sea, Matrouh and New Valley) are excluded because only 1.5 per cent of the population resides in these regions. Two sample frames were utilized in the sampling process to ensure accurate national representation. The census frame for rural and urban administrative units (villages in the former and shiakhas in the latter), as defined by CAPMAS, was used for the first sample. A second frame was used to select random housing areas as identified by the Ministry of Local Administration. The two frames overlap since random housing areas are sometimes complete administrative units within the census frame. In other cases they are secluded and self-contained settlements within or outside a larger administrative unit, or otherwise they start and proliferate around the hypothetical borders between two adjacent units.

• Sample A encompasses 132 primary sampling units (PSU) from 21 governorates and comprises 65 villages and 67 urban shiakhas. All PSU were selected with probability proportional to estimated population size in 1995. Over-representation occurred in the samples from six areas where UNICEF has area-based programme activities (Greater Cairo, Alexandria, Assiut, Sohag, Qena and Aswan) and for which estimates of different indicators are required. A total of 8,523 households were selected; 8,409 households were successfully interviewed and only 114 households (1.3 per cent of those contacted) refused to participate in the survey. Sample results are weighted to compensate for the over-representation of some governorates.
• Sample B comprises 46 random housing areas (PSU) distributed among 21 governorates. A total of 2,564 households were contacted and only nine cases of non-participation were encountered. According to the Ministry of Local Administration census, random housing areas in Greater Cairo and the three urban governorates of Alexandria, Port Said and Suez accommodate 50-60 per cent of the total population estimated in all random housing areas. Of the 46 PSU in sample B, 24 were selected from these governorates, 13 from other governorates in Lower Egypt and 9 from Upper Egypt. Sample results are weighted to match the estimated population distribution in these regions by area of residence.

As a general strategy, whenever the estimated population of a selected PSU exceeded a pre-set measure of 10,000 inhabitants in either planned or unplanned urban areas, or 5,000 inhabitants in rural areas, the PSU was divided into a number of parts that on average would have equal populations, and then only one part was selected. The selected part within the PSU or the overall PSU (if its size was less than the pre-set limit) was sketched and all buildings inside the designated boundaries were listed by block. Lists were then made to provide counts of occupied housing units, non-occupied housing units and commercial units in every listed structure. These listings defined the frame for selecting second stage units: four segments of housing units within every listed PSU. In the final stage, all households residing in all housing units in a selected segment were eligible for the survey interview.

Work on the sample design extended over three months (October to December 1995). A one-week training course for listers and supervisors was carried out at the Social Research Center in the last week of March, and implementation of the sample design (field operations necessary to obtain the listings) was completed in seven weeks during April and May 1996.

Survey instruments
The main instrument used for EMICS was a modified version of the standard questionnaire developed by UNICEF for multiple indicator cluster surveys. Except for two modules on salt iodization and vitamin A deficiency, all sections of the standard questionnaire were incorporated into the survey. Adaptation of the original instrument took two forms: changing the layout and format of some sections and introducing new sections or additional questions to existing sections.

• A roster of eligible members of the household: mothers or care-takers of children under 15 years of age and married women of reproductive age together with all infants and children under 15. Only usual residents were listed, visitors were not included. Information on age, sex and education attainment was obtained for every listed individual. School repeat and drop-out status plus data on child labour were collected for children aged 5-14 years.
• Contraceptive use among married women of reproductive age.
• Source of drinking water and sanitary means of excreta disposal.

• Source of ORS packets and awareness of mothers and care-takers of serious signs of diarrhoea and pneumonia.
• Prevalence of diarrhoea in the preceding two weeks, feeding practices during the diarrhoeal episode, awareness of mother or care-taker of the importance of increasing fluids and continuing feeding during the diarrhoeal episode, and source of medical consultation.
• Prevalence of pneumonia, treatment provided and source of consultation.
• Breast-feeding status and feeding practices.
• Information on all vaccinations the child received. Data obtained either from an official certificate or directly from the mother or care-taker if no certificate was available.
• Height and weight of each child under five.
• Tetanus toxoid vaccination among mothers of children under five.

• Hearing problems
• Loss of sight in one or both eyes
• Speech impediment
• Malfunctioning of upper or lower limbs
• Mental retardation
• Chronic health conditions such as diabetes, epilepsy, renal failure, cancer or heart conditions