Summary
Somalia - Multiple Indicator Cluster Survey 2011, Northeast Zone
The primary objective of the sample design for the Northeast Zone Multiple Indicator Cluster Survey was to produce statistically reliable estimates of most indicators for the whole Northeast Zone, for urban and rural areas, and for the three regions (Bari, Nugal and Mudug) of the Zone. There were two main sampling strata: urban and rural areas.
A multi-stage, stratified cluster sampling approach was used for the selection of the survey sample.
The target sample size for the Northeast Zone MICS was calculated as 5,179 households. For the calculation of the sample size, the key indicator used was the polio immunization coverage for children aged 12 – 23 months.
The sampling frame was the list of settlements obtained from the 2005/2006 UNDP settlement census and which was updated in preparation for the Somalia population estimation survey. For each settlement, this list contained an estimated number of households and the classification by urban and rural.
Stratification consisted of separating urban and rural settlements within each region. Settlements were then used as primary sampling units and were selected with probability proportional to size, the size being the estimated number of households. Very large settlements were selected with certainty as selfrepresenting units (that is with probability equal to 1).
In rural areas and small towns, settlements with more than 200 households were divided into segments of which one was randomly selected. All households in the selected segment were listed to create a frame for the selection of 18 households at the second stage using systematic sampling.
For very large settlements, the list of villages and sections that comprised each settlement served as frame for the second stage selection (secondary sampling units). Each selected village and section was segmented if it contained more 200 households. One of the newly created segments was then randomly selected and all of the households it contained were listed. In the final stage, 18 households were selected from the household listing. In villages and sections containing 200 households or less, a complete household listing was carried out and 18 households were directly selected from the list of households.
The sampling procedures are more fully described in "Multiple Indicator Cluster Survey 2011 - Final Report" pp.123-124.
A multi-stage, stratified cluster sampling approach was used for the selection of the survey sample.
The target sample size for the Northeast Zone MICS was calculated as 5,179 households. For the calculation of the sample size, the key indicator used was the polio immunization coverage for children aged 12 – 23 months.
The sampling frame was the list of settlements obtained from the 2005/2006 UNDP settlement census and which was updated in preparation for the Somalia population estimation survey. For each settlement, this list contained an estimated number of households and the classification by urban and rural.
Stratification consisted of separating urban and rural settlements within each region. Settlements were then used as primary sampling units and were selected with probability proportional to size, the size being the estimated number of households. Very large settlements were selected with certainty as selfrepresenting units (that is with probability equal to 1).
In rural areas and small towns, settlements with more than 200 households were divided into segments of which one was randomly selected. All households in the selected segment were listed to create a frame for the selection of 18 households at the second stage using systematic sampling.
For very large settlements, the list of villages and sections that comprised each settlement served as frame for the second stage selection (secondary sampling units). Each selected village and section was segmented if it contained more 200 households. One of the newly created segments was then randomly selected and all of the households it contained were listed. In the final stage, 18 households were selected from the household listing. In villages and sections containing 200 households or less, a complete household listing was carried out and 18 households were directly selected from the list of households.
The sampling procedures are more fully described in "Multiple Indicator Cluster Survey 2011 - Final Report" pp.123-124.