Sampling

Sampling Procedure

The principal objective of the sample design was to provide current and reliable estimates on a set of indicators covering the four major areas of the World Fit for Children declaration, including promoting healthy lives; providing quality education; protecting against abuse, exploitation and violence; and combating HIV/AIDS.  The population covered by the 2006 MICS is defined as the universe of all women aged 15-49 and all children aged under 5.  A sample of households was selected and all women aged 15-49 identified as usual residents of these households were interviewed.  In addition, the mother or the caretaker of all children aged under 5 who were usual residents of the household were also interviewed about the child.

The 2006 MICS collected data from a nationally representative sample of households, women and children.  The primary focus of the 2006 MICS was to provide estimates of key population and health, education, child protection and HIV related indicators for the country as a whole and for urban and rural areas separately.  In addition, the sample was designed to provide estimates for each of the two entities: the Federation of Bosnia and Herzegovina (FBiH) and the Republika Srpska (RS) for key indicators (owing to the constraints in the survey budget, the Brcko District is represented in the same way as other municipalities in BiH).  Each entity is subdivided into municipalities.  In addition municipalities in Federation of BiH are grouped into 10 cantons. Each municipality is divided into settlements, settlements into statistical circles and each circle into enumeration areas. In total BiH includes 154 municipalities, 12 thousands circles and 18 thousands enumeration areas.  The sample frame for this survey was based on list of enumeration areas developed from the 1991 population census. In 2006 the update of 1500 enumeration areas was done and this master sample frame was used for sample selection.

The primary sampling unit (PSU), the cluster for the 2006 MICS, is defined on the basis of the enumeration areas from the master sample frame.  A total of 455 census enumeration areas were systematically selected from the Master Sample with equal probability.  All households from 455 census enumeration areas were allocated to two household lists. The first list (type 1) consisted of all households with children under five, and the second list consisted of all other households. 3,000 households having equal selection probability were selected from each list. This meant that each household from the list had the same selection probability.  As the lists were different, the households with different sizes from different lists had different selection probability. Thus, a sample was obtained, which was self-weighted at the level of each list but is not self-weighted at the national level.

The number of households within each cluster is unequal and proportional to the cluster size.

The households in each list were implicitly stratified, i.e. sorted by entity/district, by urban/rural classification, by order of census enumeration area within the municipality, and by ordinal number within the cluster. 

No replacement of households was permitted in case of non-response or non-contactable households.  Adjustments were made to the sampling weights to correct for non-response, according to MICS standard procedures.

The sampling procedures are more fully described in the sampling design document and the sampling appendix of the final report.

Deviation from Sample Design

No major deviations from the original sample design were made.  All sample enumeration areas were accessed and successfully interviewed with good response rates.

Response Rates

At the BiH level (Table HH.1), 5,549 households were successfully interviewed and the response rate reached 93.4 percent. In the interviewed households, 4,977 women were identified within the sample range, out of which 4,890 were interviewed. A total of 3,209 children under five years-of-age were listed in the household questionnaire, and the questionnaire was completed for 3,188 children. The ratio of responses for children under five differed significantly between rural areas (89.0 percent) and other areas (95.2 percent). 

In the Republika Srpska, 2,019, out of the 2,129 households selected for the sample, were successfully interviewed, yielding a household response rate of 96.0 percent. In the interviewed households, 1,658 women aged 15-49 were identified and 1,620 successfully interviewed (97.7 percent response rate). In addition, out of the 1,086 children under the age of five listed in the household questionnaire, 1,071 had their questionnaires completed, which corresponds to a response rate of 98.6 percent. Overall response rates of 93.8 and 94.7 percent are calculated for the 15-49 women's and under-fives' interviews respectively. 

In the Federation of BiH, of the 3,744 households selected for the sample, 3,710 were available for interview, and 3,413 were successfully interviewed (92 percent response rate). In the interviewed households, 3,221 women (aged 15-49) were identified and 3,175 successfully interviewed, yielding a response rate of 98.6 percent. In addition, 2,065 children under the age of five were listed in the household questionnaire and questionnaires were completed for 2,060 of these children (99.8 percent response rate). Overall response rates of 90.7 and 91.8 percent are calculated for the 15-49 women's and under-fives' interviews respectively.

Weighting

Sample weights were calculated for each of the datafiles.  

Sample weights for the household data were computed as the inverse of the probability of selection of the household, computed at the sampling domain level (household with children under 5 and all other households).  The household weights were adjusted for non-response at the domain level, and were then normalized by a constant factor so that the total weighted number of households equals the total unweighted number of households.  The household weight variable is called HHWEIGHT and is used with the HH data and the HL data.

Sample weights for the women's data used the un-normalized household weights, adjusted for non-response for the women's questionnaire, and were then normalized by a constant factor so that the total weighted number of women's cases equals the total unweighted number of women's cases.

Sample weights for the children's data followed the same approach as the women's and used the un-normalized household weights, adjusted for non-response for the children's questionnaire, and were then normalized by a constant factor so that the total weighted number of children's cases equals the total unweighted number of children's cases.
Generated: MAR-18-2008 using the IHSN Microdata Management Toolkit