Sampling

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 2005 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 2005 MICS collected data from a nationally representative sample of households, women and children.  The primary focus of the 2005 MICS was to prodvide 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 additon, the sample was designed to provide estimates for each of the 6 regions (Vojvodina, Belgrade, West, Central, East and South-East Serbia) for key indicators.  Separate sample was designed for Roma living in Roma settlements.

Important factors which influenced the sample design of both Serb and Roma samples are the fertility rate and number of household members. 

A stratified, two-stage random sampling approach was used for the selection of the survey sample. 

In the case of the Serbia without the Roma settlements sample, 400 census enumeration areas within each region with probability proportional to size were selected during the first stage. Since the sample frame (Census 2002) was not up to date, household lists in all selected enumeration areas were updated prior to the selection of households. Owing to the low fertility rate and small household size, households were stratified into two categories. One category of households consists of households with under 5 children, while the other category consists of households without children under 5. The allocation of the sample in the category of households with children was significantly greater than the allocation of the sample in the category of households without children. Based on the updated information, selected units were divided into clusters of 18 households on average, plus 3 backup households. Backup households were interviewed only if some of the first 18 households were not found. In the event that a household refused to be interviewed, a backup household was not contacted. In each cluster, the number of households with children was selected with probability proportional to size. 

In the case of the Roma population, the universe could be defined only for Roma who live in separate settlements. During the first stage, 106 census enumeration areas were selected. The updating of household lists was done prior to household selection, but there was no need for sample stratification of households with and without children under 5. The average number of households selected in each cluster was 18 on average, plus 3 backup households. 

Secondly, after the household listing was carried out within the selected enumeration areas, a systematic sample of 7,794 households in Serbia without Roma from Roma settlements and 1,959 Roma households was drawn up, which makes a total of 9,953 sampled households.

The 2002 Serbian Population Census framework was used for the selection of clusters. Census enumeration areas (app. 100 households) were defined as primary sampling units (PSUs), and were selected from each of the sampling domains by using systematic pps (probability proportional to size) sampling procedures, based on the estimated sizes of the enumeration areas from the 2002 Population Census. The first stage of sampling was thus completed by selecting the required number of enumeration areas from each of the 6 regions by urban and rural areas separately.

Following standard MICS data collection rules, if a household was actually more than one household when visited, then a) if the selected household contained two households, both were interviewed, or b) if the selected household contained 3 or more households, then only the household of the person named as the head was interviewd.

The Serbia Multiple Indicator Cluster Survey sample is not self-weighted.  For reporting of national level results, sample weights were used, 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.

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

Of the 9.953 households selected for the sample, 9.372 were found to be occupied. Of these, 8.730 were successfully interviewed for a household response rate of 93 percent. In the interviewed households, 7.895 women (aged 15-49) were identified. Of these, 7.516 were successfully interviewed, yielding a response rate of 95 percent. In addition, 3.838 children under the age of five were listed in the household questionnaire. Questionnaires were completed for 3.777 of these children, which corresponds to a response rate of 98 percent. Overall response rates of 89 and 92 percent are calculated for the women's and under-5s' interviews respectively.

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 (urban/rural within each region).  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.  

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.

For merged Serbia and Roma samples, additional customisation of calculated weights was performed. Since according to the 2002 Census, the proportion of Roma households in the total Serbia household population is 1 percent, the proportion of women aged 15 to 49 is 1.3 percent and the proportion of children under 5 is 3.4 percent; the final weights are products of normalised weights and the ratio of proportion of both the Roma and non-Roma population.

The household weight variable is called HHWEIGHT and is used with the household data. Same weight is applied to the household listiong data, but under name HLWEIGHT. Women and children weights are called WMWEIGHT and CHWEIGHT and are used in women and children data, respectively.