Mongolia - HSES 2014

A stratified two-stage sample design was used for the 2014 HSES. The sampling frame was based on the 2013 administrative database that has information on the number of households for all administrative units. The primary sampling units (PSUs) selected at the first stage were based on the bags, which are the smallest administrative units in Mongolia. However, in the case of Ulaanbaatar, the PSUs were the enumeration areas (EAs) defined for the 2010 Mongolia Census. The sampling frame was first stratified by the 21 aimags and Ulaanbaatar. The Capital of Ulaanbaatar is an individual urban stratum. The remaining aimags were further divided into the following strata: urban, consisting of the aimag capital; and rural, divided into small town and countryside substrata. At the first sampling stage the clusters within each stratum were selected systematically with probability proportional to size (PPS), where the measure of size was based on the total number of households in the cluster from the administrative frame. Total of 1,800 PSUs or clusters were selected (900 sample urban clusters and 900 sample rural clusters). At the second sampling stage an updated administrative list of households was obtained for the sample bags enumerated each month. The sample households within each sample cluster were selected systematically with equal probability. (10 households for urban clusters and 8 households for rural clusters).

Total sample size of 16,200 households for the 2014 HSES. A total of 624 sample households was selected for each of the other aimags, except for the smaller aimags of Govisumber, with 216 sample households; Darhan-Uul, with 552 sample households; and Orhon, with 600 sample urban households there are no sample clusters in the rural stratum because of the small number of rural households in this aimag.

The use of this sampling procedure means that households living in different areas of the coun­try have been selected with different probabilities. Therefore, in order to obtain representative sta­tistics for each stratum and for the coun­try as a whole, it was necessary to use sampling weights. The weight which was assigned to each household corresponds to the inverse of the selection probability and takes the sampling strategy into account.