Household Budget Survey 2017

The formation of a sample of households is carried out in two stages.

At the first stage, the number of primary sampling units (PSUs) for each region is determined separately for the city and village. Mahallas are used as primary selection units in urban strata, and rural mahallas and kishlaks are used in rural strata.
Primary selection units are used in surveys for six months, after which they are completely replaced by new ones (100% semi-annual rotation). Thus, two samples of the PEO are taken in one year. The number of selected PEOs is 208 units for each half year, therefore, 416 PEOs per year. Similar figures for each territory (the Republic of Karakalpakstan, provinces and the city of Tashkent) are on average about 15 and 30 PEOs, respectively. When forming the list of PEOs, hard-to-reach or inaccessible PEOs are excluded. The size of each PEO selected should not be less than the number of households surveyed, taking into account refusals and other non-productive visits, as well as the insufficient quality of household lists in it.

At the second stage, after the selection of the PEO, prior to field work for all selected communities and villages, a list of all households living in them is compiled, from which a simple random selection is made by region based on urban and rural strata. When stratifying households by region, a disproportionate method is used (based on the square root of the number of households) to ensure the required minimum sample size for each region and the required accuracy of results at the territorial level.