Sample design

Sampling frame: The first flow sampling frame for Household Budget Survey was obtained from National Address Database.

Final sampling unit selection frame: Blocks were selected by the probability proportional to size sampling (pps) from the blocks (in urban areas or rural areas who have minicipality organisations) and from the villages formed by using the above mentioned frame for the selection of households. The sample units are systematically selected from each block.

Final sample unit: Household live at the address is defined as the final sampling unit.

Sampling method: Stratified two-stage cluster sampling method is used.

Non-response rate: The non-response forms were filled in case the selected households according to the sampling techniques couldn't been surveyed and the population weights were calculated considering non-responses. For 2015 Household Budget Survey, the non-response rate was 24.7% for overall Turkey.

Estimation dimension: The estimation level of annual Household Budget Survey is whole Turkey. It's possible to have estimates on SR Level 1 and Level 2 only for consumption expenditure variable by using combined 3-year-data sets of household budget surveys.

Population weights: Results of the household budget survey have been weighted and published by the most recent population projections. Until year 2009 aforesaid population projections was calculated based on the general population censuses. In 2007 Address Based Population Registration System (ABPRS) was established. There are some differences in the distribution of population by age, sex and regions in ABPRS compared to the censuses. In this direction, the new population projections were produced using the most recent population data obtained from this new system. National and regional population projections were renewed. The weights based on new population projections have been used for the Household Budget Surveys.

Top of page