World Values Survey 2006, Wave 5

Step One: Distribution of sampling points by region and urban/rural strata The sample was designed to represent the national population of voting age by: region, type of residence, gender, age and ethnicity. When conducting representative polls on a national level, BBSS employs the method of double-clustered random probability sampling, stratified by regions (28 in total). The database of ESGRAON was used as basis for the sample. This register system contains every inhabitant in terms of the following characteristics: full three names of each person; address (settlement, street and number of home of the person). Upon this basic information various other aggregates are later superimposed, such as: election precincts settlements municipalities regions. Universe was stratified by regions (28 in total); lists of election precincts in each of the regions was prepared containing the number of persons aged 18+ in each precinct; given the targeted size of the final sample (1000), the number of individuals to be interviewed in each region was determined. This size is in direct proportion to the relative share of the respective region within the universe; next stage relates to random selection of sampling points to be used in the survey. Random selection of sampling points to be used in the survey. The sampling points were chosen to represent regions and type of settlement.

Step Two: Selection of sampling points and replacement of sampling points The random selection of the sampling points was done by means of the following algorithm: Calculation of the number of clusters to be achieved in each region (number is proportional to the size of the region); Sampling points in each region arranged in a descending order based on the criterion number of persons aged 18+ in the cluster. (NB: sampling points, that is, election precincts are comparatively uniform in terms of scale. Each sampling point contains between 400 1 000 persons aged 18+.); A cumulative column was formed by the number of individuals contained in each cluster; Based on this cumulative column, the systematic selection was achieved of the necessary number of sampling points starting by using a random start-up figure, and then a step was applied for moving down the cumulative column which step is the quotient of the size of the regional sub-universe and the number of respondents in each sampling point.

Step Three: Selection of starting points within each sampling point. Starting points were chosen by BBSS Gallup International Headquarters

Step Four: Respondent Selection Selection of a respondent is carried out using last birthday method. Every interviewer is provided with a starting address. The interviewer has to follow the rules, depending on the type of settlement an interview will be carried out.

   1. Urban areas - the selected household is each third address on the left-hand side ofthe street in urban areas, applying left turn at junctions and going back to the last crossing, if one has reached a dead-end, and        further proceeding at random but not along the branching one had been through. In a block-of-flats of up to four floors, the selected household is every fifth apartment, counting from the first on the left on the ground floor. In cases of unsuitable household (e.g. an eligible respondent is not present, the person is less then 18 years old) procedure instructs to approach the next-door apartment and to contact each further till reaching the required one, from which point to resume the standard step of every fifth apartment. In a blockof-flats of 5 floors and more, the selection is every tenth apartment counting them the same manner. 
   2. In rural areas, the selected household is every fourth inhabitable house/dwelling on both sides of the interviewers route/track and where the houses are aligned or scattered over larger territory, the instruction requires applying wave-wise approach selecting the fourth, counting from the first house on the left. In compounds of several houses behind a common fence, the procedure instructs to select the fourth one from the left (counting from the gate), or if there are less than four houses behind a common fence, then the interviewer to get out of the common yard, counting the houses as if they were along the street. In compact and wellstructured villages the selection procedure follows the instructions for urban areas. Next birthday was determined when respondents asked the person they had contacted, who is the person with the last birthday in the household. They were not listed in writing as this may be considered as personal information which we are not authorized to collect. Step Five: Respondent Substitution Substitutions were made by contacting a person with the last birthday in the household next-door. The step of every fifth apartment is resumed after finding an eligible respondent. Step Six: Callbacks (rate, method, and results) Call-backs were made usually during the same or during the next day, after the eligible respondent was not found.

The sample size for Bulgaria is N=1001 and includes population aged 18 and over for both sexes.