The target population is defined as “all people residing in Belgium, regardless of their place of birth, nationality or any other characteristic” within the age group of 3-64 years.
The sampling frame is the National Register. The National Register does not cover the target population completely as some people such as homeless and illegal immigrants may be not listed and will be excluded a priori. Additionally, in concordance to the recommendations of EFCOSUM, following persons are not eligible and will be then exculded a posteriori:
- Persons who reside in institutions, elderly homes, nursing homes;
- Religious community or cloister with more than 8 persons;
- People residing abroad or people who moved out of the selected municipality;
- People in prison;
- Persons who are hospitalized during the period of the survey;
- People who do not speak Flemish/French;
- People who died;
- Persons with physical and mental disabilities.
The sample is divided between five age groups (3-5 years; 6-9 years; 10-17 years; 18-39 years and 40-64 years) and in the two genders.
The sampling was carried out in several steps (multistage sampling). Two main sample techniques were used: stratified cluster sampling and systematic random sampling.
A stratified cluster design is selected. Stratification at the provincial level was performed and the sample size within each of the 11 strata is proportional to the population size of the province (all ages)
|Brussels Capital region
Within each provincial stratum municipalities are selected, using a systematic sampling approach, with a selection chance proportional to their size. The number of municipalities to be selected is defined by the sample size of the provincial stratum. These municipalities are called primary sampling unit (PSU). One PSU is a group of 50 individuals (a large municipality has 1 or more PSU according to its size). The chance of selection of a municipality should be proportional to its population size, ensuring that larger cities have a higher selection probability to have at least one PSU or even more than one PSU (depending on its size in relation to the size of the stratum). The methods ensures that at least one of the larger cities is selected. A similar remark holds for the smaller towns and villages. Also from this group elements should be present. By grouping smaller communities by ordering the whole set of communities according to size, the representation of smaller communities out of the pool of smaller communities is ensured. The assumption is made that smaller communities of about the same size are exchangeable with respect to the items of interest.From 589 Belgian municipalities, 64 PSUs (64*50=3200) were selected.