The CentER panel was established in 1991 and comprises over 2,000 households, forming a representative cross-section of the Dutch population. It offers the combined advantages of computer-controlled surveying and panel research: high working speed, means of consistency control and reliable measurement of changes.
Many web panels are not based on probability sampling but on self-selection. This means that survey researchers have no control over the selection process. For the set-up and continued running of the CentER panel, self-selection was deliberately rejected as a method.
Instead, a random sample of addresses is drawn from the pool of national postal delivery addresses. These addresses are, where possible, enriched with telephone numbers. Owning a computer with an Internet connection is not a requirement. If a household does not have a computer with an Internet connection, they are provided with a so-called SimPC with which to complete questionnaires. This is a PC that is easy to operate, developed specifically for people with low computer skills.
The composition of the panel is permanently monitored to ensure a continued accurate reflection of the Dutch population. Any deviations are corrected for during recruitment, as much as possible.
Since questionnaires are submitted to the CentER panel every weekend, data can be collected rapidly. The data are disseminated shortly after the fieldwork, enriched with background information such as gender, position in the household, age, type of housing, income, education level, urbanity and province. The data can also be linked to other questionnaires completed by the CentER panel. One example is the longitudinal study into determinants of saving behavior (DNB Household Survey, DHS), which is conducted annually. Another example is the eight-weekly health monitor, containing questions about perceived health, use of health services and satisfaction with the personal life environment.