ln traditional offIine panels (for example consumer panels) the panel base is re­newed in regular intervals to avoid the so called "panel effect". The panel effect can be described as follows: the panel is a continuous survey, which should de­scribe the market reaction over time. This leads to the fact that people get used to the survey. As a result a bi as in answering and an unconscious change in the consumers behaviour occurs (dependent on the panel technique). This uncons­cious learning process of the interviewed persons leads to a bias of the desired results.

ln case of an online panel the initial position is basically different: The panel ­participants are not regularly interviewed to the same topic. but participate in all different kinds of surveys. Nevertheless there occur consistently concerns that the duration of the belonging to a panel might have an impact on the answering behaviour. The argument is, that a "professionalization" of the participants takes place, which leads to a bias of the answers. We would like to face this hypothesis by interviewing panel participants who just became a member of the panel and participants who have experience over some years. Both groups with the same structure are compared in respect to the willingness of participation, the average time of reaction, the duration of answering and at least the answering behaviour. The survey will give information if a panel effect exists and if so, if it can be applied for different kinds of questions on the same level.