Internet surveys are by far the fastest and cheapest way to gather data, but a disadvantage is that in general not everybody is reached when this method of data collection is used. It is not easy to try to correct this (see e.g., Groves et al., 2004). Not only people with low social economic status, people who live in urban areas or elderly people (see e.g., Feskens et al., 2006, 2007; Schmeets et al., 2003; Stoop, 2005), but also people who don’t have a computer or Internet access or don’t want this are examples of groups that are hard to reach. This paper investigates why it is important to try and recruit these special groups for an Internet panel. Our questions are threefold: Why are some groups difficult to reach? How do they differ from other people? And what are their unique contributions to an Internet panel?

Our research is conducted in the CentERdata LISS-panel, because the design of the LISS-panel combines a probability sample and traditional recruitment procedure with online interviewing. The panel consists of about 5000 households representative of the Dutch speaking population. A specialty of this panel is that people without Internet access are provided with the necessary equipment so that they are able to participate in the panel.

More specifically, we answer the first question by looking at what part of the recruitment process people fall out and by coding reasons for this behavior. For the second question, we take into account background variables and will investigate amount of time spent on the Internet, in what way people who were provided with a new Internet connection and PC differ on demographics from other panelmembers. Finally, we look at unique contributions of special groups on attitudes. Also, some preliminary analyzes on attrition will be done.

Suggestions for recruiting special groups in Internet panel research will be made.