In survey research it is important to have a good representation of the population. However, it appears that young adults, full-time workers, inhabitants of large cities and ethnic minorities are difficult to survey. This research has studied how these groups can be reached better. In an experiment, potential survey participants (respondents) could choose how they wanted to participate: by a web survey, a personal interview or a telephone interview. Respondents who chose the web viewed pre-recorded clips of an interviewer reading the questions to them for half of the survey (video-web survey). The experiment showed that young adults and full-time workers have a preference for web surveys. So offering these groups this method in future surveys could improve their representation. However, survey methods such as web or telephone can also affect the way respondents answer questions. Respondents can invest minimal effort when participating or can adapt their answers to social norms (socially desirable responding). The experiment showed that the degree of socially desirable responding and investing effort in video-web was equivalent to the web survey without video. The answers of both web surveys differed from the answers given in the personal and telephone interviews. From this it can be concluded that adding videos in a web survey does not improve data quality for web surveys, but does also not deteriorate data quality. Socially desirable responding is further investigated conducting three experiments. These experiments provided more insights about how the sensitivity of questions for socially desirable responding can be established.