Notice: the WebSM website has not been updated since the beginning of 2018.

Web Survey Bibliography

Title Relation between values and topic of a survey in internet panel research
Year 2009
Access date 22.11.2009
Abstract

Under

‐representation of specific groups in survey research is a common problem. People with a low economic status, people living in urbanized areas or lonely people are examples of groups that are hard to reach. Internet panel surveys are no exception and create even higher thresholds for participation. The recruitment of the LISS panel, an online multi purpose panel of 5000 households representative for the Dutch speaking people in the Netherlands, is developed very carefully with a probability sample, many contact attempts and the use of incentives. During the recruitment of the LISS panel many efforts and measures were taken to get the highest possible response. The response was really high, and often‐missing groups are (under‐) represented. Being confronted with under‐represented groups in (panel) survey research the following questions raise: in what way the survey results will be biased by the under‐representation; is the response rate per questionnaire of these groups lower than average. ‐western immigrants are more renters than home‐owners. These examples concern demographic variables that can be verified with data from Statistics Netherlands or from other national databases and therefore weight factors can be calculated for these variables. However, the LISS panel data also cover many subjective topics, which are not known at the population level. Do people from underrepresented groups differ in their opinions about topics, such as, emancipation, upbringing, gen‐technology or women rights? Or, are these groups homogeneous in their opinions? ‐represented groups have different values than others and that these values intermediate between the exogenous demographic characteristics, the topic involvement and the willingness to participate in surveys.

Examples of how survey results can be biased are: the oldest old make more use of health care; among non

Although there is no topic that pleases a whole panel, certain topics are more popular than others. From literature we know that sample members who are interested in the topic of the questionnaire have a higher chance to respond than those who are not interested (Martin 1994, Roose et al. 2003). Wijnen et al. (2006) found a relationship between personal values, topic involvement and intention to reply in a mail survey. For our study we will use value theory to better understand why certain groups participate less in surveys than others. Values are relatively stable motivational characteristics of persons that change little during adulthood (Rokeach, 1973; Schwartz, 1992,1997; Schwartz & Bilsky, 1987, 1990, Feather, 1971; Bardi & Schwartz, 2003) and recently, Inglehart, Welzel and Welzel have also shown that values are different for different cultures. We hypothesize that the before mentioned under

Access/Direct link

Conference homepage (abstract)

Year of publication2009
Bibliographic typeConferences, workshops, tutorials, presentations
Print

Web survey bibliography - WAPOR 62nd Annual Conference, 2009 (13)