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

Web Survey Bibliography

Title Are reluctant panelists worth the effort? Analysis of measurement error in a probability-based Internet panel.
Year 2011
Access date 02.08.2011
Abstract

Internet panel surveys based on probability samples of the population present new and exciting opportunities for gathering social science data. Yet despite their considerable appeal, particularly with respect to measurement quality (for example, the potential to reduce bias on socially sensitive questions, the avoidance of interviewer effects, and the possibility to exploit all the benefits of computer-assisted questionnaire administration), threats to data quality persist in the form of nonresponse and attrition, and the possibility that over time, the responses of even the most loyal panelists may exhibit effects associated with learning or declining motivation. As interest in the relation between different types of survey error and their implications for survey costs grows among survey methodologists, a question that arises in the context of probability-based Internet panels is whether the resources needed to recruit and retain those sample members who are least motivated to join and participate in the panel can be justified in terms of the resulting quality of the data such respondents contribute. Given that factors influencing the decision to participate in a survey may also influence the respondents’ motivation and ability to respond to the survey questions, variations in the quality of responses may simultaneously be affected by both non-response bias and measurement error. We address this question using data from the 2008-09 American National Election Studies Internet Panel Survey, a specially-designed study based on an RDD sample of the US electorate. Using interview and paradata from the initial recruitment survey, we examine the relation between the level of effort needed to recruit panelists, their willingness to participate in subsequent panel waves and the quality of responses given when taking part, using a range of indicators of data quality, including effects associated with satisficing (Krosnick, 1991).

Access/Direct link

Conference Homepage (abstract)

Year of publication2011
Bibliographic typeConferences, workshops, tutorials, presentations
Full text availabilityFurther details
Print