Web Survey Bibliography

Title Use of Don't Know and No Opinion Responses in Web Surveys
Author Tourangeau, R., Baker, R. P., Couper, M. P., Conrad, F. G.
Source RC33 6th International Conference on Social Science Methodology: Recent Developments and Applications in Social Research Methodology, 2004
Year 2004
Access date 14.09.2004
Abstract Among the many issues faced by Web survey designers is deciding whether to offer an explicit “Don’t Know” or other no opinion response. Such responses are frequently not read to respondents in interviewer administered surveys but are used at the interviewer’s discretion once he or she concludes that the respondent either cannot or will not provide a valid answer. Self-administered surveys offer no such luxury. A wide variety of factors may contribute to respondents not answering a specific survey question. They include personal knowledge, topic salience, and fatigue. Whether a no opinion response is offered also has been shown to have an effect. In self-administered surveys such as Web where visual presentation is a critical factor, how the response is presented can also influence response. This paper discusses the impacts of no opinion response presentation in Web surveys. It reports on an experiment that varied and crossed three question presentation formats in a six-cell design. The three presentations formats were: · Presentation of a no opinion response vs. no presentation · Presentation of a no opinion response in an answer set arrayed horizontally on the screen vs. presentation in an answer set arrayed vertically · Presentation of a no opinion response separated by a delimiter from the substantive answers vs. no separation. These presentation formats were varied across questions about both behavior and attitudes. Answer formats included fully labeled scales and scales labeled only at their end points. This design was tested in a Web survey conducted in December 2003, using both an opt-in panel and an intercept-based sampling approach. A total of 2,727 respondents completed the survey. Analysis of these data is focusing on the degree to which different presentation formats may encourage or discourage non-reporting either by selection of no opinion response or by simply skipping the question. A second analytical issue is the impact of presentation format on the distribution of substantive responses. For example, do formats that discourage nonresponse generate different distributions that those that don’t?
Year of publication2004
Bibliographic typeConferences, workshops, tutorials, presentations
Print

Web Survey Bibliography - Conrad, F. G. (70)

Page:
  • 1
  • 2
Page:
  • 1
  • 2