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

Web Survey Bibliography

Title The Uses of Paradata for Evaluating Alternative Versions of Web Survey Questions
Year 2004
Access date 11.06.2004
Abstract Procedures developed recently by Heerwegh (2003) make it possible to collect client-side paradata for all respondents to web surveys. It is possible through paradata to find out: 1) the time required for responding to each item, 2) whether any answers are changed by respondents, and 3) the order in which answers are provided to multiple answer questions (e.g. check-all items). In this paper we evaluate the use of paradata for understanding how alternative question designs affect respondents, and the strengths and weaknesses of each question format. Inasmuch as paradata may be collected from all respondents to a survey it has the potential of being another tool that will help researchers determine which question structures are most and least desirable for asking certain survey questions, thus complementing information about response distributions and that which might be gained from cognitive interviews. Four alternative question structures were developed for test items in an effort to understand how different visual layouts or designs affect respondent answers. Analysis of the paradata is aimed at helping to explain why and how different answer distributions are produced, and whether some formats appear easier to answer than do others. The overarching goal of this research is to be able to specify the strengths and limitations of this use of paradata for evaluating questions used in public opinion surveys.
Year of publication2004
Bibliographic typeConferences, workshops, tutorials, presentations