PropertyValue
?:abstract
  • (a) Relevance & Research Question: High drop-out rates are considered a major shortcoming of web surveys and considerably threaten data quality. However, until recently survey breakoff has received limited scholarly attention and knowledge about the reasons causing respondents to terminate surveys early is still fractional. In political science, the topic has been particularly neglected. Enhanced understanding of the complex processes leading to breakoff is needed in order to develop standard guidelines for web surveys that help in minimising drop-out rates. (b) Methods & Data: Connected to the German Longitudinal Election Study (GLES) a seven-wave campaign online tracking has been conducted with about 14,000 respondents, thereof 3,000 drop-outs. This data allows for a detailed analysis of drop-outs: Given the applied quota design, personal information is available even about those respondents who answered some questions but did not finish the survey. As many questions were included in each wave, yet being asked at different questionnaire positions and being surrounded by various items, contextual effects on drop-out can be analysed. In terms of research methods, among other things, life tables were presented to observe how breakoffs are distributed throughout the survey. Discrete-time survival models with page-varying covariates are estimated for each wave separately, including both respondent characteristics as well as questionnaire and page characteristics. (c) Results: Our main findings can be summarised as follows: In accordance with recent research, we find drop-out to be a function of both respondent characteristics and page characteristics. For instance, higher educated people are less likely to break off, whereas open questions tend to produce significantly higher drop-outs. In the course of the survey, the drop-out-risk tends to decrease. Varying the context a question is embedded in, may affect the number of drop-outs, thus breakoff in web surveys is not unchallengeable. (d) Added Value: Our findings confirm some results of previous research dealing with breakoff in web surveys. The reasons for breakoffs thus seem to be largely independent of the survey topic. Even more essential, based on the empirical results, some prospects for reducing the number of drop outs in web surveys by questionnaire and page design are provided. (xsd:string)
?:author
?:comment
  • (GLES) (xsd:string)
?:dataSource
  • GLES-Bibliography (xsd:string)
?:dateCreated
  • 1. Fassung, April 2012 (xsd:gyear)
?:dateModified
  • 2010 (xsd:gyear)
?:datePublished
  • 2010 (xsd:gyear)
?:duplicate
is ?:hasPart of
is ?:mainEntity of
?:name
  • Breakoff in Web Surveys of the German Longitudinal Election Study (GLES) (xsd:string)
?:publicationType
  • inproceedings (xsd:string)
?:reference
?:sourceCollection
  • General Online Research 10 (xsd:string)
?:sourceInfo
  • Bibsonomy (xsd:string)
  • In General Online Research 10, 2010 (xsd:string)
?:startDate
  • 26.05.-28.05.2010 (xsd:gyear)
?:studyGroup
  • German Longitudinal Election Study (GLES) (xsd:string)
?:tags
  • 2010 (xsd:string)
  • FDZ_Wahlen (xsd:string)
  • GLES (xsd:string)
  • GLES_input2011 (xsd:string)
  • GLES_pro (xsd:string)
  • GLES_version1 (xsd:string)
  • ZA5335 (xsd:string)
  • ZA5336 (xsd:string)
  • ZA5337 (xsd:string)
  • ZA5338 (xsd:string)
  • ZA5339 (xsd:string)
  • ZA5340 (xsd:string)
  • checked (xsd:string)
  • inproceedings (xsd:string)
rdf:type