PropertyValue
?:abstract
  • In order to lower nonresponse bias in a survey, Peytchev et al. (2010) suggested prioritizing low propensity cases during fieldwork. The aim of the case prioritization approach is to balance response propensities across different strata of the sample. I argue that a lacuna in this discussion is that it remains unknown which cases to select for prioritization and how specific selection criteria affect nonresponse bias across a variety of different variables. To shed light on this pressing methodological question, I relied on the Rolling Cross-Section Campaign Survey 2013 (RCS) of the German Longitudinal Election Study. The RCS 2013 was conducted as a computer assisted telephone interview (CATI) with a response rate of 15.5%. Of the 7,882 respondents that were interviewed prior to the election, 67.9% participated in a post-election re-interview. Based on these data, I simulated the application of case prioritization for the post-election interview with varying selection criteria. I further included different scenarios in the simulation that account for different methods of treating the selected cases and how these methods perform for the different propensity strata. Nonresponse bias (prior and posterior to applying case prioritization) was assessed for 120 socio-demographic, attitudinal, and behavioral variables. The findings of the simulation study suggest, first, that the effects of applying case prioritization are estimate-specific and, thus, vary between variables. Second, there are estimate-specific thresholds (i.e., optimal selection criteria) that mark the maximum reduction of bias that can be achieved by prioritizing cases. That is, the optimal selection criterion for one variable is not necessarily the optimal selection to minimize nonresponse bias for another variable. Third, selecting cases beyond that threshold does not further lower nonresponse bias for the respective variable. (xsd:string)
?:author
?:comment
  • (GLES) (xsd:string)
?:dataSource
  • GLES-Bibliography (xsd:string)
?:dateCreated
  • 6. Fassung, Januar 2017 (xsd:gyear)
?:dateModified
  • 2016 (xsd:gyear)
?:datePublished
  • 2016 (xsd:gyear)
?:duplicate
is ?:hasPart of
is ?:mainEntity of
?:name
  • Comparing different selection criteria when applying case prioritization: A simulation study (xsd:string)
?:publicationType
  • inproceedings (xsd:string)
?:reference
?:sourceCollection
  • ESA RN 21 Midterm Conference (xsd:string)
?:sourceInfo
  • Bibsonomy (xsd:string)
  • In ESA RN 21 Midterm Conference, 2016 (xsd:string)
?:startDate
  • 13.10-15.10.2016 (xsd:gyear)
?:studyGroup
  • German Longitudinal Election Study (GLES) (xsd:string)
?:tags
  • 2016 (xsd:string)
  • FDZ_Wahlen (xsd:string)
  • GLES (xsd:string)
  • GLES_input2016 (xsd:string)
  • GLES_pro (xsd:string)
  • GLES_version6 (xsd:string)
  • ZA5703 (xsd:string)
  • checked (xsd:string)
  • inproceedings (xsd:string)
rdf:type