1 / 7

Modeling Campaign Dynamics in the 2008 NAES

Modeling Campaign Dynamics in the 2008 NAES. Richard Johnston University of Pennsylvania. Prepared for the Wivenhoe House Conference on Cyberinfrastructure and National Election Studies 8-9 June 2007. Mode Considerations. Web: broader range of stimulus possibilities, including visual

rafal
Télécharger la présentation

Modeling Campaign Dynamics in the 2008 NAES

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Modeling Campaign Dynamicsinthe 2008 NAES Richard Johnston University of Pennsylvania Prepared for the Wivenhoe House Conference on Cyberinfrastructure and National Election Studies 8-9 June 2007

  2. Mode Considerations • Web: • broader range of stimulus possibilities, including visual • response tends to be more optimized: • fewer skips • less repetition of response codes • less biased by social desirability • unit cost much lower • Telephone: • not constrained by size of panel, so sample can be augmented at will • no literacy constraint • conditioning not an issue for pure RCS

  3. Panel Issues • Panel advantages: • for an event known in advance, effectively increases the sample size for pre-post comparison • facilitates correction for measurement error, given at least three waves • can add power to RCS dynamic estimations • allows for conditioning later response on earlier positions • different groups may respond to a given stimulus in different ways • facilitates unpacking causal order where there are plausible claims of endogeneity • post-election wave highly desirable: • Reports on election-day behavior • cumulative readings • (Strictly speaking, these are arguments for a post-election survey, not necessarily for a post-election wave of a panel. But the power of retrospective reports is increased if these can be linked to pre-election waves.) • NAES telephone survey reinterview rate ~ 40%  moving a significant portion of effort to Knowledge Networks

  4. Total interviews = 95640 Total respondents = 28420

  5. Case Release Matrix for mean of 222 interviews per day, week following 1st Oct (hypothetical)

  6. Instruments • Individual questionnaires equivalent to 18 minutes • some redesign per wave • scope for weekly variation • Profiles: core and public affairs

  7. Issues • How much common content between telephone and web modes? • mode comparison (for its own sake and as possible bridging to eventual complete transfer) vs. distinctive advantage of each mode • How much experimentation in a campaign study?

More Related