Originally posted by Cow Poke
View Post
Rasmussen is an IVR, i.e., Interactive Voice Response aka Robocall survey. Given the methodology, they are hugely prolific, providing the second largest sample of polls tracked by 538 after SurveyUSA, which is likewise a robocaller, but also includes live response polls and adheres to higher industry standards.
2019-09-22_07-34-46.jpg
NCPP/AAPOR/Roper
• Indicates the polling firm was a member of the National Council on Public Polls, a signatory to the American Association for Public Opinion Research’s transparency initiative or a contributor to the Roper Center for Public Opinion Research’s data archive as of May 22, 2018. This is a proxy for methodological quality.
There's a noticeable bias in Rasmussen polls, but it's the stability that matters, as a predictable bias can simply be subtracted. In aggregate, they're relatively stable, but the small sample sizes of individual polls regularly generate one-day blips.
Even after the adjustment to account for past performance, this poll is an outlier in the time frame specified:
2019-09-22_07-41-55.jpg
Comment