Rasmussen Polls Were Biased and Inaccurate; Quinnipiac, SurveyUSA Performed Strongly

On Tuesday, polls conducted by the firm Rasmussen Reports — which released more than 100 surveys in the final three weeks of the campaign, including some commissioned under a subsidiary on behalf of Fox News — badly missed the margin in many states, and also exhibited a considerable bias toward Republican candidates. Full Story »

Posted by Joey Baker - via Memeorandum, Columbia Journalism Review, FiveThirtyEight, Slatest, Ellie Kesselman (t), Josh_Young (t), Jon Mitchell (t), Willie Bido (t)

See All Reviews »

Review

Mike Carlson
4.0
by Mike Carlson - Nov. 8, 2010

Polling has been in trouble for some time. With the widespread use of screening (answering machines) and the rise of cell phones, the least expensive form of polling via phone is woefully inadequate for reaching many voters/consumers. Polsters know this when they receive the topline results and note that the demographics of the obtained sample looks nothing like population they are trying to reach. As Mr. Silver notes, the "fix" is to weight the results ... and therein lies the problem. What weights are applied and who determines the weighting algorithm? In a change environment weighting responses to replicate demographics (also adjusted from the 10 year old census data) is risky ... and that ignores the fact that the external descriptors that are demographics may have nothing to do with the internal chaos going on in an unstable environment.

See All Reviews »

Mike's Rating

Overall
4.0

Good
from 4 answers
Quality
4.0
Facts
4.0
Fairness
4.0
Popularity
4.0
Recommendation
4.0
How our ratings work »