24 polling firms have polled one or more of the 7 Presiential Primaries and Caucuses to date in 2008. Here’s how the pollsters stack-up, ranked from smallest average error to largest average error (where smaller error is better, and zero is perfect).
Additional detail, and caveats, on the jump …
Not all 24 pollsters polled on the same contests or on the same number of contests.
Certain contests, unexpectedly, appear to have presented a greater challenge for pollsters, for as yet unknown reasons, such as the New Hampshire Democratic Primary.
Other contests were anticipated to create great polling challenges, but did not, such as the Nevada Democratic Caucus.
Here is a Pollster Report Card through and including 01/19/08.
Error is calculated using the “Mosteller 5″ Measure. The Mosteller 5 Measure works this way: If “Pollster A” says Smith beats Jones by 8 points, and Smith in fact beats Jones by 5 points, then Pollster A has an “error” of 3 (8 minus 5). A 3 would be recorded in the table above for that contest. There are limitations to this and all other measures of pollster accuracy.
For a complete discussion of the different ways that election poll accuracy can be measured, including a summary of all 8 known measures, see this paper presented at the 2006 conference of the American Association for Public Opinion Research.
Any errors in this table, or omissions from this table, should be reported to firstname.lastname@example.org.
SurveyUSA will endeavor to update this chart throughout the 2008 election.