UPDATED Mar. 13, 2023, at 3:53 PM

FiveThirtyEight’s Pollster Ratings

Based on the historical accuracy and methodology of each polling organization’s polls.

B/C

University of Wisconsin Survey Center

Polls analyzed	8
Races called correctly	88%
Most recent cycle	2008
Affiliated with AAPOR/Roper	Yes
Other names

What are pollster ratings?

How accurate are the polls from University of Wisconsin Survey Center?

Polls conducted within 21 days of an election (based on the poll’s median date) are used to determine the pollster’s grade.

2008

Key

A = Adults

RV = Registered voters

V = Voters

LV = Likely voters

Dates	Sample	Race	Methodology	Poll margin	Election margin	Difference
Oct 21-29	359LV	President (WI) General election	Live Phone	Poll Obama +10.0	Actual Obama +13.9	Difference 3.9

2006

Dates	Sample	Race	Methodology	Poll margin	Election margin	Difference
Oct 18-26	266LV	Senate (WI) General election	Live Phone	Poll Kohl +57.0	Actual Kohl +37.9	Difference 19.1
Oct 18-26	266LV	Governor (WI) General election	Live Phone	Poll Doyle +12.0	Actual Doyle +7.4	Difference 4.6

2004

Dates	Sample	Race	Methodology	Poll margin	Election margin	Difference
Oct 23-27	545LV	President (WI) General election	Live Phone	Poll Bush +3.0	Actual Kerry +0.4	Difference 3.4
Oct 23-27	545LV	Senate (WI) General election	Live Phone	Poll Feingold +15.0	Actual Feingold +11.2	Difference 3.8
Feb 4-5	600LV	President (WI) Democratic primary	Live Phone	Poll Kerry +51.0	Actual Kerry +5.3	Difference 45.7

2002

Dates	Sample	Race	Methodology	Poll margin	Election margin	Difference
Oct 24-29	501LV	Governor (WI) General election	Live Phone	Poll Doyle +7.0	Actual Doyle +3.7	Difference 3.3

2000

Dates	Sample	Race	Methodology	Poll margin	Election margin	Difference
Oct 18-21	587LV	Senate (WI) General election	Live Phone	Poll Kohl +43.0	Actual Kohl +24.5	Difference 18.5

Show all

How this works

FiveThirtyEight’s pollster ratings are calculated by analyzing the historical accuracy of each polling organization’s polls along with its methodology. Accuracy scores are adjusted for the type of election polled, the poll’s sample size, the performance of other polls surveying the same race and other factors. We also calculate measures of statistical bias in the polls. When a pollster publishes multiple versions of the same survey (for example, versions with and without third-party candidates included), FiveThirtyEight uses an average of the different versions to calculate the pollster’s rating. However, all versions of these polls are listed here. Read more about our methodology»

Download this data as an Excel spreadsheet or get it on GitHub.

Design and development by Elena Mejía and Aaron Bycoffe. Research by Mary Radcliffe, Cooper Burton and Dhrumil Mehta. Copy editing by Andrew Mangan. Visual editing by Christopher Groskopf. Statistical model by Nate Silver.

Send us feedback.