|
2012 US Presidential
Election
Pollster
Ratings The following table
lists the
accuracy of pollsters in the November 6, 2012 US
Presidential election using the measure
of polling accuracy
proposed
by Martin, Traugott, and Kennedy.
An accuracy measure of 0.00 indicates
that the odds ratio of the survey results matches the odds
ratio of the actual results (the
closer to 0.00, the better).
| 2012 Presidential Election
|
|
Obama |
Romney |
Odds
Ratio |
Accuracy |
|
| Pew |
48% |
45% |
0.99 |
-0.01 |
| Democracy Corps |
49% |
45% |
1.01 |
0.01 |
| High Point University |
46% |
43% |
0.99 |
-0.01 |
| ABC/Post |
50% |
47% |
0.98 |
-0.02 |
| Angus-Reid |
51% |
48% |
0.98 |
-0.02 |
United Technologies/
National Journal |
50% |
45% |
1.03 |
0.03 |
| YouGov |
49% |
47% |
0.96 |
-0.04* |
| Ipsos/Reuters |
48% |
46% |
0.96 |
-0.04 |
| DailyKos/SEIU/PPP |
50% |
49% |
0.96 |
-0.04 |
| PPP |
50% |
48% |
0.96 |
-0.04 |
| YouGov/Economist |
49% |
47% |
0.96 |
-0.04 |
| UPI/CVOTER |
49% |
48% |
0.94 |
-0.06 |
| NBC/WSJ |
48% |
47% |
0.94 |
-0.06 |
| Purple Strategies |
47% |
46% |
0.94 |
-0.06 |
| IBD/TIPP |
50% |
49% |
0.94 |
-0.06 |
| ARG |
49% |
49% |
0.92 |
-0.08 |
Politico/GWU/
Battelground |
47% |
47% |
0.92 |
-0.08 |
JZ Analytics/
Newsmax |
47% |
47% |
0.92 |
-0.08 |
| Gravis Marketing |
48% |
48% |
0.92 |
-0.08 |
| CNN |
49% |
49% |
0.92 |
-0.08 |
Washington Times/
JZ Analytics |
49% |
49% |
0.92 |
-0.08 |
| Fox |
46% |
46% |
0.92 |
-0.08 |
| Monmouth |
48% |
48% |
0.92 |
-0.08 |
| Rasmussen |
48% |
49% |
0.91 |
-0.10* |
| Gallup |
48% |
49% |
0.91 |
-0.10* |
|
| Actual |
51.06% |
47.21% |
|
|
The accuracy measures for 22 polls
fall
within the 95% confidence intervals.
*The accuracy measures for the YouGov poll (sample
size of 36,472), the Rasmussen poll (sample
size of 1,500), and the Gallup poll (sample size
of 2,551) fall outside the 95% confidence intervals.
The absolute average accuracy for all polls
in 2012 is 0.055. The absolute average accuracy for all polls in the 2008
presidential election was 0.037.
|
