During the past year, there was almost nothing more controversial for me to do as a reporter than to report on poll numbers, as both sides routinely accused this reporter and the polling organizations of bias for either the Republicans or Democrats.
Now that the 2012 elections are over, it is time to take a look at how accurate some of the polls actually were.
At the national level, Fordham University graded all the major polling organizations and found that Public Policy Polling - which has been routinely derided as a Democratic leaning firm - was the most accurate.
The favorite polling organization of many listeners to talk radio and readers of more conservative web sites, Rasmussen, was judged to be 24th on the list of 28 organizations.
Here was the Fordham list:
1. PPP (D)*
1. Daily Kos/SEIU/PPP*
3. YouGov*
4. Ipsos/Reuters*
5. Purple Strategies
6. NBC/WSJ
6. CBS/NYT
6. YouGov/Economist
9. UPI/CVOTER
10. IBD/TIPP
11. Angus-Reid*
12. ABC/WP*
13. Pew Research*
13. Hartford Courant/UConn*
15. CNN/ORC
15. Monmouth/SurveyUSA
15. Politico/GWU/Battleground
15. FOX News
15. Washington Times/JZ Analytics
15. Newsmax/JZ Analytics
15. American Research Group
15. Gravis Marketing
23. Democracy Corps (D)*
24. Rasmussen
24. Gallup
26. NPR
27. National Journal*
28. AP/GfK
I decided to take a quick look at the last polls by multiple pollsters in nine of the 2012 swing states, just to get a better idea of how well each group did.
PPP -
* Iowa - Obama won by 5.6%, PPP's last poll was Obama +2
* New Hampshire - Obama by 5.4%, PPP had Obama +2
* Wisconsin - Obama by 6.5%, PPP was Obama +3
* Florida - Obama by 0.5%, PPP was Obama +1
* Virginia - Obama by 3.0%, PPP had Obama +4
* Nevada - Obama by 6.6%, PPP was Obama +4
* Ohio - Obama by 1.9%, PPP was Obama +5
* Colorado - Obama by 4.5%, PPP had Obama +6
* North Carolina - Romney by 2.2%, PPP was a tie
My back of the envelope review would say that PPP was not very far off the mark at the state level as well; while PPP was routinely denounced as a shill for the Democrats, the pollster actually underestimated the Obama final numbers in Iowa, Wisconsin, New Hampshire and Nevada, while going too high in Ohio and Colorado.
Let's compare that to Rasmussen, which polled in all those states as well:
Rasmussen -
* Iowa - Obama won by 5.6%, Rasmussen's last poll was a tie
* New Hampshire - Obama by 5.4%, Rasmussen had Romney +2
* Wisconsin - Obama by 6.5%, Rasmussen had a tie
* Florida - Obama by 0.5%, Rasmussen had Romney +2
* Virginia - Obama by 3.0%, Rasmussen had Romney +2
* Nevada - Obama by 6.6%, Rasmussen had Obama +2
* Ohio - Obama by 1.9%, Rasmussen had a tie
* Colorado - Obama by 4.5%, Rasmussen was Romney +3
* North Carolina - Romney by 2.2%, Rasmussen had Romney +6
Rasmussen is a controversial name in political polling, because some people charge it is biased towards Republicans. This data shows the final Rasmussen polls overestimated Romney's numbers in certain states, missing by over 5 points in Iowa, 7 points in New Hampshire, 6 points in Wisconsin, 5 points in Virginia and 7 points in Colorado.
One new pollster that got a lot of flak during the campaign was Gravis Marketing from Central Florida; let's look at some of their polls:
Gravis -
* Iowa - Obama won by 5.6%, Gravis' last poll was Obama +4
* New Hampshire - Obama by 5.4%, Gravis was Obama +1
* Florida - Obama by 0.5%, Gravis was a tie
* Virginia - Obama by 3.0%, Gravis was a tie
* Nevada - Obama by 6.6%, Gravis had Obama +1
* Ohio - Obama by 1.9%, Gravis had Obama +1
* North Carolina - Romney by 2.2%, Gravis had Romney +4
One thing that many of us relied on this election year were polling averages. I watched both the Real Clear Politics average and one done by Five Thirty Eight at the New York Times.
Here is how the RCP average and the Five Thirty Eight average compares to the final outcome in nine swing states:
During the past year, there was almost nothing more controversial for me to do as a reporter than to report on poll numbers, as both sides routinely accused this reporter and the polling organizations of bias for either the Republicans or Democrats. Now that the 2012 elections are over, it ...
