Talk Elections

Recently there have been a spate of attacks on the results of Rasmussen surveys from the left, who don't like the results.

Well, lets take a look at the numbers for Obama Job Approval in November, using the most recent (December) results from 19 survey research outfits.

The highest number was from AP of 56, and the lowest from Harris of 41.

If we break the polls down into the three categories of "Adult," "Registered Voters," and "Likely Voters," the results look a little different.

Type of Poll          Highest          Lowest          Mean          Median

Adult                      56 %              41 %           46 %            50 %
Registered             50                  46                48                48
Likely                     50                  47                48                48

So, Rasmussen's results of 47 % look pretty reasonable.

On the other hand, the AP (56), CNN (54), Bloomberg (54) and Harris (41) polls (all "adult") look pretty unreasonable.

Now, the leftists when confronted with such data tend to sputter that Rasmussen tends to have higher than average disapproval ratings, which is true.  But, it is pretty well accepted in the survey research community that: (a) robo calls tend to produce higher disapproval ratings as respondents at less likely to feel they will be accused of "racism" for disapproving of Obama's job performance in a robo call, (b) "unsure" respondents tend to lean (when pushed) to disapprove, and (c) likely voters are less prone to "undecided" than adult or registered voters.

Next, the enraged leftists will shriek that what they really dislike about the Rasmussen surveys is the use of an intensity gauge.  However, they won't tell you that Rasmussen did not invent that gauge (the Stapleometer was used forty years ago by Gallup, then known as the American Index of Public Opinion).

Then, they will alledge that the wording of the questions is slanted to achieve a particular result.  However, if there were such wording, the results would fall outside the MoE of the other polls!

If you want to run the numbers yourself, check out the data at Pollster.

Finally, surveys in December include: AP, CNN, Bloomberg, Gallup, ABC, GWU, Fox, CBS, YouGov Pew, PPP, Ispos, Zogby, Rasmussen, ARG, NBC, Quinnipiac, Marist and Harris.

Addendum:

For those of you who think that Rasmussen is the only survey research firm using intensity questions, here's one from NBC:

NBC News/Wall Street Journal Poll conducted by the polling organizations of Peter Hart (D) and Bill McInturff (R). Dec. 11-14, 2009. N=1,008 adults nationwide. MoE ± 3.1.
 results below are among all adults.

"Now I'm going to read you the names of several public figures and organizations, and I'd like you to rate your feelings toward each one as either very positive, somewhat positive, neutral, somewhat negative, or very negative. If you don't know the name, please just say so. The Democratic Party."

                    Very Positive Somewhat Positive Neutral Somewhat Negative Very Negative         
                            %                       %                  %                    %                       %
12/11-14/09        10                      25                 19                   19                       26
2/26-3/1/09         20                      29                 18                   14                       17

There are 3 pages about it on Politico:

http://dyn.politico.com/printstory.cfm?uuid=DCAD6DDB-18FE-70B2-A8986E439331DA11

Rasmussen is a pretty reasonable polling company. Their projected results were very close to the actual results in 2006 and 2008. If they appear to lean Republican, it's because they use likely voters in their samples instead of registered voters, since many young and minroity registered voters are less likely to actually vote than other registered voters (based on past experiences).

Nate Silver has a good post/article about this on fivethirtyeight.com