Both Gallup and Rasmussen Reports released surveys yesterday that gauged reactions to Sarah Palin's resignation announcement last week. While neither organization tracked questions they had asked previously about Palin, and while instant reactions can sometimes mislead, it is hard to quarrel with Gallup's conclusion that Palin's resignation "has apparently not affected Americans' basic opinions of her to a large degree."
When confronted with new polling numbers, the most important question to ask is, "compared to what?" That question is a tough with these two surveys because, with only one exception, neither pollster has asked the same questions about Palin previously. The exception is Rasmussen's favorable rating, but this time they asked it of a new and different population ("likely Republican primary voters").
Soon pollsters, including Gallup, will update their measurements of Sarah Palin's favorable rating, or better yet, the questions plotted by Charles yesterday that ask whether Palin is qualified or prepared for the job of president. When they do that we will have a more precise sense of whether perceptions of Palin have changed, and if so, by how much.
Meanwhile, the comparisons we can make today are more tenuous.
Consider the question that Gallup asked, "if Sarah Palin were to run for President in 2012, how likely would you be to vote for her, very likely, somewhat likely, not very likely or not at all likely?" The results are hard to interpret both because Gallup has never asked it before and because it specifies neither an election (primary or general) nor an opponent and includes those vague qualifiers that pollsters love, "somewhat" and "not very."
Nonetheless, if we compare the results to two recent measurements of Palin's favorable rating, we can see that reactions break down along the same rough partisan lines. Americans views of Palin were polarized before and after her resignation announcement. The percentage of Americans who say they are very or somewhat likely to vote for her now are roughly the same -- overall and by party -- as the percentage who rated her favorably earlier this spring on a survey conducted by the Pew Research Center and Fox News/Opinion Dynamics.
Here is a chart that makes the comparison more clearly, although please note that I do not consider the Gallup question equivalent or comparable to a favorable rating, and I would not expect these questions to obtain identical results. Still, the similarity of the partisan breakdowns across these questions is striking.
Rasmussen's rating, not included in the chart and tables above, shows 76% of Republican primary voters with a favorable opinion of Palin, 21% unfavorable. Their sample of likely Republican primary voters is probably different than the Republican adults and registered voters polled by Pew and Fox respectively. Rasmussen did not report an earlier measurement for a comparable population.
Rasmussen also asked a 2012 Republican primary vote preference question. While they have apparently not asked that question before, their result was very close to what CNN/ORC obtained in May.
All in all, the data support the interpretation given by Republican consultant Alex Castellanos to USA Today's Susan Page: "For independents and Democrats, [Palin's] already not their candidate, and with Republicans her support is not based on her record as governor of Alaska."
Another way to put the new results into context is to compare them to identically structured questions asked about other political figures or other issues in the past.
Gallup's report notes, for example that they asked the same "likely to support" question Hillary Clinton in May 2005 and found 52% of registered voters at least somewhat to vote for her, including 28% very likely. Palin's numbers are lower -- 43% at least somewhat likely, including 18% very likely. However, as the chart below shows, Gallup had asked the same question about Clinton twice before (in 2000 and 2003), with varying results. The percentage that said they were "not likely at all" to support Clinton hovered at roughly the same level on all three of those surveys as the percentage that says the same about Palin now.
Finally, Gallup finds that roughly twice as many Americans say the resignation announcement makes them "feel less favorably" about Palin (17%) as makes them feel "more favorably" (9%), while the vast majority (70%) say the news does not affect their opinion. That reaction is nearly identical to what Gallup found last October when asking about Barack Obama's decision to opt out of the public financing system -- not exactly a career ending move for Obama. Moreover, Palin's "less favorable" number is far lower than what the Pew Research found with a similar question about Harriet Mier's judicial experience in 2005 (38%) and Bill and Hillary Clinton's involvement in the Whitewater deal in 1994 (32%).
Again, we should treat these results as preliminary and hold off on firm conclusions until we have better, more comparable data based on more than few days' reflection, but on first blush, it looks as those Palin's resignation announcement made less of an impression on Americans than the punditry of the last few days might lead you to believe. Palin's abrupt resignation probably confirmed or deepened existing impressions, but it does not appear to have changed many minds.
PS: Apologies to Mike Huckabee. In my post on Monday, I described Sarah Palin as "the most popular potential candidate for president in 2012...among Republicans." That was true as far as the data I cited from the Pew Research Center, but their survey did not test a favorable rating for Huckabee. Yesterday, Rasmussen found that Huckabee's rating among likely Republican primary voters (78% favorable, including 41% very favorable) was roughly comparable to Palin's (76% favorable, including 45% very favorable).