Despite Some Screwy Crosstabs, NYT Polling Finds Unpopular Obama Hasn't Sunk Senate Democrats Yet

Photos courtesy of Corbis Images.

A spate of brand new surveys put out by the New York Times and conducted by the Kaiser Foundation finds once incredibly vulnerable Democratic incumbent Senators hanging on, and in some cases, leading their Republican opponents in four states carried by Mitt Romney in 2012:

• Arkansas, which Obama lost in 2012 61-37%
• Kentucky, which Obama lost in 2012 61-38%
• North Carolina, which Obama lost in 2012 50-48%, and
• Louisiana, which Obama lost in 2012 58-41%.

The Democratic resilience found by the Kaiser Foundation seems all the more miraculous considering how deeply unpopular the president is in all four states. His most positive rating is 41/51%, found in North Carolina, with his most negative rating coming from Kentucky, where just 32% of registered voters approve of the job Obama is doing, and 60% disapprove. Yet still, in those states, both incumbent Democratic Senator Kay Hagan, and Democratic challenger Allison Lundergen Grimes, are exceeding expectations considering the unpopularity of their party boss. Hagan's net job approval rating is 10 points higher than the President's, and she actually leads both of her likely primary opponents by 2. Grimes, who is saddled by a President with a 60% disapproval rating, only trails McConnell 44-43%.

The same applies in Arksansas and Louisiana. In the former, Obama is stained by a 33/60% job approval rating, a full THIRTY-SIX net points LOWER than Democratic incumbent Senator Mark Pryor's 47/38% job rating. Not only that, but Pryor leads Cotton (R) 46-36%. In Louisiana, the President's net job rating is SIXTEEN points lower than Democratic incumbent Mary Landrieu's.

All in all - this is some much-needed decent news for Democrats. But a word of caution regarding the results...

As many on twitter have already pointed out this morning: some of the demographic and political findings are a little screwy. For example, in the four states polled by Kaiser, respondents reported supporting President Obama by much larger margins than they actually did in 2012, while large portions of respondents claimed to not have voted at all in 2012. Consider the table below, produced by Logan Dobson, an employee for the highly respected pollster, The Tarrance Group:

The numbers in red represent a pretty staggering shift from just two years ago. Public Policy Polling, a pollster which frequently asks the "how did you vote in the last presidential election" question, rarely finds such huge deviations from the actual prior results. To illustrate this, consider the table below, which tracks the prior presidential vote question on PPP's last five state surveys:

As you can see, there's quite a difference between the 2012 vote findings of PPP in various states, and the the 2012 vote findings of the Kaiser Foundation. But without question, PPP's findings are more accurate. And had Kaiser's samples been more reflective of the 2012 vote in the respective states, you can only assume the Republican margin would improve.


UPDATE: The New York Times Nate Cohn has provided additional information regarding the Kaiser Foundation polls released today. The '2012 vote' question shown earlier in this post only provides data with regards to ALL ADULTS surveyed in the four states polled. Cohn has now provided a chart providing self-reported 2012 vote data among REGISTERED VOTERS in the four states. As you can see, it's still pretty far off from reality, and features much more deviation from 2012 than PPP's typical findings: