2010: The polling Carville & Greenberg only whisper about

posted at 9:31 am on October 28, 2009 by Karl
printer-friendly

Last week, I took a look at the Left’s wishful hypothesis that a tarnished GOP brand was going to save the Dems in the 2010 midterm Congressional elections. I missed the most recent Congressional Battleground survey from Democracy Corps (.pdf) — an outfit led by James Carville and pollster Stanley Greenberg — mostly because they made little effort to promote it. The money and effort put into this survey — which covered 1500 likely voters in 40 Democratic-held and 20 Republican-held target House districts — makes it the sort of project you generally would not like to put under a bushel.

Glen Bolger of Public Opinion Strategies has several blog posts explaining why the Dems might have wanted to keep this survey on the downlow. In the first post, Bolger notes that those awful Republicans have not budged from about 43 on the “feeling thermometer,” while Dems have dropped from 49 to 44.2 since April. As in other polling, the generic Congressional ballot has deteriorated for Dems. And the more significant re-elect question was not any better:

Democracy Corps also tested two key measures of ballot standing. In seats with an incumbent, they asked a two way re-elect (which is really the best wording of the myriad of re-elects out there). In April, Dem incumbents (and they gave the actual incumbent name) had a 39% re-elect/37% new person score. That’s down to 40% re-elect/45% new person now — a seven point shift against the Dem incumbents.

GOP incumbents have not had the same shift. In April they had a 39% re-elect/40% new person score, while in October it’s 40% re-elect/40% new person. These data show that Republican incumbents in swing seats are not out of the woods — but also that Dem incumbents in a similar situation are going deeper into the woods.

Bolger’s second post highlights that in the Dem districts, plurailities see their Representative as as “too liberal,” “will raise my taxes,” and “puts (his/her) party in Washington ahead of the people here,” with a majority saying their Dem “supports too much government spending.” Meanwhile, Democracy Corps got disappointing results on the negative characteristics tested aginst GOP incumbents. Bolger contends:

These data put a dent in the conventional wisdom that individual Republican incumbents are not getting some bounce off the Democratic Party’s problems. This is NOT a pox on both your houses of incumbents — instead, there are very real concerns with the Democratic party.

Democracy Corps disagrees (natch):

It is true that in this survey, the Republican incumbents have stronger numbers on some attributes than their Democratic counterparts. But that is expected, considering that nearly all of these Democrats are freshmen or sophomores while most of the Republican incumbents have long served their districts.

Yet despite this advantage, these Republican incumbents have lost significant standing since our last survey in July (much like everyone else in Washington), and show serious weakness on a number of important measures where you would expect a better performance from long-time incumbents. For example, just 40 percent of voters in the Republican districts say they will vote to reelect their member, the same result as in the Democratic seats. When we phrase this question differently, a 50-percent majority now say that they “CAN’T vote to reelect (their incumbent by name) because we need new people that will fix Washington” versus just 39 percent who say they “WILL vote to reelect (their incumbent) because he or she is doing a good job.” This represents a large drop since July and is a much worse showing than the Democratic incumbents. Finally, the Republican incumbents are only able to manage 48 percent of the vote in a named matchup with generic challengers – exactly the same level of support that the Democratic incumbents receive. And while the percentage supporting the generic challenger is lower in the Republican-held seats, our Voter Choice Scale[1] identifies 19 percent of voters as “winnable” for the Democratic challengers (against just 10 percent for the Republican incumbents), suggesting that the challengers have much more room to grow in these seats. Clearly, these Republican incumbents are very much at risk.

Note that in all of that spin, the primary piece of data is that “CAN’T vote to reelect” question on which Democracy Corps claims GOP has “a much worse showing than the Democratic incumbents.” The number for the GOP was 50%, which is bad. The number for the Dems was… 48%, which looks like margin of error. Moreover, the saliency of the “CAN’T” question compared to the standard re-elect question is (ahem) debatable. Perhaps the surest sign that not even Democracy Corps buys this spin is that they ran back out into the field to similarly poll 15 long-term Democratic incumbents in mostly very Republican seats that the NRCC may target — their vote averages only 50%.

Bolger’s third post starts by replying to Democracy Corps, but goes on to look at their numbers on the issues. In the Dem districts, the GOP not only had advantages on the economy, taxes, the budget deficit, and government spending, but were within the margin of error on healthcare and Medicare. Indeed, the GOP had a bigger advantage on the economy and Medicare in Dem districts than in GOP districts.

Bolger’s final post of the series looks at Democracy Corps’ economic message testing for the two parties:

It’s a very interesting and well-done series. The good news for Republicans is that the intensity of support for the GOP messages on the economy is stronger than the intensity of support for the Dem messages.

***

They tested a series of messages about what both the Dems and the GOPers “might say about the economy” and then asked how convincing respondents found each statement. The results of the six Dem messages point to some that have credibility with a majority of voters in these 60 districts, but those messages lack intensity. In comparison, they tested four GOP messages on the economy, and all four have higher intensity than any of the six Dem messages. Three of the four GOP messages have significantly more pop (a quick shout out of thanks to the folks at Democracy Corps for writing such well-crafted GOP messages for us to steal).

Unsurprisingly, the GOP message with the most intensity links the stimulus, the bailouts, the budgets, and rising unemployment.

Intensity in messaging matters, but intensity also matters for voter turnout, particularly in midterm elections. Democracy Corps recently conducted focus groups on that sort of intensity, which Charlie Cook summarized as follows:

Democrats would have to set up machine-gun nests to keep these people from voting, while the lethargy among Democratic voters is palpable.

No wonder the normally talkative Carville would want to stay away from this subject.

Blowback