I don't have a lot of time to do diaries, but on this relatively leisurely Saturday morning I decided to do this one, after my drafted comment on it grew so long I realized it's better as its own separate thing.
Here's the link: http://www.democracycorps.com/...
nd here's a link to the memo listing and explaining the chosen districts: http://www.democracycorps.com/...
My analysis below the jump.
THE SAMPLE
This is an early December sample and, overall, shows no serious movement from its previous counterpart in September, just a very slight tick in our direction. These are 60 handpicked "battleground" U.S. House districts all over the country. To emphasize this is not a "national" poll, the PVIs of these districts include only 12 that are Democratic and 2 more that are "zero," with the other 46 leaning R. This for a PVI calculated using elections where Dems nationally won 48% and then 53%, combined a majority of the overall vote for those cycles. So this is not a "national" poll, but a poll of handpicked districts that overall are more Republican than the country as a whole. The overall sample is plurality Republican, plurality conservative, 81% white compared to a national electorate that is estimated to likely be only 72% white next year, and only 39 of the 60 were districts we held and lost in 2010--the other 21 were seats where Republicans had survived the 2006 and/or 2008 Democratic waves. This list excludes 24 of the seats we lost last year that either were much more conservative than this set or more liberal--I'll want to separately analyze the leans of those seats. This poll also by its terms goes by old boundaries, with redistricting incomplete in too many states so that one can't use any new lines yet.
Still, even under old lines, these are the kinds of seats where we need to make double-digit gains to take back the House. So progress in our standing in these districts is essential.
But again, it's important not to treat this poll's toplines as a national poll, because it's not. It's perhaps typical of an American electorate from the 80s or early 90s, when the country was whiter and truly more conservative than today.
The sample reported voting 47-42 for GOP House candidates last year, with some refusing to answer that question but what's reported is not inconsistent with the GOP's 52-45 national popular vote win in last year's House races. It's possible to compile exact data for these 60 districts based on official results, but Democracy Corps doesn't seem to do that, which I wish they did. Any of us could do it, but I doubt I'll have the hours to spare for that. That result probably should be more Republican than the national popular vote, since these districts combined are more conservative and Republican than the nation as a whole.
THE RESULTS
The poll now says they still favor the GOP incumbents 47-42. So no change from how they actually voted. But that's a slight improvement from the September poll when even more were inclined to support their GOP incumbents.
Obama' job approval is 44-51, again not bad in context since these are more conservative and whiter districts than the nation as a whole. Named GOP incumbents overall have 41-24 job approval, not nearly bad enough to beat many of them.
The Democracy Corps write-ups spin this poll more optimistically than I think the data supports. But there has been slight movement in our direction, that's not a complete fiction. It's just nearly enough. That said, 2 years ago today the Republicans were not yet poised to take the House. Voters have some patience and typically give a party more time than this, as it's been still only less than a year that the GOP has had the House...it sure feels like forever as horrific as they've been!
Obama in these districts trails Romney 48-47, consistent with what one would expect given national polls typically show Obama leading him by low-to-mid-single digits. The big surprise is that Obama trails Newt 48-46 in these districts. I treat this as a fleeting "primary surge" that we've seen in some other state and national polls, like Quinnipiac's latest Ohio poll. Whatever flaws Romney has, he is more electable than Newt in a general election, and Newt will stop enjoying general election trial heat numbers like these at some point, whether he's the nominee or not. Even this Democracy Corps poll's complicated favorability index, asking respondents to pick a "score" instead of just saying "favorable" or "unfavorable," shows Newt underwater at 29-45, compared to Mitt's less bad 30-41 and Obama's positive 46-43 (which in his case is unsurprisingly subordinated to much worse net job approval in respondents' trial heat picks). So Newt is getting a lot of "soft" support in these trial heats that will melt among voters once they learn more about him; Newt's negatives are baked in.
My bottom line on this poll is that it shows the House GOP needs to become even more unpopular for us to take back the House. One way to accomplish that is by good messaging on what they're doing and not doing in their choices of priorities, and I actually think Democrats have done a decent job of that lately. A bigger way is to control the topic of the national conversation, which OWS has helped us do, so now deficits are on the backburner in the news in favor of jobs and income, which helps us. Obama and Dems have been smart to pick up on those themes. A third way is for Newt to win the GOP nomination, or for Romney to run a poor general election campaign which I think is very possible. Now that Mitt is on the ropes vs. Newt, it's getting discussed in political reporting that he made serious mistakes up to now, and that he should've known better than to make them. Mitt just hasn't defined himself at all, he's just coasted on attacking Obama, which is of very little benefit because he's not a resonating messenger for that among Republicans. He's going to have the same problem in the general election, where OFA will keep defining him with voters. But even then he's likely to outperform Newt in a general election, Newt being so easy to punch repeatedly that he's likely to really hurt his party, too. Still, tying Congressional Republicans to Newt will have to be a concerted project by Congressional Democrats, because we need to establish that connection to take back the House and outperform expectations in the Senate.