Healthcare: Parsing the Polls and Focus Groups

Well, we've got the first of the very preliminary poll ratings on the healthcare speech.  Rasmussen, predictibly, says that the bounce wasn't that big, and consisted almost entirely of rising support in the president's own party; Rasmussen's results are almost always considerably more negative for Democrats than other polls.    Meanwhile, my esteemed colleague has obtained an internal Democratic memo on their focus grouping:

Research conducted with 49 voters in Tempe, Arizona by David Binder, who was Obama's campaign focus group guru, suggests to Democrats that the speech was "effective at alleviating concerns of voters and impressing upon them that the President has a strong plan to reform health care," the memo says. "Even among those voters who held neutral or negative opinions of the President, substantial positive movement was shown as the proportion of these participants supporting the President's plan increased by nearly 40% after the speech.

Let's break that apart.  In the latest independent poll I'm aware of, the pre-speech support for the health care plan was at 29% among independents, 10% among Republicans, and 37% overall.  A "nearly 40% increase in those numbers" means something under 40% support among independents, 14% among Republicans, and still solidly less-than-50% overall.  Getting more support among Democrats doesn't help him--they'll mostly vote for Democrats anyway. 

Some other thoughts: 

• Focus groups are problematic; it's very hard for those who run them to keep their biases from subtly affecting the results, and the sample is necessarily very small.
  
• Democrats may not need majority support to strengthen their legislators' spines; they may just need to tip the balance from 37% in favor and 39% against to 39% in favor and 37% against, figuring the undecideds won't vote on it.   On the other hand, my sense is that independents tend to break against both incumbents and policies, rather than for.  Witness the storied history of Social Security Reform polling.  People actually got more anxious about the state of Social Security as things went on--but also became less willing to change it. 
  
• Bounces have half-lives.  The real action comes as Republicans and Democrats start their final push.  The fact that Max Baucus is scrambling on the illegal immigration issue suggests that Joe Wilson didn't hurt Republicans as much as I initially thought--or at least, he simultaneously dealt a blow to Democrats.
  

So split the difference between the Democrats and Rasmussen:  support probably rose modestly among independents, strongly among Democrats, and fell or stayed pat among Republicans.  The Republican support was so low that it really isn't a factor.  It will all hinge on what the independents do.  And indeed, though this CBS poll shows that more people support the plan than before, and Obama's approval rating has flipped on the issue, more people still think that the reforms will hurt them than help them, and more think that Obama has not clearly explained his plans, than that he has.  Which is about what we'd expect from a moderate speech bump.  But hell, in the history of political speeches, moving the dial a little bit is a rousing success.

Still, what will really matter is whether Obama manages to seize this bump and move it forward.  If he doesn't, the modest improvement will dissipate into fall busyness.

Developing . . .  

Update:  Nate Silver points out that more Democrats probably watched the speech than independents or Republicans . . . but says that their votes matter too.