Barack and the Hispanics

Back during the primaries, everyone kept formally admitting that it was wrong to engage in the form of inference "candidate X lost group A in a primary, and therefore he's likely to lose group A in a general election against candidate Y of the other party" but I often got the sense listening to and reading pundits that they didn't really believe that. But the Pew Center's latest findings on public opinion among Hispanics should remind people that this is a very important caveat. Barack Obama did quite a poor job of persuading Latinos to vote for him over Hillary Clinton, but they're backing him very strongly against John McCain.

The numbers deserve to be put in a historical context:

Obama is, in short, in solid shape with this demographic. Which I hope will serve as a reminder for us next time. The way a lot of people were interpreting the Obama-Clinton primary results led to the conclusion that neither candidate could beat John McCain because both were showing "weakness" among some key groups, even though both were polling ahead of McCain in general election trial heats at the time. But the "weakness" of both candidates simply reflected the fact that they were evenly matched with about half of Democrats preferring each of their choices -- it didn't say anything about either candidate's actual strength in November.

Comments

Obama should be hitting AZ hard with Spanish-language ads. Closing the poll numbers in McCain's home state would generate some juice for him.

Posted by owenz | July 26, 2008 11:35 AM

One correction.
No Obama supporter ever suggested Hillary Clinton would lose against McCain.
They argued Obama would get a bigger majority because he could put more states in play but noone of them EVER suggested she COULD NOT win.

That stupid "Candidate X cannot win Y demographic" came entirely from one side. ENTIRELY.
I know it does not matter now but I am sick and tired of people rewriting history as if both sides said stupid things.
Yeah, Obama was not perfect.

But all those stupid arguments that are proven wrong in hindsight ALL came from the Clintonites.

Posted by benjamin | July 26, 2008 11:43 AM

The question, as always, is turnout.

Posted by John Petty | July 26, 2008 11:48 AM

How is he doing in the vital demographic of porky, snarky Hispanics?

Posted by James | July 26, 2008 11:51 AM

Wasn't Bush's supposed 40% Hispanic vote in 2004 debunked as due to an exceptionally broad definition of the term (as in everyobody with any Hispanic ancestry at all)? I've seen even GOPer admit that it was exaggerated.

Posted by JonF | July 26, 2008 12:30 PM

Which is really all common sense. I couldn't understand during the primary season what all of the noise on this subject was about anyway so put it off to Clintonistas trying to confuse the issue and Beltway blathering.

Matt's right that there were all of these up-front caveats about how inferring from primary to general was a bad idea, but then the bulk of the story or commentary was about JUST THAT.

Another fine example, frankly, of how the press (including pundits and bloggers) lets the perceived common wisdom lead them around by their nose-rings. Or maybe it's more that something had to be written so let's write about that. That, after all, is the essence of punditry and increasingly of what can laughingly be called 'reporting' these days...everything under the sun must be opinionated on, regardless of whether it makes any sense or not.

Matt may be a little less susceptible to this than some others but out here in the real world, anyone with half a brain would hear this stuff about how Obama was having a Hispanic problem and wonder to themselves how polls measuring support for the primary vis-a-vis Clinton had anything whatsoever to do with the general.

Posted by Monica Wolf | July 26, 2008 2:39 PM

This is all off MattY's radar, but perhaps those Dems who supports our laws and who oppose ethnic nationalism might look into exactly how BHO has gotten those numbers. That includes thanking extremists and, of course, outright lying. Search for his name at my site for much more.

Posted by TLB | July 26, 2008 4:06 PM

This is confirmation, of course, of what some clever blogger said during the primary struggle: just because I prefer lobster to steak, that doesn't mean that if I can't have the lobster, I'd choose catshit over steak.

Posted by drw | July 26, 2008 4:34 PM

What Monica said.

The translation of primary election results into general election predictions has been a huge weak point in the political conventional wisdom as presented on the cable news channels this year. It makes me question the sophistication of the polical analysts.

Posted by Beth in VA | July 26, 2008 5:43 PM

"Wasn't Bush's supposed 40% Hispanic vote in 2004 debunked as due to an exceptionally broad definition of the term (as in everyobody with any Hispanic ancestry at all)? I've seen even GOPer admit that it was exaggerated."

JonF, Bush was first reported to have won 44% of the hispanic vote. *That* number was the number debunked. 35%-40% is now generally accepted, with 40% being cited most often. The Republicans that still cite the 44% number seem to be some pro-comprehensive reform Republicans that want to convince other Republicans that getting a majority of hispanic votes is just within reach and don't mind using a debunked number to do it.

For instance, just a year or so ago, long after the debunking of "44%", I read a Wall Street Journal pro-comprehensive reform Op-Ed that used the "44%" number. But just a few months later the WSJ news section cited the "40%" number. The WSJ news section is pretty objective compared to their Op-ed writers, as Brad Delong has said.

Posted by scottynx | July 26, 2008 6:25 PM

"Which I hope will serve as a reminder for us next time."

Good frickin' luck with that. The lonely voices of reason cited historical example after historical example to support this point, and it did little or nothing to dent the overall nature of the coverage. It turns out this particular brand of nonsense is just too tempting for pundits to resist--they have these numbers available, they have to come up with something to say, so they use the numbers.

Posted by DTM | July 26, 2008 10:34 PM

What the statistics really show is that there is no way that the Republians can remain a relevant political party in the long term. As the black and Hispanic populations grow relatives to whites, there is no way that the Republicans will be able to compete.

A long term question is how will black and Hispanic voters be affected by the coming one party state.

Posted by superdestroyer | July 27, 2008 6:07 AM

What Monica Wolf said. DTM too. Duh.

Posted by Rob Lewis | July 27, 2008 10:05 AM

A long term question is how will black and Hispanic voters be affected by the coming one party state.

This is an odd question to focus on. But my guess is that having officially driven the party of lingering racism out of politics, you'd see a fracture of the democratic party into two separate groups. I don't really think an alliance between culturally conservative hispanics and southerners is really feasible, so I think you'd end up with an economic policy oriented conservative party that is somewhat to the left of the current Republican party's position on economics. This is, of course, pretty much the optimal result and I won't be holding my breath.

Posted by mpowell | July 27, 2008 10:55 PM