A new national poll of Latino voters has some good news for Sen. Barack Obama.
The survey found that 60 percent of Latinos planned to vote for Obama, compared to 23 percent for [Sen. John] McCain, while 16 percent were undecided. Latino Decisions, a joint effort between Pacific Market Research and University of Washington political scientists Matt Barreto and Gary Segura, conducted the poll by telephone June 1-12.
Workers reached 800 Latino voters in 21 states. Among Democrats, the survey found that during primary contests, 57 percent had supported Democratic presidential candidate Hillary Clinton versus 35 percent who supported Obama.
[…]
Obama does well among Latinos across many states. In California, he leads 66 percent to 20 percent; in New York, 65 percent to 20 percent; in Texas, 61 percent to 22 percent.
There’s nothing particularly surprising about this, though as Kos says, it’s nice to see that bit of conventional wisdom about Obama having problems with Latino voters get the refutation it deserves. What interests me here in particular is the Texas result, since it stands in contrast to the Baselice poll from May. Now, the sub-sample here is likely to be in the 150-200 range, which means it would have a margin of error of seven or eight percent, but that makes it about the same as the number of Latino voters Baselice sampled (PDF). Here were the results he got then:
Obama 48, McCain 36, other/undecided 16
Clinton 57, McCain 30, other/undecided 13
This would represent a pretty significant shift towards Obama, presumably partly due to the post-primary unity effect. If you updated the Obama/McCain numbers to reflect this poll’s result, it would change the outcome from a 51.9-36.4 lead for McCain to 49.8-38.5, which is still significant but not nearly as intimidating; it’s also much closer to Poblano‘s projection of a 9.7-point win for McCain. Given that that projection is based in part on the May Baselice poll, the Democratic “bonanza” number of 45% looks better, too. I don’t know about you, but I can’t wait to see what numbers Baselice (and Rasmussen, who had a similar result from May) come up with for June.