Sometimes being first is better than being correct. I point you to a tidbit in this story about Republican agida over immigration for an example.
President Bush has been courting Latino voters for years, as both a gubernatorial candidate and presidential candidate, and it has paid dividends. In his 2004 re-election campaign, Bush won 44 percent of Hispanic voters nationwide and nearly 60 percent in Texas.
That “nearly 60 percent” figure came from an AP exit poll released shortly after the 2004 election – the exact number in question was 59. Only problem is, the AP issued a revised statement a day or so couple of weeks later, adjusting that number down to 49 percent. That did still represent an improvement of six points for Bush from 2000, but it’s not nearly as sexy as that 60 was.
Even before that correction, I was very skeptical of that figure, because I just couldn’t make the math work based on a county-by-county breakdown of Bush/Kerry numbers. Turns out also that the 44 percent national figure was also revised downwards, but I’m not as exercised about that one. It’s not so much to ask that a Texas newspaper pay attention to a story like this for more than a day, is it?
UPDATE: I just now realized that the date on that Mystery Pollster piece is December 1, 2004. Though he says he’s a “bit behind” in reporting on the revised exit poll data, it can’t be the case that the revision was within “a day or so” of the original report, so I’ve corrected that statement. I’ve also sent an email to the reporter on this story to point this out, as Kevin suggested. If I hear anything back, I’ll post about it.
UPDATE: I have received a gracious reply from reporter Greg Jefferson, who said that a correction is forthcoming. I will post that URL when I see it.
You should email the author. He’d probably appreciate the heads up.
And the Fort Worth Star-Telegram did an analysis that set the figure of Hispanic support for Bush at a more reasonable 32 percent.
Zangwell – If you look through my older posts on this, I put together a spreadsheet that compared Bush’s performance in heavily Hispanic counties in Texas in 2000 to 2004. He improved in those counties across the board, with total support in the 45% range (may have been a bit higher, I’d have to look). Bottom line, I’d doubt that 32% figure strongly – in fact, I’d doubt it for 2000 as well. The numbers don’t bear that out.
Now, I think it’s clear his current level of support among Texas Hispanics is in the low 30s, at best. But it was definitely better than that in 2004.