TX-KVUE/Public Strategies tracking poll: Obama and Clinton tied

[Tender Branson]

Obama: 46%
Clinton: 46%
Undecided: 8%

A week’s worth of data, culled from 2,400 people who said they either voted early or strongly intended to vote, puts the early voting results at about 56-percent for Obama, and 44-percent for Clinton. We also know that early turnout among African Americans was nearly identical to turnout among Hispanics.

This gets to the question that will determine the outcome of this election: Will turnout on March 4 be higher among African Americans or Hispanics?

Our pollsters at Public Strategies, Inc. ran some scenarios based on our polling data and turnout patterns from the 2004 election. It’s clear that turnout patterns in 2008 won’t remind of us 2004 in the slightest, but it’s a useful guide. Here’s what the pollsters found:

If turnout among African Americans and Hispanics is identical, then Obama would win the Texas primary by one point.

If more African Americans than Hispanics vote, even by 3 percentage points, then Obama would win by three points.

If Hispanics outnumber African Americans by 6-percent, then Clinton would win by two points.

http://www.khou.com/topstories/stories/khou080302_tj_poll.1827b584.html

[Verily]

All I remember about early voting turnout patterns is that the Hispanic counties overwhelmingly vote early, or at least did in 2004. I know at least that was true of Hidalgo County, the most Hispanic county in the top 15 most populous counties for which early voting turnout data is being released. (Something like 30,000 of 40,000 Democratic primary votes in Hidalgo County were cast early in 2004.)

[Verily]

A quick check of the SoS's website indicates that this is correct. In 2004, 12% of registered (Democratic) voters in Hidalgo County had cast early ballots, as had 5% in Nueces County and 4% in El Paso County. 7% had cast early ballots in relatively heavily black Jefferson County (33% black, not nearly so high as the 60%+ of the Hispanic counties) and 4% in student-heavy Travis County; numbers were very low everywhere else.

This suggests that turnout is up least in the Hispanic areas. It's up 3% in Hidalgo, 5% in Nueces, and 7% in El Paso against 10% in Travis, and, eyeballing it, 8% overall; Jefferson County is not included in the 2008 data as it has slipped off the 15 most populous counties list in favor of heavily Hispanic Cameron County.

In addition to implying that Hispanics will comprise less of the electorate in 2008 than they did in 2004, the trend for Hispanics to vote early (despite rising relatively little, Hidalgo, El Paso and newly-added Cameron County again top the early turnout along with Travis) continues, which means that, if Hispanics and blacks actually did vote equally early, blacks will outnumber Hispanics overall.

[Eraserhead]

DOWN TO THE WIRE