After two days of mostly discouraging news for the local GOP, today’s Chron poll of two area Congressional races should make them feel better.
Two Houston-area congressmen under political siege likely face opposite fates in the Nov. 4 election, according to a poll conducted for the Houston Chronicle.
U.S. Rep. Nick Lampson, D-Stafford, trailed Republican challenger Pete Olson by 17 percentage points early last week, according to the survey by Zogby International. The poll had a margin of error of plus or minus 5 percentage points.
U.S. Rep. John Culberson, R-Houston, led Democratic challenger Michael Skelly more modestly, by 7 percentage points, with virtually the same margin of error.
In both Republican-friendly districts, a key factor appeared to be the Democratic candidates’ inability to run strong among independent voters and cut deeply into the ranks of Republican voters.
In the 22nd Congressional District, represented by Lampson after the 2006 resignation of Republican powerhouse Tom DeLay, only 5 percent of Republican voters in the survey had defected to Lampson.
Lampson led Olson among independents, 45 percent to 39 percent. But only 16 percent in the poll identified themselves as independent; while 52 percent said Republican and 32 percent said Democrat.
Links to all the results and crosstabs are here. As noted, Lampson wins 83.5% of Dems while Olson gets 87.9% of Rs. Skelly is doing better on this score, getting 85.2% of Dems to Culberson’s 80.2% of Rs; he’s also collected 11.9% of Republicans, to Culberson’s 8.5% of Dems. If Skelly were doing as well among independents (he trails, 38.7 to 32.4) as Lampson, this race would be close to a dead heat.
The surveys in both districts are based on assumptions about the ethnic and geographic makeup of the electorates.
For instance, the survey in Lampson’s district drew more than 40 percent of its sample from the Harris County portion. A significantly lower turnout there, or higher turnout in the other counties, could alter the results, probably cutting into Olson’s advantage.
Curiously, unless I just missed it, I didn’t see a county-by-county breakdown of the CD22 sample. In 2004, the totals were 64,590 votes out of 150,386 from Harris County (42.9%); 58,444 from Fort Bend (38.8%); 18,159 from Brazoria (12.1%); and 9,193 from Galveston (6.1%).
The CD07 poll is easy to comprehend, and it’s well within range of other polls of that race. A Research 2000 poll from two weeks earlier had Culberson leading 48-40. Interestingly, that R2K sample was less Republican than this Zogby one – it had a 39/33 R/D split, whereas Zogby has 45R/33D. These are small subsamples, with larger margins of error, so don’t read too much into that. I’m just noting it for comparison.
As for CD22, I can believe Olson is ahead, but I would not have expected it to be by that much. I can’t think of any other polls of this race that I’ve seen – which is odd when you get right down to it – so I just don’t have a basis to evaluate it. If you were skeptical of the other polls because of Zogby’s general reputation for flakiness, you shouldn’t put that feeling aside just yet.
One last item to note about these two polls is that there is a Libertarian candidate in each race – multi-timer Drew Parks in CD07, and John Weider in CD22. I would expect each of them to get about two or three percent of the vote. Neither was mentioned by name in the poll – they’re lumped in as “Other” with the “Not sure” responses.
For whatever the reason, the Chron did not choose to poll the other hot Congressional race in the area, which is CD10 and which may end up being the closest race of the three. Research 2000 released a poll for CD10 over the weekend, and it showed incumbent Mike McCaul leading challenger Larry Joe Doherty by a slim 46-42 mark. A poll from earlier in October had it at 43-38 for McCaul. Given that the McCaul camp is worried about how early voting has gone, it might have been nice to get a result from here as well.