Tuesday, July 29, 2014

Latest YouGov: Abbott 54, Davis 37

Not the TexTrib this time, but CBS/NYT commissioning the poll.  I saw it first at DK early yesterday morning, posted about it here and Tweeted that at lunchtime, and the Houston Chronicle had it yesterday afternoon.  As disclosure, I was included again among poll respondents.  The results for the US Senate race in Texas ran similar to the gubernatorial, with John Cornyn ahead of David Alameel 51-35, and the generic Congressional Democrat coming in behind the R, 50-31 (counting leaners in both).

It's also important to note that there were no third party candidates included in the poll questions, just 'independent' in the Congressional query.  That's how you get such a small number of undecideds -- 'other' and not sure', in this case.  The most significant finding to me was the 7% who declared they would not be voting at all.

YouGov's polling methodoly has long been questioned because of its online, opt-in nature.  The best explanation of this comes from the NYT.

YouGov’s work is worthy of its own discussion because it’s the first set of data from an online panel this year. The other nonpartisan surveys have used traditional, random-digit dialing to reach a sample of adults by telephone.

Random-digit dialing has long been the gold standard for public polling, but declining response rates may be complicating the ability of telephone polls to capitalize on the advantages of random sampling. Most polls underestimated President Obama’s standing in 2012, perhaps because young and nonwhite voters were least likely to own a landline and least likely to respond to telephone pollsters. Polls may also have exaggerated Mr. Romney’s gains after the first presidential debate, because Mr. Obama’s supporters were less willing to respond after his weak performance. The phenomenon is known as “differential non-response.”

CBS/NYT had YouGov poll over 100,000 members in all fifty states.  The full tabs for the Texas races can be found at this link.

As the young voters who are less likely to respond to telephone surveys become an ever-greater share of the population over time, it is probably more important for analysts to have an ensemble of surveys using diverse sampling and weighting practices.

YouGov has emerged as a part of that ensemble. It has tracked many of its respondents over months, if not years, which gives it additional variables, such as a panelist’s voting history, to try to correct for non-response. After the first 2012 debate, YouGov showed less of a swing than many other polls, and its final pre-election polls were as good as or better than many other surveys in forecasting the results.

There are still questions about the effectiveness of web panels, which can reach only the 81 percent of Americans who use the Internet. That’s worse than the 98 percent of households that can be reached by a live interview telephone survey, although it’s better than the 63.5 percent of Americans who have a landline telephone and can therefore be contacted by automated polling firms, which are prohibited by federal regulations from calling people on their cellphones.

Non-Internet users tend to be less educated, less affluent and more likely to be Hispanic or over age 65. These concerns aren’t strictly theoretical: YouGov most likely underestimated President Obama’s share of the Hispanic vote in 2012. Its final survey showed Mr. Obama with 59 percent of the Hispanic vote, far lower than the 71 percent in the exit polls.

So the silver lining for Davis (and Alameel and other Dems down the ballot) is that younger voters and Latinos are widely under-represented in these numbers.

Another issue is that the YouGov panel does not use probability sampling, the theoretical underpinning of modern polling. In a probability sample, every voter should have an equal chance of being randomly selected, making the sample representative. Phone numbers provide a device for randomization that is impossible online.

Instead, YouGov attempts to build a large, diverse panel and then match its panelists to demographically similar respondents from the American Community Survey, an extremely rigorous probability survey conducted by the Census Bureau. This step is intended to mimic probability sampling. But it can require significant assumptions about the composition of the electorate, including partisanship. These assumptions are contestable and based on varying amounts of evidence.

All of this is controversial among survey methodologists, who are vigorously debating whether a non-probability web panel should be used for survey research. At the same time, they’re also debating whether the sharp rise in non-response is undermining the advantages of probability sampling. Only 9 percent of sampled households responded to traditional telephone polls in 2012, down from 21 percent in 2006 and 36 percent in 1997, according to the Pew Research Center.

John Zogby is the fellow who pioneered this polling model, and everyone should be familiar by now with his reputation for accuracy.  YouGov has refined its methods over time, and to its credit, has managed better outcomes in predictive analysis.

While the methodology debate rages, it’s probably best to have an eye on a diverse suite of surveys employing diverse methodologies, with the knowledge that none are perfect in an increasingly challenging era for public-opinion research.

One striking aspect of the YouGov results is that they are broadly consistent with previous data on the campaign. Republicans appear to have narrow leads in enough states to win the Senate, but only narrow leads. The Republican lead is less than two percentage points in Michigan, Iowa, Louisiana and North Carolina. In Arkansas, Tom Cotton, a Republican challenger, leads Senator Mark Pryor, the Democratic incumbent, by four points.

The panel provides its best news for Democrats in Colorado, where Mark Udall, a Democratic incumbent, has a four-point lead. That’s far better than a recent Quinnipiac poll, which showed Mr. Udall trailing by two points, but it’s about the same as a recent NBC/Marist poll, which showed Mr. Udall up by seven points among registered voters.

Real Clear Politics summarizes polling for US races in a way similar to what Nate Silver does: taking all polls and averaging them to come to a consensus.  Here's that for Texas.  Rasmussen is widely understood to oversample Republicans, PPP is run by DK so it's suspected of having a Democratic bias.

My humble O is that YouGov is weighted too far right in this poll to the tune of about 5 points, which would leave Davis, Alameel, et.al in the 49-42 range or thereabout, which makes much more sense.  Irrespective of what the polling reveals, Abbott will blitz the airwaves beginning soon, and Texpate thinks Davis should as well.  I don't agree with him on that, but we'll see what happens.

Democrats should simply ignore the inevitability meme and keep grinding away on the phones, keep walking the blocks and keep donating what they can.  Wendy Davis still has a puncher's chance against Greg Abbott, and the rest of the Democrats down the ballot may have better than that, as they won't have to contend with the multi-million dollar broadcast onslaught from a well-heeled Republican opponent.

The race isn't over, even as much as the GOP and the corporate media would like to keep saying it is.  It's entering the homestretch, however, and Davis needs to make a move.

Update: If you needed reminding about why you should be skeptical about polls... here you go.

Update II: Forrest Wilder at the Texas Observer has a terrific suggestion for something the Davis campaign could emphasize: Medicaid expansion.

It is understandable that Davis hasn’t made abortion—or even women’s health—a cornerstone of her campaign. This is Texas, after all, and it’s wise for a Democrat to run on issues that are more unifying. But why not a seven-city tour on, say, Medicaid expansion? Expanding Medicaid under the Affordable Care Act will not only save lives and put more than a million Texans on health insurance, it’s a terrific deal for the state. The feds will pay 90 percent of the cost. By rejecting the expansion, Rick Perry and Abbott are leaving $100 billion on the table, according to recent estimates.

It’s good politics too—even if Republicans start hollering about “Obamacare.” (They will anyway.) Democratic governors in some red states, like Kentucky, have made Obamacare a winning issue. In Arkansas, Gov. Mike Beebe—one of the most popular governors in the nation—got a Republican-controlled Legislature to sign off on a Medicaid model that uses federal dollars to help people buy private insurance. That’s the same basic idea touted by some Republicans in the Texas Legislature. Polls, including one by Rick Perry’s own pollster, also show that a solid majority of Texans favors expanding Medicaid.

Davis, when asked recently by MSNBC’s Chris Hayes, was unequivocal in her support (“I absolutely do”) for Medicaid expansion. And in mid-June, she unveiled her economic development plan, which included Medicaid expansion. But otherwise she’s rarely discussed health care so far. The word “Medicaid” doesn’t appear once on her campaign site. 

Democratic strategists I spoke with cautioned that it’s still early in the campaign; that the Davis grassroots effort feels and sounds different than the “messaging” in the media; and that her team has been frustrated by the media’s indifference to her policy ideas.

As Paul Burka of Texas Monthly has pointed out, if she made it a central issue she’d have the doctors on her side, the hospitals, and much of the business community, not to mention local governments and—most important—millions of Texans who would see the benefits of healthier families.

Go read the whole thing.


Gadfly said...

Right on not giving up.

That said, as for all those people saying they won't vote?

Well, if Wendy keeps tracking right, that may be part of the reason why.

PDiddie said...

It's not yet August, I've been hyper-critical, and that (criticism) has generated considerably more, at least in the Tex-blogosphere.

There's still time for her to turn it around but time grows short.

Gadfly said...

You're right. And, yes, the Whitaker piece was good. There's lots of "poor white' red staters who don't know what they're missing.