« Climate Change | Main | " ... an excuse to snuff out secular democracy" Quartz »

26 July 2016


Feed You can follow this conversation by subscribing to the comment feed for this post.


The same journalist has the following story: https://www.washingtonpost.com/news/the-fix/wp/2016/07/25/4-brutal-poll-numbers-that-greet-hillary-clinton-at-the-democratic-national-convention/. Basically, polls say that Clinton's support is tanking and half the Sanders supporters still don't like her. Personally, I trust this story no more than I do the one HankP linked to: journalists usually have no idea how polls work and draw the most sensationalistic interpretations they can from them, whether they make much sense or not. The bottom line is that, at this stage, no one has enough charge in their crystal balls to see what lies ahead with much clarity, especially since we never had a set of poll numbers like these (yes, we probably had elections like these--but reasonably good poll numbers are fairly new.)



The Pew poll should actually worry the Clinton folks shitless, not reassure them that they will get Sanders supporters.

The lack of informative cross-tabs is making it a bit harder to see what the numbers are actually saying, but there are some serious problems in the way MSM has been portraying the Sanders voters.

First, Sanders voters are not all young idealists. In fact, you are correct that it won't be the young idealists that would break the Democrats, if any significant number of Sanders voters break ranks with the Democrats. A significant chunk of Sanders voters come from age 30-49 block, and most in this cohort are working class and white. Looking at the cross-tabs in the polls that do provide them (Quinnipiac polls have been good at breaking down the respondents by sociodemographic cohorts) since the Democratic primaries wound up, these voters are especially likely to break ranks with the Democrats. The almost 10% of the "Democrats" in the Pew polls who would break to Trump among the Sanders supporters and the uncertain may be a statistical blip, or a sign of trouble, if they should happen to fall disproportionately among a particular sociodegramographic cohort. Given what I've seen from other polls, I suspect that the latter might be dangerously likely for the Democrats. Without the raw data or at least good cross-tabs, I would not think the Pew numbers should make one complacent.

2. One critique of panel-based surveys like the Pew ATP is that they tend to be dangerously biased in favor of the politically active (and underestimates the "independents"). This is one reason why polls of this type tended to underestimate support for Sanders and Trump even more than other polls so far. The relative absence of "neither" between Clinton and Trump (compared to those who actually said they'd jump to Trump outright among the Sanders supporters and the switchers) suggests that the sample is indeed skewed oddly. This does get a bit more complicated: as the primary process went on and Sanders seemed certain to lose (and Trump became the Republican nominee), panel-based polls began to be more accurate, as the more apolitical voters began to drop out of the primary process. So what these polls are telling us, rather than saying that 90% of the all Sanders voters would support Clinton, it is saying that 90% of the political active Sanders voters would--a significantly less gain than the former--and of the politically active Sanders voters, almost 10% would bolt outright (which sounds awfully high given their likely characteristics). As we do not have comparable polling methods in the past (ATP was inaugurated only in 2015, for the 2016 campaign), we don't know how big a "blip" this 10% is, of course, but given what we know, this strikes me as an awfully big number.

To repeat, there is every reason to suspect that the "loyalty" of the Sanders voters is overreported, and this is not thought about twice because of the preconceptions about the Sanders voters--that they are young and liberal, while many of them are neither. The MSM is focusing on how the young liberals are the "bitter enders." They are not. Do remember that, in 1980, Kennedy did very well in the primary contests in places like MI, and (given the limited data available), Kennedy votes were pretty good predictor of the Reagan Democrats. Surely, it was the young twinkle liberals who supported Kennedy that decided to go over to Reagan over Carter.

David Habakkuk


‘It’s not so much that they are “conservative,” but rather that they don't fit the liberal-conservative continuum.’

Exactly. What precisely is the ‘liberal-conservative continuum’ supposed to mean?

Polls have always been hard to interpret. But the breakdown in the meaning of traditional classifications may be making interpretation much harder.

In both the last British general election, and the EU referendum, people who believed them ended up with egg all over their faces.

Phil Cattar

The difference in generations can probably be explained by the fact by the mid 1700s that branch of my family tree was living in South Carolina .There were farmers having many children and staring young.Sometimes the girls marrying at 14.My mother was the 19th of 21 children,6 full siblings ,14 half.

Phil Cattar

Yes,I kinda picked up he was considered somewhat of an oddball.



Where Trump should really do well this fall is in the traditional campaigning of going out and meeting people at State Fairs, etc.


What is open to Bernie is finding an effective platform to press on with the issues and, as you say, to support candidates at the local and state levels.

I would rather be on Kauai then 30 miles from Hanford, so I regard Tulsi as my Congress critter. :~)



This is what they say between the lines. Stop with the sophistry you schmuck.



Stop with the lying and the pathetic attempt at handwaving.

The big issue around the DNC emails in this context (The fact that most of the media is the propaganda arm of the DNC is much bigger IMHO) is how DWS and the rest of the DNC swore up and down there was no collusion, when there was collusion night and day.

Guess what? TPP is still going through. All the globalist Borg agenda is still going through. Bernie accomplished nothing other than getting a cool plane for himself.

Hillary needs a better grade of intern.



Oh, I agree with you to a point. For example I was pointing out not too long ago how the pollster were oversampling Dems by a factor of +12 and then crowing about how Trump had no chance.

Well the worm has turned.

Hillary is a thoroughly unappealing candidate, and her and the Democrats vision of America as nothing more than a 3rd World bazaar/grievance mongering industry is unappealing to the majority of Americans.



Who is who?


Anyone else watching the media purposefully attempt to deceive about what Trump said about Russia handing over Hillary's emails and absolutely melting down?

I'm amazed at how well he's playing the media, even after taking down 16 challengers and the Bush Dynasty.

The comments to this entry are closed.

My Photo

February 2021

Sun Mon Tue Wed Thu Fri Sat
  1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
Blog powered by Typepad