Their sample is suggesting a much less diverse electorate than will show up at the polls:

RACE
White 75%
Black 11
Hispanic 7
Other/DK/NA 7

http://www.quinnipiac.edu/images/polling/us/us07302015_demos_U645de.pdf



RACE for 2012 electorate polls:

White 72%
Black 13%
Hispanic 10%
Asian 3%
Other 2%

http://www.ropercenter.uconn.edu/polls/us-elections/how-groups-voted/how-groups-voted-2012/


It is very, very,..unlikely that electorate shows up at the poll as the white percentage of the electorate has decreased in every national election since 1976 with the exception of the 1992 election* where it was similar to the election that preceded it.

*from memory.

I suspect (just guessing) that the electoral college may give more weight to white votes, just a slight edge, than a straight popular vote. Remember. Gore definitely won the popular vote in 2000 yet the Supreme Court was able to argue that Bush had actually won.

A lot of the less populated states are have more white voters -- Montana, Vermont, Utah, etc. (on and on actually). California has lots of Hispanic voters but because the electoral college only allots us the equivalent votes for two senators and states like Montana, etc. get the same two for senators, the value of our votes, our individual votes is reduced.

Would that make a difference in valuing race and ethnicity for purposes of presidential polls? How does that work?

Polling is as much art as science though it is rooted in science. The largest obvious problem is we don't have a hard result to compare it to. For instance Hillary Clinton could be beating Bush lll by twenty points or losing to him by twenty points though either scenario is highly unlikely...What statistical sampling tells us is that two pollsters working independent of one another using the same methodology should get roughly the same results.

Pollsters call people randomly to ensure everybody in the population has the same chance of being polled and then make inferences about them to infer how everybody will vote.

I just think it's odd that there would be more white voters as a share of the electorate than the previous election when white voters have been an increasingly smaller share of the presidential electorate in every election since 1976 with the exception of 1992.


I hope I didn't add to the confusion.

but rather to reflect popular votes in which case, weighting to more white votes than previous elections would result in a faulty estimate. Thanks.

eom

It was more like her goose is in the oven roasting, which it still is.

Look at the trends in her trustworthy, cares about people like me, leadership numbers. All are slipping and there's no head room for her to improve those from the don't know enough column.

Bernie's gains in those numbers are about 2-1 in his favor as the don't know enough number comes down.

I seem to recall how Quinnepac was deemed unreliable, but I could well be mistaken or have it confused with another poll.

...this trustworthiness is blown way out of proportion.

But, even if it was really important, the difference between Clinton's favorability/unfavorability is about 3 points. The difference between Sanders' favorability/unfavorability is about 1 point - they're both basically "equal" in those two ratings.

Clinton: at +12 "Thumps" Trump (well, that rhymes) ... Sanders: at +8, "Tops" Trump.

but, Clinton "gets" and "gets", on -1 (Bush) and +1 (Walker) ... But, Bush "edges" at -5 and Walker "slips past" at -5.

Come on Q ... you know better than that! You use dispassionate/disinterested/neutral language ... that's how polling outfits get bad names.

He or she wants to give their Headline Thesaurus a workout.

I could just see my Grad school Research Methods Professor blowing a gasket, with red pen in hand! She was quite the animated character.

volunteers attended. Bernie is attracting people who are new to political activism. It's very interesting and makes everything unpredictable. I expected to see the usual Democratic Party activists that I know so well. But no. Of at least 50 people, I only saw one face I recognized, and I know him from community activities not from politics. He has always turned up his nose at invitations to be active politically.

So, the Bernie volunteers are new people. Different races and ethnicities. But that is always the case in Los Angeles. Lots of diversity. You cannot find a "white" group of many people East of the 5 Freeway in Los Angeles whether in the South or in the North of LA.

So there are a couple of unusual factors in this election: demographics in large states and how that will affect who votes as well as a possible set of voters who are rather new to politics on Bernie's volunteer team.

Interesting. These polls are still too early. Bernie is still drawing crowds. Lots and lots of young people. Lots of enthusiasm for Bernie.

No matter how many buried retractions and corrections they make, they only care about planting that seed. Hillary and the Clinton's are over-attacked.

think about it, folks...

That IS good.

if she's running against Trump.

I'm reading a book a friend recommended! But I was thinking that the GE match-ups seemed off compared to recent polls. I think that sample disparity goes a long way to explaining it. But Quinnipiac is usually not terrible, so maybe they will adjust in future models. I know some pollsters are speculating that 2016 will skew more white than 2012, but I don't see any reason to assume that. Certainly no data support the idea.

If you remember Gallup's modeling assumed a much more homogeneous electorate and they were wrong and picked the wrong winner.

K & R for HRC! She has remain constant from various poll.

Biden? Really. He has not thrown in his hat but is usually mentioned.

But many are adding questions without him, as well.

I think when he announces he's out, Bernie will get a 5% or so bump and Clinton will get a 3-4% bump with the rest going to the also-rans.

I don't know how people can be so dismissive of him.

Everybody who follows politics are well aware that Walker's presidency will easily end America as we know it. My parents whom is very religious Republicans told me that they really liked Bernie and intend to vote for him.

Hope everyone of her supporters enjoys the "numbers" because while they're all comfortable,
100,000 of Bernie's supporters were just getting started at over 3,000 meetings, in all 50 states last night.

Just keep hope'n...'cause it ain't gonna happen, Hill.

The Sanderstorm will sweep the country.

Bernie has a passionate support base, but so far, it seems to be a distinct minority.

Remember, be passionate and supportive of your candidate, but never lie to yourself.

That needs to change drastically for him to have any credible shot at the nomination.

I suspect Bernie's will be the same.

Still running at about 3-1.

as the presumed Dem nominee. Once the clowncar has parked and we have a GOP candidate the Clinton and Dem attack machines will kick into gear.

1) The sample is a not a good representation of what the electorate will look like in 2016. It is more "white" than the 2012 electorate, and in every Presidential election since the 70's it has been LESS white. I think that makes the head-to-heads skew towards the R's more than they do in other polls. Pretty much every GOP voter is gonna voice a negative opinion of ANY Dem nominee once the general election campaign is in full swing.

2) Many Bernie supports will respond that they have a negative opinion of Hillary during the primaries. Some will miantain that opinion and hold their nose and vote for her anyway. Others will moderate their feeling and switch to a positive opinion.

I'm not at all concerned about these numbers. At least, not right now.

Disappointing, but true.