2016 Postmortem
Related: About this forumPoll: Sanders and Clinton in Nevada dead heat
by Jon Ralston Fri, 02/12/2016 - 08:12
Hillary Clinton and Bernie Sanders are in a tight race to win Nevada, according to a poll taken this week.
The survey, paid for by the conservative Free Beacon, shows a 45-45 tie. It was conducted by TargetPoint of 1,236 potential Nevada caucusgoers from Feb. 8-10, with a margin of error just under 3 percent. That's a lot of interviews -- "867 interviews were completed using automated telephone technology and 369 were conducted using mobile phones," according to the polling instrument, which I have posted below along with the crosstabs.
<snip>
►Clinton loses on trust, 53-29; on who cares about people like you, 49-36; and who is progressive, 49-36. Danger, Will Robinson!
►The sample is almost 60 percent female (about what it as in '08), which ought to worry Clinton. Sanders leads 63-16 among young voters (18-29), and if there are a lot of youngsters who register on Caucus Day.... She's also losing among independents, as the Free Beacon reported.
https://www.ralstonreports.com/blog/poll-sanders-and-clinton-nevada-dead-heat
Full poll available at link
Bagsgroove
(231 posts)[link:http://
|kristopher
(29,798 posts)She has zero credibility on anything election related.
FreakinDJ
(17,644 posts)and yes Rachel pointed that out.
Actually Rachel made a point to call them out for "Push Polling" on this poll
kristopher
(29,798 posts)Push polls don't use legitimate polling methodology, this one does.
It is a poll that is designed to explore the probable viability of potential lines of attack. The data is as solid as any polling can be. In fact, it has a very good sample size.
HereSince1628
(36,063 posts)Thinkingabout
(30,058 posts)Just a Democratic poll. This poll was shot down by Rachal Meadow on Friday.
CoffeeCat
(24,411 posts)I read the polling methodology and it's actually pretty sound. Their sample size was large as well, more than twice the size of a standard poll.
It's not the outlet conducting the poll that is key. It's the methodology.
I looked into the data and I had a leader from LULAC look into it as well. He's actually an ardent Hillary Clinton supporter and precinct captain for her. He's a data hound and he said the poll looks very solid.
I don't think Maddow made a decent case for discounting this poll.
Thinkingabout
(30,058 posts)The site is noted for its aggressive, ideologically driven reporting, modeled after liberal counterparts in the media such as Think Progress and Talking Points Memo.
Meadow explained why there have not been other polls, mostly because Nevada does not have a good enough of a pattern years to determine what the data and this is why other pollsters have not produced a poll. Wonder why they did not do a Republican poll?
orpupilofnature57
(15,472 posts)corrected Chris Mathews on the fact Most Women, not just under 50 yr old's voted for Bernie in New Hampshire, you can't " Shoot Down " what you have reported, unless you have an ax to grind .
Dustlawyer
(10,495 posts)and not towing the corporate line; Cenk Uygur, same reason; Keith Olberman; and Martin Bashir. Rachel makes too much money to give up her cushy job, besides, she only has to slant things the corporate way.
gyroscope
(1,443 posts)and asked why would a RW group conduct a poll of the dem candidates but not repubs.
my guess is they want to see who their potential competition is in the GE.
kristopher
(29,798 posts)The polling is looking to find the most effective way to attack the Dems. To do that, they have to first establish the candidate preferences of the people being polled. Then they try to find out how to sway them.
find out your opponents strengths and weaknesses. in other words, know your enemy.
Jefferson23
(30,099 posts)kristopher
(29,798 posts)Past caucus attendee 45
First time attendee 55
Judging by the extremely high registration numbers, new caucus-goers are going to be a much higher percentage than that.
Sure, it is a right wing poll trying to find out the best point of attack against the Dems, but there is no reason to question the lead in questions establishing the level of support and trust.
Jefferson23
(30,099 posts)orpupilofnature57
(15,472 posts)Jefferson23
(30,099 posts)Dustlawyer
(10,495 posts)still_one
(92,219 posts)Whether the poll is an actual representation of what is going on in Nevada remains to be seen when other pollsters start reporting.
That the following questions included in the poll are push poll questions, and that puts serious doubt into the credibility of the poll:
"While Hillary Clinton was Secretary of State, the Clinton Foundation received millions in foreign donations and did not disclose them, despite her promising President Obama that she would do so. Does this make you more or less likely to support her?
Bernie Sanders says he wants to increase taxes on corporations in order to improve income equality, while other say this money would go to pay for new government spending programs. Does this make you more or less likely to support him?
Hillary Clinton supported the Wall Street bailout and big banks have given her over $1.6 million dollars in campaign contributions and speaking fees. Does this make you more or less likely to support her?
Bernie Sanders has proposed trillions in new government spending, including $15 trillion dollars more for a government run health care program. Does this make you more or less likely to support him?
Recently several news organizations have reported that the FBI could indict Hillary Clinton over her handling of her email server. How concerned are you that she could actually face such an indictment?"
Those who are so anxious to embrace this poll would be wise to wait for more independent pollsters to to report their findings
PonyUp
(1,680 posts)kristopher
(29,798 posts)Push polls don't use legitimate polling methodology, this one does.
It is a poll that is designed to explore the probable viability of potential lines of attack. The data is as solid as any polling can be.
still_one
(92,219 posts)As an aside, Its editor in chief is Matthew Continetti. Continetti is also Bill Kristols son-in-law
blue neen
(12,322 posts)In fact, know very little about it. I have, however, been called for polls like this one, most recently for the PA Senate race. It included very particular questions about all three Democratic candidates, Katie McGinty, Joe Sestak, and John Fetterman....and the incumbent Republican, Pat Toomey.
While it may not fit the definition of a push poll, the questions are designed to lead one to certain answers. Some of the topics may even be in an area that the "pollee" has no concern about as far as voting, but the pollster still gets an answer. So, while the data may be solid, it can still be misleading.
I think there will be more reliable polls coming along.
kristopher
(29,798 posts)In one case, like the poll in the OP, you want to establish an accurate profile of the people that you are eliciting opinions from. Without that accurate profile, the data is useless. That takes a (relatively) long time on the phone and dramatically escalates the per call price of conducting the poll compared to what is acceptable in a push poll.
A push poll, on the other hand, isn't concerned at all with the person the pollster is contacting. The poll works on volume - they want to influence the election and they need to put their ear worm in as many minds as possible. Therefore they skip virtually all demo and preference questions and focus on the smears. I'd estimate that a push poll will want to contact a minimum of 10X to 20X the number of people that a legitimate opinion poll would that is looking for a good sample of about 1000 respondents.
PonyUp
(1,680 posts)6chars
(3,967 posts)Last edited Sat Feb 13, 2016, 01:04 PM - Edit history (1)