POLLS

User avatar
listeme
Posts: 3923
Joined: Tue Jul 14, 2009 11:09 am

POLLS

Postby listeme » Sat Oct 20, 2012 9:31 pm

Which states are south, according to Gallup, anyone know? I looked but didn't find it. (And please, don't do my homework for me if you don't know how to easily find it. I can look more better tomorrow :) )
Say what you mean.

User avatar
TollandRCR
Posts: 15691
Joined: Sun Mar 22, 2009 11:17 pm

POLLS

Postby TollandRCR » Sat Oct 20, 2012 10:01 pm

A major share of Gallup's volatility may be due to premature application of the likely voter model. Gallup shifts to that model on October 1. Respondents may be waxing or waning in their enthusiasm for a candidate even when their candidate preferences are unchanged. They may not firmly set their intentions until a few days before the election. A 2004 article explores this: Public Opinion Quarterly Winter 2004 [link]"Likely (and Unlikely) Voters and the Assessment of Campaign Dynamics",http://www.aapor.org/Content/aapor/Resources/PollampSurveyFAQ1/ElectionPollingResources/Erikson,_Panagopoulos_and_Wlezien%5B1%5D.pdf[/link] by Erickson, Panagopoulos, and Wlezien. The authors' advice: concentrate on advance estimations of respondents' likelihood of voting on election day, not at the moment of the poll. That advice probably helps Frank Newport not at all.

User avatar
Sugar Magnolia
Posts: 4488
Joined: Sun Apr 01, 2012 6:44 am

POLLS

Postby Sugar Magnolia » Sun Oct 21, 2012 6:18 am

Which states are south, according to Gallup, anyone know? I looked but didn't find it. (And please, don't do my homework for me if you don't know how to easily find it. I can look more better tomorrow :) )

According to Slate Gallup says they are;VirginiaNorth CarolinaSouth CarolinaGeorgiaFloridaKentuckyTennesseeAlabamaMississippiArkansasLouisianaOklahomaTexas[/break1]slate.com/blogs/weigel/2012/10/18/can_liberals_ignore_gallup_because_of_the_numbers_in_southern_states.html]http://www.slate.com/blogs/weigel/2012/ ... tates.html

User avatar
listeme
Posts: 3923
Joined: Tue Jul 14, 2009 11:09 am

POLLS

Postby listeme » Sun Oct 21, 2012 9:12 am

Thanks, Sugar.Okay, I think I agree with both Markos and Weigel on this specific issue. The south IS skewing the overall polling -- and Florida, Virginia and North Carolina are parts of it that I will continue to fight to keep in our column.But yeah, it's easy to see from that why the EV and national polls are not completely in sync.
Say what you mean.

User avatar
TollandRCR
Posts: 15691
Joined: Sun Mar 22, 2009 11:17 pm

POLLS

Postby TollandRCR » Sun Oct 21, 2012 9:44 am

Today's Gallup Daily Tracking poll has Romney at 51% among "likely voters," with President Obama at 45% on the seven-day rolling average ending 10/19. [imgwidth=480]http://images.huffingtonpost.com/2012-10-20-nationaltrackers.png[/imgwidth]

User avatar
Reality Check
Posts: 11285
Joined: Fri Feb 20, 2009 8:09 pm
Location: USA
Contact:

POLLS

Postby Reality Check » Sun Oct 21, 2012 1:37 pm

IBD Daily Tracking has the President +6:Obama: +5.7Obama 47.9% | Romney 42.2%•Obama has opened up a 6-point lead over Romney in our latest poll, his largest lead yet.•Some 8.1% of those polled said they were “not sure” whom they would vote for, also the largest percentage yet.•Obama’s lead seemed to build after the second debate, which many mainstream media pundits believe Obama won.•Self-described “moderates” now prefer Obama over Romney by 22 points, the second largest margin since we began polling.•Obama holds a hefty 35-point lead among urban dwellers, and a comfortable 5-point lead in the suburbs.[/break1]investors.com/special-report/508415-ibdtipp-poll.aspx]http://news.investors.com/special-repor ... -poll.aspxSomeone said this was rated the most accurate poll during the last two presidential election cycles.
“The reality of what we really are is often times found in the small snips, way down at the bottom of things.” – Jean Shepherd

chinacreekpj
Posts: 1347
Joined: Tue Jun 23, 2009 4:09 pm

POLLS

Postby chinacreekpj » Sun Oct 21, 2012 2:41 pm

IBD claims they are the #1 pollster and that Gallup is #7. Thoughts?

User avatar
Sterngard Friegen
Posts: 31098
Joined: Wed Aug 05, 2009 12:32 am

POLLS

Postby Sterngard Friegen » Sun Oct 21, 2012 2:52 pm

My suspicion is that there are flaws in many of the polls, that cell phone users are being undercounted, that students and Hispanics are being discounted, and that OFA's internal polls show a comfortable lead but want to make it seem close in order to drive GOTV.Just my opinion, but I haven't been worried since the Boss showed up for the second debate.The only thing we could use more of, to sway Indies and Undies, is a little more loony bomb throwing by Taitz and Drumpf. Every time Drumpf opens his mouth he sways those moderate voters who don't want to do what a fake billionaire tells them to do. Taitz? Well, when did anyone with an IQ above room temperature ever take Typhoid Orly seriously?

chinacreekpj
Posts: 1347
Joined: Tue Jun 23, 2009 4:09 pm

POLLS

Postby chinacreekpj » Sun Oct 21, 2012 3:02 pm

We're on the same page, bro. Maybe Taitz can make some national news tomorrow.

User avatar
Foggy
Posts: 18397
Joined: Tue Jan 20, 2009 12:00 pm
Location: Raleigh, NC\
Contact:

POLLS

Postby Foggy » Sun Oct 21, 2012 3:20 pm

My suspicion is that there are flaws in many of the polls, that cell phone users are being undercounted, that students and Hispanics are being discounted, and that OFA's internal polls show a comfortable lead but want to make it seem close in order to drive GOTV.

I would add to that the "Loh factor," which includes me personally.Many people are entirely unwilling to be polled, either because of the inconvenience or because they resent the intrusion into their private thoughts. Also there are a lot of intelligent people who have busy lives and just don't have the time for answering a poll. I received a polling phone call on my undercounted cell phone as I was walking on the street canvassing Obama voters. I refused to take part in the poll, but I did take the time to say I was working for the President and she could put me down as 100% voting for Obama before I hung up on her. But who knows if she did or not. I didn't have time for that nonsense; I was working.
"The only thing more dangerous than an idea is a belief. And by dangerous I don't mean thought-provoking. I mean: might get people killed." - Sarah Vowell

User avatar
Orlylicious
Posts: 2154
Joined: Mon Apr 23, 2012 4:02 pm

POLLS

Postby Orlylicious » Sun Oct 21, 2012 3:27 pm

RCP is 51% owned by Forbes.

User avatar
listeme
Posts: 3923
Joined: Tue Jul 14, 2009 11:09 am

POLLS

Postby listeme » Sun Oct 21, 2012 3:28 pm

My suspicion is that there are flaws in many of the polls, that cell phone users are being undercounted, that students and Hispanics are being discounted, and that OFA's internal polls show a comfortable lead but want to make it seem close in order to drive GOTV.

I would add to that the "Loh factor," which includes me personally.Many people are entirely unwilling to be polled, either because of the inconvenience or because they resent the intrusion into their private thoughts. Also there are a lot of intelligent people who have busy lives and just don't have the time for answering a poll. I received a polling phone call on my undercounted cell phone as I was walking on the street canvassing Obama voters. I refused to take part in the poll, but I did take the time to say I was working for the President and she could put me down as 100% voting for Obama before I hung up on her. But who knows if she did or not. I didn't have time for that nonsense; I was working.

I think that this evens out to a large extent. I see nearly identical comments on RW sites. Well, except for the voting for Obama part. :mrgreen:
Say what you mean.

User avatar
BFB
Posts: 5283
Joined: Mon Jan 17, 2011 11:48 pm

POLLS

Postby BFB » Sun Oct 21, 2012 3:39 pm

The scoring scheme is: [*:26x425jz]Thought given to election (quite a lot, some) [*:26x425jz]Know where people in neighborhood go to vote (yes) [*:26x425jz]Voted in election precinct before (yes) [*:26x425jz]How often vote (always, nearly always) [*:26x425jz]Plan to vote in 2012 election (yes) [*:26x425jz]Likelihood of voting on a 10-point scale (7-10) [*:26x425jz]Voted in last presidential election (yes)

Holy crap.Why don't they just ask, "Are you likely to vote in this year's Presidential election?"

User avatar
Slartibartfast
Posts: 4022
Joined: Mon Aug 10, 2009 2:52 pm

POLLS

Postby Slartibartfast » Sun Oct 21, 2012 3:46 pm

I think this article from Sam Wang at the Princeton Election Consortium is relevant here.

In the wake of his improved debate performance, President Obama’s recovery is now apparent. It is most clear when viewed in terms of the Meta-Analysis of state polls. Over the last four days, the Popular Vote Meta-Margin – the amount of swing it would take to create an electoral near-tie – has moved by over 1.0%. Today, the President’s effective lead, using Electoral College mechanisms, is Obama +1.8%. A rapid move like this can continue for a few days as polls catch up with the nation’s mental state.But why do national polls continue to look so close, and in about half of cases good for Mitt Romney? To answer that, let’s take a look back at the effects of all three debates so far.This is a graph showing the average national margin during a 4-week period when 50 national polls were taken. (If you wonder why poll aggregation is needed, that fact alone answers it.)[more at the link...][/break1]princeton.edu/2012/10/21/anatomy-of-a-bounce/#more-7753]http://election.princeton.edu/2012/10/2 ... #more-7753


User avatar
TollandRCR
Posts: 15691
Joined: Sun Mar 22, 2009 11:17 pm

POLLS

Postby TollandRCR » Sun Oct 21, 2012 3:46 pm

IBD claims they are the #1 pollster and that Gallup is #7. Thoughts?

They base this on the final pre-election segments of IBD/TIPP's daily tracking polls for 2004 and 2008. Investor's Business Daily is regarded as a consistently conservative publication, more so than the Wall Street Journal. They were very pleased to be able to [link]report that 45% of practicing physicians would consider closing their practices because of the Affordable Care Act,http://news.investors.com/091509-506199-45-of-doctors-would-consider-quitting-if-congress-passes-health-care-overhaul.aspx?fromcampaign=1[/link].





The crucial partner is the polling firm, [link]TIPP,http://www.tipponline.com/about-tipp/about-tipp/general/welcome-to-tipp[/link]:


For the past 12 years, TechnoMetrica [the parent of TIPP] has been the exclusive polling partner of the Investor's Business Daily, a leading national daily newspaper. TIPP is also the polling partner of the Christian Science Monitor.

Their [link]methodology,http://dl.dropbox.com/u/67204268/IBDTIPP_Methodology.pdf[/link] is conventional or even old-fashioned: finger-dialed telephone calls (no predictive dialing, thus no dead air); telephone numbers derived from a random digit dialing sample that includes randomly generated numbers for cell phone exchanges; quotas determined by Census regions and divisions; up to five call-backs when no one picks up; about one-third of interviews from cell phones; standard weighting to bring sample into conformity with latest Census age, gender, race, and region estimates.





In addition, TIPP weights by party identification, based on their own estimates from recent polls. This is highly controversial. It would serve to give less weight to persons who self-identify as Republicans, assuming that their prior polls conformed with others' estimates of Party ID. It is believed that Rasmussen's likely voter model gives higher probabilities of voting to Republicans until the final polls of an election, when he swings into conformity with other polls.





Media Matters ran [link]an article,http://mediamatters.org/research/2010/03/17/siegel-uses-not-scientific-ibdtipp-poll-to-prop/161803[/link] quoting Nate Silver (538) to the effect that the TIPP poll of physicians was not credible. Silver also notes that they badly missed the youth vote in 2008, projecting 74% for John McCain.





Silver's first objection to TIPP's physicians poll is that it was conducted by mail, "which is unusual." Not quite so, Nate. In conducting interviews of tightly targeted, "elite" populations, it is often necessary to recognize that you cannot interview them at your convenience. They will not accept telephone calls, and they will not see you if you show up in person. For that reason, it is conventional to contact such potential respondents by mail to notify them that they will be asked to be part of a survey. It has become common for the polling firm to promise to make a contribution to the charity of the respondent's choice if the respondent completes the interview. TIPP may have combined this pre-interview letter with an actual questionnaire; I cannot tell whether they offered the charity inducement.





There would still be questions about non-response; people with strong beliefs are more likely to respond to surveys about current disputed topics. There are also questions about non-disclosure of such critical aspects of the study as TIPP's definition of "practicing physician." TIPP's 45% may have been a wild over-estimate, but the mail aspect of their methodology was not that strange. Sorry, Nate.

User avatar
TollandRCR
Posts: 15691
Joined: Sun Mar 22, 2009 11:17 pm

POLLS

Postby TollandRCR » Sun Oct 21, 2012 3:49 pm

Holy crap.Why don't they just ask, "Are you likely to vote in this year's Presidential election?"

Because almost everybody will answer "yes." That is the worst of the social desirability problem. Most pollsters try to use some behavioral measures, while recognizing that self-reports are problematic.

chinacreekpj
Posts: 1347
Joined: Tue Jun 23, 2009 4:09 pm

POLLS

Postby chinacreekpj » Sun Oct 21, 2012 4:45 pm

Interesting info Tollie. I'm still not sure what to think though, lol.

User avatar
TollandRCR
Posts: 15691
Joined: Sun Mar 22, 2009 11:17 pm

POLLS

Postby TollandRCR » Sun Oct 21, 2012 6:01 pm

Interesting info Tollie. I'm still not sure what to think though, lol.

I have been suspicious of the IBD/TIPP polls because of the reputation of Investor's Business Daily. I am less suspicious of TIPP despite Nate Silver's criticism. Silver is not trained as a survey researcher, so it is not surprising that he is unaware of some of the techniques used by pollsters when the universe is not the national adult population.





The Volokh Conspiracy ran a [link]small analysis,http://www.volokh.com/posts/1225926066.shtml[/link] of the performance of pollsters as of Nov. 5, 2008.


The popular vote totals are not final. On Wednesday afternoon, the margin was 6.2% and rising [it ended at 7.2%]. The pollsters who nailed it by reporting 6-7% included Rasmussen Reports, Pew Research, FOX News, Ipsos/McClatchy, CNN/Opinion Research, and the poll aggregator 538. [IBD/TIPP estimated 8% spread according to this article; I calculate a spread of 7.2% from https://docs.google.com/spreadsheet/ccc?key=0Ak7keQ76qsvHcExLbVRnY1ZtN1BkX04tX3Q3VTJjVlE#gid=1.]





The worst performers were Gallup and Zogby with an 11% spread, followed by Battleground with a 3.5% average of its two models, and Marist, CBS News, and ABC News/Wash Post a 9% spread.





Interestingly, CNN and Fox News both outperformed the oldline networks – ABC, CBS, and NBC.


Curious Blue
Posts: 2462
Joined: Wed Feb 18, 2009 7:42 am

POLLS

Postby Curious Blue » Sun Oct 21, 2012 9:13 pm

I think that there is another issue relate to likely voter screens that may explain why the state polls tend to present a different picture than national -- and that is extreme variability from one state to another as to ease of voting. That is, if you look at Gallup's 10 point scale on the "how likely are you to vote" question, the choice of "7" is rather arbitrary. The more of a hassle it is to vote in a given jurisdiction, the greater the motivation needed for each voter. There's a long list of factors that bear on ease of voting.Thing that make it easier to vote:* early, on-site voting* mail-in voting / absentee ballots for all questioner* same day voter registration* multiple options for polling places* pre-paid postage on absentee ballotsThing that make it harder:* early registration deadlines* strict voter ID requirements* history of long lines at the polls* history of voter intimidation* inclement weather* confusion about voting requirementsA more sensible "likely" voter screen might be sensitive to the voter's state, being more lax in easy voting states.That's one reason I think that the NC polls might be undercounting Obama voters -- NC has a system of same-day registration for early voters, which probably enhances the value of GOTV efforts, especially if there is a canvasser on the doorstep who can help arrange transportation to the polling place. It might be an interesting (and rather time-consuming) exercise to do a state-by-state analysis to compare "likely" voter polling to the relative ease of voting in each state.

User avatar
TollandRCR
Posts: 15691
Joined: Sun Mar 22, 2009 11:17 pm

POLLS

Postby TollandRCR » Sun Oct 21, 2012 9:28 pm

...It might be an interesting (and rather time-consuming) exercise to do a state-by-state analysis to compare "likely" voter polling to the relative ease of voting in each state.

Newport has part of the data with which to do this, at least for the larger states. It is a fascinating idea.

User avatar
TollandRCR
Posts: 15691
Joined: Sun Mar 22, 2009 11:17 pm

POLLS

Postby TollandRCR » Mon Oct 22, 2012 9:01 am

Traders on Intrade are struggling to boost Mittens above 40%. They are losing. In the meantime, one diehard continues to insist (or pretend to insist) that Romney will resign from the race, to be replaced by Huckabee or Santorum. He is speculating about what Pres. Huckabee's tax policy will be: a big consumption tax, no income tax and no capital gains tax.

User avatar
Piffle
Posts: 6323
Joined: Sun Sep 26, 2010 12:39 pm

POLLS

Postby Piffle » Mon Oct 22, 2012 9:56 am

Traders on Intrade are struggling to boost Mittens above 40%. They are losing. In the meantime, [highlight]one diehard continues to insist (or pretend to insist) that Romney will resign from the race[/highlight], to be replaced by Huckabee or Santorum. He is speculating about what Pres. Huckabee's tax policy will be: a big consumption tax, no income tax and no capital gains tax.

Is he taking side bets? Dang, I'd dip into my retirement fund for a piece of that action.

User avatar
TollandRCR
Posts: 15691
Joined: Sun Mar 22, 2009 11:17 pm

POLLS

Postby TollandRCR » Mon Oct 22, 2012 10:25 am

Traders on Intrade are struggling to boost Mittens above 40%. They are losing. In the meantime, [highlight]one diehard continues to insist (or pretend to insist) that Romney will resign from the race[/highlight], to be replaced by Huckabee or Santorum. He is speculating about what Pres. Huckabee's tax policy will be: a big consumption tax, no income tax and no capital gains tax.

Is he taking side bets? Dang, I'd dip into my retirement fund for a piece of that action.

Everybody else thinks he is lying, as do I. They want to see his bank statement(s) showing this investment. Some diehards hang on far too long, however.

User avatar
TollandRCR
Posts: 15691
Joined: Sun Mar 22, 2009 11:17 pm

POLLS

Postby TollandRCR » Wed Oct 24, 2012 12:20 am

Over the last 48 hours somebody(ies) has spent ~ $20,000 to boost Mittens to a 45.3% chance of being elected. One school of comments says that the "Trump video" will sink President Obama, a view that has a small degree of rationality about it. Then there is the delusional:

Ticachaser wrote on Oct 24, 2012 04:06:15 #In reply to im a prophet. srsly's comment on Oct 24, 2012 03:54:13 who wrote: what news has come out in the last 48 hours to justify a ...Every credible poll has Romney up big 5-6% and over 50%, anytime you have greater than 2% margin of victory the electoral college has always fallen in line. It's quite obvious Obama is on the ropes, undecided vote goes to the challenger at 88%. Hate to tell you how this movie ends, you can refer to him as "the last black LIBERAL president"

I am equally prepared to believe that somebody has spent a rather small amount of money to place a cheap ad that somebody might pick up in the MSM.

User avatar
Patagoniagirl
Posts: 2194
Joined: Thu Jul 19, 2012 1:15 pm

POLLS

Postby Patagoniagirl » Thu Oct 25, 2012 10:43 am

AP Poll Romney Erased Obama Advantage over Women??? I don't see how this is possible??? The only major things that have happened since he held that lead were FU's by Reoublican men saying stupid things about women, rape and Gawd.[/break1]google.com/search?q=AP+Poll+Romney+Erased+Obama+advantage+over+Women&ie=UTF-8&oe=UTF-8&hl=en&client=safari]http://www.google.com/search?q=AP+Poll+ ... ent=safari


Return to “Miscellaneous”

Who is online

Users browsing this forum: chancery, DrIrvingFinegarten, Google [Bot], Maybenaut and 1 guest