Welcome, Guest. Please login or register.
Did you miss your activation email?
February 28, 2020, 05:20:51 am
News: 2020 Gubernatorial Predictions are now active.

  Atlas Forum
  Election Archive
  Election Archive
  2016 U.S. Presidential General Election Polls
  USC Dornsife/L.A. Times Daybreak National Tracking: 11/7 - Trump +3.2
« previous next »
Pages: 1 ... 13 14 15 16 17 [18] 19 20 21 22 23 ... 30 Print
Author Topic: USC Dornsife/L.A. Times Daybreak National Tracking: 11/7 - Trump +3.2  (Read 63826 times)
Wiz in Wis
YaBB God
*****
Posts: 2,522


WWW Show only this user's posts in this thread
« Reply #425 on: September 21, 2016, 08:54:56 am »

Despite it's immediately obvious shortcoming(s), this poll is actually useful.

That's what I've been trying to say this whole time Tongue

It's the classic illustration of precision vs. accuracy. The poll seems like it may be highly precise, and consistently inaccurate. It's like trying to hit the center of the dart board, but always being two inches down and to the left. But as long as you know that the bias is consistent, you can just correct for that bias and then you have highly valuable, useful information.

I tend to agree, but the fact that the recent shift towards Trump was driven by a huge swing among African Americans in this panel undercuts the "precision" argument a bit.
Logged
Mallow
YaBB God
*****
Posts: 739
United States


Show only this user's posts in this thread
« Reply #426 on: September 21, 2016, 08:59:08 am »

Despite it's immediately obvious shortcoming(s), this poll is actually useful.

That's what I've been trying to say this whole time Tongue

It's the classic illustration of precision vs. accuracy. The poll seems like it may be highly precise, and consistently inaccurate. It's like trying to hit the center of the dart board, but always being two inches down and to the left. But as long as you know that the bias is consistent, you can just correct for that bias and then you have highly valuable, useful information.

I tend to agree, but the fact that the recent shift towards Trump was driven by a huge swing among African Americans in this panel undercuts the "precision" argument a bit.

Yeah, I was saying that under the assumption that the results were consistently biased. If they're not, that's a different story. Of course, the small sample size for AA's could be the reason for the funky-looking crosstabs. It's still possible that the overall results are precise, but the specifics when it comes to racial breakdown are wonky. That's why caution is always advised when trying to "unskew" based on crosstab data.
Logged
Wiz in Wis
YaBB God
*****
Posts: 2,522


WWW Show only this user's posts in this thread
« Reply #427 on: September 21, 2016, 09:06:25 am »

Despite it's immediately obvious shortcoming(s), this poll is actually useful.

That's what I've been trying to say this whole time Tongue

It's the classic illustration of precision vs. accuracy. The poll seems like it may be highly precise, and consistently inaccurate. It's like trying to hit the center of the dart board, but always being two inches down and to the left. But as long as you know that the bias is consistent, you can just correct for that bias and then you have highly valuable, useful information.

I tend to agree, but the fact that the recent shift towards Trump was driven by a huge swing among African Americans in this panel undercuts the "precision" argument a bit.

Yeah, I was saying that under the assumption that the results were consistently biased. If they're not, that's a different story. Of course, the small sample size for AA's could be the reason for the funky-looking crosstabs. It's still possible that the overall results are precise, but the specifics when it comes to racial breakdown are wonky. That's why caution is always advised when trying to "unskew" based on crosstab data.

Oh, I agree... I just think we've all bought into the "accuracy" of this poll a bit too much. I'm not calling for unskewing, just skepticism of the idea that this method is better at certain things than we have any reason to think it is.
Logged
Seriously?
YaBB God
*****
Posts: 3,029
United States


Show only this user's posts in this thread
« Reply #428 on: September 22, 2016, 02:17:03 am »

USC/LA Times national tracking poll (through 9/21) Trump +2.4%
Trump: 45.4% (-0.9%)
Clinton: 43.0% (+0.8%)

The bounce is clearly fading.
Logged
Ebsy
YaBB God
*****
Posts: 8,001
United States


Show only this user's posts in this thread
« Reply #429 on: September 22, 2016, 02:38:34 am »

Not good!
Logged
Interlocutor
YaBB God
*****
Posts: 1,444


Political Matrix
E: -3.48, S: -5.04

Show only this user's posts in this thread
« Reply #430 on: September 22, 2016, 03:31:40 am »

Gonna bury my head in the crosstabs a bit, but a pretty sharp drop in support for Trump among voters under 34
Logged
michelle
MikeWells12
YaBB God
*****
Posts: 3,989
United States


Political Matrix
E: -4.13, S: 0.35

P

Show only this user's posts in this thread
« Reply #431 on: September 22, 2016, 05:00:35 am »

Old Formula Clinton +8
Clinton 48% (+3)
Trump 40% (-2)

New Formula Clinton +4
Clinton 46% (+2)
Trump 42% (-1)

Clinton's gains are larger than yesterday's, as the two biggest swing demographics in this poll (young people and rich people), all turn to Clinton.
Logged
Gass3268
Moderators
YaBB God
*****
Posts: 21,196
United States


Show only this user's posts in this thread
« Reply #432 on: September 23, 2016, 07:34:37 am »

No update?
Logged
Clamdick McClaw
HockeyDude
YaBB God
*****
Posts: 11,078
United States


Show only this user's posts in this thread
« Reply #433 on: September 23, 2016, 07:48:11 am »

Old Formula Clinton +8
Clinton 48% (+3)
Trump 40% (-2)

New Formula Clinton +4
Clinton 46% (+2)
Trump 42% (-1)

Clinton's gains are larger than yesterday's, as the two biggest swing demographics in this poll (young people and rich people), all turn to Clinton.

Wowie kaboodles!  Youngs are finally realizing that #FeelingTheJohnson is a flirtation with disease of the country. 
Logged
‼realJohnEwards‼
MatteKudasai
YaBB God
*****
Posts: 1,868
United States


Political Matrix
E: -6.19, S: -4.87

Show only this user's posts in this thread
« Reply #434 on: September 23, 2016, 08:04:32 am »

so, StatesPoll, what do you think? Do they oversample millenial Texans with addresses ending in "3"? Wink
Logged
Seriously?
YaBB God
*****
Posts: 3,029
United States


Show only this user's posts in this thread
« Reply #435 on: September 23, 2016, 08:21:46 am »

Oddly, no. They haven't updated yesterday's numbers. I think they tried to, but somehow swung and missed. Neither the LA Times or the USC site has updated numbers yet.
Logged
Wiz in Wis
YaBB God
*****
Posts: 2,522


WWW Show only this user's posts in this thread
« Reply #436 on: September 23, 2016, 08:33:26 am »

Oddly, no. They haven't updated yesterday's numbers. I think they tried to, but somehow swung and missed. Neither the LA Times or the USC site has updated numbers yet.

God, I hope they're not messing with their methodology or something. Fingers crossed this is just a blip.
Logged
Seriously?
YaBB God
*****
Posts: 3,029
United States


Show only this user's posts in this thread
« Reply #437 on: September 23, 2016, 09:16:16 am »
« Edited: September 23, 2016, 09:31:56 am by Seriously? »

Oddly, no. They haven't updated yesterday's numbers. I think they tried to, but somehow swung and missed. Neither the LA Times or the USC site has updated numbers yet.

God, I hope they're not messing with their methodology or something. Fingers crossed this is just a blip.

This happened once before, but the LA Times got the numbers out somehow.

USC updated the data files when they were supposed to at midnight PST, but they didn't contain the 9/22 data. I am sure once it's early enough on the West coast, someone will notice the error and they'll fix it.
Logged
Wiz in Wis
YaBB God
*****
Posts: 2,522


WWW Show only this user's posts in this thread
« Reply #438 on: September 23, 2016, 09:49:49 am »

Oddly, no. They haven't updated yesterday's numbers. I think they tried to, but somehow swung and missed. Neither the LA Times or the USC site has updated numbers yet.

God, I hope they're not messing with their methodology or something. Fingers crossed this is just a blip.

This happened once before, but the LA Times got the numbers out somehow.

USC updated the data files when they were supposed to at midnight PST, but they didn't contain the 9/22 data. I am sure once it's early enough on the West coast, someone will notice the error and they'll fix it.

Cool... I would anticipate a small shift to Clinton... yesterday was the last day that a big Trump day looked to be added a week ago.
Logged
Seriously?
YaBB God
*****
Posts: 3,029
United States


Show only this user's posts in this thread
« Reply #439 on: September 23, 2016, 03:42:22 pm »
« Edited: September 24, 2016, 02:12:59 am by Seriously? »

USC/LA Times national tracking poll (through 9/22) Trump +1.4%
Trump: 44.8% (-0.6%)
Clinton: 43.4% (+0.4%)

Better late than never.
Logged
michelle
MikeWells12
YaBB God
*****
Posts: 3,989
United States


Political Matrix
E: -4.13, S: 0.35

P

Show only this user's posts in this thread
« Reply #440 on: September 23, 2016, 07:11:48 pm »

Old Formula Clinton +5
Clinton 46% (-2)
Trump 41% (+1)

New Formula Clinton +1
Clinton 44% (-2)
Trump 43% (+1)

The race seems to be settling again. Most of the change in these formulas in recent days has been statistical noise. The next big change should be coming right after the debates.
Logged
Seriously?
YaBB God
*****
Posts: 3,029
United States


Show only this user's posts in this thread
« Reply #441 on: September 24, 2016, 02:12:42 am »

USC/LA Times national tracking poll (through 9/23) Trump +2.0%
Trump: 45.3% (+0.4%)
Clinton: 43.3% (-0.1%)
Logged
Seriously?
YaBB God
*****
Posts: 3,029
United States


Show only this user's posts in this thread
« Reply #442 on: September 25, 2016, 02:12:53 am »

USC/LA Times national tracking poll (through 9/24) Trump +4.1%
Trump: 46.4% (+1.1%)
Clinton: 42.3% (-1.0%)
Logged
StatesPoll
Sr. Member
****
Posts: 441
WWW Show only this user's posts in this thread
« Reply #443 on: September 25, 2016, 03:00:26 am »

USC/LA Times national tracking poll (through 9/24) Trump +4.1%
Trump: 46.4% (+1.1%)
Clinton: 42.3% (-1.0%)

Peryhaps Bump by Cruz's endorsement?
Logged
Hilldog
Full Member
***
Posts: 117
Show only this user's posts in this thread
« Reply #444 on: September 25, 2016, 03:01:42 am »

He was up 7 in the LA Times a week ago. 
Logged
Southern Senator North Carolina Yankee
North Carolina Yankee
Moderators
YaBB God
*****
Posts: 44,700
United States


Show only this user's posts in this thread
« Reply #445 on: September 25, 2016, 01:13:44 pm »

So Trump lost ground among those samples that were affected by the birther issue and gained ground among those post NY Bombing?
Logged
michelle
MikeWells12
YaBB God
*****
Posts: 3,989
United States


Political Matrix
E: -4.13, S: 0.35

P

Show only this user's posts in this thread
« Reply #446 on: September 25, 2016, 03:03:25 pm »

Old Formula Trump +17
Trump 53% (+5)
Clinton 36% (-7)

New Formula Trump +13
Trump 51% (+3)
Clinton 38% (-3)

Trump supporters have had a strong increase in enthusiasm (+0.9 to 84.1, Clinton supporters are at 83.3 enthusiasm, down 0.2). Young voters and rich voters are now back to Trump, and uneducated and male voters have swung hard away from Clinton. For some odd reason, black voters are having one of those pro-Trump swings again (due to enthusiasm, as I said before), but this doesn't mean we should discount the current trends.
Logged
Ebsy
YaBB God
*****
Posts: 8,001
United States


Show only this user's posts in this thread
« Reply #447 on: September 26, 2016, 12:11:39 am »

The shifts in the this tracker are almost entirely due to their weird seesawing in black support for Clinton. Really just a terrible poll.
Logged
Seriously?
YaBB God
*****
Posts: 3,029
United States


Show only this user's posts in this thread
« Reply #448 on: September 26, 2016, 02:37:13 am »

USC/LA Times national tracking poll (through 9/25) Trump +3.9%
Trump: 46.3% (-0.1%)
Clinton: 42.4% (+0.1%)
Logged
Hilldog
Full Member
***
Posts: 117
Show only this user's posts in this thread
« Reply #449 on: September 26, 2016, 03:32:14 am »

My question is what demographics is the LA Times using?  Hopefully they're not doing what alot of pollsters did in 2012 and revert back to the 2004 demographics giving Romney a false lead. 
Logged
Pages: 1 ... 13 14 15 16 17 [18] 19 20 21 22 23 ... 30 Print 
« previous next »
Jump to:  


Login with username, password and session length
Logout

Terms of Service - DMCA Agent and Policy - Privacy Policy and Cookies

Powered by SMF 1.1.21 | SMF © 2015, Simple Machines

© Dave Leip's Atlas of U.S. Elections, LLC