Welcome, Guest. Please login or register.
Did you miss your activation email?
March 21, 2019, 03:34:46 pm
HomePredMockPollEVCalcAFEWIKIHelpLogin Register
News: Please delete your old personal messages.

+  Atlas Forum
|-+  Election Archive
| |-+  Election Archive
| | |-+  All Archived Boards
| | | |-+  2016 U.S. Presidential General Election Polls (Moderators: AndrewTX, Likely Voter)
| | | | |-+  YouGov MI/CO/NH/WI/NC/FL/PA/NV/IA/OH/GA: Clinton up, tied in GA (search mode)
Pages: [1] Print
Author Topic: YouGov MI/CO/NH/WI/NC/FL/PA/NV/IA/OH/GA: Clinton up, tied in GA  (Read 8836 times)
HillOfANight
YaBB God
*****
Posts: 1,474
View Profile
« on: October 05, 2016, 07:15:30 am »

https://twitter.com/YouGovUS/status/783640247070560256
https://today.yougov.com/us-election/
https://twitter.com/williamjordann/status/783641151920300032
https://today.yougov.com/news/2016/10/03/election-model-methodology/

It's a mix of polls and "analysis"

Clinton’s lead:
National +4

MI +5
CO +5
NH +4
WI +3
NC +2
FL +2
PA +2
NV +2
IA +1
OH +1
GA TIE
« Last Edit: October 05, 2016, 08:09:55 am by HillOfANight »Logged
HillOfANight
YaBB God
*****
Posts: 1,474
View Profile
« Reply #1 on: October 05, 2016, 07:28:37 am »

Isn't this a 50-state poll? At least it's more realistic than anything Google or Reuters has released.

I'm not sure what this is... it seems a bit like a 50 state poll, but releasing the data in dribs and drabs

https://twitter.com/williamjordann/status/783643089214767106

Quote
different from a poll – based on interviews combined with analysis of voter types.

Whatever that means... They have a detailed methodology I haven't read yet.
Logged
HillOfANight
YaBB God
*****
Posts: 1,474
View Profile
« Reply #2 on: October 05, 2016, 08:12:23 am »

https://today.yougov.com/news/2016/10/03/election-model-methodology/

They predicted 4% in 2012 which was pretty spot on, and their September 2012 data only missed NC/NH. Their October 2012 data only missed NC.

Quote
The raw data for our analysis is about 50,000 respondents to YouGov’s polling over the past two weeks, however turning that into state-level estimates involves several other sources of data to ensure that we are generating a representative portrait of the electorate, not just of those who respond to our polls.

The approach we are following, which is referred to as multilevel regression and post-stratification (MRP), has three components.
...
So one type of (potential) voter in the election might be a female, age over 85, with post-graduate education, living in the 9th Congressional district of Massachusetts, who voted for Obama in 2012.  Change any of those characteristics, and you have another type.  For each of these types, there are three important quantities that we would like to know.
...
1) What proportion of people of that type will vote for Clinton, Trump, Johnson, Stein, etc, among those who do vote?

2) What proportion of the individuals in each voter type will turn out to vote?

3) How many (voting eligible) individuals are there of that type?
Logged
HillOfANight
YaBB God
*****
Posts: 1,474
View Profile
« Reply #3 on: October 16, 2016, 12:58:42 pm »

https://twitter.com/nickgourevitch/status/787713135809552384

Unlike a lot of public pollsters, YouGov keeps partisan metrics fairly stable. HRC +6.1 in their model -- the highest margin they've shown. Her median electoral vote total is 362.
Logged
HillOfANight
YaBB God
*****
Posts: 1,474
View Profile
« Reply #4 on: October 17, 2016, 08:25:46 am »

I don't think this is an actual Georgia specific poll, but part of their multilevel regression and post-stratification (MRP) project. They take lots of responses, not necessarily Georgia specific, and then model those responses back to Georgia, based on similar profiles.
https://today.yougov.com/us-election/?state=Georgia

See the 47-44 result here.
https://uselectionatlas.org/FORUM/index.php?topic=247418.msg5321100#msg5321100
Logged
Pages: [1] Print 
Jump to:  


Login with username, password and session length
Logout

Terms of Service

Powered by SMF 1.1.21 | SMF © 2015, Simple Machines