Welcome, Guest. Please login or register.
Did you miss your activation email?
June 16, 2019, 11:56:43 pm
News: Please delete your old personal messages.

  Atlas Forum
  Election Archive
  Election Archive
  2016 U.S. Presidential General Election Polls (Moderators: AndrewTX, Likely Voter)
  Morning Consult polls every state (search mode)
Pages: [1] Print
Author Topic: Morning Consult polls every state  (Read 3630 times)
cinyc
YaBB God
*****
Posts: 11,701


« on: July 14, 2016, 02:49:23 pm »

Apparently this mds loser is adding every single poll from this to the database. Sad!

How is that sad? Please you guys add every Democratic pollster you can on there.

I would be fine with these polls being entered -- and grateful that some states are finally being polled -- if there were any evidence that those states were actually polled.

From Morning Consult's methodology: "Morning Consult asked more than 44,000 registered voters nationally over the past three months who they would support in a series of hypothetical general election matchups. We used a statistical technique called multilevel regression and poststratification
(MRP) to construct state-level estimates from the national survey data."

estimates from the national survey data is the operative phrase there. It's an election model from national polling. Not state polling.

Where are you getting the number of respondents for these "polls" you're entering? Are you just making them up?

I don't think these are polls in the tradition sense, either.  Morning Consult is using regression on a large, multi-state poll response database to try to model what the results might be in each state.   The variables they chose include state age, education and income along with state-specfic economic variables, like unemployment.  Morning Consult claims this is more accurate than actual state-based polling, based on a 2012 study, but it is not polling in the traditional sense.  It might be the wave of the future, though.

Back in April, Morning Consult also separated state-specific interviews and stratified the data to create something that would approximate a state poll with special sauce, albeit one taken from January to April.   But doesn't appear that they did that here.  The April state special sauce poll results differed from the MRP estimates in some states.

I'm not sure that we should be including these "polls" in the database, but that's not up to me.
Logged
cinyc
YaBB God
*****
Posts: 11,701


« Reply #1 on: July 14, 2016, 02:52:58 pm »

Test
Logged
cinyc
YaBB God
*****
Posts: 11,701


« Reply #2 on: July 14, 2016, 03:05:09 pm »

I think this thread is broken.  It ate two responses I had, explaining why these Morning Consult "polls" aren't polls in the traditional sense.  It's multivariable regression, using nationwide responses from January to June, then controlling for state demographic factors like age and income, and economic factors like unemployment.  Morning Consult claims, based on one study, that this is more accurate than actual state-by-state polling - but it is not state-by-state polling. 

In fact, when Morning Consult did this exercise in April, they actually also reported results of in-state only respondents, stratified by some special sauce - and those results differed in some states.
Logged
cinyc
YaBB God
*****
Posts: 11,701


« Reply #3 on: July 14, 2016, 06:29:27 pm »

Some responses, including mine, aren't publishing on the first page of this thread.  I'll try this again for the third time.  If the responses on the prior page start showing up, please forgive me:

These Morning Consult state "polls" aren't polls in the traditional sense.  As I understand it, Morning Consult took subsamples of their national January-June nationwide poll responses and used multiple regression to reweight them for state-specific demographic factors, like age and income, and economic data, like unemployment rate, to come up with a figure for the Trump, Clinton and Other percentage in each of the 50 states.  This MRP method is allegedly more accurate than actual state polls according to one study they cite - but is not a traditional poll.  Perhaps this is the wave of the future, but again, it's not a traditional poll.

Back in April, they not only did MRP regression, but also took subsamples of respondents from January to April in each state, weighed them for demographic factors, and put out statewide special-sauce poll numbers for those states.  The results of these special-sauce state polls differed from the MRP results in some states.  It's not clear that Morning Consult did that this time.

Given that, I don't think these "polls" should be entered into the database - but that decision is not made by me.
Logged
Pages: [1] Print 
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