Welcome to DU! The truly grassroots left-of-center political community where regular people, not algorithms, drive the discussions and set the standards. Join the community: Create a free account Support DU (and get rid of ads!): Become a Star Member Latest Breaking News General Discussion The DU Lounge All Forums Issue Forums Culture Forums Alliance Forums Region Forums Support Forums Help & Search
6 replies = new reply since forum marked as read
Highlight: NoneDon't highlight anything 5 newestHighlight 5 most recent replies
The USC/LA Times National Panel finally has Drumpf and Hill tied (Original Post) titaniumsalute Oct 2016 OP
This LA TImes poll has been so far off from all other polls, how does it get an "A" from 538? BlueStreak Oct 2016 #1
The problem is it is a panel type poll titaniumsalute Oct 2016 #3
I didn't know that. That explains a lot. Thanks. BlueStreak Oct 2016 #4
Is this specific poll getting the A? Renew Deal Oct 2016 #5
AFAIK, 538 rates pollsters, not individual polls BlueStreak Oct 2016 #6
Funny Freepr article. Johnny2X2X Oct 2016 #2
 

BlueStreak

(8,377 posts)
1. This LA TImes poll has been so far off from all other polls, how does it get an "A" from 538?
Wed Oct 12, 2016, 08:49 AM
Oct 2016

I understand that consistency is something, but being consistently off by 5% doesn't sound like a poll that should be given much credence.

titaniumsalute

(4,742 posts)
3. The problem is it is a panel type poll
Wed Oct 12, 2016, 10:12 AM
Oct 2016

Same people are polled. Depending on the feelings of people when the panel was created it doesn't change much. (Who really has been changing their minds on who they are going to vote for in the election?)

 

BlueStreak

(8,377 posts)
4. I didn't know that. That explains a lot. Thanks.
Wed Oct 12, 2016, 10:28 AM
Oct 2016

We see variations of that. NPR keeps going back to this same set of Ohio voters who claim to be "undecided" But they are obviously very entrenched because while the Ohio numbers have moved about 10% in the past 30 days, the Trump supporters that NPR interviewed haven't given an inch.

This becomes the most worthless, disgusting false equivalence king of reporting. Pick an equal number of people on each side and give them all equal time, ending up with "everything is equal". A complete waste of airtime, and an insult to anybody who tuned in expecting to learn anything relevant.

Considering that the LA poll is a fixed panel, then any movement is extremely significant.

 

BlueStreak

(8,377 posts)
6. AFAIK, 538 rates pollsters, not individual polls
Wed Oct 12, 2016, 10:36 AM
Oct 2016

I believe the "A" goes to LA Times. 538 certainly categorizes the results of each poll, and that factors into their algorithm. For example, they treat registered voters different from likely voters, and give weight to larger sample sizes.

Johnny2X2X

(19,139 posts)
2. Funny Freepr article.
Wed Oct 12, 2016, 08:52 AM
Oct 2016

Over the weekend this poll went from like Trump +3.8 to Trump 3.4 and then back to Trump +3.7, they did a huge article about it saying how,"the Tape made Trump did a little, but the debate he gained it all back." A total ignorance of how polls work and what noise is. Now thy it's tied in sure they're ignoring it and going back to their online Breitbart poll that has Trump +30.

Latest Discussions»Retired Forums»2016 Postmortem»The USC/LA Times National...