2016 Postmortem
Related: About this forumClinton Opens Up Huge Lead in Michigan Clinton 66% (+11) Sanders 29% (-8)
EAST LANSING, Michigan --- In a poll conducted Sunday afternoon and evening before the start
of their CNN debate in Flint, former Secretary of State Hillary Clinton has opened up a huge lead
over Vermont U.S. Senator Bernie Sanders among Democratic Party Presidential Primary voters
in Michigan. Clinton leads Sanders 66%-29%, up from 55%-37% on Thursday night. Only 6% of
the voters remain undecided.
http://www.realclearpolitics.com/docs/2016/Mitchell-FOX_2_Detroit_MI_Poll_DEM_Primary_3-6-16.pdf
CoffeeCat
(24,411 posts)Clinton and Sanders have remarked that the race has tightened significantly in Michigan.
I think Clinton and Sanders would both agree that this poll is a bit goofy.
A robo poll?
Who in the hell doesn't hang up on a robo poll? That has to be one nutty demographic!
DemocratSinceBirth
(99,711 posts)CoffeeCat
(24,411 posts)shows the race within 11 points.
That poll is in line with what the Clinton/Sanders campaigns are both saying about the race tightening and being close.
The RCP average included the crazy outlier robopoll. LOL!
DemocratSinceBirth
(99,711 posts)They missed South Carolina by twenty nine points!!!
I bet you think I made that up.
Well I didn't:
http://www.realclearpolitics.com/epolls/2016/president/sc/south_carolina_democratic_presidential_primary-4167.html#polls
Twenty nine points!!!
As someone who has actually worked with survey data I don't know why CBS attaches their good name to such flawed research.
BTW, I don't hold the CBSYOUGOV national polls in the same low esteem as their state polls because the methodology of the former is more rigid.
CoffeeCat
(24,411 posts)as evidence that Clinton is creaming Bernie. Yet, you are now telling me that the CBS poll--which is included in the RCP average--is unreliable.
Polls...are crazy. And that robopoll is laughable. And it's included in the RCP average. Do they allow any poll into their average?
It's nutty!
I think I'll just rely on what both camps are saying--that the race has tightened significantly. Nothing more reliable than internal polling, right?
DemocratSinceBirth
(99,711 posts)The Law Of Large Numbers suggests the more polls you include the better, because presumably the "good" ones will cancel out the "bad" ones, and we will get a truer picture of what is really occurring.
Cui bono... All campaigns spin, even the ones of the candidates I like. I consequently ignore most of what they say
P.S I will concede your point that robo calls are not the most robust pollsters but it is a data point, nothing more and nothing less, and some robo callers do a fairly decent job. If you are really concerned about methodology please check the yougov Battleground poll methodology. It's crazy!!!
CoffeeCat
(24,411 posts)in the beginning of the primary. I didn't know that they missed the mark that bad on SC.
If I really want to know what is going on in a state, I look at RCP and read the methodology of the 2-3 most recent polls.
Wouldn't you agree that the latest poll with the soundest methodology is going to give you the best indication (or even prediction) of the final outcome? States that have impending primaries are moving targets. The numbers are changing so quickly.
A 5-day old poll can be stale and outdated--as voters firm up their choices and fall into their candidate camps at a furious pace--as they prepare to vote/caucus.
It's too bad that we can't have a "gold standard" poll doing frequent and consistent polling throughout the campaign.
Too bad Ann Selzer doesn't do more polling. I'm kind of tired of sifting through the junk.
DemocratSinceBirth
(99,711 posts)Marist and MRG have the soundest methodologies. They both call landline and cell phone users. Nate Silver gives the former an A-.
Here is the CBS/YOU GOV Battleground Poll methodology. Their methodology should give anybody pause:
...
Respondents were selected from YouGovs and two other online panels. These are opt-in panels which are open for anyone to join. However, YouGov also randomly selected persons from voter registration lists who had previously voted in primary elections and contacted them by phone. A total of 22,517 registered voters were contacted by phone and the YouGov sample includes 1,753 phone recruits.
https://today.yougov.com/news/2015/09/13/methodology-2016-cbs-news-battleground-tracker/
I don't have a problem with online polls if they are properly weighted and designed to approximate a random sample. My problem with the YOUGOV Battle Tracker Poll is it combines an online poll with a traditional phone poll, and interviews the same people more than once to compound the problem. They are introducing more complexity than what is necessary and allowing error to seep in.
leftynyc
(26,060 posts)That's a matter of opinion (they blew SC HUGE). And certainly not the latest poll (CBS was 3/2-3/4):
Fox Poll from 3/6 has Hillary up +37
ARG from 3/4-3/5 has Hillary up +24
MoonRiver
(36,926 posts)DemocratSinceBirth
(99,711 posts)livetohike
(22,163 posts)NurseJackie
(42,862 posts)I mean, I'm thrilled for Hillary and everything, but this is TOTALLY DEVASTATING news for Bernie's campaign. I don't believe he can recover after last night's debate.
Go, Hillary! We love you!
Attorney in Texas
(3,373 posts)Bluenorthwest
(45,319 posts)nt
Lizzie Poppet
(10,164 posts)Bluenorthwest
(45,319 posts)that really builds a reputation!
Hubris. It's what's for dinner!