Profile picture
, 3 tweets, 1 min read
My Authors
Read all threads
The composition of the Selzer/DMR/CNN poll is a mystery, but methodologically it seems pretty similar to ours. The frame is the full Iowa voter file, with a two-party likely voter screen
One difference is that the Selzer/DMR/CNN poll appears to exclude all unlikely voters, while we give anyone that passes the first screen some chance of voting.
This doesn't make a huge difference. If we had done so, Biden would have lost a point and
Buttigieg would have gained 1
Another difference is that the Selzer/CNN/DMR poll doesn't weight by education. If we drop the education weight as well and keep the tighter screen, we go to Sanders 23, Buttigieg 19, Biden 16, Warren 16, which is within 3 points of their result on each
Missing some Tweet in this thread? You can try to force a refresh.

Enjoying this thread?

Keep Current with Nate Cohn

Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

Twitter may remove this content at anytime, convert it as a PDF, save and print for later use!

Try unrolling a thread yourself!

how to unroll video

1) Follow Thread Reader App on Twitter so you can easily mention us!

2) Go to a Twitter thread (series of Tweets by the same owner) and mention us with a keyword "unroll" @threadreaderapp unroll

You can practice here first or read more on our help page!

More from @Nate_Cohn see all

Profile picture
Nate Cohn
@Nate_Cohn
Without wading into this specific debate, the feud over MRP gets at a pretty fundamental question about survey research right now:
do polls work because they have a design that yields a representative sample, or do polls work because they have a good model to correct error?
The design-based view is the classical view. Here, a probability sample = a good poll. If everyone has a known chance of selection, then the results should fall within a theoretical margin of error due to sampling
The model-based view says design doesn't do it anymore; that's why they all use extensive weighting.
Instead, polls work because of the model: we basically just make inferences about the population based on an unrepresentative sample by controlling for demog. used for weighting
Read 11 tweets
Profile picture
Nate Cohn
@Nate_Cohn
Happy caucus day, and welcome to the 2020 Democratic presidential nominating season
There will be four tonight nytimes.com/2020/02/03/ups…
As an aside, the NYT--and most news orgs--will project a winner based on the state delegate equivalency results. Doesn't mean the other metrics will be ignored, and all will be available, but it'll be SDEs for the purpose of the coveted projection 'check-mark'
Read 4 tweets
Profile picture
Nate Cohn
@Nate_Cohn
Most of you know that there are *four* different sets of results this year.
They could all split.
nytimes.com/2020/02/03/ups…
Some of the splits are obvious: second choices could break one way, shifting the result from first to final alignment.
And the tabulated vote counts and pledged delegates have split before (NV '08, for ex.) and could again under some fairly unlikely and arcane circumstances
The split between the final alignment and state delegate equivalency, however, is unique to Iowa and worth some discussion.
Read 10 tweets

Related threads

Profile picture
Yossi Gestetner
@YossiGestetner
CNN DID NOT KNOW THIS THE PAST 60 MINUTES???
DMR DID NOT KNOW THAT THEY WILL DO THIS? Why did they wait until now to announce? To get clicks clicks clikcs the past few hours/days?

Degenerate hacks.
The TOP news item on Politico (and many websites) this eve was the pending DMR/CNN poll. This gave DMR (and CNN) free PR, clicks and views on a thing that they all knew the past how long (hour? Two hours? Half day?) will not be released.

What a FARCE. Took everyone on a ride!
Read 7 tweets
Profile picture
Nate Silver
@NateSilver538
So in terms of Iowa polls it looks like:

* Selzer/DMR/CNN at 9pm-ish tonight
* YouGov/CBS, probably tomorrow morning
* Emerson College, tomorrow night

We've heard reports from readers that make us wonder if there's a 4th poll in the field, but may have been a false positive.
I'd keep in mind that there's actually been quite a bit of Iowa polling, much of it from high-quality firms. So these polls will matter in our forecasts/average, but not as much as if there had been a total dearth of polling.

The most consequential thing would be if... (cont.)
...there's a clear pattern of someone rising or falling across the final ~2 weeks. Our model has been pretty conservative about Klobuchar so far but she's probably closest to showing this pattern and a highly-rated poll showing her gaining could make it more aggressive.
Read 4 tweets
Profile picture
Nate Silver
@NateSilver538
This thread on the Selzer/DMR/CNN poll is worth reading but one further meta-thought:

This is the rare instance (usually I think this theory is overrated/under-evidenced) where a poll itself could have an impact on the race by influencing media and voter behavior.
For voters—with several candidates sitting near the 15% threshold and/or needing a good Iowa night to be viable going forward—there's the chance of tactical voting if someone seem to have momentum. If it's a good poll for Amy K. for instance, she could gain at Pete B.'s expense.
For the media, it will heavily affect expectations. And the funny thing is that if you're a frontrunner (Bernie, Biden) you would rather expectations be *lower*. You tend to get a bigger bounce that way if you win, and a smaller "anti-bounce" if you *don't* win.
Read 5 tweets
Profile picture
Steve Herman
@W7VOA
"I worked so hard for this state," says @POTUS, opening political rally in #Iowa.
@POTUS "It's going to be a historic landslide," predicts @POTUS at @DrakeUniversity rally. "If we don't win your farms are going to hell."
@POTUS @DrakeUniversity After disparaging @ewarren, @SenSanders and @JoeBiden, @POTUS remarks: "Boot-edge-edge...presides over a city that failed,.

"What happened to mini-Mike? Where's he?"
Read 22 tweets
Profile picture
postcardsforAmerica
@postcards4USA
Hey #Iowa! You get to vote in Primary Election on JUNE 5, 2018
Register by May 25: sos.iowa.gov/elections/pdf/…
And here’s a THREAD of All your Democratic Candidates with photos & website links.
Also info & links for Registration, Absentee & Poll Worker
#IA01 #IA02 #IA03 #IA04
REGISTER TO VOTE in Iowa
Voter Registration by May 25
Form: sos.iowa.gov/elections/pdf/…
Mail or Deliver to County Auditor. Addresses here: sos.iowa.gov/elections/audi…
Or Iowa DMV office. Addresses here: dmv.org/ia-iowa/dmv-of…
Register Online through DMV here: mymvd.iowadot.gov/Account/Login?…
Democratic Candidates #IA01 Iowa
Abby Finkenauer
abbyfinkenauer.com
@Abby4Iowa
Thomas Heckroth
thomasheckroth.com
@HeckrothForIowa
George Ramsey
ramseyforuscongress.org
@RamseyforIowa
Courtney Rowe
courtneyroweforcongress.org
@RoweCourtneylr7

vs GOP Incumbent Rod Blum (R)
Read 12 tweets
Profile picture
postcardsforAmerica
@postcards4USA
Be a GUARDIAN OF DEMOCRACY!
Be a POLL WORKER!
As well as voting & registering others, becoming a Poll Worker (or Poll Watcher) is one of the best ways to help Democracy & ensure fair elections in your State
THREAD of
State-by-State Requirements for ALL STATES & links to sign up
Each state has a different set of laws and rules governing the selection of #PollWorkers #PollWatchers

POLL WORKERS
Clerks, Judges, Inspectors, Checker, Moderator, Selectman, Precinct Worker
Minority Inspector, Majority Inspector

POLL WATCHERS
Observers, Monitors, Challengers
Following are the requirements
to be a Poll Worker
or Poll Watcher
in your State
with Links to where to sign up.
THREAD

Tip: Look at the Replies to your State Tweet to see further Info & Links for your State.

Here's a Medium article with this info: medium.com/@postcards4USA…
Read 148 tweets

Trending hashtags

#SoS #CommonCore #Indiana #Economy #Kentucky #WhiteHouse #Iowa #TomMcAlpin #Nevada #iacaucs #PollWatcher #FamilyFued #PrimaryElection #ImpeachmentTrial #BlackstoneGroup #RTL #VoterRegistration #caucus #AG #Wehrmacht #IA01 #Poll #polls #Election2020 #ILLINOIS
Follow Us on Twitter!

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just three indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3.00/month or $30.00/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!