Deleted tweet detection is currently running at reduced
capacity due to changes to the Twitter API. Some tweets that have been
deleted by the tweet author may not be labeled as deleted in the PolitiTweet
interface.
Showing page 185 of 729.
Nate Cohn @Nate_Cohn
Biden leads the final Monmouth poll of Florida Biden 51, Trump 45 ('high turnout') Biden 50, Trump 46 ('low turnout') https://t.co/ZjQBH8EBJ8 — PolitiTweet.org
Nate Cohn @Nate_Cohn
I wouldn't do this if we hadn't already collected such a large amount of data. But we have four prior polls of WI, this poll of WI, and we all the other polls we've done this month. We can do something smart and unbiased (or at least as unbiased as other NYT/Siena polls) — PolitiTweet.org
Nate Cohn @Nate_Cohn
Finally, and unlike every other state, we might use self-reported 2016 vote for weighting. The rationale is similar: we don't have party registration, so we're looking for alternatives. If we do, we'd weight to self-reported vote in prior NYT/Siena polls of WI--not actual results — PolitiTweet.org
Nate Cohn @Nate_Cohn
At the same time, we're not going to incorporate this same data on early voting into our turnout model. It's a little weird, but basically we have no equivalent way of boosting the likeliest GOP voters--so letting the Dems bank their vote would create a risk of bias — PolitiTweet.org
Nate Cohn @Nate_Cohn
In other states, we're not incorporating the absentee data into our estimates, since it creates various risks of biasing our turnout. Here, I think the advantage of improving our models outweighs, since we don't have the alternative measures of partisanship we do elsewhere — PolitiTweet.org
Nate Cohn @Nate_Cohn
One different thing in Wisconsin: incorporating early voting into our partisanship models. Unlike other states, WI doesn't have party reg. As a result, we weight on a model of partisanship. These models aren't great! But early voting helps a lot. https://t.co/uqGuNxdwk6 — PolitiTweet.org
Nate Cohn @Nate_Cohn
We've conducted four polls of Wisconsin over the last year, and we can go back and join them up to some of the abse… https://t.co/0C2GVVz8l1
Nate Cohn @Nate_Cohn
A couple of methodological notes on our final polls in AZ, FL, PA, WI, which are just now getting under way --We're targeting some big N here--including 1500 in Pennsylvania --I do think we might do a couple things differently in Wisconsin than we've done elsewhere — PolitiTweet.org
Nate Cohn @Nate_Cohn
RT @jmartNYT: NEW: @tedcruz + @BetoORourke agree on something: TEXAS is in play My report from a trip last week to three crucial regions… — PolitiTweet.org
Nate Cohn @Nate_Cohn
@AaronJonesStats if we have their phone number, i don't see why they wouldn't be — PolitiTweet.org
Nate Cohn @Nate_Cohn
NC party ID is R+3, NC recalled 2016 vote is Trump+6; party reg is D+4 Close NYT/Siena watchers will note our sample sizes have been steadily escalating, and we're up to 1034 here in NC https://t.co/Mra4IAOByY — PolitiTweet.org
Tyson Whelan @tyson_whelan
@Nate_Cohn Recalled vote, party ID etc. pls
Nate Cohn @Nate_Cohn
Today, around 1PM: NYT/Siena in North Carolina Finally, we'll conclude with our last surveys of Arizona, Florida, Pennsylvania and Wisconsin https://t.co/FuxTuaNwLR https://t.co/EazxjTSWQS — PolitiTweet.org
Nate Cohn @Nate_Cohn
RT @bencasselman: U.S. gross domestic product rose 7.4% in the third quarter (33.1% annualized), by far the biggest quarterly gain on recor… — PolitiTweet.org
Nate Cohn @Nate_Cohn
@81_eagles In this case, the problem's the methodology and you can see the symptoms in the crosstabs. But I don't think the problem is the crosstabs, in itself — PolitiTweet.org
Nate Cohn @Nate_Cohn
In this case, the problem's the methodology and you can see the symptoms in the crosstabs. But I don't think the problem is the crosstabs, in itself — PolitiTweet.org
Nate Cohn @Nate_Cohn
Your crosstabs don't look 'right' if you don't weight by education. Your nonwhite voters get more conservative. Your white voters get more liberal. And that shows up in just about every set of Marist tabs, especially nationwide https://t.co/cS4IKD8x6Y — PolitiTweet.org
Dave Goldberg @askaphysicist
Paging @Nate_Cohn , @Redistrict et al. Do any of you find the Marist Florida poll a bit weird? I realize that Biden… https://t.co/42Pjk8iq2j
Nate Cohn @Nate_Cohn
RT @mmurraypolitics: NBC/Marist poll of Florida Biden 51% (+3 from Sept) Trump 47% (-1) Oct 25-27, +/- 4.4% https://t.co/yXpAqSu5lt — PolitiTweet.org
Nate Cohn @Nate_Cohn
We also have a series of Univision/University of Houston/Latino Decisions/North Star polls, showing pretty-much-average results across the board: Biden+3 in FL Biden+5 in PA Biden+5 in AZ Trump+3 in TX — PolitiTweet.org
Nate Cohn @Nate_Cohn
Today, our trusty 'if the polls are wrong' table gets a slight makeover: it shows what would happen if the polls are as wrong as they were over the final week of 2016, rather than the final three weeks https://t.co/6qVQn80pjT https://t.co/q4vs73SyI1 — PolitiTweet.org
Nate Cohn @Nate_Cohn
Another possible factor: voters are ready to make up their minds, just because it's about time--even if they don't know who these folks are. We're seeing a big drop in undecided voters in the NC Senate race as well, which you'll get tomorrow. — PolitiTweet.org
Nate Cohn @Nate_Cohn
One possible factor: partisanshp. Our last poll was R+1 on party ID and Clinton+1 on recalled vote. This poll is D+5 on party ID, and Trump+3 on recalled vote. Seemed to cancel out in presidential race? But maybe that Dem PID edge matters more down ballot — PolitiTweet.org
Nate Cohn @Nate_Cohn
The result in the presidential race is basically the same as our prior result (48-40). The Senate is quite a bit different: we had Peters up just 1, with far more undecided voters. Interestingly, no real change in favorability ratings. Voters still don't know who these guys are — PolitiTweet.org
Nate Cohn @Nate_Cohn
We asked a straight ticket voting question after the horse race questions. It didn't have much of an effect. With straight ticket voting added: Biden 50, Trump 41 Peters 50, James 42 — PolitiTweet.org
Nate Cohn @Nate_Cohn
Joe Biden leads in Michigan, according to the final Times/Siena survey of the state Biden 49, Trump 41 Peters 49, James 41 https://t.co/5UlYPlcqeg — PolitiTweet.org
Nate Cohn @Nate_Cohn
RT @MULawPoll: In new Marquette Law School Poll, Joe Biden is the choice for president of 48% of likely WI voters, Donald Trump is the choi… — PolitiTweet.org
Nate Cohn @Nate_Cohn
Democrats takes the lead in Georgia, according to Monmouth Biden+4, 50-46 (previously Trump+2) Ossoff+2, 49-47 (previously Perdue+5) Warnock well ahead in both runoffs https://t.co/OcWCN7yX2Q — PolitiTweet.org
Nate Cohn @Nate_Cohn
@Redistrict lol no of course not. it's just a secret indulgence — PolitiTweet.org
Nate Cohn @Nate_Cohn
@Redistrict tbh ... i definitely look at our xtabs by district lol — PolitiTweet.org
Nate Cohn @Nate_Cohn
@IChotiner we're bringing back live polling for WA08 to close things out — PolitiTweet.org
Nate Cohn @Nate_Cohn
NYT/Siena in Michigan, 1PM North Carolina is tomorrow https://t.co/FuxTub58ar https://t.co/hBxASuOt4x — PolitiTweet.org
Nate Cohn @Nate_Cohn
No, just presidential https://t.co/yIWTN9Wia3 — PolitiTweet.org
Shaun Eli Breidbart @ShaunEliComedy
@Nate_Cohn Is this about the presidential election or to try to win down-ballot to prevent the next batch of gerrymandering?
