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 13 of 729.
Nate Cohn @Nate_Cohn
The polling error 'warning sign' from September has turned off https://t.co/noNbCNhzou https://t.co/2ZkmPcDUKR — PolitiTweet.org
Nate Cohn @Nate_Cohn
On a personal note, I personally wouldn't play it. If there's not a path to generating *good enough* data to allow for unbiased estimation controlling for stable and knowable characteristics, I'm out — PolitiTweet.org
Nate Cohn @Nate_Cohn
It's worth emphasizing, again, that #1 may help the averages this year. But the idea that this is a game people should play and prove they're right... is stupid — PolitiTweet.org
Nate Cohn @Nate_Cohn
The 'barrier to entry' is low, if you want data that's a combination of 1) broken clocks subject to darwinain electoral selection; 2) people overfitting to their own priors, since everyone knows you can't trust the underlying data — PolitiTweet.org
Nate Cohn @Nate_Cohn
But this next step--the idea that the trash data is fine and anyone can do it, produce good numbers and look smart is just comically out of touch with what's actually happening here — PolitiTweet.org
Nate Cohn @Nate_Cohn
There is an element of the FiveThirtyEight macro view that has been vindicated: a lot of 'gold standard' type polls have reduced output; GOP firms have flooded the zone in some key states. This may improve the averages, even if the underlying challenge remains the same! — PolitiTweet.org
Nate Cohn @Nate_Cohn
The disparity between the FiveThirtyEight macro/empirical view on polling and the reality of conducting high-quality research right now is outright comical at this point https://t.co/jSLpyy0CDN — PolitiTweet.org
Nate Silver @NateSilver538
The barriers to entry right now in polling are fairly low. If someone thinks the averages are wrong because other f… https://t.co/ge4IkVOuj8
Nate Cohn @Nate_Cohn
What do you think the results will show: --Evidence and explanations/possible solutions for D nonresponse bias --Evidence of D nonresponse bias, but no explanation/solution --No major D nonresponse bias/evidence things are better in 202 — PolitiTweet.org
Nate Cohn @Nate_Cohn
@mattyglesias @jbarro @nytimes well, we do have a baghdad bureau and we are paying for this study, so idk what more there is to say — PolitiTweet.org
Nate Cohn @Nate_Cohn
@mattyglesias @jbarro @nytimes idk, is there a financial upside to having a baghdad bureau? — PolitiTweet.org
Nate Cohn @Nate_Cohn
@jbarro @nytimes if it was just the financial incentive, it would be cheaper. but there's a lot of expense on mailing. priority mail envelopes are like $9 bucks each. there's overhead on printing and assembling and mailing and data collection / processing etc — PolitiTweet.org
Nate Cohn @Nate_Cohn
@jbarro @nytimes it does not, at least not yet — PolitiTweet.org
Nate Cohn @Nate_Cohn
TBH you should love the @nytimes for it. It's a lot of money for an experiment -- one that might not even be worth a news alert on your phone. I was very clear that we might not find anything. The folks who count still supported it https://t.co/Tpkj4TOT9M — PolitiTweet.org
Otis Reid @otis_reid
Absolutely love @Nate_Cohn for this one (I can quibble about stuff -- why not do identical sampling frames? -- but… https://t.co/jcGHyDKrnq
Nate Cohn @Nate_Cohn
Previewing our Wisconsin experiment https://t.co/7EOF6Dqlpr — PolitiTweet.org
Nate Cohn @Nate_Cohn
https://t.co/7EOF6DIuDz — PolitiTweet.org
Nate Cohn @Nate_Cohn
@kwcollins @otis_reid yes — PolitiTweet.org
Nate Cohn @Nate_Cohn
@otis_reid (and we can do the voter file match on an ABS sample which, while imperfect, gets us a lot of the benefits) — PolitiTweet.org
Nate Cohn @Nate_Cohn
@otis_reid the voter file here is bad and could be part of the issue — PolitiTweet.org
Nate Cohn @Nate_Cohn
@MattGrossmann i don't think the online panel will involve any recontacting, but that's on ipsos' side / haven't asked — PolitiTweet.org
Nate Cohn @Nate_Cohn
@LPDonovan i do think there's something to the red poll wave in the state polling — PolitiTweet.org
Nate Cohn @Nate_Cohn
Pretty clear split between cheap online panel and everything else on the generic congressional ballot right now, with 'Big Village' and Morning Consult taking on the world https://t.co/WSSvYg0s4T — PolitiTweet.org
Nate Cohn @Nate_Cohn
Pretty clear split between cheap online panel and everything else right now on the generic congressional ballot right now, with Big Village and Morning Consult taking on the world https://t.co/WSSvYg0s4T — PolitiTweet.org
Nate Cohn @Nate_Cohn
RT @MULawPoll: In new Marquette Law School Poll, 50% of likely voters in WI support Republican incumbent Ron Johnson in the Senate race and… — PolitiTweet.org
Nate Cohn @Nate_Cohn
RT @JonHaidt: In real conversations, speakers who frequently spout anger, contempt, insults, and obscenity are shunned. On Twitter, they ar… — PolitiTweet.org
Nate Cohn @Nate_Cohn
In the newsletter this AM, a deep dive on weighting on recalled 2020 vote -- including the results of our recent polls with alternative weights. https://t.co/vfLG2cOZH5 — PolitiTweet.org
Nate Cohn @Nate_Cohn
RT @SeanTrende: More Republican pollsters flooding the zone to show Republicans ahead. https://t.co/hqeKOECroD — PolitiTweet.org
Nate Cohn @Nate_Cohn
@LoganDobson though, right or wrong, we had laxalt ahead despite those issues — PolitiTweet.org
Nate Cohn @Nate_Cohn
@JustSchmeltzer @billscher we ignore that stuff — PolitiTweet.org
Nate Cohn @Nate_Cohn
@SeanTrende i'm not even sure it's a lie -- they don't want to do it! — PolitiTweet.org
Nate Cohn @Nate_Cohn
@SeanTrende based on personal '16 experience, i also think there's another layer -- not wanting to admit to it yourself bias — PolitiTweet.org