Laura Bronner
@laurabronner
Followers
9K
Following
4K
Media
737
Statuses
2K
Scientific Director, Public Discourse Foundation & Senior Applied Scientist, @ETH | past: Quant Editor, @FiveThirtyEight | PhD, @LSEGovernment | she/her
Vienna / Zurich / Berlin
Joined August 2015
For anyone who works with data, or relies on people who work with data: Here's a quick guide on "quant editing" as developed at @FiveThirtyEight, essentially just a list of some of the things you can (and probably should!) be asking yourself and others.
7
55
321
Interested in how people talk to each other online? Care about causal inference and/or NLP? Want to design and implement field experiments? . Come do a PhD with Dominik Hangartner, me and a bunch of awesome people at IPL in Zurich: .
0
0
8
This election is uncertain not just because the polls are close, but also because a decisive victory for either candidate is within a normal-sized polling error - and since the polls seem to be herding, a polling error may be likelier than usual. Great piece by @lennybronner:.
0
1
9
Anhand aktueller Umfragen und historischer Abweichungen haben wir simuliert, wie die Wahlen ausgehen könnten. Da gleich mehrere Parteien diesmal an der 5%-Hürde scheitern könnten, ist die Unsicherheit besonders hoch. Mit @christianbangel, @c_endt, @colorfuldata und David Schach.
0
1
3
Welche Koalitionen könnten nach den Wahlen in Sachsen und Thüringen anstehen?. Zusammengefasst: in Sachsen könnte es sehr schwer werden, eine Regierung ohne BSW zu bilden - und in Thüringen beinahe unmöglich. Auf @zeitonline mit simulierten Wahlausgängen:
4
10
20
RT @c_endt: Nach den Landtagswahlen in Sachsen und Thüringen wird die Regierungsbildung extrem kompliziert. Eine Zusammenarbeit von CDU und….
0
14
0
RT @lennybronner: Our polling average topline takeaway: Trump is leading in 5 of the 7 battleground states that are most likely to determin….
0
9
0
Working with any kind of user-generated data.
0
3
38
2. So instead of looking at hypotheticals in past polls, look at the polls coming up. But keep in mind: high-profile news can lead to partisan non-response bias - people unhappy with the news may avoid responding to polls for a bit, possibly leading to a brief pro-Biden bias.
0
2
7
A couple notes on polling after Trump's conviction:.1. Some earlier polls asked people how a conviction would change their vote, finding that it would help Biden. Important to keep in mind, though: people aren't great at evaluating how something would change their attitudes.
1
3
13
Wie verbessert man den Onlinediskurs? .Heute um 16:15 bei der @republica #rp24 : unser Panel zu Stop Hate Speech, mit @HateAid und Public Discourse Foundation. Schon vorbei: wie wir zusammen mit @Blickch Email Feedback eingesetzt haben, um Hasskommentare zu verringern.
0
3
12
The Eurovision/UK parliament crossover content🧵I didn't know I needed:.
The only serving member of either House of Parliament to have performed at the #Eurovision Song Contest was Lord Lloyd-Webber (@OfficialALW), who accompanied Jade Ewen on stage at the 2009 contest:
0
0
7
Until it becomes so ubiquitous that we all just write like ChatGPT.
Pro tip: If you've used "delve into" and "intricate" in the same sentence, your reader knows with 100% it's ChatGPT.
1
0
10
Realizing that the additional skepticism that used to be necessary on the April Fools' Day internet is now just. normal. The internet in 2024: All fools, all the time.
0
1
14
RT @lennybronner: Super cool thread on racial depolarization (what I'd call it, rather than realignment) in US politics. A bit of additiona….
0
3
0
