Econ AP @UofT, Y Combinator Research PI, Social Science Prediction Platform PI, AidGrade founder. Likes effective altruism & math. Ex-World Bank, Berkeley grad.

San Francisco, CA
Joined March 2012
Eva Vivalt retweeted
The Toronto Data Workshop on Reproducibility will be held online on 25/26 February 2021 (two full days!). Free. All welcome. Register: forms.gle/LKhHvDPpAC9auoox6 Details: rohanalexander.com/reproduci… Keynotes: @evavivalt (practices) @minebocek (teaching) @rianaminocher (evaluating).
1
68
111
Show this thread
Eva Vivalt retweeted
Now at #bitssAM9: @sdellavi, @evavivalt, @arunadvaniecon, and Nick Otis reflect on the early growth of the @socscipredict Platform -- 2,275 users, 13 projects, 8,225 #forecasts!
0
6
12
Join Nick Otis, @evavivalt, @arunadvaniecon, and @sdellavi 11:25 am-12:25 pm PT tomorrow, Jan. 7, at the @UCBITSS Annual Meeting, for a discussion of the #SocSciPrediction platform and results from a couple of the early SSPP projects. Registration link below.
Join us for the 9th BITSS Annual Meeting, which will be hosted online on Jan. 7-8! #BITSS2020 #OpenScience See the agenda + RSVP for a Zoom link: bitss.org/events/9thbitssam A few highlights from the agenda👇
Show this thread
0
4
3
If the new strain does spread more easily (higher R0), a lot more people will need to be vaccinated to reach herd immunity. Perhaps pushes out a return to normal by some months.
1
0
5
Department holiday party on Zoom had a game, organized by the grad students, where you saw photos of faculty members from early childhood and had to guess who was who. My pic:
1
0
71
This was, btw, a really fantastic idea for a holiday party, fully endorse.
0
0
0
If the study was partially unblinded due to side effects and the treatment group guessed it worked and took more risks, the efficacy could be slightly underestimated. ("Severe side effects included fatigue in 9.7% of participants, muscle pain in 8.9%...") sciencemag.org/news/2020/11/…
3
1
6
(Unless it's 100% effective among those who have strong side effects, in which case it doesn't matter if they change their behaviour. Seems plausible that side effects would be correlated with efficacy though I haven't read anything about it.)
0
0
0
Perfect response! :)
Replying to @JohnHolbein1
show them using data from socialscienceprediction.org/ that what is surprising is subjective/in the eye of the beholder -- cc: @evavivalt @sdellavi
0
0
10