Absolutely thrilled to be joining the University of Chicago as an Assistant Professor in 2021! Thanks to all who have supported me in making this possible.
Incredibly honored to join the 2024
#CarnegieFellows
!
The fellowship will support my research on how people update their views and attitudes in response to the conversations they take part in and the information they engage with online.
@CarnegieCorp
My dad, Peter Westort, passed away on New Years Eve, at home, surrounded by family. He had been diagnosed with stage IV pancreatic cancer four weeks earlier. I am heartbroken to have lost him, but so grateful for the time I got to spend with him over the last several weeks.
New paper on adaptive experiments out with
@aecoppock
and Don Green in
@AJPS_Editor
! We try to tackle settings where you have multiple treatments, care most about estimating the treatment effect btwn the best performing arm and control, but don't know ex ante which arm is best.
The
#PrincetonU
Board of Trustees has voted to remove Woodrow Wilson’s name from the School of Public Policy and International Affairs and residential college.
Can we come up with an alternative to grad student poster sessions at conferences--one which isn't inherently regressive? Grad students often have no/limited conference funds, so, in addition to travel, fees etc., they have to pay another ~$100 out of pocket.
So cool to see an experimental design course with undergrads use adaptive experiments! Interesting findings, too. Just a great way to have students get involved with research design.
In my experimental design course, we developed messages to improve scientific trust and implemented an adaptive experimental design to identify the most promising messages. What works? Neutral videos that emphasize the everyday benefits of science.
Excited to see this project launch! It has taken a LOT of work behind the scenes, but our bot is now live, chatting with users in Kenya/Nigeria to collect survey responses and adaptively learn which interventions are most effective at curbing the spread of misinformation online.
@Susan_Athey
@mofferw
and I currently using
#facebook
#chatbot
to deliver adaptive experiment testing interventions to curb spread of COVID misinfo [PAP here: ] Results coming soon....
@paulgp
@instrumenthull
Maaaaybe, from Tukey, on the computational revolution for statistics: "The tool that is so dull that you cannot cut yourself on it is not likely to be sharp enough to be either useful or helpful."
My alma mater
@GrinnellCollege
is engaging in really shameful union busting. I worked in the d hall all 4yrs; collective bargaining would not "impede learning & diminish educational opportunities" available to me. How cynical can you get?
#grinnellvalues
We are hiring at
@UChicagoPoliSci
! This is a senior search, with an emphasis on quant methods, formal theory, and political economy. Please get in touch with questions.
EOE/Vet/Disability
I am so saddened to hear of Nuno Monteiro’s passing yesterday. He was always kind to me as a grad student at Yale, and I know he will be remembered warmly by friends and colleagues there and here at Chicago. My thoughts are with his family.
Apply for this postdoc at UChicago and come work with me! (and/or any of the other faculty affiliated with
@UChicagoCEG
and
@DSI_UChicago
)
This is a great program with a lot of support.
More info here:
We're hiring! Apply to the UChicago Data & Democracy Postdoctoral Fellowship—a data-focused, interdisciplinary postdoc with competitive salary & incredible research opportunities w/
@UChicagoCEG
ad
@DSI_UChicago
Apply today:
@betsylevyp
@reblitt
Joining the political science dept at
@UChicago
as an asst prof in 2021! Very excited for this opportunity to critique Chicago style pizza.
I think this is such a good idea—as a student, solving the problems with the solutions in the back of the book was a really useful exercise. Checking in on how students are doing with the method, rather than the answer, seems like more useful info for instructors, too.
@adam_thal
lol Adam I want to write the inverse of this textbook—all math, but accessible to *non* math/stats majors and covering the components of probability, estimation, and inference that are relevant for doing social science, with minimal assumptions.
Thanks to the quick work of colleagues, here is an open letter from faculty against the
#StudentBan
. Please share widely! In it we call on ICE to rescind its decision, and on our university leaders to join us in advocating for all of our students.
@JaeJaeykim2
Our pre-registration on our contextual adaptive experiment on OSF w/
@LeahRosenzweig
and
@Susan_Athey
has information on our adaptive algorithm, how we used simulations to select algorithm hyperparameters, and how we would learn policies from our data
@aeggers
I don't disagree with the first part, but I think maybe the solution would be to spend more time on digging into this fundamental confusion wrt probability, and what we/our students are trying to get out of it.
"Many of my colleagues express surprise upon learning that my research focuses on violence in armed conflict...It’s as if they believe that just because I’m a black woman, I need to filter everything through the lens of my identity."
Excited to be chairing two very compelling panels at
#MPSA2021
tomorrow 4/18! At 10.40 CT, New Insights in Causal Inference, with papers from
@Gusvalo
, Xiao Lu, Idriss Bennis/Alexis J Diamond/Mikulas Plesak, and
@xuyiqing
/
@EddieYang_
New piece up by
@petertinti
: "As the U.S. and Europe systematically build barriers to seeking asylum, a broader question is emerging: Is the right to seek asylum disappearing?"
.
@DSI_UChicago
hosts this amazing two-day workshop for early career data scientists (late stage grad students, post docs, and research scholars). Really a fantastic opportunity to present your work and meet a lot of great people in the field. Deadline to apply is September 30!
Apply to the UChicago Rising Stars in Data Science workshop! This is a great opportunity to give a talk, network with faculty, and learn about post-grad career opportunities:
Coming up Saturday morning at 8am PT/11am CT, our Experiments in Methodology panel. Looking forward to discussing these great papers. The session is listed as in-person, but we'll be ~virtual~, check the APSA platform or DM for the Zoom link.
Saturday at 8am PDT/11am EDT "Experiments in Methodology" with
@ScottClif
Geoffrey Shelley Kevin Easterling David Brady and the great
@mofferw
@tara_slough
@Matt__Graham
I’ve had good success with having students “spin” reports—same benefits of documented reports, but you can do it on a basic .R script with no formatting.
"It’s already a hard sell to get talented high school students...to move to a tiny town they’ve never heard of...That gets significantly harder if you’re trying to pitch your school’s history of social activism while you’re making national headlines for crushing unions."
.
@GrinnellCollege
is heading down a dangerous path by fighting their students workers union in this manner. Does a university with a very liberal student population really want to be known for causing a national union-busting decision?
1 year since the chronicle article about Jorge Dominguez. 18 separate accusations of sexual harassment or assault. 40 years of institutional failures. 124 government graduate students calling for an external review. 0 accountability.
#timesupharvard
#externalreviewnow
#TimesUp
New NBER paper summarizing what we know about UBI in developing countries from Banerjee, Niehaus, & Suri. Basically, evidence from *targeted* transfers has lots of positives...but it's hard to know what things would look like w/ truly universal transfers.
We’re hiring! Postdoc opening to work on projects revolving around ed tech in the GC Lab at Stanford Graduate School of Business with faculty director Susan Athey. Exp w empirical projects and experimentation needed. More details:
@Apoorva__Lal
Keen to try this once I update to MacOS 12. The automatic changes excel makes to large numbers/dates makes it basically unusable even for just viewing for data analysis projects.
"Hundreds of students from both Yale and Harvard flooded the field and sat around the fifty-yard line, locking arms and chanting that both of their Universities divest from fossil fuels — along with private prisons and Puerto Rican debt."
I wanted to re-up this announcement since the deadline is *January 10* and we are hoping to start looking at applications right away.
The position is for one year with the possibility to renew for a second year.
I stand in righteous fury beside fellow women in science in filing a lawsuit against Dartmouth. We want our voices to be heard. We demand to be taken seriously. We demand change. Thread (1/10)
#metooSTEM
#MeToo
Truly a pleasure to share new work on adaptive experiments at UC Berkeley RWAP yesterday, including this paper and ongoing projects at
@GSBsiLab
--thoughtful and useful feedback, thanks for the invite
@dbroockman
@Matt__Graham
@JaeJaeykim2
Mostly just base r for variable creation and selective indexing. I admit that aggregate is inferior to group_by, summarize, and I do use dplyr, but try to be selective in which packages I use for specific tasks (dplyr, lubridate) rather than loading in tidyverse as a whole.
Does anyone have best practices for collaborating on public AND private versions of something like class files, using github and/or Dropbox? Some files + directory structure should be public, while we collaborate privately on, e.g., solution sets, that can live locally nearby.
“Yale often claims that the university cannot punish the fraternities because they are unregistered, off-campus organizations...while Yale permits them to use the Yale name, Yale email addresses, Yale bulletin boards and campus facilities for recruitment."
A lawsuit by 3 Yale students against the university claims that Yale has often looked the other way while parties rage and women from Yale and surrounding colleges are routinely sexually harassed and abused
At 5.20 PM CT, Deep Learning Applications in Political Science (and more!), with papers from Xuan Luo/Sota Kato/Budrul Ahsan/Takafumi Nakanishi,
@Patrick_Y_Wu
/ Walter R. Mebane, and
@doug_rivers
/ Stephanie Nail
"Because democratic erosion is piecemeal and involves the use of ostensibly legal tactics, it can be difficult to know when the line has been crossed from democracy to non-democracy. This is what makes democratic erosion so insidious."
Experts agree: American democracy is at risk, and we must take action to save it. Members of the Democratic Erosion Consortium have signed on to an open letter laying out the threats and what should be done to address them
Ugh, yet again, not a good look,
@facebook
. "The revenue Facebook earned off children had such large chargeback rates...that it far exceeded what the Federal Trade Commission has said is a red flag for deceptive business practices."
Come see my colleagues at the
@Yale
Political and Economic Development Conference, focusing on the political consequences of structural transformation in developing countries. Conf May 10. Papers due February 15.
It is such an honor and I am so moved to be on this year’s
#TIME100
alongside one of the people who influenced me the most -
@tourmaliiine
. Thank you
@Lavernecox
for your beautiful words and friendship and Naima Green for the beautiful image.
THREAD
In July 2018, a horrifying video began to circulate on social media.
2 women & 2 young children are led away by a group of soldiers. They are blindfolded, forced to the ground, and shot 22 times.
#BBCAfricaEye
investigated this atrocity. This is what we found...
Meanwhile, professors assigned session presentations can do so for free. At the very least, offer discounted rates for standard printing at the conference location.
If you’re a PhD thinking about alt-ac options, consider applying to this job at Google marketing. They’re looking for PhD behavioral economists or quantitative social scientists from related fields. It’s with a great team.
As an alum of
@WilsonSchool
, it's shameful that Princeton trustees' decision against changing the WWS name in 2016 is being used as justification to continue honoring a eugenicist.
Another person said that the lecture name should not change because, in 2016, Princeton University did not change the name of the Woodrow Wilson building, despite his racist views.
Princeton is a leader in a lot of ways, but I don’t think this is one of those ways. 14/
This was an incredible conversation with incredible scholars. Thoughtful and motivating, the highlight of my day. If you weren't able to watch it live, the video is linked below.
#raceandjusticeconvo
For those interested in the video from today's
#RaceAndJusticeConvo
, here's the link to a folder that includes the video, audio, and transcript. Let me know if you have trouble accessing. Thanks for all the interest! Share widely.
I’m a mother of six, and a Mormon. I have a good understanding of arguments surrounding abortion, religious and otherwise. I've been listening to men grandstand about women's reproductive rights, and I'm convinced men actually have zero interest in stopping abortion. Here's why…
"We’re often told that we should make a plan to vote to increase the likelihood of fulfilling this civic duty...we also suggest making a plan to protest — ideally as part of an organized, nonviolent group — should the need arise to ensure that our votes actually mean something."
@monika_nalepa
@a_strezh
@austinlwright
Absolutely agree; the most important thing students take from methods class is how to be thoughtful about assumptions they invoke in analysis (quant or otherwise). W/ causal inf/missing data we always need to make assumptions, and like Anton says, can discuss w/out a lot of math.
@StommesDrew
This is the problem that Groundhog attempts to address; I don’t think it’s quite as explicit as a python virtual environment, and I’m not sure what uptake will look like in the R community. But if used consistently, would help.
@JaeJaeykim2
Also we would all be doing ourselves a favor by documenting which versions of R packages we're using in a given project (I also don't systematically do this).