Keith Chen

#KamalaHarris!

My daughter wanted to dress like Kamala this Halloween!

Happy Birthday, Internet! 55 years ago, the 1st internet message was sent from @UCLA @UofCalifornia by a team of computer scientists led by Dist. Prof. Leonard Kleinrock @kleinrock @uclaconnection1 @CS_UCLA . Join us for his lecture today to learn more: samueli.ucla.edu/upcoming-event…

@BeSciPol 2024 Online Conference - May 2-3 is less than a month away! Featured behavioral science experts include Team Scientists @MKeithChen, @DollyChugh, @wdlrosa, @TatianaHomonoff, @m_sendhil, @elkeweber, & David Yeager. Register at: behavioralscienceconference.com

Smartphone data reveal how much more Black Americans are policed. The reason why varies by city. Just Accepted new paper by M. Keith Chen @MKeithChen, Katherine L. Christensen @katechristens, Elicia John, Emily Owens @ProfEmilyOwens, Yilin Zhuo zurl.co/pkyq

Yes... I had to censor my work without which my committee wouldn't sign it... But the great thing with Internet is that you can post the original version and even add a disclaimer explaining why it differs from the published version…

1/Today I am thrilled to announce, after years of work, our team @fervoenergy has developed a major advancement in geothermal, proven it at the field scale, and is set to deliver commercial next-gen geothermal decades ahead of schedule. THREAD. bloomberg.com/news/articles/…

Oh, nothing, just a robot programmed to rapidly identify all members of the unwanted species and kill them with lasers.

@morewedge @ProfKMEricson @katy_milkman @HarvardHBS @francescagino @DataColada @OSFramework But I suspect we agree on more than we disagree; we probably both appreciate recent open science changes in our field / hope they accelerate / think there is significant room for improvement.

@morewedge @ProfKMEricson @katy_milkman @HarvardHBS @francescagino @DataColada @OSFramework I think that's probably right! But I worry if that's our bar.

If you’re in or adjacent to the field of #BehavioralScience & aren’t talking about the new evidence of data fraud by ⁦@HarvardHBS⁩ ⭐️ Prof. ⁦⁦@francescagino⁩, read this immediately 🤢👇🏻 ⁦⁦@DataColada⁩ We have to do better. datacolada.org/109

@morewedge @ProfKMEricson @katy_milkman @HarvardHBS @francescagino @DataColada @OSFramework Maybe this is too low-bandwidth of a channel to really understand this though, and lots is currently happening. Would love to hear your thoughts next time we run into each other.

@morewedge @ProfKMEricson @katy_milkman @HarvardHBS @francescagino @DataColada @OSFramework I'd like to understand where we disagree Carey, I take it that you don't think low replication rates and cases of fraud are real evidence? Maybe we disagree on what the claim is? I think we have evidence of problems with the reliability of experimental behavioral science

@morewedge @ProfKMEricson @katy_milkman @HarvardHBS @francescagino @DataColada @OSFramework and yes, i don’t think behavioral science is unique in this, (but we do seem worse than most?)

@morewedge @ProfKMEricson @katy_milkman @HarvardHBS @francescagino @DataColada @OSFramework i agree we shouldn’t rely on intuitions, though we do have data on low levels of replication, and that is as telling as cases of fraud, maybe more so.

@ProfKMEricson @morewedge @katy_milkman @HarvardHBS @francescagino @DataColada @OSFramework As examples, only one author collecting and touching the data seems common, and failures of replication seem to have no negative repercussions for the original authors. Both seem to suggest the field doesn't prioritize / incentivize reliable science.

@ProfKMEricson @morewedge @katy_milkman @HarvardHBS @francescagino @DataColada @OSFramework I think this take is wrong @morewedge, I don't think behavioral science appears to have taken failures of replication and the worse cases of fraud seriously enough; the field doesn't seem to be to be self-correcting.

Why Remote Work Could Lead to Less Innovation A new study by Professors @MKeithChen (@uclaanderson), David Atkin & Anton Popov (@MITSloan) suggests that when employees from one company run into employees from another company, creative sparks fly. wsj.com/articles/remot…

@ColebRob @KricheliKatz @regev_tali I agree we don't know how much of the increased spread is between vs within subjects, but that distinction doesn't affect the main prediction: weak-FTR will decrease the value of future rewards either way.

@MKeithChen @KricheliKatz @regev_tali That’s right but the precision mechanism hypothesises individual weak-FTR speakers # represent future dates less precisely. Higher variance between subjects doesn’t necessarily entail this. You’d have ask individuals the same question multiple times.

Are cross linguistic difference in future grammar all about time or all about probability? In contrast to conventional wisdom, we show the latter may be true. What do you think @MKeithChen? onlinelibrary.wiley.com/doi/full/10.11…

@ColebRob @KricheliKatz @regev_tali Between subjects, FTR appears to have temporal effects: weak-FTR treatments decrease temporal precision. That's the temporal-precision effect that I think the experiments support.

@ColebRob @KricheliKatz @regev_tali We should hop on a zoom and talk at higher bandwidth, but in these experiments, the amount of future rewards is not uncertain; so there is no role for reward uncertainty