David Hardisty

🚨 NEW PAPER IN @NatureHumBehav 🚨 Across contexts, we find a counterintuitive #FormalityEffect: Formal government communications are more effective than informal ones at shifting residents’ behavior. With @chris_e_larkin, @lindsay_m_moore, @Karminker rdcu.be/drRnl

Minister Freeland Wastes $10 Billion on Big Oil’s Carbon Capture Fantasy 🌍💔 Breaking rules, greenwashing, and more oil? The proposed CCUS tax credit is a $10B step back, not forward. 📢 Read more: bit.ly/3ReA58J #ClimateCrisis #StopFossilFuelSubsidies

In 2022, @cafreeland said the #CCUS tax credit wouldn't subsidize projects that use carbon for oil production A broken promise: the tax credit will subsidize projects that sell up to **90%** of captured carbon for EOR Great reporting @benshingler cbc.ca/news/climate/c…

Can the engineers just take a little break from developing AI and maybe implement a usable search function for Microsoft Outlook?

Taylor Swift and Travis Kelce as figures in the new U.S. National Climate Assessment report, a thread

@J_R_Schroeder Bravo!!! 👏👏👏 A critical project, both for science and for the people (especially ECRs) doing the science. Thank you!!!

Loved the Circle Practice at #BigDifferenceBC!! Thanks @StephPapik!!!

📢 New working paper alert! With the help of more than 700 volunteer reviewers, we (@MilosFisar, @chrhuber_, @elena_katok, @ozkesali and myself) assessed the reproducibility of ~500 articles published in the journal Management Science. Some results. 📊🔍 #Reproducibility

@andre_quentin I once reported concerns in my comments to the editor in peer review. Paper was rejected, and the editor alluded to anomalies in the decision letter to the authors. I don't think anything else happened.

This is a damning claim. Does anyone know of a counter-example (i.e., someone's fraud being detected during peer-review)? ... I only know examples of the opposite: Fraud was *ignored* during peer-review (despite reviewer waving crimson flags), then confirmed some time later.

👀Yikes, a huge amount of social science (and market research) relies on surveying humans on platforms like Mechnical Turk... except over 30% of human workers on both platforms are just using AI. This new paper finds the same size problem on the more curated Prolific platform.

I wrote a post-mortem on the Gino-Ariely scandal: How I came to suspect Gino’s work, the resistance I met during my Ph.D., my experience working with Data Colada, and the lessons I hope business academia will learn from it. theorgplumber.com/posts/statemen…

We are just two weeks away from #BIGDifferenceBC! We have an excellent line up including our keynote speaker @betsylevyp, as well as a special session focusing on #behaviouralinsights and communications!

"b-hacking" revealed in the FTX trial yesterday. Could be a way to spot potential fraud in business! Look for numbers just over $1m or just over $1b. coindesk.com/policy/2023/10…

@PsychScience @siminevazire Wow this is amazing news! Thank you for your service, @siminevazire !!!

Congratulations, @siminevazire! #psychtwitter #psychologicalscience psychologicalscience.org/publications/p…

Weird ad for pre-registration in my timeline. Open science advocacy has taken a strange turn. "Together, we brave the 'Storm' together" is a great slogan, though!

@AaronCharlton @stephaniemlee Yeah that's what I thought -- if you delete "inconvenient" data, it makes your results stronger of course. So her refutation here seems illogical to me.

@dhardisty @stephaniemlee She's saying that if you rerun the analysis without the observations that are tampered with the hypothesis still holds. This argument is weak because (a) the fishy data is massively in support of the hypothesis and (b) there could be other tampering that DC wasn't able to find.

This is jaw-dropping. It seems that everything I thought I knew about the cost of college is wrong. nationalaffairs.com/publications/d…

@stephaniemlee @thecrimson Keep in mind that IP addresses can easily be manipulated, through VPNs, for example. Also, as @andre_quentin mentioned, this doesn't explain the discrepancy between the original Qualtrics data and the publicly posted data.

DJ Paisley Eva rockin a set at @ChanCentre Come Towards The Fire