Sign in

Uri Simonsohn

@urisohn.bsky.social
3.3K followers 569 following 840 posts

Pragmatic methodologist and behavioral scientist at Esade business school. Barcelona website: urisohn.com Blog: DataColada.org

PostsRepliesMedia
Uri Simonsohn @urisohn.bsky.social · 19/09/2026
Just shows up in my (paid) Spotify account
000
Uri Simonsohn @urisohn.bsky.social · 17/09/2026
Enjoying the spotify AI list maker "studio" e.g., I asked "make a list with the two most listened to songs from every jazz artists visiting barcelona in the next 2 months"
140
Uri Simonsohn @urisohn.bsky.social · 13/09/2026
After a few years of being just fine, the 99% invisible podcast is again amazing in its History of the US in 100 objects. 99percentinvisible.org/100-objects/ The M43 field Jacket The wheel of fortune dalmatian The Sand letters Too good! Just great stuff. Niche history through cool object stories
99percentinvisible.org
A History of the United States in 100 Objects - 99% Invisible
100 objects. 100 stories. A new history of the US hiding in plain sight. America’s 250th birthday calls for a history as sprawling and contradictory as the country itself. A History of the United Stat...
040
Reposted by Uri Simonsohn
simine vazire @simine.com · 02/09/2026
Psych Science has retracted Ariely & Wertenbroch (2002). The second author requested the retraction. Both authors agreed to it. journals.sagepub.com/doi/full/10.... So grateful to @datacolada.bsky.social for these blog posts and the work behind them: datacolada.org/138 and datacolada.org/139 1/2
journals.sagepub.com
Sage Journals: Discover world-class research
Subscription and open access journals from Sage, the world's leading independent academic publisher.
426794
Uri Simonsohn @urisohn.bsky.social · 02/09/2026
Like others, I have long understood recylcing to have a marginal benefit (except for aluminium i think?) But what i find most intersting and don't know, is who is responsible for having decided *reciclying* was the thing to spend all efforts persuading people to focus on. How did that happen?
120
Reposted by Uri Simonsohn
Data Colada @datacolada.bsky.social · 02/09/2026
Colada[139] Artificial Deadlines Part 2 Evidence of fraud in the 25 year old gradesheet behind the key finding in an influential paper on deadlines and procrastination. datacolada.org/139
816857
Reposted by Uri Simonsohn
Data Colada @datacolada.bsky.social · 31/08/2026
An influential paper by Ariely & Wertenbroch (2002) reported two studies with tampered data. Post 1 of 2. datacolada.org/138
datacolada.org
[138] Artificial Deadlines (Part 1): Evidence of Fraud in an Influential Study About Procrastination
A new paper in Psychological Science (.htm) reports a failure to replicate Study 2 of Ariely and Wertenbroch’s influential article entitled, “Procrastination, Deadlines, and Performance: Self-Contr…
723799
Reposted by Uri Simonsohn
Catherine E. de Vries @catherinedevries.bsky.social · 31/08/2026
We’re hiring! Poli Sci & IR department at IE University in Madrid is hiring three tenure-track Assistant Professors, starting 1st of September 2027, in: Comparative Politics: lnkd.in/ePDfwjJB International Relations: lnkd.in/eDu5vGTf Political Economy: lnkd.in/eciDukTA 1/3
28084
Uri Simonsohn @urisohn.bsky.social · 16/08/2026
Yes. Researchbox has blind version which hides author names. Even after public you can anonymize for exta 6 months (renewable). RB even warns you when uploaded code scripts include author names or emails
010
Uri Simonsohn @urisohn.bsky.social · 16/08/2026
Did they say where they got it from,when, how and who?
100
Uri Simonsohn @urisohn.bsky.social · 16/08/2026
I think requiring results provenance in AsCollected.org is the only protection
ascollected.org
Home | AsCollected
141
Uri Simonsohn @urisohn.bsky.social · 13/08/2026
4/4 One possible long-term solution involves getting univ. libraries to buy in and build infraestracture for permanent storage. But asking to store crazy # of TB, with high % of junk makes that difficult. This a real problem for all, but least so for the Credibilty Lab.
020
Uri Simonsohn @urisohn.bsky.social · 13/08/2026
3/4 I think at the Credibility Lab we have the best plan. We already have public-facing, ready backups at archive.org and Zenodo (and soon Clockss). Thinking today about your end, makes you realize how expensive it is to accept junk or irrelevant files. Neither the OSF nor Zenodo did/do
110
Uri Simonsohn @urisohn.bsky.social · 13/08/2026
2/4 More important that who funds today, is: does the platform have a real ready-to-go plan for permanence. I used to (and still do) worry about OSF's supposed 50 year $ for file storage. OSF is complex, who will update to new OS and hardware? I worry about Zenodo's vague '20 year' funding promise
110
Uri Simonsohn @urisohn.bsky.social · 13/08/2026
1/4 It's not "my" lab the way a professor has a lab, it's a cost unit at Wharton that is called "lab" instead of "center", and I am one of the directors. Wharton funds it. The platforms are well designed so they cost a trivial, truely trivial amount per year The issue of funding is interesting->
120
Uri Simonsohn @urisohn.bsky.social · 13/08/2026
OK, added to table on DataColada.org/137 ResearchBox and Zenodo store in Europe ResearchBox but not Zenodo stores also in USA
010
Uri Simonsohn @urisohn.bsky.social · 13/08/2026
RB is *definitely* designed for people. Specifically, for people who wish to re-analyze data from a paper they are reading or evaluating. I think Zenodo is a distant 2nd for this. But, the metadata you ask for, aimed at bureaucrats & metascientists, is easy to add. Put on the RB to-do list. Thanks
000
Uri Simonsohn @urisohn.bsky.social · 13/08/2026
(added it to the table in datacolada.org/137)
010
Uri Simonsohn @urisohn.bsky.social · 13/08/2026
Hm. I see a bunch of download 1 buttons or a single download All Am I missing something? To download say N data files you need to click on N download 1 file buttons
100
Uri Simonsohn @urisohn.bsky.social · 12/08/2026
i am actually on vacation, when i get back I will email you to discuss further
010
Uri Simonsohn @urisohn.bsky.social · 12/08/2026
Private components were part of the problem with the OSF. RB doesn't have antying designed to be private, all files are accessible, if code protected and anonymous for peer review For the 2nd,i pitched something to qualtrics with Gino case: Column level MD5 hashes that are public,allow verifiability
010
Uri Simonsohn @urisohn.bsky.social · 12/08/2026
OTOH , I think uploading straight to repository is not a good idea. We want curation, privacy protection, flow control. But OTOH, i think two very closely related uses are very valuable 1) Automatic human-readable codebook generation 2) Data integrity verification (e.g.,for AsCollected)
110
Uri Simonsohn @urisohn.bsky.social · 12/08/2026
@joshdeleeuw.bsky.social @giladfeldman.bsky.social I'm skeptical of API for uploading, quality control would just suffer and the end goal, making it easy to evaluate published empirical claims suffers API for getting content, maybe, though ResearchBox already does everythingn Gilad's OpenOSF did
000
Uri Simonsohn @urisohn.bsky.social · 12/08/2026
For uploading I am skeptical but open minded. 1) What's the advantage of programatically uploading a dataset vs drag-and-dropping it? 2) How would you deal with the required codebooks for data? 3) What about other files like stimuli, pre-reg, etc. For downloading, i am curious. What's a use case?
200
Uri Simonsohn @urisohn.bsky.social · 12/08/2026
They aren't searchable. Terms of use indicate they may become, but currently they aren't. AsPredicted seeks to address p-hacking not file-drawer We have protections against parallel pre-registrations (same study) Even with access to all pre-reg, figuring out file-drawer would be very hard.
120
Uri Simonsohn @urisohn.bsky.social · 11/08/2026
Good point.RB doesnt have an API will add to the table. What actions did you use from API and from what program?
100
Uri Simonsohn @urisohn.bsky.social · 11/08/2026
Here is what i think the head-to-head comparison of ResearchBox vs Zenodo look like as OSF replacements. If I am missing an important attribute, let me know, will update accordingly.
97232
Uri Simonsohn @urisohn.bsky.social · 11/08/2026
Do you think this comparison is accurate? Am I missing something you find valuable?
220
Reposted by Uri Simonsohn
Jan Drugowitsch @jdrugowitsch.bsky.social · 13/07/2026
🚨 Big news 🚨 After 10 years at Harvard, my lab is moving to Barcelona in Spring 2027 to join @cbc-upf.bsky.social and the Department of Engineering at @upf.edu. 1/4
68818
Uri Simonsohn @urisohn.bsky.social · 07/06/2026
Took taxi from lisbon airport and was overcharged 2.5 times price and threatened with violence when I complained. Make sure to use Uber or Bolt in Portugal.
010
Uri Simonsohn @urisohn.bsky.social · 04/06/2026
If it helps, syllabus intro:
011
Uri Simonsohn @urisohn.bsky.social · 03/06/2026
I will be visiting Yale SOM for AY 2026/2027 The visit includes me teaching an interdisciplinary PhD methods course. If you often experience New Havenness, ask your Dr (advisor) whether "Methods Stumblers" may be right for you.
1102
Reposted by Uri Simonsohn
Paul Goldsmith-Pinkham @paulgp.com · 13/04/2026
First pass at a website that keeps track of the data availability status of various published papers: paulgp.com/replication-... Very much in Alpha mode, so data will be updated soon.
109123
Uri Simonsohn @urisohn.bsky.social · 06/03/2026
Ah, just figured it out. It's an IP block by manuscriptcentral. I reproducibly get in with a VPN and not without.
010
Uri Simonsohn @urisohn.bsky.social · 06/03/2026
I login to manuscriptcentral, and when i try to access anything i get logged out. Password change does not help. Happens at two journals (psychological science and management science) Anybody else? @silverchairnews.bsky.social do you have a contact for support?
100
Reposted by Uri Simonsohn
Data Colada @datacolada.bsky.social · 02/03/2026
Answering the most popular question in a PhD methods course datacolada.org/133
datacolada.org
[133] Heterofriendly: The Intuition for Why You Always Need Robust Standard Errors - Data Colada
When I taught my first PhD-level methods course, I invited students to submit questions about any topic in statistics or methodology. Six out of 10 students asked about the same topic: robust & cluste...
12611
Reposted by Uri Simonsohn
Data Colada @datacolada.bsky.social · 16/02/2026
A more user friendly t-test regression variable description frequency plots, and more. datacolada.org/132
2153
Reposted by Uri Simonsohn
de’on @justkeepskeeting.pro · 25/11/2025
this is mesmerizing
4748001787
Reposted by Uri Simonsohn
The Onion @theonion.com · 11/11/2025
Study Finds Most Americans Can't Find Where They Are Being Deported On Map
302963583
Reposted by Uri Simonsohn
Katy Milkman @katymilkman.bsky.social · 24/10/2025
Wouldn't it be great to gather top social science journal editors + experts on fraud-prevention to discuss better ways to fraud-proof our field @ the National Academies? This is happening! Step 1 is creating an organizing committee. Submit nominees by 11/7: www.nationalacademies.org/our-work/enh...
nationalacademies.org
1137
Reposted by Uri Simonsohn
alvaro-le-fou.bsky.social @alvaro-le-fou.bsky.social · 09/10/2025
Tidal. Slightly cheaper, hugely better sound quality, artists are paid better, no investments into military AI.
031
Uri Simonsohn @urisohn.bsky.social · 03/10/2025
Which restrictions do you have in mind?
100
Uri Simonsohn @urisohn.bsky.social · 29/09/2025
why may chatgpt give better answers than Cursor, even when i set gpt as the engine for the latter and it has the whole project as background instead of just one script?
100
Reposted by Uri Simonsohn
Nicola Gaston @nicgaston.bsky.social · 26/09/2025
First time I’ve seen the rabbit/bird illusion with an actual photo. Took me just a half second to process the word in all caps…
08025
Uri Simonsohn @urisohn.bsky.social · 24/09/2025
I don't think it is fair to characterize this as a "both sides' situation. I tried *for months* to engage in open dialogue They lacked the common-decency to reply to the emails asking for the paper, after I answered emails from them for years If they write a reply, happy to have a follow-up post
010
Uri Simonsohn @urisohn.bsky.social · 24/09/2025
2) Clintin & Richard say we claim to test if p-curve is skewed but we don't do that I see this as semantics. Indeed. We don't test whether *observed* p-values are skewed. Rather. We test whether data generator produces skewed ps e.g. p=.0001, p=.0002 are not skewed but suggest skewed generator
010
Uri Simonsohn @urisohn.bsky.social · 24/09/2025
They say, I think 2 things that fall in this category. 1 ) Clintin & Richard say "evidential value" is not a stats concept. I agree, but this is because we invented it. We needed a term for "significant after accounting for p-hacking", It didn't exist, so we came up with evidential value.
110
Uri Simonsohn @urisohn.bsky.social · 25/08/2025
A feature i really like is custom intro and "outro" skipping so you skip the yada yada and get to the actual episode
110
Uri Simonsohn @urisohn.bsky.social · 25/08/2025
Podcast addict
110
Reposted by Uri Simonsohn
Cloudflare @cloudflare.social · 04/08/2025
Perplexity is repeatedly modifying their user agent and changing IPs and ASNs to hide their crawling activity, in direct conflict with explicit no-crawl preferences expressed by websites. blog.cloudflare.com/perplexity-i...
blog.cloudflare.com
Perplexity is using stealth, undeclared crawlers to evade website no-crawl directives
Perplexity is repeatedly modifying their user agent and changing IPs and ASNs to hide their crawling activity, in direct conflict with explicit no-crawl preferences expressed by websites.
36844