Skip to primary content

Statistical Modeling, Causal Inference, and Social Science

Search

Main menu

Home
Authors
Blogs We Read
Books
Sponsors

Post navigation

Judea Pearl on why he is “only a half-Bayesian” Lessons learned from a recent R package submission

A counterfeit data graphic

Posted on January 21, 2012 3:18 PM by Andrew

Kaiser Fung discusses. It’s a good sign when statistical graphics are so popular that people feel the need to fake them!

This entry was posted in Statistical Graphics by Andrew. Bookmark the permalink.

Art
Artificial Intelligence
Bayesian Statistics
Causal Inference
Decision Analysis
Economics
Jobs
Literature
Miscellaneous Science
Miscellaneous Statistics
Multilevel Modeling
Obituaries
Papers
Political Science
Public Health
Sociology
Sports
Stan
Statistical Computing
Statistical Graphics
Teaching
Zombies

Andrew on Ideological bias in “Ideological bias in the production of research findings”October 26, 2026 7:02 AM
Jules: I can't remember when I wrote the two posts but their content is different, no? The posts discuss the…
Jules on Ideological bias in “Ideological bias in the production of research findings”October 26, 2026 5:10 AM
You posted on this in June, so is this now rewritten or different? https://statmodeling.stat.columbia.edu/2026/06/22/one-star-is-p-0-20-two-stars-is-a-set-of-steak-knives-three-stars-is-youre-fired/
Anonymous on “In the belly of the whale: Portraits from a fractured Israel”October 26, 2026 1:04 AM
Joshua, I understand we’ve spent long on the military service issue, so I’ll try to summarize – I’d be more…
Joshua on “In the belly of the whale: Portraits from a fractured Israel”October 25, 2026 8:48 PM
Anonymous - Thanks for the detailed reply. Fair points on reserve duty - I didn't mean to suggest that service…
Anonymous on “In the belly of the whale: Portraits from a fractured Israel”October 25, 2026 4:38 PM
Hi Joshua, Thanks for your reply. Regarding education investment - it's largely pork barrel politics. The most favored sector (in…
Joshua on “In the belly of the whale: Portraits from a fractured Israel”October 25, 2026 1:46 PM
Anonymous - Thanks for the reply. You say your Arab friends mentioned, briefly, lesser investment in education and infrastructure. That’s…
Jukka on CmdStanR 1.0 release candidate is now available!October 25, 2026 9:57 AM
"For example, cmdstanr has commands to download binaries from the Stan project and then execute them. It doesn’t get much…
Seren Dipity on My exchange with Yann LeCun about killer robotsOctober 25, 2026 12:19 AM
Anonymous on October 23, 2026 10:48 PM at 10:48 pm said: ... "The lunatic fringe in politics is endlessly varied, but they…
Joey on Ideological bias in “Ideological bias in the production of research findings”October 24, 2026 10:04 PM
Just as a sidenote, social cohesion (or social trust) can be measured as Putnam did by: political efficacy, confidence in…
Pedro Franco on Ideological bias in “Ideological bias in the production of research findings”October 24, 2026 8:52 PM
Briefly, Andrew, just popping in to say that I remember this study and there's serious critiques about the methodology (and…
gec on Ideological bias in “Ideological bias in the production of research findings”October 24, 2026 10:34 AM
I admit to not understanding the aim of the original study. As you say, it is not surprising that people…
Will on Ideological bias in “Ideological bias in the production of research findings”October 24, 2026 9:27 AM
Hearing people like Donald Trump and Clarence Thomas described as "conservative" (no! They're distinctly un-Burkean! Saying "the meaning of conservative…
AAAnonymous on “I write to seek your help and advice regarding a published article that my group and I believe is invalid and should be corrected or retracted.”October 24, 2026 2:57 AM
Quote from above: "The norms of science do not matter for the “successful” ones." When reading the above quote, I…
Seth Finkelstein on I am shocked, shocked to learn that this 2002 paper by Dan Ariely is based on dubious data and doesn’t replicate.October 23, 2026 11:08 PM
Hmm, to be fair, that comparison strikes me as a bit misleading. The S&P 500 minus the "AI" companies has…
Milind Watve on “I write to seek your help and advice regarding a published article that my group and I believe is invalid and should be corrected or retracted.”October 23, 2026 11:08 PM
I flagged at least 25 papers in the last two years with serious statistical or some other flaws that challenged…
Anonymous on My exchange with Yann LeCun about killer robotsOctober 23, 2026 10:48 PM
The definition of a herrenvolk democracy is a type of government where elections exist, but voting rights and the ability…
Anonymous on “In the belly of the whale: Portraits from a fractured Israel”October 23, 2026 9:37 PM
Joshua, Let me be very clear that my comment above is a direct reply to Andrew’s use of the term…
Andrew on I am shocked, shocked to learn that this 2002 paper by Dan Ariely is based on dubious data and doesn’t replicate.October 23, 2026 7:40 PM
It was probably another one of his unnamed research collaborators doing something with fake data. Worst. Luck. Ever.
Matt Skaggs on I am shocked, shocked to learn that this 2002 paper by Dan Ariely is based on dubious data and doesn’t replicate.October 23, 2026 7:25 PM
Andrew wrote: "Dan Ariely [...] has had pretty much the worst professional luck that any scientist could have." It happened…
Daniel Lakeland on My exchange with Yann LeCun about killer robotsOctober 23, 2026 4:57 PM
The US is also a herrenvolk democracy. It was set up that way (with outright slavery written into the constitution…
paul alper on A big problem with conventional experimentation or A/B testing is that the treatment effect can vary over time and it can depend on the entire history of the treatment. Usual comparisons miss this point.October 23, 2026 1:24 PM
Whenever the topic of placebo effect and "time" comes up, I turn to the famous https://www.google.com/search?client=firefox-b-1-d&q=israeli+researcher+who+looked+at+prayer+after+the+fact&sei=g4bbasTIDfW8mLQP9M6J8QU By now, Leibovici's famous…
jd on A big problem with conventional experimentation or A/B testing is that the treatment effect can vary over time and it can depend on the entire history of the treatment. Usual comparisons miss this point.October 23, 2026 12:42 PM
I understand all that, but I also think it’s reasonable that regression to the mean and selective reporting need not…
Jeff on A big problem with conventional experimentation or A/B testing is that the treatment effect can vary over time and it can depend on the entire history of the treatment. Usual comparisons miss this point.October 23, 2026 12:05 PM
I like the noise explorer. When I was consulting I made something like this (though much less sophisticated) in Excel…
Bob Carpenter on A big problem with conventional experimentation or A/B testing is that the treatment effect can vary over time and it can depend on the entire history of the treatment. Usual comparisons miss this point.October 23, 2026 11:59 AM
Aren't most of our studies of dynamic systems based on static snapshots? For example, the pharmacokinetic/pharmacodynamic models I've worked on…
Jesse O. on A big problem with conventional experimentation or A/B testing is that the treatment effect can vary over time and it can depend on the entire history of the treatment. Usual comparisons miss this point.October 23, 2026 11:44 AM
> the treatment effect can vary over time and it can depend on the entire history of the treatment Simply…
Anoneuoid on A big problem with conventional experimentation or A/B testing is that the treatment effect can vary over time and it can depend on the entire history of the treatment. Usual comparisons miss this point.October 23, 2026 11:24 AM
Yea you cant effectively study a dynamic system with static snapshots. But studying dynamics requires calculus/programming skills that are missing…
Andrew on A big problem with conventional experimentation or A/B testing is that the treatment effect can vary over time and it can depend on the entire history of the treatment. Usual comparisons miss this point.October 23, 2026 11:11 AM
Jd: Not at all! When people argue---convincingly, in my view---that placebo effects aren't such a big deal, what they're saying…
jd on A big problem with conventional experimentation or A/B testing is that the treatment effect can vary over time and it can depend on the entire history of the treatment. Usual comparisons miss this point.October 23, 2026 10:58 AM
Parts of the health supplement and non-traditional medicine industry would seem to substantiate their existence
Andrew on A big problem with conventional experimentation or A/B testing is that the treatment effect can vary over time and it can depend on the entire history of the treatment. Usual comparisons miss this point.October 23, 2026 10:57 AM
Dale: Nick Brown was telling me about this; it came up in our recent paper.
Dale Lehman on A big problem with conventional experimentation or A/B testing is that the treatment effect can vary over time and it can depend on the entire history of the treatment. Usual comparisons miss this point.October 23, 2026 10:48 AM
Andrew, you are starting to sound like some politician who will remain nameless. Some people have said, many people have…
Andrew on A big problem with conventional experimentation or A/B testing is that the treatment effect can vary over time and it can depend on the entire history of the treatment. Usual comparisons miss this point.October 23, 2026 10:36 AM
Dale: Many people are saying that placebo effects don't really exist.
Dale Lehman on A big problem with conventional experimentation or A/B testing is that the treatment effect can vary over time and it can depend on the entire history of the treatment. Usual comparisons miss this point.October 23, 2026 10:27 AM
Very interesting. I found myself thinking about placebo effects. It is true that the treatment and control groups are randomly…
shira on Survey Statistics: textbooks adding non-probability sampling chaptersOctober 23, 2026 10:22 AM
Thanks, Joe ! I certainly would too :)
Joshua on “In the belly of the whale: Portraits from a fractured Israel”October 23, 2026 9:48 AM
Anonymous – I don't doubt your description growing up alongside Arab neighbors, classmates and colleagues, and I suspect many people…
Joshua on “In the belly of the whale: Portraits from a fractured Israel”October 23, 2026 9:44 AM
Anonymous – You say my death-penalty point is “a bit irrelevant” because Palestinians are subject to different laws in almost…
John Kane on My exchange with Yann LeCun about killer robotsOctober 23, 2026 8:34 AM
Nothing to worry about. Cartels already have MANPADs and anti-tank weapons, a few drones won't be any big deal. https://english.elpais.com/international/2026-10-18/fiber-optic-drones-and-cartel-members-at-the-front-ukraines-imprint-on-mexicos-narco-groups.html.
Andrew on My exchange with Yann LeCun about killer robotsOctober 23, 2026 6:28 AM
Eli: I agree that there are different views on the matter. I consider it fairly uncontroversial to characterize Israel as…
Eli Nelken on My exchange with Yann LeCun about killer robotsOctober 23, 2026 3:40 AM
I would like to comment on the use of 'herrenvolk democracy' with respect to Israel. You used that a few…
Joe Paxton on Survey Statistics: textbooks adding non-probability sampling chaptersOctober 23, 2026 3:11 AM
I'd definitely contribute, and I know a couple dozen people off the top of my head who would as well.
Seren Dipity on My exchange with Yann LeCun about killer robotsOctober 22, 2026 11:50 PM
To Yann LeCun. 1)Musk & Palmer Lucky are now deciding weapons. 2. "Any sufficiently advanced punditry is indistinguishable from bollocks"…
Anonymous on “In the belly of the whale: Portraits from a fractured Israel”October 22, 2026 9:12 PM
Joshua, Your discussion of the death penalty law is a bit irrelevant – Palestinians are subject to different laws from…
Anonymous on “In the belly of the whale: Portraits from a fractured Israel”October 22, 2026 8:54 PM
Andrew, You speak of "aspect of Jim Crow" and "segregation" in Israel itself. I have to wonder how it was…
Matt Skaggs on OK, here’s a new scam. But I can’t figure out how it’s supposed to work.October 22, 2026 6:48 PM
It seems that the 1st Annual Chen Tingzuo Immunological Science Award Ceremony did occur last year and there is a…
Anon2 on “In the belly of the whale: Portraits from a fractured Israel”October 22, 2026 5:42 PM
"Even the Palestinian account of the 2000 Camp David Summit claims that Israel would control ~20% of the West Bank…
Jesse O on How to get the most out of Bayesian Data AnalysisOctober 22, 2026 3:33 PM
Typo: "Partial pooling for estimating the rats of a rare cancer" should surely read "rate".
Tiffany Weinkopff on OK, here’s a new scam. But I can’t figure out how it’s supposed to work.October 22, 2026 2:14 PM
I got the same email. It has scam written all over it. Thanks for sharing these posts to confirm my…
Joshua on “In the belly of the whale: Portraits from a fractured Israel”October 22, 2026 1:16 PM
Anonymous, The history of actual cases, including one in which Israel’s Supreme Court ordered a community to admit an Arab…
Jonah on CmdStanR 1.0 release candidate is now available!October 22, 2026 11:14 AM
Yeah, remotes::install_github("stan-dev/[email protected]") will get the release candidate plus any fixes since it was announced.
Bob Carpenter on CmdStanR 1.0 release candidate is now available!October 22, 2026 10:56 AM
Thanks, Jonah---I should have been patient and asked before posting!
Bob Carpenter on CmdStanR 1.0 release candidate is now available!October 22, 2026 10:55 AM
You can use install_github on a branch if you want to live dangerously.

Proudly powered by WordPress