Skip to primary content

Statistical Modeling, Causal Inference, and Social Science

Main menu

  • Home
  • Authors
  • Blogs We Read
  • Books
  • Sponsors

Post navigation

Cheap talk and nukes 1, 2, 3, Many Tea Parties?

Lucia de Berk found not guilty

Posted on April 14, 2010 3:35 PM by Andrew

Maarten Buis writes:

A while ago you blogged on the case of Lucia de Berk:

Maybe you would like to know that her conviction has been overturned today.

This entry was posted in Miscellaneous Science 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
  1. rmb on Columbia University says, “We believe in difficult conversations at Columbia—in a broad embrace of freedom of expression. It’s in our DNA.”October 11, 2026 8:49 PM

    "...In some circumstances they trimmed their authoritarianism, or at least its public expression, and even posed as defenders of freedom…

  2. paul alper on Columbia University says, “We believe in difficult conversations at Columbia—in a broad embrace of freedom of expression. It’s in our DNA.”October 11, 2026 7:27 PM

    Indeed, as Seth writes, "It’s not an easy task for anyone to navigate this." Fortunately, it is a piece of…

  3. Andrew on Thinking more seriously about the design of exploratory studies: A manifestoOctober 11, 2026 4:36 PM

    Elin: I agree that these issues don't just arise with exploratory analysis. I also think that exploratory analysis is important…

  4. Andrew on Columbia University says, “We believe in difficult conversations at Columbia—in a broad embrace of freedom of expression. It’s in our DNA.”October 11, 2026 4:34 PM

    Ziggy: A difficulty here is that it can be hard to tell if someone's speech is in good faith. For…

  5. Anonymous on Columbia University says, “We believe in difficult conversations at Columbia—in a broad embrace of freedom of expression. It’s in our DNA.”October 11, 2026 3:54 PM

    Maybe they could try giving up on that whole 'putting identities into the oppressor/oppressed framework' business?

  6. Ziggy on Columbia University says, “We believe in difficult conversations at Columbia—in a broad embrace of freedom of expression. It’s in our DNA.”October 11, 2026 3:28 PM

    Let me introduce a heterodox opinion. As a normative matter, I can't see the point to tolerating bad faith speech.…

  7. Seth Finkelstein on Columbia University says, “We believe in difficult conversations at Columbia—in a broad embrace of freedom of expression. It’s in our DNA.”October 11, 2026 3:19 PM

    Matt: I don't want to turn this comment thread into an Israel/Palestine debate. That never works. But ... to me,…

  8. Jonathan Baron on Columbia University says, “We believe in difficult conversations at Columbia—in a broad embrace of freedom of expression. It’s in our DNA.”October 11, 2026 2:59 PM

    Columbia played a special role in the history of academic freedom. In 1917, it fired James McKeen Cattell for his…

  9. Anonymous on Columbia University says, “We believe in difficult conversations at Columbia—in a broad embrace of freedom of expression. It’s in our DNA.”October 11, 2026 2:43 PM

    “And, although our president did not say so in this message, this policy of protecting and welcoming dissent, even when…

  10. D on Columbia University says, “We believe in difficult conversations at Columbia—in a broad embrace of freedom of expression. It’s in our DNA.”October 11, 2026 2:22 PM

    I don't think that it's "simply inconceivable that [the Crimson] would publish a Palestinian student’s op-ed that began 'from the…

  11. elin on Thinking more seriously about the design of exploratory studies: A manifestoOctober 11, 2026 1:49 PM

    It seems like you are saying that statisticians should be learning research methods, which I definitely agree with. Introduction to…

  12. Andrew on Columbia University says, “We believe in difficult conversations at Columbia—in a broad embrace of freedom of expression. It’s in our DNA.”October 11, 2026 1:38 PM

    Oncodoc: I was pleased to see the nice words from the university president because, in recent years past, we weren't…

  13. elin on The problem is noise, not N: why increasing the sample size isn’t all that.October 11, 2026 1:37 PM

    There are three factors in power: sample size, "true" effect size, and the alpha level used. Two of these are…

  14. oncodoc on Columbia University says, “We believe in difficult conversations at Columbia—in a broad embrace of freedom of expression. It’s in our DNA.”October 11, 2026 12:43 PM

    I don’t think that the nice words of a university president matter very much. We live in a world where…

  15. Andrew on Columbia University says, “We believe in difficult conversations at Columbia—in a broad embrace of freedom of expression. It’s in our DNA.”October 11, 2026 12:16 PM

    Matt: I agree with you on the merits of the case: both "from the river to the sea" and "proud…

  16. Larry on The problem is noise, not N: why increasing the sample size isn’t all that.October 11, 2026 11:55 AM

    You might get the right direction on the effect size anyway, but in null hypothesis testing the effect size is…

  17. Matt Skaggs on Columbia University says, “We believe in difficult conversations at Columbia—in a broad embrace of freedom of expression. It’s in our DNA.”October 11, 2026 11:26 AM

    "Our statutes lay out quite clearly and sensibly that expression, on our campuses, does have limits, in particular as regards…

  18. Dale Lehman on Columbia University says, “We believe in difficult conversations at Columbia—in a broad embrace of freedom of expression. It’s in our DNA.”October 11, 2026 9:46 AM

    Interesting to think of organizational DNA as "some equivalent of self-replicating code." I think many universities can be characterized in…

  19. Dale Lehman on Digital twins version 2.0: The skeptics speak upOctober 11, 2026 7:20 AM

    I hadn't heard of those, but they sound interesting. I've always preferred focus groups to surveys. This removes one variable…

  20. Raphael K on Digital twins version 2.0: The skeptics speak upOctober 11, 2026 4:18 AM

    @Dale Lehman: You might be interested in qualitative AI-led interviews. They have been around for a few years now —…

  21. Don Vicendese on Digital twins version 2.0: The skeptics speak upOctober 10, 2026 11:29 PM

    I meant to add that in my work with prediction models, I have created models on projects that had the…

  22. Don Vicendese on Digital twins version 2.0: The skeptics speak upOctober 10, 2026 11:19 PM

    LLMs can be viewed as powerful data compressors: this helps them learn broad patterns from enormous amounts of information, but…

  23. Anoneuoid on The problem is noise, not N: why increasing the sample size isn’t all that.October 10, 2026 10:59 PM

    Larger sample size typically means a more meaningless mean. The exception is when the variation is actually "noise" (eg, apparent…

  24. AAAnonymous on The problem is noise, not N: why increasing the sample size isn’t all that.October 10, 2026 5:29 PM

    I don’t understand statistics very well. When numbers and formulas are present I mostly don’t understand things. But I try…

  25. Matt Skaggs on The problem is noise, not N: why increasing the sample size isn’t all that.October 10, 2026 2:35 PM

    I'm a little out of my depth on this one but this seems to be an observational approach to something…

  26. Jukka on The problem is noise, not N: why increasing the sample size isn’t all that.October 10, 2026 1:05 PM

    So, one of my papers under review has n = 200,000 out of a noisy moving target population of around…

  27. Pau on Getting better causal inferences by using more information (my talk at Stanford this coming Wed)October 10, 2026 12:55 PM

    Is there an available write up of the paper?

  28. AAAnonymous on The problem is noise, not N: why increasing the sample size isn’t all that.October 10, 2026 12:50 PM

    The following paper seems relevant for those interested: "Challenging the N-Heuristic: Effect size, not sample size, predicts the replicability of…

  29. paul alper on The problem is noise, not N: why increasing the sample size isn’t all that.October 10, 2026 12:23 PM

    My family is on holiday in Mexico City and I am not so I decided to determine in a definitive…

  30. Dale Lehman on Digital twins version 2.0: The skeptics speak upOctober 10, 2026 9:36 AM

    One of the papers you link to deserves more discussion, I believe. It notes "public opinion surveys may not capture…

  31. Jukka on What’s the deal with the RMET (“the Reading the Mind in the Eyes test”)?October 10, 2026 5:44 AM

    https://www.skeptic.com/article/what-is-it-like-to-be-a-dolphin-i-dont-know-neither-do-you/

  32. Olaf Zimmermann on What’s the deal with the RMET (“the Reading the Mind in the Eyes test”)?October 9, 2026 11:38 PM

    To err is human, ...

  33. Andrew on Digital twins version 2.0: The skeptics speak upOctober 9, 2026 7:29 PM

    Matt: Good point about the darts. I've always thought of "dart throws" as a synonym for randomness, but, yeah, some…

  34. Matt Skaggs on Digital twins version 2.0: The skeptics speak upOctober 9, 2026 7:25 PM

    An expert can throw a dart pretty accurately. My reference to 20 years goes back to Tom Tango's Marcel system.…

  35. Alain on What’s the deal with the RMET (“the Reading the Mind in the Eyes test”)?October 9, 2026 5:28 PM

    You may. It would be more helpful, though, for you to say exactly what you mean instead of posting a…

  36. Andrew on Digital twins version 2.0: The skeptics speak upOctober 9, 2026 4:54 PM

    Bob: The usual survey, whether it be for public opinion or public health or marketing, is about change. Something is…

  37. Andrew on What’s the deal with the RMET (“the Reading the Mind in the Eyes test”)?October 9, 2026 4:50 PM

    OK, so you retract your statement that the person who informed me that I had Tourette's syndrome should be punched?

  38. Olaf Zimmermann on What’s the deal with the RMET (“the Reading the Mind in the Eyes test”)?October 9, 2026 3:22 PM

    https://www.bbc.co.uk/news/articles/crn8wyl4n42yo

  39. Olaf Zimmermann on What’s the deal with the RMET (“the Reading the Mind in the Eyes test”)?October 9, 2026 3:21 PM

    If you think that Freud was silly - well, he wasn't, but he found a way to make money, because…

  40. Olaf Zimmermann on What’s the deal with the RMET (“the Reading the Mind in the Eyes test”)?October 9, 2026 3:06 PM

    May I suggest that you read the entire book before commenting on the merits of the short quote I used?…

  41. Bob Carpenter on Digital twins version 2.0: The skeptics speak upOctober 9, 2026 2:25 PM

    Even though I can see Andrew through my office window, I'll ask this here for everyone's benefit. Andrew---I know you…

  42. Andrew on Digital twins version 2.0: The skeptics speak upOctober 9, 2026 12:54 PM

    Matt: I disagree with you regarding individual baseball predictions. Yes, they're noisy, but they're a lot better than dart throws.…

  43. Matt Skaggs on Digital twins version 2.0: The skeptics speak upOctober 9, 2026 12:50 PM

    " LLMs may do a decent job of predicting population averages for well-represented populations, but they have a long way…

  44. Robin Blythe on Digital twins version 2.0: The skeptics speak upOctober 9, 2026 12:24 PM

    I worry that this push for "silicon samples" misses a fundamental point of why we do research, and survey/opinion research…

  45. Andrew on It’s fine to use computer-generated survey responses—if you don’t care about the data anyway!October 9, 2026 11:49 AM

    Jessica: Agreed. Your point #2 is similar to my reasoning above, that if the purpose of the survey is to…

  46. Jessica Hullman on It’s fine to use computer-generated survey responses—if you don’t care about the data anyway!October 9, 2026 11:45 AM

    I previously wrote a bit about mixed subjects designs here: https://statmodeling.stat.columbia.edu/2025/06/12/better-and-worse-ways-to-mix-human-and-llm-responses-in-behavioral-research-but-you-still-have-to-figure-what-youre-measuring/ There are some more related papers here at my…

  47. Andrew on Digital twins version 2.0: The skeptics speak upOctober 9, 2026 11:26 AM

    I wrote something about this topic last month in the post, It’s fine to use computer-generated survey responses—if you don’t…

  48. Alain on What’s the deal with the RMET (“the Reading the Mind in the Eyes test”)?October 9, 2026 9:48 AM

    I don't think James' point was that classification and individuality can't coexist--it was that classification shouldn't supersede individuality, or, minimally,…

  49. somebody on What’s the deal with the RMET (“the Reading the Mind in the Eyes test”)?October 9, 2026 9:41 AM

    Sure, I'm open to the idea that overdiagnosis is a problem these days. But how could you be so confident…

  50. Benjamin Cowley on What’s the deal with the RMET (“the Reading the Mind in the Eyes test”)?October 9, 2026 6:35 AM

    The RMET is quite dubious "almost science", if not outright pseudoscience. I've never used it or looked into its literature.…

Proudly powered by WordPress