Skip to content
 

On deck through the rest of the year

July:

  • The Ponzi threshold and the Armstrong principle
  • Flaws in stupid horrible algorithm revealed because it made numerical predictions
  • PNAS forgets basic principles of game theory, thus dooming thousands of Bothans to the fate of Alderaan
  • Tutorial: The practical application of complicated statistical methods to fill up the scientific literature with confusing and irrelevant analyses
  • All of Life is 6 to 5 Against
  • He wants to know what to read and what software to learn, to increase his ability to think about quantitative methods in social science
  • Divisibility in statistics: Where is it needed?
  • Joint inference or modular inference? Pierre Jacob, Lawrence Murray, Chris Holmes, Christian Robert discuss conditions on the strength and weaknesses of these choices
  • He wants to model a proportion given some predictors that sum to 1
  • BD reviews
  • The persistence of bad reporting and the reluctance of people to criticize it
  • “Bayesian Meta-Analysis with Weakly Informative Prior Distributions”
  • What happens to your career when you have to retract a paper?
  • The course of science
  • The “Carl Sagan effect”
  • The statistical checklist: Could there be a list of guidelines to help analysts do better work?
  • Data-based ways of getting a job
  • “The idea of replication is central not just to scientific practice but also to formal statistics . . . Frequentist statistics relies on the reference set of repeated experiments, and Bayesian statistics relies on the prior distribution which represents the population of effects.”
  • Where that title came from
  • “A Headline That Will Make Global-Warming Activists Apoplectic”
  • Of statistics class and judo class: Beyond the paradigm of sequential education
  • From no-data to data: The awkward transition
  • Recently in the sister blog
  • Journals and refereeing: toward a new equilibrium
  • How to think about an accelerating string of research successes?
  • What makes Robin Pemantle’s bag of tricks for teaching math so great?
  • Of Tennys players and moral Hazards
  • Revisiting “Is the scientific paper a fraud?”
  • The file drawer’s on fire!
  • Is it really true that babies should sleep on their backs?

August:

  • “Seeding trials”: medical marketing disguised as science
  • China air pollution regression discontinuity update
  • The replication crisis and the political process
  • Don’t call it a bandit
  • Response to Rafa: Why I don’t think ROC [receiver operating characteristic] works as a model for science
  • Let’s be open about the evidence for the benefits of open science
  • “The most important aspect of a statistical analysis is not what you do with the data, it’s what data you use” (survey adjustment edition)
  • A ladder of responses to criticism, from the most responsible to the most destructive
  • “Either the results are completely wrong, or Nasa has confirmed a major breakthrough in space propulsion.”
  • Jeremy Freese was ahead of the curve
  • Discussion of the value of a mathematical model for the dissemination of propaganda
  • “Usefully skeptical science journalism”
  • How feminism has made me a better scientist
  • It was the weeds that bothered him.
  • It should be ok to just publish the data.
  • No, I don’t think it’s the file drawer effect
  • I think there may be some overlap in this Venn diagram.
  • The fallacy of the excluded middle — statistical philosophy edition
  • Let’s get hysterical
  • The competing narratives of scientific revolution
  • The scandal isn’t what’s retracted, the scandal is what’s not retracted.
  • Who spends how much, and on what?
  • Problems in a published article on food security in the Lower Mekong Basin
  • When anyone claims 80% power, I’m skeptical.
  • In statistics, we talk about uncertainty without it being viewed as undesirable
  • Bayesian model comparison in ecology
  • “To get started, I suggest coming up with a simple but reasonable model for missingness, then simulate fake complete data followed by a fake missingness pattern, and check that you can recover your missing-data model and your complete data model in that fake-data situation. You can then proceed from there. But if you can’t even do it with fake data, you’re sunk.”
  • Old school
  • In Watergate, the saying was, “It’s not the crime, it’s the coverup.” In science reporting, it’s not the results, it’s the hype.
  • 3 recent movies from the 50s and the 70s
  • “Identification of and correction for publication bias,” and another discussion of how forking paths is not the same thing as file drawer

September:

  • John Hattie’s “Visible Learning”: How much should we trust this influential review of education research?
  • What should JPSP have done with Bem’s ESP paper, back in 2010? Click to find the surprisingly simple answer!
  • Isaac Newton : Alchemy :: Michael Jordan : Golf
  • Robert Heinlein vs. Lawrence Summers
  • Journalists not running corrections
  • The gaps between 1, 2, and 3 are just too large.
  • Bothered by non-monotonicity? Here’s ONE QUICK TRICK to make you happy.
  • “It’s Always Sunny in Correlationville: Stories in Science”
  • “Check out table 4.”
  • What if a big study is done and nobody reports it?
  • Mouse Among the Cats
  • Narcolepsy Could Be ‘Sleeper Effect’ in Trump and Brexit Campaigns
  • High-profile statistical errors occur in the physical sciences too, it’s not just a problem in social science.
  • Don’t get fooled by observational correlations
  • What to do when your measured outcome doesn’t quite line up with what you’re interested in?
  • The hot hand—in darts!
  • A couple more papers on genetic diversity as an explanation for why Africa and remote Andean countries are so poor while Europe and North America are so wealthy
  • Post-publication peer review: who’s qualified?
  • A psychology researcher uses Stan, multiverse, and open data exploration to explore human memory
  • Multilevel data collection and analysis for weight training (with R code)
  • “Tweeking”: The big problem is not where you think it is.
  • Don’t calculate post-hoc power using observed estimate of effect size
  • You’ve got data on 35 countries, but it’s really just N=3 groups.
  • A potential big problem with placebo tests in econometrics: they’re subject to the “difference between significant and non-significant is not itself statistically significant” issue
  • Yes on design analysis, No on “power,” No on sample size calculations
  • “Imaginary gardens with real data”
  • Statistical Modeling, Causal Inference, and Social Science Regrets Its Decision to Hire Cannibal P-hacker as Writer-at-Large
  • What do you do when someone says, “The quote is, this is the exact quote”—and then misquotes you?

October:

  • A corpus in a single survey!
  • “Moral cowardice requires choice and action.”
  • David Weakliem points out that both economic and cultural issues can be more or less “moralized.”
  • “Six Signs of Scientism”: where I disagree with Haack
  • Not Dentists named Dennis, but Physicists named Li studying Li
  • “Fudged statistics on the Iraq War death toll are still circulating today”
  • Bayesian inference and religious belief
  • Rising test scores . . . reported as stagnant test scores
  • Perhaps you could try a big scatterplot with one dot per dataset?
  • The Golden Rule of Nudge
  • Why are functional programming languages so popular in the programming languages community?
  • From the Stan forums: “I’m just very thirsty to learn and this thread has become a fountain of knowledge”
  • Understanding Chicago’s homicide spike; comparisons to other cities
  • Claims about excess road deaths on “4/20” don’t add up
  • Toward better measurement in K-12 education research
  • The AAA tranche of subprime science, revisited
  • Fitting the Besag, York, and Mollie spatial autoregression model with discrete data
  • Predicting spread of flu
  • An actual quote from a paper published in a medical journal: “The data, analytic methods, and study materials will not be made available to other researchers for purposes of reproducing the results or replicating the procedure.”
  • He’s a history teacher and he has a statistics question
  • Multilevel models with group-level predictors
  • Mister P for surveys in epidemiology — using Stan!
  • “The dwarf galaxy NGC1052-DF2”
  • What to think about this new study which says that you should limit your alcohol to 5 drinks a week?
  • A study fails to replicate, but it continues to get referenced as if it had no problems. Communication channels are blocked.
  • Can we do better than using averaged measurements?
  • Debate about genetics and school performance
  • MRP (or RPP) with non-census variables
  • What does it mean to talk about a “1 in 600 year drought”?
  • Facial feedback: “These findings suggest that minute differences in the experimental protocol might lead to theoretically meaningful changes in the outcomes.”
  • “Evidence-Based Practice Is a Two-Way Street” (video of my speech at SREE)

November:

  • “Boston Globe Columnist Suspended During Investigation Of Marathon Bombing Stories That Don’t Add Up”
  • “Simulations are not scalable but theory is scalable”
  • “We are reluctant to engage in post hoc speculation about this unexpected result, but it does not clearly support our hypothesis”
  • Cornell prof (but not the pizzagate guy!) has one quick trick to getting 1700 peer reviewed publications on your CV
  • The purported CSI effect and the retroactive precision fallacy
  • “Statistical and Machine Learning forecasting methods: Concerns and ways forward”
  • “35. What differentiates solitary confinement, county jail and house arrest” and 70 others
  • Watch out for naively (because implicitly based on flat-prior) Bayesian statements based on classical confidence intervals! (Comptroller of the Currency edition)
  • “Recapping the recent plagiarism scandal”
  • Matching (and discarding non-matches) to deal with lack of complete overlap, then regression to adjust for imbalance between treatment and control groups
  • Hey! Here’s what to do when you have two or more surveys on the same population!
  • “Law professor Alan Dershowitz’s new book claims that political differences have lately been criminalized in the United States. He has it wrong. Instead, the orderly enforcement of the law has, ludicrously, been framed as political.”
  • Chocolate milk! Another stunning discovery from an experiment on 24 people!
  • Robustness checks are a joke
  • The State of the Art
  • “Heckman curve” update: The data don’t seem to support the claim that human capital investments are most effective when targeted at younger ages.
  • “Using numbers to replace judgment”
  • Graphs and tables, tables and graphs
  • Tom Wolfe
  • “The hype economy”
  • A Bayesian take on ballot order effects
  • “She also observed that results from smaller studies conducted by NGOs – often pilot studies – would often look promising. But when governments tried to implement scaled-up versions of those programs, their performance would drop considerably.”
  • Hey! There are mathematicians out there who’ve never read Proofs and Refutations. Whassup with that??
  • The evolution of pace in popular movies
  • These 3 problems destroy many clinical trials (in context of some papers on problems with non-inferiority trials, or problems with clinical trials in general)
  • “Economic predictions with big data” using partial pooling
  • $ vs. votes
  • Multilevel models for multiple comparisons! Varying treatment effects!
  • “And when you did you weren’t much use, you didn’t even know what a peptide was”
  • Stephen Wolfram explains neural nets

December:

  • “James Watson in his own words”
  • The p-value is 4.76×10^−264
  • In which I demonstrate my ignorance of world literature
  • Bayes, statistics, and reproducibility: “Many serious problems with statistics in practice arise from Bayesian inference that is not Bayesian enough, or frequentist evaluation that is not frequentist enough, in both cases using replication distributions that do not make scientific sense or do not reflect the actual procedures being performed on the data.”
  • Niall Ferguson and the perils of playing to your audience
  • A parable regarding changing standards on the presentation of statistical evidence
  • How to think scientifically about scientists’ proposals for fixing science
  • My footnote about global warming
  • Should we be concerned about MRP estimates being used in later analyses? Maybe. I recommend checking using fake-data simulation.
  • Prior distributions for covariance matrices
  • “Do you have any recommendations for useful priors when datasets are small?”
  • Olivia Goldhill and Jesse Singal report on the Implicit Association Test
  • Oh, I hate it when work is criticized (or, in this case, fails in attempted replications) and then the original researchers don’t even consider the possibility that maybe in their original work they were inadvertently just finding patterns in noise.
  • A couple of thoughts regarding the hot hand fallacy fallacy
  • “My advisor and I disagree on how we should carry out repeated cross-validation. We would love to have a third expert opinion…”
  • The Global March of Extrapolation
  • Why do sociologists (and bloggers) focus on the negative? 5 possible explanations. (A post in the style of Fabio Rojas)
  • Classifying yin and yang using MRI
  • When “nudge” doesn’t work: Medication Reminders to Outcomes After Myocardial Infarction
  • Exploring model fit by looking at a histogram of a posterior simulation draw of a set of parameters in a hierarchical model
  • The causal hype ratchet
  • Carol Nickerson explains what those mysterious diagrams were saying
  • “When Both Men and Women Drop Out of the Labor Force, Why Do Economists Only Ask About Men?”
  • Zak David expresses critical views of some published research in empirical quantitative finance
  • “Thus, a loss aversion principle is rendered superfluous to an account of the phenomena it was introduced to explain.”
  • What is probability?
  • Back to the Wall
  • Using multilevel modeling to improve analysis of multiple comparisons
  • “Check yourself before you wreck yourself: Assessing discrete choice models through predictive simulations”
  • Combining apparently contradictory evidence
  • Vigorous data-handling tied to publication in top journals among public heath researchers

If you want the full immersive experience, just turn off this blog for six months, then come back on 1 Jan 2019 and read all these posts at once.

12 Comments

  1. Andrew sure is the fastest thinker I’ve come across. Excluding myself of course. Just kidding you all.

    Seriously, great topic lineup. I’ve been stimulated intellectually. That’s when I’m the happiest.

    Keeping broader themes in motion: Measurement & Theory.

  2. Brian Blais says:

    I’d be very curious to know what your writing process is like for this blog. Have you actually written all these posts ahead of time and automatically post? Do you have outlines for all of them, and fill them in along the way? Do you have a method for queuing them up and automatically posting? It’s an impressive list!

    • Andrew says:

      Brian:

      All but one of the above posts are completely done. When I write a new post and it does not seem time-sensitive, I add it to the queue. The blog software allows me to assign dates to future posts.

    • One of the aspects of Andrew’s blog that is interesting and effective is how he acts kind of as a moderator for an ongoing discussion. When I write blog posts, they flesh-out some particular idea I’ve had, and they often take me a couple hours of total writing time. when Andrew writes blog posts many of them, probably the majority, are a little bit of an opinion, and his invitation for his commenters to discuss a topic brought to his attention by a third party. I’m sure the prolific nature of the blog is in a major way due to this. Andrew is smart and clearly writes quickly, but the time he’d have to spend to read through every branch of every science in order to even know about half the topics would be superhuman.

      We do have one of the best comment sections on the internet here at the blog, and this is in large part because the format is so participatory. Thanks Andrew!

  3. jrc says:

    “Problems in a published article on foot security in the Lower Mekong Basin”

    Wow – I thought people in the Lower Mekong Basin had it bad not having enough to eat. But even their feet aren’t safe? What is this world coming to…?

  4. Mikhail Shubin says:

    Where is Deborah Mayo’s book’s discussion?
    is is scheduled next year? :(

  5. Bob says:

    This list stimulates a variety of emotions:
    1. Impatience (“Robert Heinlein vs. Lawrence Summers”, “3 recent movies from the 50s and the 70s”, Why do we have to wait?)
    2. Wonder—How does he crank all this out and still keep his day job?
    3. Envy—How does he crank all this out at such a high level?

    Bob

  6. Excited about this! I look forward to about fifty specific posts, including “Hey! There are mathematicians out there who’ve never read Proofs and Refutations. Whassup with that??” But on this blog it’s fun to read a post just because it’s been posted, learn from it and the comments, and come back to it over time.

    In my forthcoming book (due October 15) I briefly discuss Singal and Goldhill on the IAT (in my chapter on the phrase “Research has shown”). I look forward to your take.

Leave a Reply