New post: Diagnosticity and SDT's d′ & c can rank confession techniques differently, but because they answer different questions. Before declaring one metric “better,” check whether they were entered into the same competition. #statistics #SDT #estimands open.substack.com/pub/mzlotean...

"also, do I need to buy a snake to be a coder? 🐍"

@markrubin.bsky.social I'm curious what you think of this paper, as they highlight: "p<α procedure all but compels researchers to misinterpret the p-value as an ‘observed’ or ‘data-dependent adjustable’ Type I error rate" 1/2

I'm in the weights, and seems Claude knows me best :p

#statstab #548 Checking model assumption {easystats} Thoughts: The {performance} package is great at a one-function plot for assunptions. Good explanations also (bug theory limited). #rstats #assumptions #linearity #linearmodel #r #modelselection easystats.github.io/performance/...

How the authors of that paper interpret stats:

🚨 Come to iIIRG 2026 to hear our talk about better methods to analysing interview data! @drcodyporter.bsky.social @mattansb.msbstats.info #belgium #conference #methods #investigativeinterview

#statstab #522 Bayes Rules! Different priors, different posteriors Thoughts: Nice illustration of how uninformative and informative priors change your posterior. #bayesian #dataviz #bayes #inference #sensitivity #priors www.bayesrulesbook.com/chapter-4#ch...

#statstab #503 Reaction time distributions: an interactive overview Thoughts: Might be useful when you start off with this type of research. But I am curious what people think for no-nos with this approach. #reactiontime #rt #distributions #guide #r lindeloev.github.io/shiny-rt/

#statstab #494 {ggsurvfit} Thoughts: "eases the creation of time-to-event (aka survival) summary figures with ggplot2" #hazard #survival #ggplot #dataviz www.danieldsjoberg.com/ggsurvfit/in...

PSA: What google says the formula for Cohen's d with pooled SD is wrong! This formula is only a valid approx when you have equal groups & homogeneity of variance! Otherwise, it gives you a different standardizer: SD_av Don't panic though, often this is what you should report for a Welch's test.

Gym is good to do regardless of your goal. Mine is to not have to buy larger pants every year.

#statstab #485 Bayesian ANCOVA and the ATE Thoughts: Still grappling with the implications of using the causal inference approach to randomized experiments. But it's interesting. #ATE #causalinference #ancova #ANOVA #rstats #estimand #counterfactuals solomonkurz.netlify.app/blog/2025-07...

#statstab #481 Getting over ANOVA: Estimation graphics for multi-group comparisons Thoughts: Complex designs are harder to visualise, but with Estimation Statistics you get some perks over simple bar charts. #design #estimationstatistics #ANOVA www.biorxiv.org/content/10.6...

Compositional data (proportions that sum to 1) behave in ways standard models aren’t built for I walk through why Dirichlet regression is often the right tool & what extra insight it gives using a real ex of eyetracking #Dirichlet #r #brms #guide #eyetracking open.substack.com/pub/mzlotean...

#statstab #465 Linear Regression vs. Logistic Regression Thoughts: Not a very good article but the illustration is nice. Shows why drawing a straight line of best fit is an issue. #logisticregression #linearmodel #regression #MachineLearning #dataviz pub.towardsai.net/machine-lear...

#statstab #463 One-and-a-half sided test Thoughts: Alberson has a more different take on the issues from #461 and #462: human behaviour. (excerpt from the book) #pvalue #onesided #NHST #directional #hypothesis #testing #logic www.routledge.com/Statistics-A...

#dataviz #rstats #ggplot What are people's thoughts on UpSet plots? Easier way to show multiple comparisons in one plot? Or just more plot clutter? share.google/ggr8YOogsnBt...

New clout metric to chase just dropped for #2026 (I guess I need to post more controversial stuff to get my engagement up) Bluefacts.app

#statstab #454 A very ggplot2 Christmas Thoughts: Tis the festive season. 🎄 #ggplot #rstats #xmas #xmastree t-redactyl.io/blog/2016/12...

#statstab #453 Gradient plots {ggdist} Thoughts: Like these as an alternative to half eye plots. #bayesian #dataviz #plots #figures #rstats #r #tidybayes #ggdist mjskay.github.io/ggdist/artic...

🎉🍻🥳 Thank you all for subscribing. More posts to come this year. Subscribe now: mzloteanu.substack.com #substack #statstab #statistics #FiguringStuffOut

R: How many cores do you want to use? Me: Yes. #R #brms #bayesian #rstats

I meant, if researchers plot their results (like in the image) they will more easily understand their effects. You can just see how wide the CI is, what effect thresholds it spans/excludes.

#stats tip: Want to quickly approx the CIs for a reported effect? Goodman & Berlin ('94) to the rescue: - take the sample estimate and use the formulas for 80% and 90% power E.G. If authors anticipated a true δ=.50, obtained d=.40, then for 80%: 0.40 +- 0.7x0.5

Finished presenting at #eurocrim2025, with my partner in crime (research) Josh Bullock We expand our 2024 findings on the relationship bw #TrueCrime media, fear of victimhood, anxiety, and distrust in the Criminal Justice System using an experimental design. #greeece #athens

#statstab #407 Choosing the Causal Estimand for Propensity Score Analysis of Observational Studies Thoughts: Doing obs. research? Do you know the difference b/w the ATE, ATT, ATU, and ATO? #observational #estimand #ATE #biomedical #clinical #effect arxiv.org/pdf/2106.105...