Dr Mircea Zloteanu 🍁 🍃
LightNews LightNews
banner
mzloteanu.bsky.social
Dr Mircea Zloteanu 🍁 🍃
@mzloteanu.bsky.social
3.1K followers 690 following 1.2K posts
Lecturer Psych & Crim @kingscollegelondon.bsky.social | Deception Detection; Emotions; JDM | Open Science; R; Bayes | @ukrepro ReproTea & StatsTea | #statstab | 🇷🇴 🇬🇧🌍

Stats blog: https://mzloteanu.substack.com/
Posts Replies Media Videos
mzloteanu.bsky.social
#statstab #469 Significance tests, p-values, and falsificationism

Thoughts: An "interesting" thread of philosophy of science and debating.

#discussion #debate #pvalue #falsification #nhst

discourse.datamethods.org/t/significan...
Significance tests, p-values, and falsificationism
This thread takes its inspiration from the recent discussions in social science and statistics about significance tests, what they’re good for, whether p-values should be banned, and what all of that ...
discourse.datamethods.org
November 27, 2025 at 7:35 PM
mzloteanu.bsky.social
#statstab #468 Why Summaries of Research on Psychological Theories are Often Uninterpretable

Thoughts: psychology uses significance testing backwards relative to hard sciences.

#meehl #theory #criticism #falsification #popper #methods #nhst #psychology

scienceplusplus.org/metascience/...
scienceplusplus.org
November 26, 2025 at 6:19 PM
mzloteanu.bsky.social
#statstab #467 Hypothesis testing, model selection, model comparison some thoughts

Thoughts: An excellent (but too short) discussion on bayesian inference.

#bayesian #bayesfactor #modelselection #inference #NBHT #BF #ROPE #primer

discourse.mc-stan.org/t/hypothesis...
Hypothesis testing, model selection, model comparison - some thoughts
EDIT: This was an attempt to write guidance. It turns out I stepped quite far from my depth and the text sounded much more conclusive than it should. I think it is correct to currently just classify i...
discourse.mc-stan.org
November 25, 2025 at 6:58 PM
mzloteanu.bsky.social
#statstab #466 Bayesian workflow: Prior determination, predictive checks and sensitivity analyses

Thoughts: Having a good bayesian work flow can be challenging with complex models.

#priors #bayesian #sensitivityanalysis #posterior #ppc #brms

pablobernabeu.github.io/2022/bayesia...
Bayesian workflow: Prior determination, predictive checks and sensitivity analyses | Pablo Bernabeu
This post presents a run-through of a Bayesian workflow in R. The content is closely based on Bernabeu (2022), which was in turn based on lots of other references, also cited here.
pablobernabeu.github.io
November 24, 2025 at 4:34 PM
Reposted by Dr Mircea Zloteanu 🍁 🍃
reproducibilitea.org
✒️ New blog post by Mircea Zloteanu @mzloteanu.bsky.social

💡 Post-publication reviewing - some reflections
reproducibilitea.org/jc/2025/11/2...
Post-publication reviewing - some reflections
There is a pedagogic and pragmatic reason to consider post-publication reviews, which I believe are often dismissed. Science is an ever-correcting process, but also a slow process. Academia is a socia...
reproducibilitea.org
November 24, 2025 at 8:38 AM
mzloteanu.bsky.social
#r #bayes Q: is there a way to easy plot a contrast density using just the {easystats} package library? I just want a simple stats_halfeye() plot from estimate_contrasts() 🤷‍♂️
November 23, 2025 at 9:23 PM
Reposted by Dr Mircea Zloteanu 🍁 🍃
steamtraen.eu
Moral: Be careful what you ask for.
November 15, 2025 at 10:14 PM
mzloteanu.bsky.social
#statstab #465 How to embrace variation and accept
uncertainty in linguistic and
psycholinguistic data analysis

Thoughts: An accessible paper on communicating your results with nuance.

#bayes #bayesian #uncertainty #error #bias #guide #tutorial

sites.stat.columbia.edu/gelman/resea...
sites.stat.columbia.edu
November 21, 2025 at 5:19 PM
Reposted by Dr Mircea Zloteanu 🍁 🍃
tomstafford.mastodon.online.ap.brid.gy
A new paper shows that you can use LLMs to fake survey data - not just successfully bypassing all reasonable checks used by survey researchers but also allowing you to instruct the LLM to fake public opinion in any direction you choose. It’s a strong challenge to any trust we want to have in […]
Original post on mastodon.online
mastodon.online
November 21, 2025 at 8:26 AM
mzloteanu.bsky.social
#rstats #dataviz Q: Does anyone have any code they can share for easily making a plot like this (sensitivity to priors for your model)?
stephenjwild.bsky.social Stephen Wild @stephenjwild.bsky.social · 19d
That's why you need things lkle this (this comes from a paper that reported positive results. Some of them are clearly driven by the prior)
A plot of effect estimates. While most are positive, a few are only positive *because of the prior*
November 20, 2025 at 9:19 PM
mzloteanu.bsky.social
#statstab #464 Plotting p-check interaction {brms}

Thoughts: Annoyingly #brms doesn't natively allow plotting for interactions (that I know of). The forum has a solution.

#ppc #posterior #bayesian #modelfit #diagnostic #rstats #r #stan

discourse.mc-stan.org/t/plotting-p...
Plotting pp_check interactions
I am trying to model ordinal data with an interaction of two predictors. However, I have issues with plotting interactions and individual data with the BRMS functions. My model is of the type ratin...
discourse.mc-stan.org
November 20, 2025 at 8:15 PM
mzloteanu.bsky.social
I spent 3 days meticulously trying to specify and justify my priors, only to end up with normal(0,1) being very appropriate...
a cartoon character is standing in the rain looking up at the sky
ALT: a cartoon character is standing in the rain looking up at the sky
media.tenor.com
November 20, 2025 at 3:33 PM
mzloteanu.bsky.social
#statstab #463 {modelbased} Understanding your models

Thoughts: A deceptively simple case study on how to understand and report your model.

#rstats #modelling #easystats #r #reporting

easystats.github.io/modelbased/a...
Case Study: Understanding your models
easystats.github.io
November 19, 2025 at 8:08 PM
mzloteanu.bsky.social
It was an interesting and (semi-) fun experience! Based on the feedback most teams provided so far, it does seem there has been an improvement in how well research is documented and computationally reproduced.
ukrepro.bsky.social UK Reproducibility Network @ukrepro.bsky.social · 8d
Playing Replication Games
www.ukrn.org/2025/11/18/t...
Last week, 40 researchers from 6 countries joined the 2025 UKRN Replication Games with @i4replication.bsky.social

Huge thanks to Lenka Fiala, Abel Brodeur, Derek Mikola, and all our participants.

#researchintegrity #openresearch
November 19, 2025 at 10:51 AM
mzloteanu.bsky.social
#statstab #462 Factorial Plots

Thoughts: Not the most modern of plots, but nice to have a guide on what to show based on your design.

#plots #dataviz #ANOVA #interaction

www.processma.com/resource/fac...
Factorial Plots | ProcessMA
Factorial Plots can be used to help you visualise how the response relates to one or more factors.
www.processma.com
November 18, 2025 at 6:25 PM
mzloteanu.bsky.social
I used to be proud that I know how to use the em-dash, but now I worry when I do ppl will think "he clearly used an LLM because his 1st language isn't English!" 🙃
talyarkoni.com tal yarkoni @talyarkoni.com · 13d
the internet has decided that em-dashes are a hallmark of llm writing, which is *extremely* annoying to me, as someone who uses em-dashes all the time.

dear internet please consider the possibility that LLMs use em-dashes a lot because they're, like, good
November 18, 2025 at 11:42 AM
mzloteanu.bsky.social
#statstab #461 Interpreting Ordinal and Disordinal interactions

Thoughts: Interactions are not simple things. Their shape can determine many things (including sample size and effect size)

#design #ANOVA #interaction #effectsize #ordinal #crossover

www.jolley-mitchell.com/Appendix/Web...
RDE Ordinal Interactions
www.jolley-mitchell.com
November 17, 2025 at 7:18 PM
mzloteanu.bsky.social
#stats Q: for a bayesian negative binomial (in #brms), who should I think about the prior on phi (shape)? How can I connect it to either theory or stat desiderada?

Examples, tutorials, resources appreciated

(it all seems a bit abstract to me)
November 15, 2025 at 6:44 PM
mzloteanu.bsky.social
#statstab #460 {permuco} permutation tests in linear models with nuisances variables

Thoughts: Supports ANOVA, ANCOVA, t-tests and more.

#permutation #randomization #ANOVA #rstats #r #pvalues #ancova #ttest

jaromilfrossard.github.io/permuco/inde...
Permutation Tests for Regression, (Repeated Measures) ANOVA/ANCOVA and Comparison of Signals
Functions to compute p-values based on permutation tests. Regression, ANOVA and ANCOVA, omnibus F-tests, marginal unilateral and bilateral t-tests are available. Several methods to handle nuisance var...
jaromilfrossard.github.io
November 14, 2025 at 5:34 PM
mzloteanu.bsky.social
#statstab #459 Getting Comfortable with Expressing Beliefs as Distributions

Thoughts: Bayesian stats requires a good understanding of priors, but these are often unintuitive. Plots help.

#bayesian #priors #ggplot #r #dataviz #learing #education

brian-lookabaugh.github.io/website-bria...
Getting Comfortable with Expressing Beliefs as Distributions – Brian Lookabaugh
Thinking about our beliefs as distributions is not super intuitive for most people, which creates a stumbling blog for getting into Bayesian statistics. Check this blog out to break down the mystique!
brian-lookabaugh.github.io
November 13, 2025 at 6:22 PM
Reposted by Dr Mircea Zloteanu 🍁 🍃
mattansb.msbstats.info
This year I'm teaching an advanced stats course for our psych grad students, and I want to squeeze as much causal stuff as I can - but there's just too much!

ATE, DAGs, confounder selection, table 1 & 2 fallacies, collider bias, ...

What else should I squeeze in there?
November 13, 2025 at 11:30 AM
mzloteanu.bsky.social
#statstab #458 Causal inference for observational data using {modelbased}

Thoughts: IPW, g-computation, and more. Learning OS and ways to compute ATE for (more accurate, but still not great) inference.

#gcomputation #ipw #iptw #observational #inference

easystats.github.io/modelbased/a...
Case Study: Causal inference for observational data using modelbased
easystats.github.io
November 12, 2025 at 8:23 PM
mzloteanu.bsky.social
#statstab #457 Using color packages

Thoughts: Some package ideas to make your {ggplot2} figures more interesting.

#dataviz #ggplot #r #figures #plots

bookdown.org/hneth/ds4psy...
D.4 Using color packages | Data Science for Psychologists
This book provides an introduction to data science that is tailored to the needs of students in psychology, but is also suitable for students of the humanities and other biological or social sciences....
bookdown.org
November 11, 2025 at 8:07 PM
mzloteanu.bsky.social
#statstab #456 Shall we count the living or the dead?

Thoughts: survival ratio -> if the intervention increases risk of the outcome
risk ratio -> if the intervention reduces risk of the outcome

#causalinference #riskratios #survivalanalysis #estimand

arxiv.org/abs/2106.063...
Shall we count the living or the dead?
In the 1958 paper "Shall we count the living or the dead?", Mindel C. Sheps proposed a principled solution to the familiar problem of asymmetry of the relative risk. We provide causal models to clarif...
arxiv.org
November 10, 2025 at 7:01 PM
mzloteanu.bsky.social
#statstab #455 {plotROC} Generate ROC Curve Charts for Print and Interactive Use

Thoughts: I dislike ROC curves, just check out the way you show uncertainty on them.

#roc #auc #roccurve #specificity #sensitivity #signaldetection #sdt
#ggplot #r

sachsmc.github.io/plotROC/arti...
Generate ROC Curve Charts for Print and Interactive Use
sachsmc.github.io
November 7, 2025 at 8:43 PM