iiinova — where your digital self keeps evolving

Reposted by Tu

Dr Mircea Zloteanu 🌺🌞🍃 @mzloteanu.bsky.social · 23/09/2026

#statstab #623 Six Lemmas Concerning Heterogeneity and Nonlinearity in Causal Inference Thoughts: A very clear treaty of Causal inference problems. #causalinference #ATE #MachineLearning #inference #bias #heterogeneity #lemma #nonlinear #observational math.la.asu.edu/~prhahn/six-...

math.la.asu.edu

six-lemmas-narrative

122

Reposted by Tu

Aki Vehtari @avehtari.bsky.social · 13/08/2026

All four books I've co-authored are freely available online for non-commercial use: - Bayesian Workflow at avehtari.github.io/Bayesian-Wor... Links to other three books are in the quoted post 👇 (too many books to fit in one post!)

avehtari.github.io

Bayesian Workflow book: Website – Bayesian Workflow book

Website for the Bayesian Workflow book by Gelman, Vehtari, McElreath, et al. — case studies, code, and exercises in R and Stan.

8285134

Reposted by Tu

Danielle Navarro @djnavarro.net · 18/07/2026

And so after a very brief hiatus, version 0.7.0 of "learning statistics with R" exists. Full rebuild in quarto, stylistic fixes, "epilogues from 2026" to comment on how the world changed since original publication, and as an added bonus, no longer misgenders the author learningstatisticswithr.com

learningstatisticswithr.com

Learning Statistics with R

217146

Reposted by Tu

Lisa DeBruine @debruine.bsky.social · 14/07/2026

The PsychDS format of codebook could serve as this. I’d love to have a way to read in data in tabular formats like CSV with validated and sensible factor ordering and labels. It would be straightforward to code with this. psych-ds.github.io

psych-ds.github.io

Psych-DS

A specification for psychological datasets. JSON metadata, predictable directory structure, and machine-readable specifications for tabular datasets.

162

Reposted by Tu

Grant McDermott @gmcd.bsky.social · 03/07/2026

𝐭𝐢𝐧𝐲𝐩𝐥𝐨𝐭 v0.7.0 now available on CRAN 🎉 This is big release, with loads of new features. Same tiny footprint, though ;-) Short highlight thread below, borrowing examples from our gallery: grantmcdermott.com/tinyplot/vig...

https://github.com/grantmcdermott/tinyplot/blob/main/vignettes/gallery_figs/bubble-quakes.R

https://github.com/grantmcdermott/tinyplot/blob/main/vignettes/gallery_figs/simpsons-paradox.R

https://github.com/grantmcdermott/tinyplot/blob/main/vignettes/gallery_figs/barplot-meat.R

https://github.com/grantmcdermott/tinyplot/blob/main/vignettes/gallery_figs/spineplot-titanic.R

48726

Reposted by Tu

Solomon Kurz @solomonkurz.bsky.social · 04/07/2026

At a basic level, I think this depends on what you care about. If you want quick pubs for a grade or a promotion, just do t-tests. If you are interested in your data, learn how to fit and interpret ordinal models.

66114

Reposted by Tu

R Works @rworks.bsky.social · 01/06/2026

An incredible 376 #RStats packages made it on to CRAN in April. Do you want to know which had the most interesting applications, thorough documentation, innovative functions? No need to ask Claude, @revojoe.bsky.social already looked through them all meticulously for you: rworks.dev/posts/april-...

rworks.dev

April 2026 Top 40 New CRAN Packages – R Works

An idiosyncratic, unabashedly biased, time-constrained attempt to capture the depth and breadth of the new packages submitted to CRAN in a single month entirely without the use of AI: clearly unsustai...

0124

Reposted by Tu

The BMJ @bmj.com · 21/07/2025

This article provides an overview of the current state of handling continuous variables in healthcare research. It discusses the potential limitations of assuming a linear relationship between independent and dependent variables www.bmj.com/content/390/...

Linear predictor plot for three modelling approaches to analyse continuous variables in a case study of cerebrospinal fluid glucose and acute bacterial meningitis

33918

Reposted by Tu

Richard Riley (R²) @richarddriley.bsky.social · 07/05/2026

⭐ Now published: trilogy of articles proposing how to calculate the sample size for building a prediction model that targets precise & fair predictions at individual level. part 1: binary outcomes - pubmed.ncbi.nlm.nih.gov/40624575/ part 2: time-to-event outcomes - pubmed.ncbi.nlm.nih.gov/41402895/

lnkd.in

This link will take you to a page that’s not on LinkedIn

1166

Reposted by Tu

Institute of Formal and Applied Linguistics @ufal.mff.cuni.cz · 27/03/2026

On the weekend, #EACL2026 continues with workshops and @ufal.mff.cuni.cz folks present their research 👇 Also, don't miss @tuetschek.bsky.social's keynote talk on How (Not) to Find Errors in LLM Outputs at the LowResMT workshop in the morning.

182

Reposted by Tu

Society for Scholarly Publishing @scholarlypub.bsky.social · 31/03/2026

AI is already reshaping how research is done - not just in theory, but in daily workflows. Join us for "How AI Is Transforming Research: From Idea to Impact" to explore real-world use cases, changing behaviors, and what it all means for publishing and research.

sspnet.org

Webinar Preview | How AI Is Transforming Research: From Idea to Impact

Expert speakers share real-world use cases that illustrate how AI is reshaping discovery, reading behavior, and decision-making across disciplines and career stages.

002

Reposted by Tu

R in Pharma @rinpharma.bsky.social · 06/04/2026

📣 The Call for Talks for R/Pharma Virtual 2026 is now open! If you're working with R, Python, or open-source in Pharma, we’d love to hear from you. Submit by May 2nd: sessionize.com/rpharma-2026/ www.linkedin.com/company/open... #RStats #Python #Pharma #DataScience #ClinicalTrials #AI #LLMs

sessionize.com

R/Pharma 2026: Call for Speakers

075

Reposted by Tu

Chenxin Li, PhD @chenxinli2.bsky.social · 12/11/2024

Here is a thread showcasing my GitHub repositories: 1) Friends Don't Let Friends Make Bad Graphs. An opinionated essay on good and bad graphs. My popular one by a long shot with 6.4k stars and 248 forks. github.com/cxli233/Frie...

1222871

Reposted by Tu

Big Book of R @bigbookofr.com · 07/03/2026

Complex Surveys: A Guide to Analysis Using R by Thomas Lumley #RStats bigbookofr.com/chapters/social%20sc…

122

Reposted by Tu

Health Nerd @gidmk.bsky.social · 26/11/2025

The stuff you find when you actually read the RCTs in a systematic review... This paper is one of the foundational studies on vitamin D to prevent respiratory infections in kids. Cited 1,400 times as per Google Scholar.

58925

Reposted by Tu

ArXiv Paperboy (Stat.ME+Econ.EM) @paperposterbot.bsky.social · 27/07/2024

Efficient Causal Graph Discovery Using Large Language Models arxiv.org/abs/2402.01207 We propose a novel framework that leverages LLMs for full causal graph discovery. While previous LLM-based methods have used a pairwise query approach, this requires a quadratic number of queries which q 📈🤖

011

Reposted by Tu

Phileas 🌿🔥🦦🔥🌿 @phileasdg.bsky.social · 13/10/2025

I think one difference between beneficial uses of ai and a cognitively detrimental ones is often who the expert is in the interaction. If he human is the expert, then the ai can be a helpful tool for ideation assistance, sanity checks, etc. If the ai is the expert, you have to be much more careful.

183

Reposted by Tu

Julia M. Rohrer @dingdingpeng.the100.ci · 17/09/2025

You're very welcome @vincentab.bsky.social P.S. Check out our preprint on an alternative to staring at coefficients: j-rohrer.github.io/marginal-psy...

Manipulated theatrical release poster of "The men who stare at goats" which now reads "the men who stare at coefficients"

Top shows the profiles of George Clooney, Jeff Bridges, Ewan McGregor, Kevin Spacey and a goat.

Below the text: the men who stare at coefficients with the silhouette of a man sitting in front of a computer screen, and the silhouette of a goat

28218

Reposted by Tu

W. Joel Schneider @w-joel-schneider.bsky.social · 11/09/2025

Recreating each of the tables in Chapter 7 of the Publication Manual of the American Psychological Association (7th Edition) tables in R with apa7 (Part 1 of 24) #rstats #apastyle wjschne.github.io/posts/apatab...

An APA-formatted table showing descriptive statistics

011

Reposted by Tu

W. Joel Schneider @w-joel-schneider.bsky.social · 20/08/2025

Now on CRAN, ggdiagram is a #ggplot2 extension that draws diagrams programmatically in #Rstats. Allows for precise control in how objects, labels, and equations are placed in relation to each other. wjschne.github.io/ggdiagram/ar...

Trigonometric functions and a unit circle

A bivariate change model with structured residuals

A hierarchical model of cognitive abilities

1018074

Reposted by Tu

Vilgot Huhn @vilgothuhn.bsky.social · 05/09/2025

Following up on this, I spent yesterday evening writing a blogpost with DAGs and R code (yes, I postponed playing Silksong) about mediator-outcome confounding. I chose to call this scenario "mediation by tautology" but would still be happy to find some established term. #stats #rstats

vilgot-huhn.github.io

Establishing mediation is difficult even in the best of all worlds – Vilgot’s website

3174

Reposted by Tu

Julia M. Rohrer @dingdingpeng.the100.ci · 25/08/2025

Ever stared at a table of regression coefficients & wondered what you're doing with your life? Very excited to share this gentle introduction to another way of making sense of statistical models (w @vincentab.bsky.social) Preprint: doi.org/10.31234/osf... Website: j-rohrer.github.io/marginal-psy...

Models as Prediction Machines: How to Convert Confusing Coefficients into Clear Quantities

Abstract
Psychological researchers usually make sense of regression models by interpreting coefficient estimates directly. This works well enough for simple linear models, but is more challenging for more complex models with, for example, categorical variables, interactions, non-linearities, and hierarchical structures. Here, we introduce an alternative approach to making sense of statistical models. The central idea is to abstract away from the mechanics of estimation, and to treat models as “counterfactual prediction machines,” which are subsequently queried to estimate quantities and conduct tests that matter substantively. This workflow is model-agnostic; it can be applied in a consistent fashion to draw causal or descriptive inference from a wide range of models. We illustrate how to implement this workflow with the marginaleffects package, which supports over 100 different classes of models in R and Python, and present two worked examples. These examples show how the workflow can be applied across designs (e.g., observational study, randomized experiment) to answer different research questions (e.g., associations, causal effects, effect heterogeneity) while facing various challenges (e.g., controlling for confounders in a flexible manner, modelling ordinal outcomes, and interpreting non-linear models).

Figure illustrating model predictions. On the X-axis the predictor, annual gross income in Euro. On the Y-axis the outcome, predicted life satisfaction. A solid line marks the curve of predictions on which individual data points are marked as model-implied outcomes at incomes of interest. Comparing two such predictions gives us a comparison. We can also fit a tangent to the line of predictions, which illustrates the slope at any given point of the curve.

A figure illustrating various ways to include age as a predictor in a model. On the x-axis age (predictor), on the y-axis the outcome (model-implied importance of friends, including confidence intervals).

Illustrated are
1. age as a categorical predictor, resultings in the predictions bouncing around a lot with wide confidence intervals
2. age as a linear predictor, which forces a straight line through the data points that has a very tight confidence band and
3. age splines, which lies somewhere in between as it smoothly follows the data but has more uncertainty than the straight line.

461000285