Daniel 🕹️
LightNews LightNews
banner
strengejacke.de
Daniel 🕹️
@strengejacke.de
580 followers 110 following 410 posts
He/she/it - 's' muss mit.
We're lower than the world!

R easystats project:
https://easystats.github.io/easystats/
Posts Replies Media Videos
Reposted by Daniel 🕹️
andrewjdbell.bsky.social
Want to know how to use MAIHDA to examine heterogeniety in policy outcomes? Here's a nice worked-through example!

How does intersectional identity impact preference for in-person vs online GP appoinments?

Has been great working with @healthfoundation.bsky.social colleagues on this :)
A caterpillar plot of slope variability in the effect of appointment type. Statistically significant differentials are coloured in blue, of which there are 4 positive and 8 negative. The figure is taken from the linked blog figure 5
October 16, 2025 at 8:38 AM
strengejacke.de
These are my #Positron extensions I've currently installed. Are there any "must haves" missing in my list, what other extensions would you recommend etc.? (answers including "python" will be ignored... 😎) #rstats #rstudio #vscode
October 16, 2025 at 9:15 AM
strengejacke.de
St. Pauli
September 27, 2025 at 8:08 AM
strengejacke.de
St. Pauli
September 27, 2025 at 8:06 AM
strengejacke.de
Is there a term for this kind of approach, which maybe also include "simpler" analyses, like the general use of EMMs or comparisons and alike? I always use "pairwise comparisons" as a synonym, because that is what more people know. What I'm actually referring to is a "post-estimation framework" /1
dingdingpeng.the100.ci Julia M. Rohrer @dingdingpeng.the100.ci · Aug 25
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.
September 23, 2025 at 7:24 AM
strengejacke.de
St. Pauli
September 19, 2025 at 5:44 AM
strengejacke.de
St. Pauli
September 19, 2025 at 5:44 AM
strengejacke.de
When you start setting the zoom in Word or Acrobat to 125%, you know it's time for an eye doctor's appointment again.
September 11, 2025 at 12:27 PM
strengejacke.de
This weekend I dug out my very old #starwars #ccg cards (decipher) because my son really wanted to play the game. Now I'm thinking about checking out Star Wars Unlimited, because it's no longer easy to get cards for the CCG. Anyone experience with Star Wars Unlimited? Would you recommend it? #tcg
September 8, 2025 at 5:30 AM
strengejacke.de
R is not Nintendo. #rstats
September 5, 2025 at 9:10 PM
strengejacke.de
Capitalism/populism in a nutshell
scottmccloud.bsky.social scott⚡️mccloud @scottmccloud.bsky.social · Jan 5
Time for my annual repost of this banger by Australian artist Denis Lushch.

Ever-more relevant with each new horrible year.
Three men sit at a table. At left, a destitute man sits with no plate. At right, a working man sits with a plate containing one cookie, listening to the man in the center. Sitting in the center is an old man in a business suit (looking suspiciously like right-wing media tycoon Rupert Murdoch) who is pointing at the destitute man, while speaking to the working man, saying: "Careful Mate... that foreigner wants your cookie!" Sitting in front of the man at center is his own plate, which contains MANY MANY COOKIES.
September 2, 2025 at 10:12 AM
strengejacke.de
This is how table printing in #easystats look like - nice tables out-of-the-box thanks to #rstats packages like {gt} or {tinytable}, which is now fully supported across easystats📦
September 1, 2025 at 6:21 AM
strengejacke.de
Nice thread that gives examples how many research questions can be answered by some kind of estimated marginal means, contrasts/comparisons or marginal effects.

Check out the recent release from the #rstats {modelbased} 📦 and the cool examples shown in the #easystats thread!
easystats.github.io easystats @easystats.github.io · Aug 31
Okay, so you've crunched your numbers and got some awesome statistical models? Sometimes, just knowing "X predicts Y" isn't enough to really get to the juicy bits. That's where the cool post-hoc stuff comes in – think estimated marginal means, contrasts, pairwise comparisons, or #marginaleffects.
August 31, 2025 at 12:54 PM
strengejacke.de
I've discovered that AI can't help you untangle your daughter's knotted necklaces yet.
August 30, 2025 at 10:45 AM
strengejacke.de
St. Pauli
August 29, 2025 at 8:52 AM
strengejacke.de
Am I the only person who really *never* uses ChatGPT? (I use other LLMs 😎 - it's of course always cooler to be a fanboy of the less fanboyed tools)
August 27, 2025 at 8:17 AM
strengejacke.de
That's pretty cool seeing the #easystats 📦 in teaching and daily work beyond your own little cosmos #rstats
profandyfield.com Prof Andy Field @profandyfield.com · Aug 20
I’m about halfway through this update (first 11 tutorials are done). I think they’re a lot better. Using a consistent @easystats.github.io workflow throughout will - I think - massively reduce the cognitive load for students. Looking forward to road testing in autumn term.
profandyfield.com Prof Andy Field @profandyfield.com · Feb 10
Probably no-one except me uses my R tutorials in their teaching, but if you do, I'm re-writing them over the next 6-9 months. My goal is to streamline them based on 5 years of using them in class, but if you have (polite) requests/suggestsions let me have them. www.discovr.rocks/discovr/
August 21, 2025 at 9:17 AM
strengejacke.de
I still don't get how people like more than 3 colours on their smartphone home screen...
August 20, 2025 at 10:52 AM
strengejacke.de
I'm a Bayesian or a frequentist depending on the sample size. As long as the paradigm can answer my research question, and as long as I don't think that absence of evidence is evidence of absence, I'm fine with either paradigm.
August 9, 2025 at 10:01 AM
strengejacke.de
Post-hoc wizardry in #rstats! It's fascinating how many statistical challenges can be solved or research questions can be answered by some form of predictions, marginal means and effects, or comparisons and contrasts. Things you can easily do with packages like {modelbased} or {marginaleffects}!
easystats.github.io easystats @easystats.github.io · Jul 28
How to summarize the total effect of a categorical variable like education? A new vignette shows how to compute absolute and relative inequality with the #easystats {modelbased}📦in #rstats. Get a single, interpretable number to quantify overall group disparities!
easystats.github.io/modelbased/a...
Case Study: Measuring and comparing absolute and relative inequalities in R
easystats.github.io
July 28, 2025 at 8:49 AM
Reposted by Daniel 🕹️
jebyrnes.bsky.social
Modelbased for Quick and Beautiful Model Visualization in #rstats imachordata.com/2025/07/25/m... Thanks, @easystats.github.io!
Modelbased for Quick and Beautiful Model Visualization · I'm a Chordata! Urochordata!
imachordata.com
July 25, 2025 at 8:38 PM
Reposted by Daniel 🕹️
tjmahr.com
bayestestR::describe_posterior() works on rvar columns
screenshot showing the row dataframe with a column of rvars and the markdown-formatted-table output of describe_posterior() + print_md()
July 14, 2025 at 8:52 PM
Reposted by Daniel 🕹️
zx8754.bsky.social
#rstats base ?`$`
library(dplyr) library(purrr) library(bench) set.seed(42) create_df <- function(rows, cols) { out <- replicate(cols, runif(rows, 1, 100), simplify = FALSE) out <- setNames(out, rep_len(letters, cols)) as.data.frame(out) } results <- press( rows = c(1000, 10000, 100000), cols = c(100, 1000, 10000), { d <- create_df(rows, cols) mark( d$f, d %>% .$f, d %>% pull(f), d %>% pluck("f"), min_iterations = 100 ) } ) ggplot2::autoplot(results) bench::mark output plot
May 29, 2025 at 10:58 AM
strengejacke.de
After several years, I noticed that the first author of a co-authored article had corrected what he believed to be a ‘spelling mistake’ in the name of an R package.
July 17, 2025 at 1:15 PM
strengejacke.de
@dingdingpeng.the100.ci Maybe of interest for you, aren't you working on a (more or less) similar topic?
gscollins.bsky.social Gary Collins @gscollins.bsky.social · Jul 16
NEW PAPER in @bmj.com "Dealing with continuous variables and modelling non-linear associations in healthcare data: practical guide"

--> www.bmj.com/content/390/...

#methodologymatters #StatsSky #EpiSky
July 16, 2025 at 3:17 PM