I can't be the only one who hears this, right? #statsmeme

Looks like {ggcube} plots work with {patchwork} syntax, too. Kruschke and I have a few 3D plots in the upcoming DBDA3, and we might just use this. Here's a quick and dirty mockup:

This is a good piece. Julia got a little optimistic in middle, but she really brought it home at the end: “So what would I recommend if a friend actually wanted to run a moderated cross-country regression? I’d probably suggest they spend their time on something more fruitful.” Many such cases...

New blog post: Evaluating Dr. Cuddy’s Claim that the Debunking of Power Posing is a Myth. daniellakens.blogspot.com/2026/05/eval... On an AI generated description of a non-existent study, incorrectly citing findings from studies, and the importance of scientific criticism.

I posted about this Roast Pork Loin with Pineapple recipe a few weeks ago as a troll comment to another thread. But I actually made it this weekend and it's really quite good. memoriediangelina.com/2025/11/07/a...

Happy #ClinicalTrials Day Trial design and conduct is no place for novices with best intentions. Patient participants make sacrifices. Medical progress holds its breath. Large sums of public and private money is invested. So get it right, from the start. statsepi.substack.com/p/dealing-wi...

Smartphones can do a lot, but apparently they cannot keep you from analysing your data in an incredibly wacky manner to feed a certain narrative about smartphones.

#StatsPubs No matter how many times you read it, this paper on hierarchical GAMs keeps giving. Clear model explanations together with R code (mgcv). So useful! doi.org/10.7717/peer... Thanks for writing @ericjpedersen.bsky.social, D. Miller, @gsimpson.bsky.social @noamross.net

woohoo filing away in my “good in-class dag example” folder

I'm going to pass this talk along to the folks in my center. We had a brief workshop on DAGs a few weeks ago, and this is a nice follow-up. Descriptive research is totally cool, but vague causal, JK not causal, but maybe causal LOL language is no longer okay.

Social science researchers need to face up to the fact that a substantial number of their colleagues are just idiots (or grifters) who learned very little during their training and are now winging it on a regular basis.

Done with my last talk before leave! This one was fun and I'm glad I added Columbo last minute. Slides here, in case you're curious: juliarohrer.com/wp-content/u...

#rstats visualization question: This plot shows multilevel shrinkage by comparing intercepts and slopes three models. I'm trying to highlight the effect by emphasizing a few of the groups by displaying their group numbers, rather than as dots. Does this work?

Imagine a short book (~200 p.) introducing Bayesian statistics in the context of clinical trials and drug development. Scope would be introductory -- sort of "your first short course on Bayes". But practical enough to be applied out of the box. What would you like to see covered in such a text? 👇

Do you know a good framework to document data analysis choices? Ex: dropping obs, recoding a var, tuning pars, model + test statistics, etc. Ideally human-readable so people don't have to read my entire codebase. Low-maintenance+ machine-readable for bonus points! #RStats #PyData

Many Labs 5 was such a cool project. Authors who argued that a study from the Reproducibility Project: Psychology was 'too different', leading to a null result, could propose their own replication. Mostly null results again. Blaming auxiliaries is easy. Replication is hard.

some Bayesian cognitive modeling students are having trouble seeing the industry-relevance of the methods (my worked out examples are research based, since it's what I do). Tips for good industry examples? @ajordannafa.com @natehaines.bsky.social @rmcelreath.bsky.social @avehtari.bsky.social

NHS data should be driving innovation & improvement – it’s an amazing resource We just published the sequel to our 2021 "Road to Hell" paper on using NHS data for research... things are worse! Buckle up for a long 🧵about *how COMPLETELY mad* system is for accessing NHS data for research 1/20

Does anyone know of a tidy dataset with a large number of common multi-item self-report psychology scales? Ideally clinical. The data does not have to be within-subject, just a variety of common scales with decent N. I have: - Bainbridge et al - AIID dataset - Ideologies 2.0 dataset

And a new post- on the process of making comprehensive indexes in a Quarto book friendly.github.io/blog/posts/2...

Garofolato (Roman pot roast) is remarkably good. The cloves add something special to the sauce. I threw in a couple bay leaves for good measure, and they fit right in. 9/10 memoriediangelina.com/2015/01/16/g...

Anyone know the history behind this infamous phrase? Who was the first to use it? How did it catch on??

Reminder: If researchers find Cohen's d = 6, no they didn't. trustworthy.scientific.claims/posts/if-res...

I've made an R package for Bayesian Rasch #psychometrics with brms models, easyRaschBayes (on CRAN), implementing simple functions to create figures and tables with model fit metrics, etc. Attaching figures from conditional item infit, item-restscore with GK gamma, and the log-likelihood criterion.

New paper— everyone is collecting intensive longitudinal data but recent review studies report that less than half of studies consider measurement/psychometrics. The paper covers a few foundational psychometric methods for ILD and provides a shiny app to apply them link.springer.com/article/10.3...

Anybody got relevant experience doing marginal effects in the context of within-between/“hybrid”/Mundlak models?

Ahead of next week Insight meeting knocked out a #epinowcast release. package.epinowcast.org/news/index.h... It adds print, summary, and improved plotting to both data and fitting objects...

Several folks rightly pointed out clothing advice depends on context, so I’ll rephrase: What are your clothing do's and don'ts for an ECR man interviewing for a psychology professor gig in Western Europe? [Again, this isn't for me. I really am asking for a friend.]

Ian is smart, and he has integrity. You want to work with him.

Academic Bsky: What are your recommendations on do's and don'ts for clothing in an academic job interview? Suit obviously, but any specifics beyond that? [Asking for a male friend.]

I've been reading a lot of papers on making data AI-ready and what I'm finding is that most of them report practices that were best practices before AI. Good documentation, standardization, interoperable formats, clear naming conventions, etc. Is there something new we should be doing though?

Is there any good empirical work on what colors one might use to depict values (a) below, (b) at, and (c) above zero? My personal default would be something like blue, white/gray, red. But I'd like to move beyond personal defaults to best practices.

Why it’s good to make your data openly available reason #42: Authors emailed saying they need some stats not reported in our manuscript (shame on us) for their meta-analysis. Wrote back essentially right away saying here’s the OSF link. Done!

#stats help: I saw a thread/post here (I think) on how there was this bad practice where in meta-analyses researchers were *inverting* effects so all would have the same interpretation, but those effects would obviously not have the same structure (variance, range etc) Anyone know a source?

I'd be interested in more girl typology. What are the different types of girls? Girl bosses, highlighter girls, horse girls, ..., wine moms, tiger moms, soccer moms. What else? As someone who enjoys personality quizzes, I'd also like to know which one(s) I am.

Well posted. This is the behavior of a mature scientist.

Very pleasantly surprised to see my recent preprint being discussed in this NYT article on silicon sampling and its implications for opinion polling. More forthcoming work on this coming soon! (H/T @rohanalexander.bsky.social) www.nytimes.com/2026/04/06/o...

I recommended this book to a colleague just last week.

To lead in open science efforts, we must engage in open science practices. These papers reporting on the reproducibility et al of science projects aren't even accessible. At least one of the commentaries was also closed access. This is a huge lost opportunity and sets a bad example to all.

Part 2 of my shrinkage estimator series is out! Part 1 covered the univariate case, but now we dive into multivariate shrinkage 🤓 We cover Spearman's classic correlation disattenuation formula, multivariate James-Stein estimators, and hierarchical methods too haines-lab.com/post/how-to-...

Dear causal inference people: Has anybody seen or produced a missingness-graph with different missingness mechanisms *for a reflective measurement model*? Any pointers would be appreciated!

Updating my mixed-effects modelling slides... I have opinions... #statsmeme

Happy "large enough n" day

This is correct, but the problem as always is separating the wheat from the chaff. So with that in mind, what is your favorite niche youtube content from an extremely nerdy perso who excels in their field?

I was rejected from all 15 programs I applied to in 2010. In 2011, I was admitted to 1 of another 15 programs. In both cases, I had high grades, a high GRE score, a developmentally appropriate CV, and several mentors helping with my materials. Mourn the rejection, get a consult, and try again.

Is there some open source spreadsheet that enforces and supports all these recommendations? I'd love to read in a methods section: "We entered and managed all data using OpenSheet with Broman-Woo guardrails."

This is a good reminder that following the demise of bookdown.org, all of my ebooks are now hosted by quarto.pub. You can find the new links here: bsky.app/profile/solo...

Is there a nice way to do this with SPSS? library(marginaleffects) fit <- lm(data = mtcars, mpg ~ disp) predictions(fit, newdata = datagrid(disp = c(100, 200, 300)), hypothesis = difference ~ pairwise)

New blog just dropped! This one is all about estimators—we cover James-Stein, classical test theory, empirical Bayes, penalized regression, and hierarchical models, showing how they all can be used to do a better job than sample stats alone 🤓 haines-lab.com/post/how-to-...