Adam Pettitt, PhD 🏳️‍🌈

A log likelihood is shown. The quantities for the score, wald, and likelihood tests are illustrated with respect to the log likelihood.

You will be visited by three ghosts

Folks who are doing survey analysis (e.g., poststratification, etc.) in R using {survey}, how are you doing fancy multilevel stuff? there is no svylmer

Hot damn if this is as useful as it sounds 💘

I would suggest finding some starter packs and following them!

Still figuring it out myself

📊New study published in Emotion sheds light on affect dynamics in depression. Analysis of 842 adults over 56 days shows negative affect becomes more 'sticky' rather than unstable. This suggests emotional predictability, not instability, may be key. psycnet.apa.org/record/2025-...

📊 #dataviz A couple of years old, but still worth reading--
The Science of Visual Data Communication: What Works

A very thorough review of ideas and evidence
by @SteveFranconeri, @LacePadill and others.
journals.sagepub.com/doi/epub/10....

Doing a deeper dive into reflective vs. formative modeling, thanks to some wonderful reviewers.

#StatsSky #AcademicSky

pdfs.semanticscholar.org/d156/8a44bb3...

Why you are not allowed to say that your 95% confidence interval contains the true parameter with a probability of 95% A shibboleth is a custom, such as a choice of phrasing, that distinguishes one group of people from another. The term goes back to the Hebrew Bible, in which the inhabitants of Gilead identify members...

New post on the 100% Christmas Interval -- after almost 8 years, it's time that we finally talk about CIs.

www.the100.ci/2024/12/05/w...

🚨 New preprint alert! 🚨

Our latest study focuses on identifying adolescents at heightened risk of developing depression by examining how emotion regulation dynamics during this critical developmental period predict future symptoms. osf.io/preprints/ps...

What would a “validity movement” look like in your area? In mine, I’d list these top 4:

- chase down the meaning of established signals with many different methods

- better description

- explore why diff methods converge or diverge on same questions

- question narratives for current signal

You?

📌

Data ellipses (both classical and robust) for Sepal.Length and Sepal.Width in the iris data, by species. The grey ellipses show the pooled, with-group covariance matrix.

HE plot for the multivariate effect of Species, Sepal.Length and Sepal.Width in the iris data, together with hypothesis tests for two contrasts (which plot as lines)

All pairwise HE plots for the variables in the iris data

HE plot for iris data in canonical space. Differences among species nearly entirely confined to one dimension.

#rstats #StatSky Want to understand/explain the results of multivariate tests for linear models?
The {heplots} 📦 visualizes these as ellipsoids for hypothesis end error matrices for an lm(). V 1.7.1 now on #CRAN
{candisc} 📦 shows these in simpler canonical space
🔗See: friendly.github.io/heplots/

At first, I thought you meant you pick a side every night randomly and that sent me into a panic

On today's edition of "Sh*t I wish I knew 8 years ago"

Simple "isolate" function from blml package to get within- and between-person centered versions of a list of variables.

H/t to kzee.github.io/Centering_De...

#Rstats #AcademicSky

Four raincloud plots: a normal distribution, bimodal, skewed, and one riddled with outliers. Popular sample statistics are identical so that mere plotting of the identical means and confidence intervals would miss these qualitative differences.

abstract of pre-print

The Raincloud Quartet (new pre-print)

All N = 111, M = 0.04, SD = 0.27.
One-sided t-tests vs. 0 yield: t(110) = 1.67, p = .049.

Only raincloud plots reveal the qualitative differences, preventing the drawing of inappropriate conclusions.

How did we create the quartet?
Where are data & code?
🧵👇

Sex, sexual orientation, and gender identity data collection across electronic health record platforms: a national cross-sectional survey AbstractObjectives. To assess the current state of sex, sexual orientation, and gender identity (SSOGI) data collection options in US electronic health rec

Despite recommendations & requirements to collect sexual orientation & gender identity #SOGI data in electronic health records #EHR there is WILD variability across health care systems and EHR platforms. This will be critical in the coming years...

academic.oup.com/jamiaopen/ar...

In case no one has told you, go watch Wicked.

Goose chase meme. Top Panel: Suspicious goose asks "biased relative to what?" Bottom Panel: Goose is chasing a man, yelling "Biased relative to what?"

When a study talks about various things that could bias estimates but never explicitly addresses the analysis goal in the first place

$150-$250/hr as the other reply says

We’ve got everything here

Do you like network psychometrics? Are you interested in Exploratory Graph Analysis and related techniques? Check out our new website: r-ega.net

Great to see this paper out!
'Simulation studies for methodological research in psychology: A standardized template for planning, preregistration, and reporting.'

Journal: doi.org/10.1037/met0...
Accepted: osf.io/preprints/ps...

Line plot showing Main effect of Organization-level Inclusion Signaling on LGBT and non-LGBT Employees in Study 1

More inclusive companies are seen as having a better culture for both LGB and non-LGB people

Bar graph with experimental results showing that inclusive signals benefit both non-LGBT and LGBT participants

Using a mix of observational and experimental methods, organization-level inclusion signals had positive effects for both LGBT and non-LGBT employees

osf.io/preprints/ps...

#socialpsyc #psychology #PsychSciSky #prejudice

You need a better key to translate probability to words.
Then, the logit = log (p/1-p) is just another y axis on this chart

1 for sure

Me!

An embedding-based distance for temporal graphs - Nature Communications Temporal graphs can be applied to represent time-evolving complex natural systems, describing the evolution of the interactions between system elements, however quantifying the similarity between temp...

New paper out in Nature Communications: www.nature.com/articles/s41...
With Lorenzo Dall'Amico and @alainbarrat.bsky.social, we define a distance between temporal networks and explore its sensitivity to complex features in empirical temporal network data. @isifoundation.bsky.social

🚨New Preprint on Affective Stress Responses in Daily life (N=248) - w/ @vizecolin.bsky.social and @aidangcw.bsky.social! We used high-density sampling after stressful events to closely examine the dynamic affective response following daily hassles by estimating latent microtrajectories. 1/3