Siwei Liu

Happy to share that our large-scale network analysis is now out in @nathumbehav.nature.com

We show that networks are often supported by too little evidence from the data for results to be reported with confidence, not meaning that results are flawed but rather suggests caution in interpretation.

Great study! A general implication is that when we infer effects of retrospectively measure variables on outcomes, we’re largely just seeing the effects of how people are currently feeling.

Wow, congrats!

Rethinking measurement invariance causally Highlights: It is preferable to work with a causal definition of measurement invariance A violation of measurement invariance is a potentially substantively interesting observation Standard tests for measurement invariance rely on strong assumptions Group differences can be thought of as descriptive results

Conceptual graph illustration the central points of the manuscript. A group variable is potentiall connected to a construct of interest which affects items. Measurement invariance is violated if the group variable directly affects the items, for example by modifying the loadings from the construct to the items, or by directly affecting an item

To make this less abstract, consider a scenario where students take an exam, R, meant to capture some ability, T, and then are admitted to a program, V, depending on their exam results: R → V. This is sufficient to result in a violation of the statistical definition of measurement invariance. Exam results and admission are not independent given ability because exam results have a direct effect on admission. Even if we know somebody’s ability (e.g., we know it’s very high), learning about their admission status (e.g., they were not admitted) can tell us something about their exam result (e.g., it may have been worse than expected). According to the causal definition, this in itself does not constitute measurement bias, which seems a sensible conclusion here. After all, the scenario does not involve any reason to believe that the measurement process varied systematically by admission status. Admission happens after the exams took place, it cannot retroactively influence the measurement process (and, for example, lead to unfair treatment depending on admission status).

New paper out with @boryslaw.bsky.social 🥳 In which we sketch out how to rethink measurement invariance causally for applied researchers. And provide a causal definition of measurement invariance!

www.sciencedirect.com/science/arti...

Assistant Professor, Psychological Sciences - Flagstaff, Arizona, United States About the Department/College The Department of Psychological Sciences is located on the Flagstaff Mountain Campus of Northern Arizona University (NAU), situated at the base of San Francisco Peaks. NAU...

Tenure-Track Quant Psyc job opening at Northern Arizona University in Flagstaff. Areas of interest are pretty broad (SEM, multilevel, or psychometrics), the deadline to apply is coming up soon (Sept 15) if you're interested!

careers.nau.edu/jobs/assista...

AMPPS Call for Papers: Replicability and Reproducibility in Methodological Research. Proposals due September 15. @jkflake.bsky.social 

Please add me.

Great post! I just read this paper by @drewhalbailey.bsky.social and colleagues that shows the RI-CLPM also performs better than CLPM when there are unmeasured time-varying confounders:

psycnet.apa.org/record/2025-...

This work was officially accepted for publication today!

This is a required reading in my Intro to Research Methods class.

Love it!

Figure 1. Trends in Biannual US Infant Mortality Rates, 2012-2023

🧵 US states that implemented abortion bans saw higher than expected infant mortality rates, with larger increases among Black infants and those in southern states, according to this analysis of US national vital statistics data from 2012–2023.

ja.ma/4aVchPn

#MedSky

😭

Are psychometric networks sufficiently supported by data such that one can be confident when interpreting its results? We analysed 294 psychometric networks from 126 papers with the Bayesian approach to address this question @jmbh.bsky.social Sara Ruth van Holst @maartenmarsman.bsky.social 🧵

I don’t know if it’s an addition, but drinking too much milk tea (aka bubble tea, which is not just milk and tea but usually high in sugar) is a common problem in Chinese teenagers and a legitimate public health concern. It’s the same problem that Americans have with soda.

7 steps to junk science that can achieve worldly success
statmodeling.stat.columbia.edu/2025/01/17/7...

It’s fascinating to watch the exchanges between “TikTok refugees” and Chinese netizens on Rednote (Xiaohongshu)! Never imagined the “wall” would start to collapse in this way. Now I just worry that Rednote will get banned here if it becomes too popular😕

@dmcneish.bsky.social has a recent paper on this topic in which he uses a data set on binge eating behavior. It’s published in Psychological Methods.

Non-native English speakers need 50% more time to write a paper

When they do, they face a 2.5 times higher chance of being rejected because of language

journals.plos.org/plosbiology/...

Wow. I’ve expressed concerns about Frontiers for a long time and won’t submit or review for them, but this suggests things are really bad.

Congratulations!

Hi all! I'm looking for a short, accessible reading for 1st year UGs that covers:
1) contributors to the replication crisis(e.g., p-hacking, publication bias, researcher degrees of freedom)
2) initiatives to address it (e.g., preregistration, sharing data and code, registered reports).

Any tips?

Introducing PowerLMM.js!

A new tool for power analysis of longitudinal linear mixed-effects models (LMMs) – with support for missing data, plus non-inferiority and equivalence tests.

powerlmmjs.rpsychologist.com

Would really appreciate your feedback as I refine this app! Details below 🧵👇

2025 PQPC RFA Request for Applications 2025 Pacific Quant Psych Conf (PQPC) June 26-27 2025 UC Davis Center for Mind and Brain in Davis, CA We would like to build a community of researchers on the west coast (and...

West-coast (ish) Quant-Psych(ometric) friends! Ben Domingue and I are organizing a small conference to be held in Davis in June. Please apply to attend, we're going to have so much fun.
docs.google.com/document/d/1...

😅

That’s my impression too. But most students are already familiar with R so it’s really tempting to switch. Maybe I’ll give it a try and see how it goes.

That’s interesting. I have found that nlme is less likely to produce convergence warnings. Combined with what Aidan said, it seems nlme is preferable.

I am considering switching from SAS to R in my graduate-level multilevel longitudinal data analysis class. Pro: R is free! Con: R is less flexible in terms of specifying the residual covariance matrix. Anything else I need to be aware of?