An invalid analysis is, well, invalid.

Curious what we're covering in our curriculum? Check out a list of topics to see what we're talking about with all this rigor stuff! c4r.io/project-update…

🏋🏻Resistance training decreased valgus & internal rotation moments during unplanned #sidesteps in elite #female athletes ❓we compared #dynamic vs #isometric resistance training and its effect on knee joint kinetics associated with #ACL injuries read.qxmd.com/read/39186066

A sample of articles on natural experiments, calculating and interpreting prediction intervals, the misuse of the bayes factor, analyzing ecological momentary assessment data, and much more. psychologicalscience.org/news/releases/…

This reanalysis is unserious and low quality: - No confidence intervals or p values to base conclusions on, just point estimates - Averages effect sizes without weightings - No test of moderation, which is the claim they state they want to test 🚩 🚩 🚩

My blog post critical of this re-analysis is up: matthewbjane.com/blog-posts/blo… The re-analysis conducted by Rausch & Haidt does not follow a principled statistical approach. I re-do their re-analysis with appropriate statistical methods

An under thought about point in meta-analysis

Go Michael Roeger !!! #Paris2024 AUS Paralympic Team

Arguing with public intellectual types is a waste of time because they have no interest in rigor. They are trying to sell a narrative and are functionally no different from used car salesman.

Sometimes I see a post trying to use AI to do something and all I can think is that there is nothing stopping you from using a saxophone to slice a cake. Nothing.

#statstab #171 Guideline of Selecting & Reporting Intraclass Correlation Coefficients for Reliability Research Thoughts: "There are 10 forms of ICCs." Are you reporting the correct one? Find out! #ICC #modelcomparison #reliability #interraterreliability ncbi.nlm.nih.gov/pmc/articles/P…

The re-analysis of social media's impact on mental health is a perfect example of how flawed methods can lead to misleading conclusions. Kudos to MatthewBJane for redoing the analysis with proper statistical rigor!

"Science loses credibility when authors draw the wrong conclusion from their data" - Kristin Sainani

First, as sort of a meta-discussion — this is how science 𝘴𝘩𝘰𝘶𝘭𝘥 be done in my view. This is a question that arguably matters a lot for society. It's important we get it right, and we're transparent in our methods.

If studies are independent, the rules of garbage math suggest you get slightly better garbage. Garbage nonetheless though 😟

sciencedirect.com/science/articl… Stage 1 of the INSPECT-SR project, now published in Journal of Clinical Epidemiology

1) a "random" effect is a statistical tool for performing shrinkage/smoothing/regularization. 2) it's called "random" because you assume that the effects are draws from distribution and you want to estimate the hyperparameters of that distribution

📢 Calling all thermoregulation & hydration researchers! University of Arkansas and UARK Health, Human Performance and Recreation is seeking an Assistant Professor to join their team. Apply now & share widely please! uasys.wd5.myworkdayjobs.com/en-US/UASYS/de…

Our paper has now been published (open access)! rdcu.be/dS1p5 Bottom line: More effort is needed to clarify the rationale for and improve the reporting of multiple imputation for estimation of causal effects from observational data. Clinical Epi & Biostatistics Unit (CEBU) ViCBiostat