Scot Morrison

Would be fun to place multiple sensors and use a lactate test and met cart to see if there’s a good threshold for # of sports to estimate a system level response

Thanks Jem, appreciate the reference. I still mostly use this during a step test to identify thresholds and/or compare with lactate. The bilateral variance made me wonder about the error that adds to the estimate.

Interesting thread on measuring SmO2 using NIRS devices and bilateral differences.

Isometric torque ≠ strength. Strength mediates recovery if recovery includes restored function & force is a limfac. Dynamometer testing, done well, adds validity evidence to the inference that a patient is adapting towards goals. It’s a proxy that helps us iterate ExRx faster.

A human bioactuator inspects the entire factory and turns a single screw, fixing the problem. The next day, he bills the company $10,000.

“$10,000? My AI could’ve figured out the problem in an instant!”

The bioactuator relied: “It’s $1 for knowing which screw to turn, and $9,999 for turning it”

While we often believe that the number we get from our test represents the construct we care about this is rarely true. Eg. Measuring “strength” as example

One of the best

Do Therapists Know When Their Clients Deteriorate? An Investigation of Therapists' Ability to Estimate and Predict Client Change During and After Psychotherapy In routine outcome monitoring, psychotherapists receive feedback from their clients about their self-reported progress during therapy. This practice is based on research indicating that therapists ov....

More evidence that "therapists cannot rely on their clinical judgement alone to assess client progress and outcomes and will depend on routine outcome monitoring to detect client deterioration".

Sample size for developing a prediction model with a binary outcome: targeting precise individual risk estimates to improve clinical decisions and fairness When developing a clinical prediction model, the sample size of the development dataset is a key consideration. Small sample sizes lead to greater concerns of overfitting, instability, poor performanc...

Developing a prediction model?

Ever wondered how to target sample size to improve model fairness and precision of risk estimates?

Check out our new pre-print & software package pmstabilityss
arxiv.org/abs/2407.09293

Interested in learning/teaching psychometrics with Rasch Measurement Theory and Item Response Theory? I’ve made CC-BY intro materials (3h lecture) using #rstats and #quartopub with slides here (code repo link on slide 4): pgmj.github.io/RaschIRTlect...

I recently made a comic complaining that NASA refuses to listen to my good ideas for improving the Solar System (xkcd.com/2750).

To my delight, NASA’s Science Mission Directorate has sent me an actual expert panel evaluation of my “flatten the planets” proposal! Sadly, they decided not to fund it.

STORK (storkinesiology.org) are back with a journal club on Feb 7th!

We will kick off with a discussion of chpt 1 on philosophical concepts in sport and exercise science from the forthcoming STORK book. Available: doi.org/10.51224/SRX...

Register: mcmaster.zoom.us/meeting/regi...

All welcome!

Delightful. Enjoy!

Cooked in an iron skillet for the corn bread, with a ham hock or bacon & hot sauce for collards? Sounds like my childhood.

For the most part people seem to find @jamesheathers.bsky.social foul language the least offensive aspect of listening to him talk.

"We need less research, better research, and research done for the right reasons." --- Doug Altman

The Scandal of Poor Medical Research
www.bmj.com/content/308/...

#metascience #openscience

🧵 Thread 1/n

I suspect much of the proxy failure we see in sports science is due to our failure to identify proxies that map reasonably well onto the constructs we care about. Maybe we should reduce discussions on goodharts law and just get better at using validity evidence to modulate our inferences

2 types of editor responses when submitting measurement work to clinical journals:

1) If editor > 95 years: “Desk-reject. This is well known, see my PhD from 1931.”
2) Else: “Desk-reject. Measurement has no clinical relevance and should be published in specialized measurement journals.”

This is a depressing observation. It depresses me most because there's a bunch of people who focus on the HARKing/p-hacking to the exclusion of other problems and their cause

Yes exactly! I've previously distinguished between "active" and "passive" HARKing to try to get at some of the issues you raise.

doi.org/10.1037/gpr0...

There Are No Magic Outcome Variables I'm very sorry but I am going to write about statistics and causal inference again. I'd much rather be doing science. I'll make it brief. I was reading a preprint about the relationship between kinsh...

All week I was planning to write a new blog about selection bias. But no time it seems. So in the meantime, here is older blog on important problem of not using back-alley statistical control but rather thinking causally. https://elevanth.org/blog/2022/09/02/there-are-no-magic-outcome-variables/

I’m on the other extreme now mostly drinking very mild coffees. That’s the fun of the drink - so much variability

It’s a sweet lens from my exploration of it. Recently switched to a Z mount from my D810 and so far impressed but now begins the decisioning about replacing my glass or just sticking with the adapter…

Honestly, Chmess by Dennett is so worth the 13 minutes it takes to listen to it in our latest podcast (or the 8 to read it yourself). It completely changed which papers (or social media accounts for that matter) I engage with. https://nulliusinverba.podbean.com/e/prologus-13-chmess-d-c-dennett/