@fndportal.bsky.social @fndsociety.bsky.social @donbps.bsky.social @fndhope.bsky.social @jonstoneneuro.bsky.social @alancarson15.bsky.social @timnicholson.bsky.social

Thanks so much Norah - this has been really helpful

Thanks Bryan - All really helpful

I would like to find a way to link a database to a html application, allowing users to deposit local normative data into an online repository and scoring application. Shouldn’t be too complicated in principle. Might be considerations for ensuring quality and transparency of data but still doable.

You guys talk about availability of regression based norms, and I do think that the this is one of the primary barriers.
Clinicians haven’t the time to be continuously reviewing literature, and even then it depends on the availability of a scoring tool or skill developing models from published data

Thanks Andrew for the clarification 🙂 very helpful.
I agree, this seems justifiable, although in cases where minor change represents a statistically significant decline in someone with a high baseline, I wonder if the effect-size for change (e.g., z) becomes more telling 🤔

Thanks Bryan - helpful answer and just the reference I needed 🤝

@bryancephalon.bsky.social @andrewk-phd.bsky.social @jeffbrowndyke.bsky.social

I know that some formulas include regression to the mean however, this does not account for the fact the formula doesn't recognise what the achievable limits for each test are.

Does anyone have any references? Or thoughts

In the UK, when training as a neuropsychologist, there is very little teaching and discussion devoted to this. In my experience, an hour or so within a larger day on applied statistics. What is the experience in the US and other countries?

For some I suspect additional factors of superficial understanding, not seeing the value, dogmatic practice, and anxiety towards more sophisticated stats are also bridges to cross

Not as much as I would wish. For me I use the ones that are easiest to integrate into my workflows because they are prompted or immediately available to me (e.g., CVLT in QI).
Learning which others are available, where to find, and how useful they are is hard to find time for when practice is busy.

I would love to have a deeper understanding of this Bryan, hopefully see lots of discussion here on blue sky 🙂.
Initial thoughts…
1) avoids annoying discontinuities between age brackets.
2) superior (I imagine) precision and reduced standard error.
3) adjust for important covariates.

Thanks 🙂♥️

Still new to all this, feedback is welcomed. And if you think this might be of interest to others, please share,

p.s.
Lots of upcoming deadlines, so it might be a while before the next post!