Shooting from the hat with AI driven precision and accuracy to understand reliability and validity: UDLCO CRH

Conversational Transcripts:

[06/06, 09:18]AY: See the beauty of AI's language, where it constructs the sentence "Experts know what they don't know"

[06/06, 09:56]cm: Knowing what's unknown is the crux of "Meta cognition"?

[06/06, 11:19]ay: How do you guarantee validity (val idi _iti_) of a PaJR?

[06/06, 11:27]cm: The word validity is etymologically derived from the Latin validus meaning strength which uses the Sanskrit root Val aka Bal or in Bengali Ball!

So validity is in a way "might is right!"

[06/06, 11:28]ay: And idi

What about iti? 😃

Iti is Sanskrit and derived.

Choose idi from whatever is normatively cognate

[06/06, 11:29]cm: Eta ota

Iti is shesh

[06/06, 11:30]ay: Will tackle this over the weekend 🙂

[06/06, 11:32]cm: In Western theory of research evidence generation, the concept of the power of a study often determines the bal (validity) of the study but unfortunately study power itself currently probably is simply restricted to sample size?

Perhaps something more about this can be worked out over the weekend?

[06/06, 11:33]ay: That's because our (geolocated and geotagged linguaphonetically) research often reaches the waste urn!

[06/06, 12:31]si: Nice representation of validity & reliability: 

[06/06, 16:07]cm: The same is used as diagrammatic markers for precision and accuracy!

Does that mean precision medicine is all about reliability and not so much about accuracy and perfect medicine could be about both?

https://www.researchgate.net/figure/Precision-versus-accuracy-The-bullseye-represents-the-true-value-eg-the-true_fig6_304674901

[06/06, 18:09]ay: 😭 this is not what and how we teach shooting. 

Grouping is important. It's Valid! 

All that is needed is a minor adjustment of "sights".

PS: Grouping is accuracy and precision combined.

Accuracy is of the posture and breath, and precision is of "sighting" and reaction.

[06/06, 18:19]ay: How AI explains it...

The infographic labelling tight groupings offset from the bullseye as “high precision but low accuracy” may be visually appealing but is conceptually misleading, especially when applied to research. In reality, tight groupings indicate both control and consistency—hallmarks of sound method and trained execution. If the shots are consistently clustered, the issue is not with accuracy of posture or breath (i.e., the research design and methodology), but with the sight alignment—akin to a minor misalignment in hypothesis framing or measurement scale. Dismissing such groupings as "inaccurate" is analogous to rejecting nuanced, repeatable research outcomes simply because they don’t conform to a pre-assumed center—often a stereotype or oversimplified standard. It mirrors the problematic practice of discarding data as “outliers” when in fact they may reflect expertise navigating edge cases or reveal important deviations. True accuracy emerges from refining the aim, not questioning the shooter’s competence when they’re grouping tight. Let’s not confuse deviation from norm with deficiency in method.

https://chatgpt.com/share/6842e394-8404-800f-85e8-ff6d9f97f88f