'Nother quick and pointless socionics question

[Cerelict]

Which element(/s) would be most hesitant to dismiss an outside theory/system entirely just because one part doesn't always apply? Which position would said element take in the stack?


"Accounting for other factors, this system may apply"
vs.
"Factors vary too much for such a system to ever be necessary, so it's not relevant even if it does apply"

[Chakram]

Well if a part of it doesn't work, then we should just make a new system where that point doesn't matter right? Or at least say it's new. But then again systems aren't always 100% correct and we need to be able to account for flaws in them. My answer to your question is I have no idea.

[Sol]

By 2 factors it's possible. If practical compatibility of duals will fail - all IR theory is bullshit what is 50% of Socionics. Then we may check correlation of abbilities and dominating function - better in math should be T types, not F, and should not be random, then something for S-N, - another 50%. The problem - everybody types with low match, so even if the experiment will fail we can't be sure - theory is wrong or skew eyes of the typer.
If I've understood your question correctly.