> To me, MBTI or Big 5 (not the actual tests, but their framing of aspects of personality) are mental tool kits for trying to make better predictions from limited data (i.e., the interview process). As a manager, I've found them incredibly helpful for avoiding problems (i.e., assigning the wrong task to a person).
I think of MBTI as a framework for categorizing human tendencies. But the real power of MBTI is in cognitive functions as vocabulary to classify how people will react in situations.
Given a person without major forcing functions changing their behavior, what tendencies do they have?
Cognitive functions defines a set of tendencies that you can have a discussion around, assuming you agree on what each cognitive function really means.
For instance, someone with Ti as a primary function subconsciously looks for correctness that can be validated. Whereas someone with Fi as a primary function subconsciously evaluates whether someone is expressing their true self or not. This can be used as a manager tool to do behavioral risk management.
But everyone has forcing functions that change their behavior, and people with enough practice can get good at anything. As an evaluation tool MBTI probably accounts for at most 10-25% of the outcome. And the older you get, the more well rounded your cognitive stack becomes, making it even less of a predictor of your behavior.
Additionally, cognitive functions are abstract enough to not correlate highly to specific tasks, although one can certainly argue a correlation (I've heard arguments where programming suits Ti and Ni the best).