@anoni
Jung typology isn't about MBTI, MBTI is a resellable remake of it, like Socionics. I am referencing classic Jung typing, which is nonetheless classed as MBTI now.
Just because something is sold doesn't make it wrong, most science papers are expensive to get a hold of and what is noteworthy legally is registered under a paid government-licensed roll. Statistics are a result of people, many answer tests as the ideal self which I agree is inconsistent and mood dependent. But I know my functions have never changed, they simply can't. Order is my issue. How worthwhile MBTI is is indeed a concern but for the New Age and as a community thing there is nothing wrong with it, it's become to be about self improvement.
If you answer the questions wrong each time, that is your inconsistency, tests are about preference - how you've always been. Forer effect is affected by the dynamics of changing action, yet it is extremely rare for people to change their cognitive style considerably, even through trauma. Minds don't change, philosophy and wants can. I don't know what convinces people that the psychology that has survived the test of time is more defining to psychological reality than the words of a few recent people, even neurology shows we have cognitive bias which in fact demonstrates what Jung expressed.
If anything, typology is more subject to the Galatea effect, for example those who get typed as thinkers and see themselves as thinkers started to express more of this expectation and excel better at it when working, feelers started to believe they were more people persons. However, T and F is not actually to do with what kind of people you are, everyone thinks and everyone feels. MBTI isn't about self-efficacy, it's about finding what you will enjoy better, career-wise, etc.
No theory is 100% verified, that said, most modern particle and cosmological theories are constantly debunked then reaffirmed. When done right, MBTI is actually very accurate. I'm not say it is right 100% but it is substantial.
Contrast it with Model A Socionics, based on a reinvention of Sigmund Freud's egos, both theories are compatible but states mode of thought in quadra. If you check out some official archives and communities around the theories, you'll understand that MBTI is so dependable in most cases, that it is ever increasingly mainstream. You can't disprove MBTI on people's misuse anymore than you can disprove mathematics by those who all keep getting their paper wrong.
For the most part I'm not overly bothered by the legitimacy of the MBTI so much as that is an interesting thing, to see what adds up from the inside out, I've always enjoyed people and how they relate to each than rather than math, people are far more complex.
Certainly, I do agree, however, a test simply isn't reliable due to not so much cognitive bias or idealism as philosophical, post-test belief and situational ones.