Is MBTI Fake And Gay? The FAQ

Q: Why use MBTI when it's pseudoscience like Horoscope, especially when the superior Big Five exists?
A: Both psychological typologies are valid in their ways each. And MBTI, despite mainly being useful as a tool for self-reflection, has statistical validity to it as well.
>In this paper we provide for the first time a comparison of Big Five and MBTI from a personality computing perspective. To do so we use two multilingual Twitter datasets, one annotated with Big Five classes and one with MBTI classes. For the first time, we provide an evidence that algorithms trained on MBTI could have better performances than trained on the Big Five, although the Big Five is much more informative and has great variability in performance depending also on the algorithm used for the prediction
https://www.researchgate.net/publication/332298018_Is_Big_Five_better_than_MBTI

Q: The science is in a replication crisis and the MBTI fields can't be an exception, what now?
A: Indications of validity of MBTI that come from results showing consistent correlations have been replicated through different scientific methods each, which are all publicly available for learning.
>This paper presents a generic evaluation framework for Large Language Models to assess human personalities based on Myers-Briggs Type Indicator (MBTI) tests.
>Experiments show the feasibility of using ChatGPT and InstructGPT to assess human personalities. It was found that ChatGPT can achieve more consistent and fairer assessments with less gender bias.
https://arxiv.org/pdf/2303.01248.pdf

Q: Results can be manipulated just for researchers' portfolio-making purposes and personal profit, how is this legit?
A: In total, different countries' institutions and teams of authors were involved. Given the ease of amateur replication of some methods and poor potential to profit from MBTI through large official systems, a conspiracy is highly unlikely due to inefficiency, low gratification value and whatnot.
>Myers-Briggs personality classification from social media text using pre-trained language models
>text examples illustrate correctly classified instances for each MBTI type, in which the most relevant (word) features are highlighted
https://arxiv.org/pdf/2207.04476.pdf

Q: Hasn't MBTI been debunked as pseudoscience by officially verified experts?
A: The hivemind of officially certified experts has yet to even approach debunking the statistical studies in any form or merit.
>Personality Type Based on Myers-Briggs Type Indicator with Text Posting Style by using Traditional and Deep Learning
>this study was able to predict personality utilizing social network data with a machine learning algorithm that we applied
https://arxiv.org/pdf/2201.08717.pdf

Q: But isn't the available research like in the examples above has been based mostly on 16p, which is a grotesque simplification of MBTI that omits focus on cognitive functions?
A: By extension of logical and rational reasoning, introspection-requiring precursors of 16p - cognitive function-based MBTI and Jung's Psychological Types - would be even better in making consistent sense and being useful for self-contemplation.