The astrology comparison is dumb as fuck because the Myers Briggs personality rating is based on quantitative data, even if the data is flawed. Many in the academic community (and wannabe nerds) dismiss it due to its lack of empirical basis and use of self-reported personal information, which admittedly is a problem for a number of reasons and especially so for academic purposes.

What they don't acknowledge however is that the data is useful for understanding someone's self-perception, which is still an important thing when assessing personality type. Yes, you might not be able to provide a completely objective or external analysis of a person, but for someone wanting to find out the personality type they perceive themselves to be at a certain point in time, and/or wanting some insight into how they might be responding to their current situation, it can still be useful.

I'm a very consistent INFP :-)

Edit: Speaking as someone in an academic career doing lab research, on a very general level people need to stop blindly fetishising science, its been consistently used to completely fuck indigenous and religious people.