@Tim:

I always cheer for contrarians so I enjoyed your comment, though it would have been much stronger without links to the Journal of Personality Assessment. The JPA is published by the Society of Personality Assessment. The SPA still offers training in such scientifically dubious methods as rorshach tests.

A big part of the problem with type indicators is that they tend to become cult-like, to the point they feed themselves. If I administered Myers-Briggs (and got paid well to do so), I'd have a strong reason to find data which says I'm not a fraud!!!