>>11497165To be honest, I have a somewhat unconventional view. I work with organisational psychologists, but personality isn't my area of expertise.
I see the MBTI (and derivatives) are reasonably valid, and in some cases even the interpretations (e.g.,
16personalities.com) are sensible. The issue is how we define what we really mean by valid? Is it a scientifically accurate model of exactly how an individual person will operate in a given situational context? No.
However, it should be acknowledged that overall, the categories do tend to split people up into sensible and discrete socio-behavioural patterns. The scale has decent test-retest reliability. Personally, I do find that with a bit of judgement and thought, it tends to accurately describe my behavior quite well. In terms of a model, I would say that for self-understanding, it is relatively useful.
However, if we are talking about the index being useful for organisational psychology (or for any serious application), then there are a whole host of issues...
One core problem is that it assumes each scale is binary. This arbitrary splits individuals, and this is quite problematic when people are on the border. If you look at personality reads they can be dramatically different for just 1-5 points scale difference on say T/F. Most people end up hybrid.
More broadly, many of these personality tests are completely bullshit if people aren't answering honestly (e.g., for self interest). My view, and that of many organisational psychologists is that faking on job-related personality tests is fine. It shows ambition for the job and enough anti-autism human insight to know what traits would be required for the job.
This leads to the biggest issue from an academic perspective. My understanding is the MBTI is extremely questionable in terms of predictive/criterion/external validity. It's not actually linked with typical measures of job performance, social effectiveness, etc. It's basically just a hyped up mirror.
