<p>
[quote]
I had no idea this test existed, and in general I'm sceptical of any type of personality test.
[/quote]
</p>
<p>Yeah - it is good to be skeptical of psychometric tests in general. They're based on initial theories that cannot be experimentally verified (and they also suffer from vague definitions - and are easy to get caught up in their own definitions as well). The power in a psychometric test lies in its ability to correlate with particular variables. Of course correlation does not imply causation, but correlation can help produce questions that can help isolate the source of causation. I think one problem is that psychometric tests are inherently inflexible (and often stick to their own assumptions). IQ tests, for example, are not completely BS. But correlations between IQ and other variables are usually fairly weak, peaking out at 0.6. In fact, the highest correlation between IQ and anything else is merely the correlation between the IQ of identical twins reared apart (that does not mean that IQ measures what it's intended to measure - it's just that IQ measures something that is biological, but that could be highly irrelevant to success).</p>
<p>It's very unlikely that you can really categorize people in 16 different types. Some people find the MBTI to be amazingly accurate in describing themselves - whereas others do not find it so accurate. Personality tests have arbitrary distinctions (the Big 5 is used more for psychological research - that is the type I see when I read books on psychological research). But they certainly do correlate with a lot of factors - and hence this is where their value lies.</p>