I don't know though, I don't think the type of people who you like/annoy you should be such a big part in determining your OWN personality type. That's the big flaw of this test.
When I re-read some of those questions they seem to be gettign at core ideas (I see the focus as being on "how" and "what" rather than specific other people) that certain types or functions would have a probability of prefering one over the other, but I'll agree that those type of questions making up 40% of the questions does put a lot of weight on them and having more questions of different varity to dilute the weight of those questions would be a way to improve it.
I also don't like the fact that you have to adjust your own meter for the questions in the test, it's not just straight up multiple choice answers. Yeah, I think the following tests are far more accurate than this MMDI one:
PersonalDNA | Your True Self Revealed - Fast Fun Free Personality Tests
Ah, I forgot about that one, thats another favorite of mine,
great test.
A fun and interestign test to take and one where I do agree with the results.
It won't help you determine your true MBTI type, but its still a test I'd recommend to anyone.
Since it is a rather time consuming test, I did not re-take it, but I kept my
results from a year or so ago.
Personality Test and Types (Jungian Psychological Typology)
That one requires an account to be created before taking the test. It may be free to do so, but does it offer something different or interesting that all the similar free no-account tests do not (a different approach, more innovative questions etc)?
I may go back to it later, but for the moment, I'll skip it and refrain from additional comment (if I take it before my edit time expires I'll go back and revise this).
Personality test based on Jung - Myers-Briggs typology
According to this test I score as...
INFJ - Introverted - 78, Intuitive - 75, Feeling - 12, Judging - 44
It is VERY common for me to score a moderate to high J preference on a dichotomy based test (ones that score based off of I/E, S/N, T/F and J/P values). There is enough in the INFJ (and even INTJ) descriptions that fit me that for quite a while I did not have a good reason to examine other options and for a long time did beleive myself to be an INxJ of some kind. From time to time I still feel compelled to go back and re-read INFP and INFJ descriptions (specifically Fi and Fe descriptions) and talk to others of both type to re-evaluate myself, but the conclusion thus far has always been the same, so I'm left to conclude that the type sorter tests are inherently limited in their accuracy - they may get you within 1 letter of your type with fairly high accuracy, but I think thats the best one can expect from them.
Its not a bad test as far as dichotomoy sorters go, but for tests like these to have value, they should stress the importance of following up the results by reading multiple type descriptions (specificaly all the "one-letter off" variants) and ideally (tho it may be asking too much) explaining functions, function orientation (Ti vs Te etc) and function order and stacking.
Interestingly, this is something the MMDI does well, but if its giving you a very skewed answer to begin with, all the extra stuff like that still makes it less useful and meaningful to you than a generic result that is overall more accurate.
Free Jung Personality Test
My results from this one.
Introverted (I) 83.33% Extroverted (E) 16.67%
Intuitive (N) 70.59% Sensing (S) 29.41%
Feeling (F) 54.05% Thinking (T) 45.95%
Judging (J) 55.88% Perceiving (P) 44.12%
Your type is: INFJ
Again not a bad test as far as dichotomoy sorters go, but the repeated same false results for me may help illustrate why I find these tests to be more flawed than the MMDI. If I got equaly flawed results from the MMDI, I would probably consider it a novel approach to doing the same thing as other tests, but since it consistantly gets me more correclty than other tests I can't help but have higher regard for it than basic dichotomoy sorters.