Rarest MBTI type and most common: where the percentages come from

The rarest MBTI type is INFJ, at 1.5 percent. The most common is ISFJ, at 13.8 percent. You can find those two figures on a few thousand websites, nearly always without a word about where they came from.

They came from 45 people and 414 people respectively, in one American sample published in 1998. That is not a reason to throw the numbers out. It is a reason to know what they can carry.

Where the percentages come from

The source is the third edition of the MBTI Manual, published in 1998 by Isabel Briggs Myers, Mary McCaulley, Naomi Quenk and Allen Hammer. For that edition the publisher gave the questionnaire to a national sample of 3,009 adults in the United States, chosen to resemble the adult population rather than to resemble the sort of person who signs up for a personality workshop.

That last part is what makes it the figure everyone quotes. Almost every other count of types is a count of volunteers. This one was built to be a count of the general public, and it is still the baseline later research is compared against.

I have not handled the manual itself. The figures below are as reproduced in a 2025 review of the instrument by Bradley Erford and colleagues in the Journal of Counseling and Development, which reprints the national sample to compare later studies against.

All sixteen, from most to least common

• ISFJ: 13.8 percent
• ESFJ: 12.3 percent
• ISTJ: 11.6 percent
• ISFP: 8.8 percent
• ESTJ: 8.7 percent
• ESFP: 8.5 percent
• ENFP: 8.1 percent
• ISTP: 5.4 percent
• INFP: 4.4 percent
• ESTP: 4.3 percent
• INTP: 3.3 percent
• ENTP: 3.2 percent
• ENFJ: 2.5 percent
• INTJ: 2.1 percent
• ENTJ: 1.8 percent
• INFJ: 1.5 percent

Three types at the top, ISFJ, ESFJ and ISTJ, account for 1,131 of the 3,009 people between them: more than a third of the sample in three of sixteen boxes.

One letter does most of the work

Look down the list and the pattern is hard to miss. The seven most common types include six with an S in second place. The seven least common include six with an N.

Add the counts up and 2,202 people in the sample came out on the sensing side against 807 on the intuitive side, about 73 percent to 27. That one split explains most of the table: an intuitive type starts with roughly a quarter of the population to draw from before the other three letters have divided it further.

The other three pairs are nowhere near as lopsided. Introverts and extraverts were almost a tie, 1,527 to 1,482, which surprises people who assume introverts are a small minority. Feeling led thinking by about 60 to 40. Judging led perceiving by about 54 to 46.

The four rarest types are the four that combine N with J: INFJ, ENTJ, INTJ and ENFJ. Together they were 237 people, a little under 8 percent. Sensing vs intuition explained covers what that second letter is measuring, which is worth knowing before reading anything into belonging to the smaller group.

The answer changes with gender

The sample held 1,478 men and 1,531 women, and the two halves do not rank the types the same way.

Among the men the rarest type was INFJ at 1.3 percent, followed by ENFJ at 1.6. The most common was ISTJ, at 16.4 percent.

Among the women INFJ was not the rarest at all. INTJ and ENTJ tied for that place at 0.9 percent each, 13 women and 14 women. The most common were ISFJ at 19.4 percent and ESFJ at 16.9, so that more than a third of the women in the sample had one of two types.

The third letter is the one doing this. Add up the counts and about three quarters of the women came out on the feeling side, against a little under half of the men, and why T and F is not logic versus emotion explains why that letter in particular is tangled up with how people were raised to describe themselves. So "which type is rarest" has at least three correct answers, depending on who is asking.

How much weight 45 people can bear

The ranking at the bottom of the table rests on very small counts. INFJ was 45 people. ENTJ was 54. The whole difference between the rarest type and the second rarest is nine people in a room of three thousand.

Rerun the survey with a fresh sample of the same size and those two could easily swap. What the data supports is the broad statement: the intuitive judging types are each a few percent of the population, and the sensing judging types are each above ten. What it cannot support is a confident ordering inside the bottom four, or a decimal place.

Two more limits are worth stating plainly. The sample is American, so it says nothing direct about any other country. And it was published in 1998, which makes it a generation old.

Why everyone you meet online is an INFJ

If INFJ is one person in sixty-seven, forums about personality type should be nearly empty of them. They are not, and the reason is not that people are lying.

The people who seek out a personality questionnaire are not a cross-section of the public. Erford's review shows this in the research literature itself. Pooling the participants of twenty-five years of published studies that used the instrument, it found INFJ at 3.3 percent, more than double the national figure, and ENTJ at 6.2 percent against 1.8. ISFJ, the most common type in the general population, fell to 5.8 percent. Across all the comparisons the review ran between its pooled samples and the national one, about two thirds came out significantly different.

That is among people recruited into studies. A website that anyone can visit is selected more heavily still, toward whoever finds this subject interesting enough to spend an evening on.

There is a second reason, and it is mechanical. Each letter is assigned by which side of a midpoint a score falls on, and a different questionnaire with different questions puts that midpoint in a slightly different place. Move the line on one scale by a little and everybody who was sitting near it changes letters. Why your result changes between tests goes into that. So the manual's percentages describe the manual's questionnaire. They do not transfer to a free test, including the one on this site.

What a rare type does not mean

It does not mean gifted, deep or misunderstood. It means that on four scales you landed on the less crowded side of at least a couple of them. Uncommon is a fact about a crowd. It says nothing about the person.

It does have one practical consequence. If your type is uncommon, most of the people around you make at least one of those four calls differently than you do, and it helps to know which one before assuming they are being difficult.

MBTI is a trademark of The Myers-Briggs Company, which has no connection to this site. I run it by myself and I am not a psychologist, and none of this is a clinical or diagnostic measure. The sixteen type pages describe each type without ranking them, and the test will give you four letters whether or not they turn out to be the fashionable ones.