Choosing one option withholds the other
On the 16-type test, every item adds a point to one pole and withholds it from the other. After 70 items, each letter is the side with more points. The score compares you with yourself: your Extraversion answers against your Introversion answers. It is not an amount of extraversion you can set beside someone else’s amount.
The Enneagram test does the same thing across nine piles. An option credits one or more types. The winner is the largest pile. A person with 14 points on Type 5 and 13 on Type 6 is not “a stronger 5” than a person with 9 on Type 5 and 4 on the rest. The first result is a near tie. The second is a clear sort. The result page shows the runner-up so you can tell which kind you got. A form that keeps only the winning label throws the only useful part away.
The scoring rules, including the tie-break on the 16-type test, are written out on the methodology page.
Two INTJs are not the same result
Both codes read INTJ. One person won each letter by a large margin. The other won Thinking over Feeling by a single item and would print INTP on a slightly different afternoon. Pittenger’s reviews, which the guide on the 16-type test summarizes, found that roughly half of people get at least one different letter when retested after about five weeks. The retest post is the mechanism: many scores sit near the cut, so a small shift changes the letter.
A spreadsheet cannot see the margin. It sees INTJ, INTJ, ENTJ, and starts to look like a column of data.
The margins on this site’s 70 items
The four letters are not the same kind of close call. Extraversion/Introversion and Judging/Perceiving are 17 items each, so one side always leads. The closest finish is 9 to 8. Sensing/Intuition and Thinking/Feeling are 18 items each, so a 9 to 9 tie is possible. The scorer breaks that tie toward S and toward T. The methodology page states the tie rule. The result page prints the eight tallies, which is the only way to see whether Thinking was 16 to 2 or 10 to 8.
Two candidates can both print INTJ and still be different measurements. One won Thinking 16 to 2. The other tied Thinking and Feeling 9 to 9 and received T from the tie-break. A spreadsheet that sorts on the letters treats those as the same cell. The second person is one item away from INTP. Pittenger’s retest figure, about half of people changing at least one letter after five weeks, is mostly people in that second situation.
What a job does with the letters
The publisher of the official MBTI has long told practitioners not to use that instrument for hiring or selection. Our questionnaire is not that instrument. It is 70 items written for this site, in the same four-dichotomy tradition, with no norm group and no validity study of its own. There is little evidence that a four-letter type forecasts job performance, which is the only job a selection tool has. Personality tests in hiring covers the evidence that does exist for trait measures, and why a type code is the wrong shape for a hiring decision.
DISC percentages from this site fail a different test. They look like ranks and are not. That argument is the norms post. Forced-choice codes fail one step earlier. They were never amounts. Lining two candidates up by letter treats a within-person sort as a between-person score, and the arithmetic does not survive the trip.
Use the code for a conversation about how two people prefer to work. Do not use it to drop a resume. If you need a number that was at least built as an amount, the Big Five test reports five scores, and even those are not percentiles. A hiring decision needs a better instrument than either.
If you are the one hiring
Ask for the tallies, not the code. If you cannot get the tallies, you do not have the result this test computed. A within-person sort answers which side of one person’s answers won. A job asks who will do the work better. Those are different questions, and the second one needs a sample of the work.
Barrick and Mount’s 1991 meta-analysis, which the hiring post summarizes, is about trait scores and job performance. It is not a license to rank people by a four-letter code that was never an amount. Conscientiousness in that literature is a score with a norm group behind it. An INTJ cell in a spreadsheet is a label. Lining the labels up does not turn them into the score the meta-analysis was about.
