Data · 2026-07-06
Ask what percentage of people share your type and you will get a confident answer, usually with a decimal point. That answer is a point estimate from a sample somebody else recruited. Two of the largest published tables for the same instrument disagree about which type is rarest, and for eleven of the sixteen types the gap between them is wider than their own sampling error allows. The disagreement is not noise. It is a measurement of who was counted.
Method · 2026-06-01
A dichotomy test asks you to compare yourself to an average person. An average person is not a fixed reference, so your answer moves with your mood, your week, and the comparison you happen to be holding in mind. Function-level questions ask about behaviour you can actually recall, which is far more stable.
Interpretation · 2026-05-04
When all eight functions land between 45 and 60, most tests will still hand you a confident four-letter label. That label is arithmetically meaningless. A flat profile usually means one of three things: you are mid-development, you answered from a role rather than a baseline, or you genuinely sit near an axis midpoint.
Theory · 2026-04-06
In the Grant stack the four functions not in your stack are the attitude-flipped versions of the four that are. INTJ runs Ni–Te–Fi–Se, so its shadow is Ne–Ti–Fe–Si. Under sustained stress the inferior function does not merely weaken — it takes the wheel, which is why stress behaviour often looks like a caricature of your own weakest function.
Data · 2026-03-02
Function-level testing consistently disagrees with dichotomy testing for a substantial minority of people. The disagreement is not random: it clusters around specific pairs — INTJ/INTP, INFJ/INFP, ENTP/ENFP — because those pairs differ mainly in which function leads, and a comparative questionnaire cannot see that.