How to read a psychometric test report

A psychometric report is a comparison, not a measurement. Nothing in it says how much of something you have in absolute terms; every number says where your answers sat relative to other people who answered the same questions. That single fact changes how almost every line should be read.

Percentiles and stanines: the same idea, different resolution

A percentile says what share of the reference group scored at or below you. At the 70th percentile, roughly seventy in a hundred scored at or below your answer. It is not a mark out of a hundred, and a report that lets you read it as one is doing you a disservice.

A stanine compresses that same scale into nine bands. It looks less precise, and that is the point: the difference between the 61st and 66th percentile is noise, not signal, and a scale that displays it invites you to act on a difference that would not survive taking the test again next week. Bands 4 to 6 are the middle, where most people sit. Reading a band 5 as a middling grade is the most common misreading there is.

Both ends of a dimension are real positions

Personality traits are dimensions, not scores. Extraversion runs from reserved to outgoing, and neither end is the good end. Someone low on extraversion has not failed at being extraverted; they have a preference that suits some working conditions and costs them effort in others. The same is true of every one of the five.

This matters practically. If a report describes one pole as a shortfall to be fixed, it is telling you about the writer rather than about you. What a good report does instead is name the conditions under which each preference is an advantage and the conditions under which it is a tax — because both are true, and which one you get depends on the job.

Interests are not aptitudes, and neither is a prediction

Interest inventories ask what you would enjoy. Reasoning tests ask what you can currently do. They answer different questions and they routinely disagree, which is useful information rather than a contradiction: enjoying something you are not yet good at is the ordinary condition of every skill worth having.

Neither one predicts a career. What they narrow is the search, by telling you which directions are worth the cost of investigating properly and which you would be investigating out of obligation.

When a score is missing, read the reason

A responsible report withholds scores it cannot stand behind. If a timed section was rushed, if the same option was selected down a whole page, or if two questions that ask the same thing in opposite directions got the same answer, the honest output is a note explaining what happened — not a low number that will follow the reader around.

So a blank section is not a fault in the report. It is the report declining to guess. The right response is to look at what the note says and, if it matters, take that section again in conditions where it can be answered properly.

What to do with it in the first week

• Read the whole thing once without deciding anything. First reactions to the lowest bar are reliably wrong.
• Find the two or three places where the report tells you something you did not already believe. Those are the parts worth your attention; the rest is confirmation.
• For every direction that interests you, write down what would have to be true for it to work — grades, exams, money, time. The report can rank fit, but only you know the constraints.
• Talk it through with somebody who knows you. A report is a starting point for that conversation, not a substitute for it.

If you want the specifics of which instruments we use, how norms are built and what each response-quality check does, it is all written down: How our scores are computed.