About the project

Why this exists

Search for a temperament test and you get twelve questions and a one word verdict. You are a choleric. Nothing about how strongly, nothing about what else is in the mix, nothing you could not have guessed before you started.

The four temperaments deserve better than that, because the old typology was never meant to sort people into four boxes. It was a language for describing how differently people run: how fast they start, how long they hold, how hard things land and how quickly they let go. Those are measurable things.

So this site does one thing. Eight scales, measured with enough questions to be worth trusting, turned into a profile that is almost always a mixture of two temperaments rather than a pure type, and written out in plain language. Not a personality zoo, not a badge for a profile page, and not a second test bolted on to sell the first.

How the test is built

There are 166 items: 160 that belong to the eight scales and six consistency checks scattered through them. Twenty items per scale, half of them worded the other way round, so that agreeing with everything does not quietly produce a profile.

The eight scales are grouped on two axes. One is about activation, meaning how much energy goes out and how fast. The other is about steadiness, meaning how hard things land and how quickly the ground comes back. Where you sit on those two axes decides the shares of the four temperaments in your result.

That mixture is what you get instead of a verdict. Most people come out as one leading temperament with a clear second, which is why there are thirteen profiles rather than four: four pure ones, eight pairs where the order matters, and one for a profile that genuinely sits near the middle.

The order of the items is fixed and deliberate: neighbouring questions come from different scales, the consistency checks land at set points along the way, and a reworded pair never appears back to back. A questionnaire that groups its scales together teaches you what it is measuring, and then it measures that.

What the numbers mean, and what they do not

Each scale is reported as a position between its two ends. High and low are both ordinary. Neither end is the good one, and the descriptions are written so that reading the low end out loud does not sound like a complaint.

We do not currently claim percentile norms. Your number says where you landed on the scale itself, not what percentage of some population you beat. That is an honest limitation rather than a shortcut, and it is stated wherever a number appears.

When there is enough data collected here to build comparison groups that are actually ours, we will publish how they were built and what they cover, and only then start comparing people to them. Borrowing somebody else's sample and calling the numbers percentiles would be the faster route and the dishonest one.

How the texts are written

Every explanatory line here was written for this site: the scale descriptions, the thirteen profiles, the four temperament pages, the pair readings and the material behind the paywall.

The rules we hold ourselves to are short. A claim about a trait has to trace back to what the test actually measures. A description of a low score has to be something a person would recognise rather than something they would resent. Anything the test cannot support gets cut, however well it would read.

When something is wrong we fix it and move the date at the top of the page, so a returning reader can see the text moved. If you spot an error, tell us. A correction from a reader is worth more than a paragraph about how careful we are.

What this test cannot do

It is a self report questionnaire about ordinary traits in ordinary people. It is not a medical or psychiatric instrument, it says nothing about disorders, and it does not replace a conversation with a doctor or a psychologist.

It is not a screening tool either. Do not use a result to hire, admit or grade anyone. Once something depends on the answers, a self report is trivial to answer the way it needs to be answered.

And it is a snapshot. Temperament is stable across years rather than fixed forever, and a result taken in a hard month is a result taken in a hard month. Take it again later and compare the two: that comparison usually says more than either reading alone.

How the site pays for itself

The test is free and the base result is free with it: the formula, the portrait, the eight scales, what stands out and the two axes, with no account and no payment.

What is sold is depth on the same result. One payment opens six readings on the areas of life where temperament shows, the scale library written against your own numbers, and a written analysis produced for your profile. The price and the contents are shown before payment, and the conditions live in the Terms of Sale.

There are no advertising networks here, no cross site trackers and no answers for sale. The site is paid for by the people who choose to open the extended part, which is the arrangement that keeps the test itself free of anybody else's interests. What is collected and for how long is written out in the Privacy Policy.

Who is behind this and how to reach them

The site is run independently, not on behalf of a publisher and not as an adaptation of anybody's commercial instrument.

Questions, corrections, research and press all go to one address, and a person reads it. It is on the contact page, together with what to include so the answer comes back the first time.