Our method
How the questionnaires on this site are chosen, translated, scored and compared, what the AI-written report does, and what these tests cannot tell you.
Who writes the pages
The pages and questionnaires are prepared by Jérôme Stéphan, founder of the site, from the publications cited on the test pages. The most visited pages show their sources and the date of their last update; the others are being brought up to the same standard.
Three kinds of questionnaires
- Published scales with validation studies, whose questions and scoring follow the original publication: for example the WHO ASRS-v1.1 for adult ADHD, the AQ-10, the PHQ-9 or the GAD-7.
- Questionnaires adapted from a published scale, for example the highly sensitive person test, adapted from the scale of Elaine and Arthur Aron.
- Questionnaires written for the site, such as the learning style test inspired by the VARK model or the reasoning and IQ tests: they are not validated instruments.
The test page says which kind it is; if you find a page where this is not clear, please tell us.
Translations
Unless stated otherwise, the translations of the questions were made for this site. They have not been validated like the original versions, which can shift results a little.
Scores and cut-offs
When a cut-off comes from the original study (for example 4 out of 6 in Part A of the ASRS), we cite it. When an intermediate level is a landmark chosen by the site, we say so. A score above a cut-off is not a diagnosis, and a score below it does not rule anything out.
Comparison with the other test-takers
For some trait questionnaires (high sensitivity, giftedness, sensory perception, self-esteem...), the result shows where your score stands among the people who took the same test on this site, once there are at least 100 of them. Ties count for half, and the figures are refreshed every few hours; no one's answers are shown. These people came by themselves, often because they were asking themselves questions: it is not a population norm. We do not do this for questionnaires that have a recognised cut-off for a disorder or a difficulty (ADHD, autism, face blindness, depression, anxiety...): for those, only the validated cut-off counts. The IQ tests use norms computed from the site's test-takers (100 = the median of the participants).
The AI-written report
The optional paid report is written by Claude, an AI model made by Anthropic, from your answers and from instructions written for each test: it must not make a diagnosis or recommend medication, and it must stick to what the questionnaire can say. It can still contain mistakes. If it does not help you, you can ask for a refund within 30 days.
What these tests are not
They are self-assessments, not diagnoses. Only a qualified professional can make a diagnosis, after an interview that looks at your history and at other possible explanations. If you are in distress or thinking about suicide, contact the emergency services or a helpline in your country now.
Your data
Answers to health questionnaires are health data: they are kept only with your consent, and you can ask for them to be deleted at any time. Privacy policy · Delete my data
Corrections and licences
A mistake, a missing source, an awkward translation? Write to us: we correct it and date the update. The authors and licences of the questionnaires are listed on the test licences page. Contact · Test licences
Last updated
