How these tests are scored
The exact arithmetic behind every percentage on this site, what the numbers can support, and the specific claims they cannot. Nothing here is a validated psychometric instrument, and this page explains why that matters when you read a result.
Last updated · changes to questions or scoring are logged at the bottom of this page
Two separate tests, and why they are not one
KinkBDSMTest runs two educational self-reflection questionnaires for adults: a kink test covering 8 broad interest families, and a BDSM test covering 26 community-derived power-exchange archetypes. Both exist to help people find language for current interests. Neither is a clinical, psychological, medical, or diagnostic assessment.
Kink is the umbrella; BDSM is one region of it. A 26-archetype power-exchange engine cannot honestly be relabelled as a complete map of kink, so the kink test has its own question set, its own answer model, and its own scoring rather than reusing the archetype engine.
| Property | Kink test | BDSM test |
|---|---|---|
| Questions | 40 statements | 52 statements (Text Mode) or 12 choices (Image Mode) |
| What it scores | 8 broad interest families | 26 community-derived power-exchange archetypes |
| Items per category | 5 per family | 2 per archetype (Text Mode) |
| Answer model | Five distinct answers that separate curiosity, fantasy, and real-world interest | A five-point agreement scale, Strongly Disagree to Strongly Agree |
| Extra output | A fantasy-versus-practice ratio of your own positive answers | An optional profile comparison between two saved results |
| Covers | Kink broadly — the umbrella | Power exchange specifically — one region of that umbrella |
How a percentage is calculated
Both tests use the same arithmetic: the points you selected in a category, divided by the points that category could have awarded, rounded to a whole percentage. Nothing is weighted secretly, and no category is scaled differently from another.
Worked example · one kink-test family
Each family has 5 statements, and each answer is worth 0 to 4 points:
- 0Not for me
- 1Curious, but unsure
- 2Fantasy only
- 3Yes, with the right person and conditions
- 4Actively interested
Answer those 5 statements 4, 3, 2, 1 and 0 and you have 10 points out of a possible 20.
10 ÷ 20 = 50%
That 50% means half the available points in that family. It does not mean a 50% chance that a label applies to you, and it says nothing about anyone else’s answers.
The BDSM test runs the same division with its own numbers: 2 statements per archetype on a five-point agreement scale, so each archetype has 8 points available. Do not apply the 5-item example to a role result.
The fantasy-versus-practice reading
Alongside the 8 percentages, the kink test reports what share of your positive answers were fantasy only rather than something you would consider in practice. This is a descriptive ratio of your own answers, not a psychological measure.
It exists because most quizzes silently collapse “haven’t tried it”, “like imagining it”, and “want to do it” into a single agreement score — which is the most misleading thing a kink questionnaire can do.
What Image Mode can and cannot tell you
The 12-choice Image Mode is a forced-choice, exploratory experience. Each choice compares unlike options and cannot assess all 26 archetypes independently, so its output is a playful snapshot rather than an equivalent or more precise version of Text Mode.
What a percentage is not
The formula is simple and it runs correctly. That is a much smaller claim than the four things people most often read into a score.
Not a percentile
80% does not mean you are more interested than 80% of people. There is no comparison group, because there are no population norms behind either test.
Not a probability
50% in a family does not mean a 50% chance that a label applies to you. It means you selected half of the available points on those statements.
Not a diagnosis
Neither questionnaire is a clinical, psychological, or medical instrument. A score is vocabulary for reflection, not an assessment of you.
Not experience or readiness
A high score says nothing about technique, risk awareness, a partner’s agreement, or whether an activity is safe for you to try.
The profile comparison has the same ceiling: it summarises complementary and shared archetype patterns, and it cannot measure hard limits, communication quality, experience, health considerations, consent, or real-world compatibility. It is a conversation starter.
Result links and privacy
Scoring happens in the browser for both tests. The resulting percentages are then encoded into the result URL so the page can be reopened and shared without an account.
Raw answers are never included in that link, but the score profile is: anyone with the URL can view it. Result and comparison pages are noindex, and noindex does not make a link private. Read the privacy policy for what is and is not collected.
Known limitations and revision log
Neither questionnaire has representative norms, a reliability study, factor analysis, or external validation. The BDSM test uses only 2 positively worded items per archetype, and its Image Mode is inherently comparative. The kink test uses positively worded items throughout, so it measures stated interest rather than aversion. Preferences also change with context, relationships, knowledge, and time.
Claims about reliability, clinical accuracy, or comparison with a population would require evidence beyond a working formula. A calculation that runs correctly is not an instrument that has been shown to measure what it claims.
Revision log
Three kink test items reworded
No change to scoring, dimensions, or item counts. One item asked readers to judge how rope work looks, which cannot be answered without seeing it; two others did not say whether the sensation or the language was being given or received, so the same answer meant different things to different people. Rewording changes what an item measures, so scores from before and after this date are not strictly comparable.
Material changes to questions or scoring are added here with the date they took effect. Safety-critical content is reviewed against reliable educator and community resources.
Frequently Asked Questions
Is the BDSM test accurate?
It is consistent, not validated. The same answers always produce the same percentages, and every archetype gets the same number of items, so nothing is weighted secretly. But there are no population norms, no reliability study, no factor analysis, and no external validation — so "accurate" in the psychometric sense is a claim this site does not make. Treat the output as organised vocabulary, not a measurement.
Does 80% mean I am more interested than 80% of people?
No. These are not population percentiles. The number describes how your own answers interact with this site’s question set and scoring model, and nothing about anyone else.
How is the kink test scored?
Each of the 8 interest families has 5 statements, and each answer is worth 0 to 4 points. Your points in a family are divided by the 20 available and rounded to a percentage. The same arithmetic runs for every family, so families can be compared with each other — but not with any outside benchmark.
Why do the kink test and the BDSM test give different results?
Because they are different questionnaires measuring different things. Kink is the umbrella; BDSM is one region of it. The two tests do not share a question set, a set of categories, or an answer model, so a high score on one does not translate into a score on the other.
Is the three-question role reflection scored the same way?
No. The short exercise on each role page counts four kinds of response and quotes your current interests back to you. It does not calculate an identity score and is not part of either test.
Will my results change if I take the test again?
Often, yes. The questionnaire records answers, not a fixed trait. Mood, recent experience, and how you read an ambiguous sentence all move individual responses, and small shifts can reorder categories that were already close together.
Are my answers stored or sent anywhere?
Scoring happens in your browser. The resulting percentages are encoded into the result URL so the page can be reopened or shared without an account. Raw answers are never included in that link, but the score profile is — anyone holding the URL can read it.
What would it take to call this validated?
Evidence beyond a working formula: representative norms, a reliability study, factor analysis, and external comparison. A calculation that runs correctly is not the same as an instrument that has been shown to measure what it claims. Until that evidence exists, no score here is presented as a diagnosis or a percentile.
Read a profile without overreading it
Now that you know what the numbers are, the results guide shows how to turn overlapping percentages into specific questions instead of an identity.