24-question dark triad test: separate Machiavellianism, narcissism and subclinical psychopathy scores. No combined score, and no diagnosis.
Dark Triad Test
Twenty-four statements, about three minutes. They are deliberately unflattering — the result is only worth anything if you answer them honestly. You will get three separate scores: Machiavellianism, narcissism, and subclinical psychopathy. There is no combined score.
The three dark-triad traits are only moderately intercorrelated — which is exactly why this test refuses to give you one combined number.
How the Dark Triad Test Works
Answer as you are, not as you would like to be
This is the only instruction on this test that really matters. Every item is transparently unflattering, so most people's scores are pushed down by self-presentation. You cannot game your way to a good result, because there is no good result — only an accurate one and an inaccurate one.
Rate 24 statements
Eight per trait, four worded each way. The balance is there to defeat straight-line answering: agree with everything and all three scores land exactly on the midpoint, and the page will tell you so.
Read the three scores separately
Each is a sum of eight items on a published 8–40 scale, not a percentile. They are never added together, and they never will be — averaging them would destroy the combination that carries the meaning.
Read "higher" as a description, not a verdict
Higher describes where a tendency sits on a scale. It is not a moral grade and it is not a diagnosis. Each trait page covers what the pattern looks like day to day, where it genuinely helps, and what it costs.
About the Dark Triad — the Science, Honestly Told
Where the Framework Comes From
The "dark triad" grouping comes from a 2002 paper by Paulhus and Williams. They noticed that three concepts that had grown up separately in personality research — Machiavellianism, subclinical narcissism and subclinical psychopathy — were moderately correlated in ordinary populations while clearly not being the same thing, and proposed studying them together. Note the word subclinical: this research tradition is about personality dimensions found across the general population, not about clinical groups. Reading it as "find out whether you are a psychopath" is the most common and most serious misunderstanding of the framework, and this page will not encourage it.
The twenty-four statements you will rate are an original RECATOOLS item set: eight per trait, balanced keying, scoring published openly on this page. We wrote our own because the established brief measures — the SD3, the Dirty Dozen — are licensed for non-commercial academic research, and this is an ad-supported commercial site. Rather than borrow ambiguously, we built transparently. That honesty runs both ways: the three constructs are well studied, but our specific items carry no validation studies, so treat your scores as structured self-reflection rather than measurement.
"Every item here is transparently unflattering. That means scores are pushed down by self-presentation — and it means an all-low result is genuinely ambiguous."
What the Research Does and Does Not Support
Three things worth carrying. First, these traits do predict real differences — negotiation style, counterproductive behaviour at work, exploitation in relationships — at meaningful but modest strength. They are a current, not a cage. Second, they behave differently from one another: high Machiavellianism is planned and patient, high psychopathy is impulsive and short-horizon, and the two frequently contradict each other in behaviour. That is the concrete reason there is no composite. Third, and most important — this page cannot diagnose anything. Clinical psychopathy is assessed by trained clinicians using structured interviews and collateral information, never by a web questionnaire. If impulsivity or conflict is causing real harm to you or the people around you, that is worth talking to a professional about, and it is a separate question from where a 24-item test put your score.
The Three Traits
Machiavellianism
Treating interaction as a board: strategic disclosure, thinking three moves ahead, reasoning backwards from what the other person wants.
Narcissism
A grand self-assessment, a need for recognition, and a sense that special treatment is simply owed.
Psychopathy (subclinical)
Callousness and impulsivity: not much moved by other people's distress, acting ahead of consequences. A personality dimension, not a clinical diagnosis.
Frequently Asked Questions
-
Because the three traits are only moderately correlated, so summing them asserts a single underlying "darkness" factor that the evidence does not support. The practical reason is sharper: high Machiavellianism (planned, patient) and high psychopathy (impulsive, short-horizon) are close to behavioural opposites, and an average would turn two completely different people into the same number. Our scoring kernel returns no total, and the test suite carries an assertion specifically to stop one being added later.
-
No. These measure where tendencies sit on a scale, not moral worth. High Machiavellianism is genuinely useful in negotiation and organisational politics; the confident component of narcissism is associated with emerging as a leader; even lower empathic reactivity is not purely a liability in roles that require cold decisions. Equally, each has real costs, and the costs usually land on the people nearby. Description and verdict are different things.
-
No, and there is no softer way to put it. This measures subclinical personality dimensions in the general population. Clinical psychopathy and antisocial personality disorder are assessed by trained clinicians using structured interviews and information from multiple sources — never by a 24-item web questionnaire. Any test claiming to diagnose these online is misleading you.
-
This is the one result where we openly admit we cannot read it. An all-low profile is what an honest low score looks like, and it is also what answering the way you would like to be seen looks like — every item is transparently unflattering, so that downward pressure is systematic. We cannot distinguish the two, so the page says so rather than congratulating you.
-
To catch straight-line answering. Some people tend to agree with statements regardless of content (or disagree with all of them), and on a one-directional bank that habit becomes a high score. With half the items reversed, the habit cancels itself out: agree with all 24 and every trait lands exactly on the midpoint of 24. The page tells you when it detects this, because it means the result is not worth much.
-
Yes. Personality traits are relatively stable in adulthood but not fixed, and a self-report questionnaire also picks up your current state and self-image. More to the point, this item set has no test-retest reliability study, so we cannot tell you how stable its results are. Treat it as a snapshot.
-
RECATOOLS did — twenty-four original statements, with the scoring rule published openly on this page and an authorship attestation in the tool's provenance record confirming that no existing dark-triad instrument's item text was consulted. The established brief measures are licensed for non-commercial research, so rather than borrow, we built.
-
Because that is the name in the research literature, and renaming it would disconnect the page from the work it rests on while making it vaguer rather than kinder. What we do instead is mark it "subclinical" everywhere it appears and state repeatedly that it is not a diagnosis. Using the accurate word and then being clear is more responsible than using a fuzzy one.
-
No. Your answers are scored entirely in your browser and are never uploaded or stored by RECATOOLS. While a test is in progress they are kept in your browser's local storage so you can resume if the tab closes, and they are deleted when you finish. Viewing pages on this site works like any other website and is covered by our Privacy Policy.
-
That is your call, but it is worth a thought: these three traits carry heavy reputational freight, and a label from an unvalidated web test can stick to you — or to whoever you send it to — in ways you did not intend. Sharing the test is a good deal safer than sharing the scores.
Related News
You may be interested in these recent stories from our newsroom.
No related news yet for this tool. Our editorial team publishes new pieces every week.
Browse all news →Pick up where you left off
Stored only in this browser — never sent to our servers.