How PopTypes Tests Are Built and Scored
PopTypes is made and maintained by the Curious Bytes studio. This page explains where our tests come from, how your answers become a result, and what we check before a test is published. If something here is unclear or you think we've got a framework wrong, email contact@poptypes.com.
The frameworks behind our personality tests
Our framework tests are independent, simplified versions of well-known ideas. They are written for self-reflection, not as replacements for the original or official instruments.
- MBTI-style 16 types. Based on Carl Jung's psychological types and the four-letter system popularised by Isabel Briggs Myers and Katharine Cook Briggs. Our test compares your preferences on four dimensions. It is not the official MBTI assessment.
- Big Five. The five-factor model of personality (openness, conscientiousness, extraversion, agreeableness, neuroticism), the most widely used trait model in personality research, associated with researchers such as Lewis Goldberg, Paul Costa, and Robert McCrae. Results are shown as relative trait scores rather than a type.
- Enneagram. The nine-type system developed in the 20th century by Oscar Ichazo and Claudio Naranjo and later popularised by writers such as Don Riso and Russ Hudson.
- DISC. Based on William Moulton Marston's 1928 model of four behavioural styles: dominance, influence, steadiness, and conscientiousness.
- Four temperaments. The ancient Greek idea of sanguine, choleric, melancholic, and phlegmatic temperaments. We present it as a historical, metaphorical way of describing energy and focus; it is not a scientific classification.
- Love languages. Based on the five love languages described by Gary Chapman in 1992. Our quiz is independent and not affiliated with the author or the official assessment.
- Attachment styles. Based on attachment theory from John Bowlby and Mary Ainsworth, its extension to adult relationships by Cindy Hazan and Phillip Shaver (1987), and the four-category model of Kim Bartholomew and Leonard Horowitz (1991): secure, anxious, avoidant, and fearful-avoidant.
- Career interests. Based on John Holland's RIASEC theory of vocational interests (Realistic, Investigative, Artistic, Social, Enterprising, Conventional), the same model used by the US O*NET Interest Profiler. Our test is independent of O*NET.
Character and lifestyle quizzes
Character quizzes are based on how characters think and behave in the original books, shows, and games. We write each character's result from canon, avoid major plot spoilers, and never use official artwork. Lifestyle quizzes such as aura colours, elements, and flowers use these images as personality metaphors; they make no spiritual, energetic, or scientific claims.
How your answers become a result
- Points per choice. In most quizzes, each answer adds two points to the result it most reflects and one point to a related result. After the last question, the result with the most points is yours.
- Dimensions. Type-based tests such as the MBTI-style test score each dimension separately and combine them into your type.
- Ties. When two results score the same, a fixed order decides, so the same answers always give the same result.
- Profile charts. Many results include a chart. For frameworks where the results are the dimensions (such as DISC, love languages, or Holland codes), the chart ranks all of them so you can see your second and third. For other tests, it shows how strongly your answers leaned on four themes.
- Scoring on our server. Your choices are sent to our server to calculate the result. If you save an attempt in an account, its answers and totals are retained as explained in our Privacy Policy. The full scoring keys are not included in the public test page.
Entertainment matching and psychological measurement
A character or district quiz is an authored comparison. It can make a conversation or a moment of reflection enjoyable, but it has not established a psychological category. Using a recognised framework also does not make our particular short implementation equivalent to a validated instrument. Reliability, validity and population norms require evidence beyond a functioning scoring system.
How to read a score or a tie
A score total counts the contributions of the answers you selected. A chart percentage usually expresses that total relative to the largest available score on that angle. It does not mean that you scored above that percentage of people. For example, half the available points is not evidence that you are at the population median.
A fixed tie-break makes repeated identical answers return the same headline. If two portraits have equal points, the rule does not establish that the first one fits better. Likewise, a narrow lead should not be treated as a strong claim about identity. Compare the description with a specific choice you remember making.
What we check before publishing
- Balance. We use random and deliberately consistent answer patterns to look for unreachable results, unexpected ties and score mistakes. The number of runs depends on the check. These simulations describe the scoring design, not the expected distribution of real personalities.
- Question quality. We review whether a question asks one thing, whether two alternatives can apply at once, and whether an ordinary response is missing. Preference questions ask for one priority among the alternatives. A fictional or playful question is not presented as proof of a real-world ability.
- Result writing. Each result has a summary and a deep dive written for that result alone, with concrete examples. We check length, repetition across results, and tone, so no result reads as an insult.
- Accuracy and spoilers. Framework descriptions are checked against their sources; character quizzes are checked for spoilers and canon accuracy.
- Automated checks. Automated checks inspect structure, references, score ceilings and other implementation details. They cannot determine whether every option set is realistic or whether a result is psychologically valid; those require different kinds of review and evidence.
Limits of online personality tests
Short online quizzes can be a useful mirror, but they have real limits. Your result can change with your mood, your situation, or how you read a question. Many popular frameworks, including type-based models and love languages, have mixed scientific support; the Big Five has the strongest research base. PopTypes tests are for self-reflection and entertainment. They are not clinical, diagnostic, or hiring assessments, and they do not replace advice from a qualified professional.
Updates and corrections
We revise tests when we find a problem: a result that is too hard to reach, a question that reads ambiguously, or a factual error. Material changes to scoring are versioned internally so results stay consistent. To report an issue, email contact@poptypes.com.
November 20, 2026: Bird question clarity
We revised choices that mixed when someone spoke with whether they had been asked to speak, and options that made declining help require another commitment. Other revisions separated an already completed share from the work still remaining. The test keeps 30 questions and its existing score weights; the revised question meanings have a new scoring version.
November 20, 2026: Hunger Games District question review
The published test has 28 questions. We removed repeated prompts from the report design, replaced vague dramatic wording with specific choices, and made sharing amounts cover the range from none to all without a gap. Its optional story is labelled as an original PopTypes scenario rather than an event from the books or a prediction of survival.