How the test works
Everything we measure, every formula we use, how sure each result is, and the honest limits of a short self-report test.
What we measure
40 scenes, 8 per scale. Each one sets the two poles of a scale against each other: you lean toward one side of a choice, or a puzzle's outcome leans for you.
| Scale | Poles | Big Five match | An example scene |
|---|---|---|---|
| EnergyWhere do you get your energy? | Extraverted (E) or Introverted (I) | Extraversionhigh = Extraverted | A free evening after a long week. Which would recharge you?“A lantern festival, packed with people” or “A quiet room, a good book and the rain” |
| PerceptionWhat kind of information do you trust first? | Intuitive (N) or Observant (S) | Opennesshigh = Intuitive | A new cabinet arrives in pieces, with a sheet of instructions. How do you start?“Picture the finished cabinet, then work out how it fits” or “Lay out every piece and check it against the list” |
| DecisionsWhat carries the most weight when you decide? | Thinking (T) or Feeling (F) | Agreeablenesshigh = Feeling | Which teacher would you rather learn from?“One who encourages you and keeps you loving it” or “One who points out every flaw, so you improve fast” |
| StructureHow do you like to organize your life? | Judging (J) or Perceiving (P) | Conscientiousnesshigh = Judging | A big order is due in two weeks. How do you tackle it?“Start right away and work steadily, so it's done early” or “Let it take shape, then finish it in a focused burst near the end” |
| ComposureHow strongly do stress and emotion move you? | Steady or Sensitive | Neuroticismhigh = Sensitive | Rain is forecast for the festival you've spent weeks organizing. How does the week go?“It weighs on you until the day comes” or “You shrug: rain or shine, you'll make it work” |
Scenes, not statements
Most personality tests hand you a sentence about yourself and ask how far you agree. Ours puts you in a moment instead, drawn as a woodblock print, and asks which way you would go. Picturing a real situation is easier than judging a sentence about yourself, and it changes what can go wrong with an answer.
This or that15 of 40
Two prints of the same moment going two ways, and you say how far you lean. With two good things to choose between, there is no statement to simply agree with and no side that makes you look better.
Dilemmas20 of 40
A situation and two reasonable things to do. Choosing between the two poles of a scale is how type questionnaires have always asked (Myers and colleagues, 1998), and a concrete situation helps everyone read a question the same way (Lievens and colleagues, 2008).
Puzzles5 of 40
One per scale: find your spot at a festival, name a print, share a bonus, plan a trip, light a lantern for each worry. None has a right answer, on purpose: asking what people would do tracks their personality, while asking what is best tracks reasoning ability (McDaniel and colleagues, 2007).
The rules every scene follows
Pictures cannot fix everything. You might pick the quiet room because you are tired tonight, not because you are introverted: behavior shifts with the situation (Mischel and Shoda, 1995). That is why each scale averages 8 different scenes, and why every score comes with a likely range.
- Both sides are worth wanting
- Neither side of a scene is the kind, grown-up or clever answer. A forced choice only resists the pull to look good when its options are balanced like this (Cao and Drasgow, 2019).
- One scale at a time
- Each question sets the two ends of a single scale against each other, never two different scales, so your five scores stay independent. Choices between different traits give scores that can only be compared within one person (Hicks, 1970).
- The sides swap
- On every scale, each pole sits on the left in about half the questions and on the right in the rest, so a habit of picking one side cancels out instead of tilting a score.
- Prints show the moment, not the answer
- A dilemma's print shows the situation, never what to do. The two prints of a pair share a time of day and a distance, so neither looks like the better evening.
How a score is made
Every answer is a whole number from -3 to +3. On a two-sided scene it is how far you lean and which way: Strongly, Clearly or Slightly (3, 2 or 1), or Torn (0).
Puzzles land on the same range. In Share the bonus each coin is one step, from all six to one apprentice to all six to the other. In Plan the trip and Lanterns at bedtime each box you tick is one step, from none to all six. In Find your spot and Name the print each choice carries a value set in advance, 1 or 3 either way.
Where a question's positive end points toward the second pole, the sign is flipped. A scale's 8 answers are added up and placed on a 0-100 range:
score = 50 + 50 × (sum of keyed answers) ÷ (3 × 8)
- 100 means leaning all the way toward the first pole on every question, and taking every puzzle to that end.
- Scores are rounded to whole numbers, and your letter is the side of 50 you land on. An exact 50 goes to I, N, F or P, the convention in the type literature.
- Distance from 50 sets the label: under 6 points is balanced, 6-15 slight, 16-29 clear, 30 or more strong.
How sure is each letter?
A score is an average of 8 answers, and an average is only as firm as the answers behind it. So every score comes with a margin: the standard error of your answers on that scale, in score points. It treats the 8 scenes as a sample of every scene that could be asked about the scale (Nunnally and Bernstein, 1994).
Your likely range is the score plus or minus that margin, drawn as a pale band on every scale. In simulated respondents it held the true score about two times in three, as a range of one standard error should.
When the range reaches the midpoint, the letter is marked as a close call: a retake could plausibly flip it, so the report points you to the neighboring type as well. A share link carries the margins with the scores, so a shared report shows the same ranges.
8 answers on Energy. Their average gives a score of 40 with a margin of 3 points.
8 answers on Energy. Their average gives a score of 40 with a margin of 16 points.
Checks on how you answered
Research on careless responding uses the same signals (Meade and Craig, 2012). These are advice, not gates: you always get your report.
Rushing
If your typical answer comes in under 1.5 seconds, faster than most people can take in a scene, the test suggests reviewing before you see the report.
One lean for everything
Each scale puts its poles on the left and the right about equally often, so giving the same lean to 75% or more of the two-sided scenes cancels out. The test tells you so.
Leaning on Torn
Choosing Torn on 40% or more of the two-sided scenes pulls every score toward the middle. The test points it out, because even a slight lean sharpens a result.
From five scores to everything else
The report never asks you anything extra. Every view is calculated from the same five numbers.
- Your type
- The four letters come from Energy, Perception, Decisions and Structure. Your archetype, family and cognitive function stack follow from the letters, as in Myers' and Beebe's models.
- Your Big Five
- Each scale is read as its Big Five partner: Energy as Extraversion, Perception as Openness, Decisions as Agreeableness (Feeling end), Structure as Conscientiousness (Judging end) and Composure as Neuroticism (Sensitive end). These pairings are the ones found by McCrae and Costa (1989). They are estimates, not normed percentiles.
- Your compass region
- Stability is the average of Agreeableness, Conscientiousness and 100 minus Neuroticism; Plasticity is the average of Extraversion and Openness (Digman, 1997; DeYoung, 2006). Their distance from 50 gives a compass bearing. Within 7 points of the center you are at the Crossroads; otherwise the nearest of eight directions names your region. More on the model.
- Your Mindprint
- A height map with one hill per letter, placed by direction and raised by strength, on ground smoothed or roughened by Composure. Contour lines are traced through it. It is a picture of your numbers, not an extra measurement.
A yardstick for any short test
We have not yet run a retest study of our own. The most useful benchmark comes from the publisher of the MBTI, whose 92-question assessment is longer than ours and scores .88 to .91 on internal consistency for each letter pair.
When 236 people took it twice within 15 weeks, 51% got the same four letters and 90% kept at least three (Schaubhut et al., 2018). A shorter test is unlikely to do better, which is why your report marks close calls instead of hiding them.
Kept each letter
- E or I87%
- S or N86%
- T or F82%
- J or P84%
Letters kept out of four
- All four51%
- Three39%
- Two9%
- One2%
How common each type is
Type shares on this site come from the same publisher's US sample of 3,578 adults, recruited to match census targets (Schaubhut et al., 2018). Every letter and family share is added up from the sixteen type counts, so they always agree. They describe one country, one instrument and one point in time, and Koseiha results are not normed against them. See every type's share.
What it cannot tell you
It measures how you see yourself
Self-report reflects self-image as well as behavior, and even the puzzles ask what you would do, not what you did. People who know you well might place you differently, and both views are informative.
Close calls are genuinely close
A score of 53% is barely a preference, and your likely range shows how far it could move. Retest studies of type instruments find that people often change a letter between sittings, mostly on scales where they scored near the middle (Pittenger, 2005).
Types are a simplification
The traits underneath are continuous: most people score near the middle, not at the poles (McCrae and Costa, 1989). That is why we show every score, not just the letters.
It has not been independently validated
Our 40 scenes were written and drawn for this site to sit close to well-studied dimensions, in formats with a long research record. The scenes themselves have not been through a peer-reviewed validation or retest study. Margins describe how consistent your answers were, not whether the scenes measure what they aim to.
It is not a clinical or hiring tool
Nothing here diagnoses anything, and type should never decide who gets a job, a team or a relationship. Personality explains a modest share of how people actually behave.
Profiles describe tendencies
Type descriptions are written to fit most people of a type most of the time. Keep what rings true, discard what does not, and be wary of any description that seems to fit everyone (Forer, 1949).
Your privacy
The test runs entirely in your browser. While you take it, your answers and how long each one took are saved to your own device (local storage) so you can resume, along with which boxes you ticked in the puzzles. When you finish, they are replaced by your result: five scores and their margins, kept so you can revisit your report and see what has changed when you retake it. Your answers are never sent to us or anyone else. A shared link carries only your five rounded scores and their margins, never your individual answers.
To count visits and see which pages people read, Koseiha uses Google Analytics and Cloudflare Web Analytics. Google Analytics sets cookies in your browser and sends Google the address of each page you open, with your browser, device and rough location. Cloudflare, which delivers the site, counts visits without cookies and also measures how quickly each page loads. Both leave out the scores in a page's address (a report or compatibility link), so neither records your result, and neither ever sees your answers. A content blocker or your browser's privacy settings can turn them off, and the site works the same without them.
You can erase your saved results here at any time. Clearing this site's data in your browser settings does the same and also removes the analytics cookies.
References
- Jung, C. G. (1921/1971). Psychological Types. Collected Works, Vol. 6. Princeton University Press.
- Myers, I. B., with Myers, P. B. (1980). Gifts Differing: Understanding Personality Type. Davies-Black.
- Schaubhut, N. A., Thompson, R. C., Morris, M. L., and Arneson, J. J. (2018). United States (North American English) Supplement to the MBTI Manual for the Global Step I and Step II Assessments. The Myers-Briggs Company.
- McCrae, R. R., and Costa, P. T. (1989). Reinterpreting the Myers-Briggs Type Indicator from the perspective of the five-factor model of personality. Journal of Personality, 57(1), 17-40.
- Goldberg, L. R. (1993). The structure of phenotypic personality traits. American Psychologist, 48(1), 26-34.
- Digman, J. M. (1997). Higher-order factors of the Big Five. Journal of Personality and Social Psychology, 73(6), 1246-1256.
- DeYoung, C. G., Peterson, J. B., and Higgins, D. M. (2002). Higher-order factors of the Big Five predict conformity: Are there neuroses of health? Personality and Individual Differences, 33(4), 533-552.
- DeYoung, C. G. (2006). Higher-order factors of the Big Five in a multi-informant sample. Journal of Personality and Social Psychology, 91(6), 1138-1151.
- DeYoung, C. G., Quilty, L. C., and Peterson, J. B. (2007). Between facets and domains: 10 aspects of the Big Five. Journal of Personality and Social Psychology, 93(5), 880-896.
- Myers, I. B., McCaulley, M. H., Quenk, N. L., and Hammer, A. L. (1998). MBTI Manual: A Guide to the Development and Use of the Myers-Briggs Type Indicator (3rd ed.). Consulting Psychologists Press.
- Hicks, L. E. (1970). Some properties of ipsative, normative, and forced-choice normative measures. Psychological Bulletin, 74(3), 167-184.
- Paulhus, D. L. (1991). Measurement and control of response bias. In J. P. Robinson, P. R. Shaver, and L. S. Wrightsman (Eds.), Measures of Personality and Social Psychological Attitudes (pp. 17-59). Academic Press.
- Mischel, W., and Shoda, Y. (1995). A cognitive-affective system theory of personality: Reconceptualizing situations, dispositions, dynamics, and invariance in personality structure. Psychological Review, 102(2), 246-268.
- McDaniel, M. A., Hartman, N. S., Whetzel, D. L., and Grubb, W. L. (2007). Situational judgment tests, response instructions, and validity: A meta-analysis. Personnel Psychology, 60(1), 63-91.
- Lievens, F., De Corte, W., and Schollaert, E. (2008). A closer look at the frame-of-reference effect in personality scale scores and validity. Journal of Applied Psychology, 93(2), 268-279.
- Cao, M., and Drasgow, F. (2019). Does forcing reduce faking? A meta-analytic review of forced-choice personality measures in high-stakes situations. Journal of Applied Psychology, 104(11), 1347-1368.
- Nunnally, J. C., and Bernstein, I. H. (1994). Psychometric Theory (3rd ed.). McGraw-Hill.
- Meade, A. W., and Craig, S. B. (2012). Identifying careless responses in survey data. Psychological Methods, 17(3), 437-455.
- Quenk, N. L. (2002). Was That Really Me? How Everyday Stress Brings Out Our Hidden Personality. Davies-Black.
- Roberts, B. W., Walton, K. E., and Viechtbauer, W. (2006). Patterns of mean-level change in personality traits across the life course. Psychological Bulletin, 132(1), 1-25.
- Soto, C. J. (2019). How replicable are links between personality traits and consequential life outcomes? Psychological Science, 30(5), 711-727.
- Pittenger, D. J. (2005). Cautionary comments regarding the Myers-Briggs Type Indicator. Consulting Psychology Journal: Practice and Research, 57(3), 210-221.
- Forer, B. R. (1949). The fallacy of personal validation. Journal of Abnormal and Social Psychology, 44(1), 118-123.
- Malouff, J. M., Thorsteinsson, E. B., Schutte, N. S., Bhullar, N., and Rooke, S. E. (2010). The five-factor model of personality and relationship satisfaction of intimate partners: A meta-analysis. Journal of Research in Personality, 44(1), 124-127.
Myers-Briggs Type Indicator, Myers-Briggs and MBTI are trademarks or registered trademarks of The Myers-Briggs Company. Koseiha is independent and is not affiliated with or endorsed by The Myers-Briggs Company.
Take the test