How this personality test works
Implementation notes from MockMemo. Updated 5 October 2026.
MBTI Atlas offers an independent questionnaire for exploring personality preferences. This page describes what our code does and what the resulting scores can support.
Three versions, the same four preference pairs
The quick test contains 20 questions: five for each of E/I, S/N, T/F and J/P. The 70-question version contains 18 E/I, 18 S/N, 17 T/F and 17 J/P questions. The full version contains 21 E/I, 26 S/N, 24 T/F and 22 J/P questions. The shorter versions select questions from the same bank; they are not independent instruments.
How answers become a type
Each question offers two choices assigned to opposite poles of one preference pair. Each answer adds one count to its chosen pole; questions have equal weight within their pair. The pole with more answers supplies the corresponding letter. On an exact tie, our implementation chooses I, N, F or P. This is a deterministic display rule, not evidence that a tied preference has been established.
What a percentage means
The displayed percentage is the share of your answers within that pair, rounded to a whole number. The other side is its complement, so each pair totals 100%. It is not a percentile, confidence interval, probability of a type, or clinical score. A 60% I result means 60% of your E/I answers favored I in this session.
What the test does not establish
We have not published reliability, validity, population norms or a study comparing these versions with an official assessment. We cannot substantiate an accuracy percentage. The four-letter scoring does not directly measure eight cognitive functions; the function material is an explanatory framework to explore after the test.
Progress, results and privacy
While this tab remains open, session storage can keep answers and your current question for up to 24 hours, if browser storage is available. Restarting clears that draft. Submission creates a result with its preference percentages, type, timestamp and a random identifier; individual answers are not saved in the result database. Anyone with a private report link can view it. The default share action links to the public type guide without sharing your scores. Optional analytics, after consent, measures quiz milestones without answers, scores or result identifiers.
Questions before you go further
Is this the official MBTI assessment?
No. MockMemo operates this independent questionnaire. It is not affiliated with or endorsed by the Myers & Briggs Foundation or the official assessment provider.
Will 93 questions guarantee a more accurate result?
No. More questions sample more situations, but no published validation establishes an accuracy advantage for these versions.
Why can my result change?
How you interpret a question and your current context can affect answers. Near ties are especially sensitive to small changes. Read the neighboring type descriptions and compare concrete experiences.
For the official assessment and its interpretation, see the Myers & Briggs Foundation. This site provides a separate questionnaire for self-reflection.