Personality test sites usually keep their mechanics vague, partly to protect commercial IP and partly because vagueness photographs well. We don’t have IP worth protecting and would rather you know what you’re taking. This page explains how the four tests on TraitTally are built, exactly how each is scored, and what corners a free test cuts.
Where the questions come from
We wrote all the questionnaire items ourselves. For each framework, we started from its published structure (the four MBTI dichotomies, the five OCEAN dimensions, the nine Enneagram types, Marston’s four DISC factors) and wrote items to cover each dimension in everyday language, then edited them against two rules: no item should require a specific job or life situation to answer, and no item should telegraph its “good” answer. We did not license items from the commercial publishers, which is why the site says “in the MBTI tradition” rather than claiming to be the official instrument, and it’s also why our results and the paid versions’ won’t match item for item.
The answer format differs by test on purpose. The 16-type and Enneagram tests use forced choices between two options, because those frameworks claim to sort people into categories, and forcing a choice is the honest version of that claim. The Big Five and DISC tests use five-point agree/disagree ratings, because those models treat traits as continuous and a rating scale preserves the shades.
How each test is scored
The scoring code runs in your browser, and this is all of it in prose:
16 Types (70 items). Every question adds one point to one pole of one dichotomy. After 70 questions, each of the four dimensions is a simple tally, say E 11 / I 6, and your letter is the side with more points. A tie goes to the first-listed side, which is a design compromise you should know about: an 11-6 split and a 9-8 split print the same letter with very different confidence. The results page shows the tallies so you can see which kind of letter you got.
Big Five (50 items). Ten items per dimension, rated 1 to 5. About half the items are reverse-keyed (“I leave my belongings around” counts against conscientiousness), which is a standard defense against straight-line agreeing. Your ratings per dimension are averaged, then rescaled linearly to 0-100. So a 75 means your average answer was 4 out of 5. It does not mean you scored higher than 75% of people; see the section below.
Enneagram (36 items). Each option credits one or more of the nine types with a point. Your type is the one with the most points at the end, and the runner-up scores are shown, because with nine candidates and 36 questions, close finishes are common and you deserve to see them.
DISC (20 items). Five items per style, rated and averaged like the Big Five, then rescaled. Your profile is all four scores with the top two labeled primary and secondary.
What we deliberately don’t do
There is no norm sample behind these scores. A professionally normed instrument compares you against a reference group (“higher than 82% of working adults”); building one requires collecting thousands of people’s answers, and we made the opposite choice — answers never leave your device, which means we couldn’t build a norm sample without breaking the privacy design. That trade is worth stating plainly: you get privacy and a free test; you give up percentile rankings.
For the same reason, we can’t publish reliability coefficients for our own items the way a test manual would. What we can tell you is what published research says about each framework itself, and it says different things: the Big Five’s factor structure has replicated across dozens of cultures, while the MBTI’s retest stability and the Enneagram’s evidence base have both drawn serious criticism in the literature. The test guides carry those receipts.
Sources
The claims on our test pages trace to these, all real and checkable:
- Jung, C. G. (1921). Psychological Types.
- Marston, W. M. (1928). Emotions of Normal People.
- Barrick, M. R., & Mount, M. K. (1991). The Big Five personality dimensions and job performance: a meta-analysis. Personnel Psychology, 44.
- Costa, P. T., & McCrae, R. R. (1992). NEO PI-R Professional Manual.
- Goldberg, L. R. (1993). The structure of phenotypic personality traits. American Psychologist, 48.
- Pittenger, D. J. (2005). Cautionary comments regarding the Myers-Briggs Type Indicator. Consulting Psychology Journal: Practice and Research, 57.
- Hook, J. N., et al. (2021). The Enneagram: a systematic review of the literature and directions for future research. Journal of Clinical Psychology, 77.
If you find an error on any page, write to contact@traittally.com. Corrections ship as site updates, same as code.