The open color study

We plan to publish what our quiz results look like in aggregate: how common each color is, how often people sit near the middle of an axis and how gut answers compare with considered ones.

Status: planned. We have not collected any study data yet, so there are no numbers on this page. We'll publish the first report only once we have enough opt-in responses to be meaningful.

What we'll look at

How we'll protect privacy

What it won't be

This won't validate the model as a psychological test. It's a descriptive look at how people answer a fun quiz. We'll publish our method and limitations with the data.

Questions or want to collaborate? Email hello@whatsmypersonalitycolor.com.

Personality colors are a popular framework for self-reflection and conversation, not a clinical or scientific diagnosis. Disclaimer · How the quiz works

Why we are planning a study at all

Most personality color sites never publish anything about their own results. We think that is a missed opportunity to be transparent. If many people choose to share their anonymous results, we can report simple, honest facts: how common each color is, how often people land near the middle of an axis, and whether fast answers from Under Pressure mode differ from considered ones. None of this would validate the model as science, but it would show readers how our results behave in practice.

How participation will work

  • Participation will be opt-in only, with a clear checkbox at the end of the quiz. The default will be not to share.
  • We will collect only anonymous answer patterns and results, not names, emails or precise locations.
  • Small groups will not be reported separately, so no individual can be identified from a published table.
  • The first report will appear only when there are enough responses for the numbers to be meaningful. Until then this page will keep saying that no data has been collected.

What we will not claim

Even a large sample of quiz results would not show that personality colors are real types. Research on personality, summarized in reviews such as Goldberg's work on the Big Five, finds that traits vary continuously rather than falling into neat boxes. Our report will describe our tool, not human nature. We will also publish the method, including the exact questions, so anyone can check our work.

Questions we hope to answer

  • Are some colors much more common than others, or are they spread fairly evenly?
  • How many people are close to the middle on at least one axis, and so receive a mixed shade?
  • Do Under Pressure results tend to shift in a consistent direction compared with the full quiz?
  • Do friend mirror results agree more with self-results on pace than on focus, as research on observable traits would predict?

If you have a suggestion for the study design, use the contact page. We would especially welcome comments from researchers.

Sources

  1. Goldberg, L. R. (1990). An alternative "description of personality": the Big-Five factor structure. Journal of Personality and Social Psychology, 59(6), 1216-1229. doi.org/10.1037/0022-3514.59.6.1216

What the numbers would mean, and what they would not

If we ever publish a table, it will be a description of this quiz, not of human nature. Each person who opts in would contribute three axis totals. Six items are scored on each axis, from −2 to +2, and the result page turns each total into a percent. A future report could say how often each of the eight colors occurred, how often a percent fell near 50, and how often the nine-question timed version picked a different color from the full 18. Those are facts about answer sheets. They do not show that the colors are natural kinds. Lewis Goldberg's 1990 paper, already our reference for the Big Five, is the reminder: the stable patterns in personality language are dimensions, not eight bins. A lopsided color, say one name appearing far more often than the others, would tell us our items are easier to agree with on that side. It would not tell us that most people "are" that color.

We have not collected the opt-in file. There is no sample size to report, and this page will keep saying so until there is. Kids pages will never be in the file. We will not publish a cell smaller than 50 people. We will not claim the study validates hiring, grading or diagnosis, because a pile of anonymous quiz rows cannot do that. If you are a researcher and the design is missing a safeguard, write through the contact page before any switch is turned on. The commitment is the point of publishing the plan while the dataset is still empty.