datawithaplot.org · a demonstration of construct validity
Answer a few quick questions and our model will generate a profile tailored to you. Then you'll tell us how accurate it is. Be honest; no one is grading you, and there are no wrong answers.
Takes about a minute. Nothing you enter leaves your browser.
The key point: every visitor to this page sees the exact same profile. It's a fixed set of statements, identical for everyone, written to feel personal. And
A typical spread of accuracy ratings on a 0 to 5 scale. Your rating is highlighted; the dark line marks Forer's 1949 average of 4.26.
This is the Barnum effect (or Forer effect): people rate vague, generic descriptions as highly accurate when they believe the descriptions were made just for them. In 1949 the psychologist Bertram Forer gave his students a personality "test," then handed every single one of them the same profile. They rated it 4.26 out of 5 for accuracy.
The profile you read is built from five subtle techniques. Here is the full profile, with the mechanics laid out:
A profile that fits everyone equally well tells you about no one. A measurement that can't tell people apart isn't measuring anything. That's construct validity: the question of whether a tool captures the real thing it claims to. The next time a dashboard, an engagement score, or a "data-driven" personality assessment feels strikingly accurate, ask the Barnum question: would it feel just as true if it were about someone else?