Hard-Easy Effect

What is the Hard-Easy Effect?

The hard-easy effect is a pattern in confidence judgments: people tend to be overconfident on harder question sets and underconfident on easier ones. Their confidence exceeds their accuracy on the hard questions and falls below it on the easy questions.

An example of the hard-easy effect

Imagine answering two quizzes. On the difficult quiz, you average 60% confidence but get only 40% right. On the easy quiz, you average 80% confidence and get 95% right. Those hypothetical results illustrate the pattern. They do not explain why it happened.

Why the explanation matters

It is tempting to turn this into a story about ego: people overestimate themselves when a task is difficult. But Juslin and colleagues (2000) found that statistical artifacts can account for much of the hard-easy pattern. How questions are selected and how confidence is measured can change the result. A striking graph is not, by itself, evidence of self-deception.

If you want to examine your own calibration, record confidence before seeing the answers, then compare it with actual results across many comparable questions. That gives you something to inspect. A general feeling that you are “good at difficult things” does not.

The hard-easy effect is distinct from the worse-than-average effect, which concerns comparisons with other people. Being wrong about your score and being wrong about your rank are different judgments.