A strict preference task found a confidence gap.
Gershoff and Johar studied how people estimated a friend’s knowledge of their preferences. In one study, 109 friend pairs rated 40 movie categories on seven-point scales. Targets expected their friend to reproduce 64% of those ratings exactly; exact agreement was 22% in that task.
Greater relationship involvement increased expected accuracy, but it did not increase actual exact accuracy there. That is a calibration result: confidence about item-specific knowledge can outrun item-specific performance.
Read the evidence literally.
Exact friend predictions of movie-category ratings on a seven-point response task.
A large gap between expected and actual exact agreement in one 109-pair study.
A universal percentage for friendships, closeness, empathy, or a ViewsMeet score.
Use a miss as a prompt: “What made you expect that?”—never as a friendship grade.
Familiarity can still carry useful information.
Funder and Colvin found that close acquaintances’ personality ratings agreed more with each other and with targets’ self-ratings than stranger ratings did. That study measured personality judgment, not exact everyday preference prediction. Taken together, the studies support a cautious inference: familiarity can reveal patterns while an exact choice remains easy to misread.
What a quiz cannot tell you.
An exact prediction score cannot measure friendship strength, loyalty, honesty, care, or whether someone is a “real friend.” Different questions, moods, response scales, and contexts create different tasks. A high or low result belongs to one frozen set and one round.
TRY THE LOW-STAKES VERSION
Five picks, then ask why.
ViewsMeet Quick Read lets you lock five choices and send one private link. Your friend predicts yours, adds theirs, and sees Knowledge + Match. The result is deterministic, but it is not a validated relationship measure.
Make a best-friend Quick Read →Original sources.
- Gershoff & Johar (2006), DOI 10.1086/500479
Consumer calibration of friends’ knowledge across three studies. - Funder & Colvin (1988), DOI 10.1037/0022-3514.55.1.149
Acquaintanceship, agreement, and personality judgment.