Percentile rank converts scores into position—"better than 80% of the group"—with the small-sample traps to avoid.
Percentile rank is the share of the group at or below a score. It survives any distribution shape, which makes it the most honest way to position a score in a cohort—80th percentile means ahead of 80% of the group, not 80% correct.
PR = (# below + 0.5 × # equal) / N × 100
In a 12-person workshop, moving one question flips several percentiles—small-cohort reports show raw score + band, not PR.
Benchmarking, admissions-style ranking and any client question of "how good is this, really"—with n ≥ 30 for stable readings.
The classic confusion: percentile ≠ percentage correct. Also, never average percentiles across dimensions—average the z-scores, then convert back.
It's position, not quality—if the cohort is elite, 90th can be thin. State the reference group every time.
Report bands (top/middle/bottom third) instead; exact PRs wobble with every response.
Turn the concept into a live assessment：Percentile Rank
Browse templates →