Outlier Test
A statistical procedure for deciding whether a suspect result differs enough from the others to be rejected.
Formula / equation
Q = |suspect − nearest| / range
Unit
Dimensionless
In depth
Dixon's Q test compares the gap between the suspect value and its nearest neighbour with the total range, rejecting it if the ratio exceeds a tabulated critical value. Results should never be discarded merely because they are inconvenient, and any rejection must be reported.
Examples in the real world
Applying the Q test to titres of 24.55, 24.57, 24.56 and 25.10 mL to justify rejecting the last.