Grubbs' Test 2D: Dot-Plot Outlier Rejection

Grubbs' test is a classic statistical procedure for deciding whether the most extreme value in an approximately normal sample is a genuine outlier rather than ordinary variation. This 2D simulator lays a synthetic sample out as a stacked dot plot — every value bucketed into a bin and drawn as a column of dots — and pairs it with a live trace chart that plots the G statistic and its shrinking critical threshold across every iteration of the test. Step through the test one rejection at a time or run it to completion: rejected points fall out of the dot plot into a rejection tray while the trace chart's G_crit line steps down as N shrinks, making the iterative logic used to screen extreme lab readings and QC measurements directly visible as it happens.