A statistical method used to determine if two samples were taken from different populations, by comparing the proportions of a characteristic that appears in both samples. If the difference in proportions is large enough, the samples are judged to have come from different populations.