Some aspects of the statistics of small numbers of triangular taste tests

Abstract
Summary. The binomial distribution of random selections of odd samples in small numbers of triangular taste tests has been applied to estimate: (a) scores establishing a difference between samples, (b) the extent of true discrimination shown by any significant score, (c) the fixed‐panel size needed to establish difference or similarity, and (d) decisive scores in a sequential test scheme. It is shown that conclusions may be reached with fewer tests than the published literature suggests, thereby affording a saving of tasting effort.

This publication has 2 references indexed in Scilit: