Measurement of Discrimination Ability in Taste Tests: An Empirical Investigation
比较重复配对比较、三角测试和偏好排序三种任务在测量消费者辨别产品配方差异能力上的效果，发现重复配对比较最敏感，偏好排序最不敏感。
For purposes of product testing, several tasks have been used to measure subjects’ discrimination ability (i.e., their ability to distinguish between two slightly different product formulations). Three of the more common tasks are repeat paired comparisons, triangle tests, and preference rankings. In this empirical study, the properties of these three tasks are compared. The repeat paired comparison test is found to be the most sensitive discrimination task, in that subjects demonstrate the greatest ability in distinguishing between formulations, and preference ranking is the least sensitive. The finding that discrimination ability measured by triangle tests is significantly correlated with that measured by paired comparisons lends support to the validity of consistent preference discrimination testing.