Abstract:

 

A recent survey found that, when using discrimination methodologies, sensory and consumer research scientists in CPG companies focus predominantly on the issue of sensory equivalence.  They want supporting evidence that a change to an existing brand (e.g., formulation, process, raw materials) will not be rejected by consumers.  While the issue is equivalence, the typical approach in discrimination testing is to run a statistical test against the Null Hypothesis of no difference.  This is done even though two samples will never be identical and thus a conclusion of difference will always be reached provided that the test power (method, sample size) is high enough.

In this webinar, we will describe the basic principles behind testing for sensory equivalence and illustrate how using the incorrect Null Hypothesis of no difference will result in missed opportunities for product changes when an experiment sample size is increased.

​

Equivalence testing requires setting a consumer-relevant action standard against which future product changes will be compared.  We will review three standard indices for an action standard:  Proportion of tests correct, proportion discriminators in the population, and Thurstonian delta.  This webinar will show you how the Thurstonian approach is the only one providing consistent scaling across methodologies, an essential feature for reliable decision-making since discrimination paradigms differ in power (e.g., tetrads vs. triangle and duo-trio methods). It also permits the creation of a trained panel/consumer sensitivity link to ensure that the action standard is consumer relevant.

​

This webinar is intended for a general audience of industrial and academic sensory professionals and graduate students.

​

A digital recording of this webinar is available for purchase ($289).