site
stats
Making Count-Based Claims from Sample Data (2019) | Ifpress.com
top of page

TECHNICAL REPORTS

Proper Task Instructions and the Two-out-of-Five Test

Ennis, J. M., Rousseau, B., and Ennis, D. M.  (2012). IFPress, 15(4) 4-5.

Abstract:

 

Difference tests continue to play a prominent role as businesses investigate the sensory effects of ingredient or process changes. Even so, the importance of correct instructions in these tests is not always appreciated - incorrect instructions can reduce the ability of a test to detect real differences. The effect of instructions on the outcome of sensory difference testing was recognized when the difference between the 3-AFC and Triangle test was explained and when the relative power of difference testing methods was explored. This topic has resurfaced as other difference testing methods, such as the Two-out-of-Five test, have become more commonly used. In this report, we explain how incorrect instructions can sabotage a method, and we provide guidance for the optimal set of test instructions for the Two-out-of-Five test.

15-4.jpg

Figure 1. Three cases illustrating the difference between expected results under two instructions - “group the samples into a group of two and a group of three” (Grouping) and “choose the same pair” (Same Pair). The Same Pair instructions are marked by circling the selected responses and the Grouping instruction by underlining the groups.

bottom of page