Revision of the performance evaluation methods of sensory panelists performing descriptive analysis

Thursday, June 15, 2017

Authors: László Sipos, Márta Ladányi, Zoltán Kókai, Attila Gere


1.     Summary

Sensory tests form the basis for sensory science. Sensory science uses human senses as measurement tools. During sensory tests, the properties of a product are evaluated by sensory panelists and by a sensory panel consisting of them. Decisions made after sensory tests are fundamentally determined by the quality of the data experienced, therefore, the quality of sensory data is determined by the trained and expert sensory sensory panel and its members. In our work, revision of the correlation and regression methods recommended by the standard titled „MSZ ISO 11132:2013 Sensory analysis. Methodology. Guidelines for monitoring the performance of a quantitative sensory pa­nel” are described, and corrections are suggested.

2. Introduction and literature review

According to Kermit és Lengard [1], a good sensory sensory panel should provide precise, discriminative and accurate results. An ideal group performance can be achieved if the products are differentiated by each member of the group (“great product variety”), and the same results are obtained several times (“small variation for the individual panelists”). However, there should be agreement with the other panel members regarding the sensory property, within a given tol­erance (small variation among panelists) [2],[3],[4]. The task of the sensory panel leader is to collect the necessary information about members of the panel during sensory tests. By monitoring and following up performance it can be ensured that panel members and panel are capable of distinction, their results are constant, repeatable and free of error [5].

Based on their training, sensory panelists are clas­sified into three categories by the literature: naive assessors (consumers), trained panelists, expert panelists. Panelists of different training levels are re­quired for different tasks [6], [7], [8].


To view more contant please login or if you are not registered, register, please.

Subscription fees for Journal of Food Investigation:


for 1 year (4 issues): 4200 HUF (4000 HUF + tax)

1 issue: 
1260 HUF (1200 HUF + tax)

For further information, you are kindly asked to contact us at contact or write to


Thank you