Abstract:

In sensory evaluation, panels and panelists are used as instruments that are the key tool to obtain reliable sensory
data. Assessment techniques for their performance can provide an important approach for managing them effectively. In this
paper, the validity and reliability are used as the main indexes for performance assessment, which have reached a plateau of
research since the beginning of the 21st century. More than 85% of the research papers have been reported in the top journals
in sensory science, such as Food Quality and Preference and Journal of Sensory Studies. Especially, the majority of the
research reports come from the USA and Europe and have tended to be frequent and mature in assessing the performance
of descriptive sensory panels-panelists. The various methods include univariate or multivariate variance analysis, and
multivariate statistical methods such as principal component analysis (PCA), generalized procrustes analysis (GPA), etc. At
the same time, softwares such as PanelCheck, Compusense Five, and so on have been developed, while the International
Organization for Standardization and the American Society for Testing and Materials are also formulating the relevant
standards on quantitative descriptive analysis. By contrast, relatively insufficient research has been done on performance
assessment in ranking and discrimination sensory panels-panelists. Six key research contents are proposed for future study.
They include the number and property of sample, the form of data presentation, the design of sensory experiment, expected
value and confidence interval, and the selection of sensory analysis methods and statistical methods. This paper will be
helpful to establish the unified and systemic framework for assessing performance in panels and panelists.

Key words: sensory evaluation, panel, panelist, performance assessment