ABSTRACT

Following is a discussion of the words that I have italicized in this definition.

Analyzing and interpreting relate to statistics, and I would say that 50% or more of the science sustaining sensory evaluation is statistical science. Some people just do not like statistics or are cagey about believing in its methodology; coping with sensory evaluation will be a hard task for people thus inclined. As will be seen in future chapters of this book, statistics play a fundamental role in predicting SSL. One aspect that is not contained

in the above definition and is related to statistics is design. The first part of the definition could read: “Sensory evaluation is a scientific discipline used to design, evoke, meas ure, analyze and interpret…” Statisticians are forever warning practitioners to design their experiments with anticipation to ensure maximum efficiency and analyzable results. Taking this advice is particularly necessary in SSL studies, as will be discussed in Chapter 3. For example, the simple exercise of calculating the number of samples to be stored at a desired temperature is fundamental to ensure sufficient material is available for trained panels, consumer panels, physicochemical and/or microbiological studies. Very often, SSL studies can extend for weeks or months. If careful and proper designs are not considered, the experiment can come to an end with no meaningful results; or halfway through the study the experimenter suddenly realizes that he is running out of control samples due to not having considered that at each time point a control and blind control are necessary for correct sensory evaluation.