Abstract
Two categories of quality assessment of foodstuffs by panels of judges may be distinguished. In 'grading', an absolute assessment representative of the generality of consumers is sought. In 'analysis', maximum sensitivity is desirable and emphasis is shifted from absolute to relative assessments. In four series of 'grading' tests, individual ratings were most variable in the quality region close to the lower limit of acceptability, thus increasing to about 30 the calculable size of panel required to distinguish differences of the order of 5% from an assigned standard. The threshold concentration of primary taste substances detectable varies considerably between individuals, but except in extreme cases no consistent relation between taste acuity alone and palatability judgments was indicated. However, the judging characteristics of individuals may be investigated numerically by computing the correlation coefficients and regression equations relating their assessments to the average of those of all other members of the same panels. In this way a range of sensitivity of the order of 40% was demonstrable in the tests under review, making possible an objective evaluation of the suitability of individuals for 'grading' or 'analytical' tests.

This publication has 1 reference indexed in Scilit: