Multiple comparison procedures involving Friedman rank sums for the analysis of ranked data have recently been proposed as alternatives to tests developed by Kramer. Expanded tables for these multiple comparison procedures (involving the "all treatments" and "treatments versus control" comparisons) were presented to accomodate the range of panelists and samples usually encountered in sensory evaluation experiments involving ranked data.

The use of ridge regression is proposed to overcome the problem of multicollinearity associated with the use of multiple linear regression in food technology. This technique is discussed and illustrated with an example from food technology.

A new procedure for the analysis of sensory evaluation data is discussed. This procedure involves the linear logistic model and does not require the use of any category scaling methods such as the 9-point hedonic scale, but utilizes the ordinal nature of the data. This procedure is illustrated with data from sensory evaluation experiments.

Food science requires the ability to measure the amounts of substances under study. As new methods are developed, it is necessary to compare these methods in terms of their precision and accuracy. In most cases, since in error‐free or referee method is not available, there are no “exact” values with which these new method measurements can be compared. To overcome these difficulties, a procedure for comparing methods, which incorporates principal component analysis, is discussed. This procedure is illustrated with an example where methods for the determination of the cholesterol content of cod muscle are compared.

No abstract