Abstract
ABSTRACT.Three methods of standardizing magnitude estimation data, external calibration, modulus normalization, and equalization were examined using a sensory evaluation data set arising from an incomplete block experiment testing five gels of varying firmness. Both the original data and a logarithmic transformation of the data were analyzed. Instrumental data were also collected. When untransformed data were analyzed the method of standardization profoundly affected tests of significance, coefficients of variation (%CV), and estimation of the power function relating the sensory data to the concentration of the underlying gel. The logarithmically transformed data lead to results independent of the standardizing technique and with higher F‐ratios, lower %CV's and normally distributed errors.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
