Abstract In this paper a possible strategy for selection of a reduced sample-set for preference testing is discussed. The sensory variation within commercially available dry fermented lamb sausages in Norway and their relationship to standard chemical measurements are also studied. Fourteen dry fermented lamb sausage samples were analysed by conventional sensory profiling (quantitative descriptive analysis — QDA) using 11 trained assessors. Principal component analysis (PCA) indicated that juiciness, colour tone, acidic flavour and rancid flavour were closely related to the first principal component, and spiciness and flavour of smoke to the second component. Partial least squares (PLS) regression indicated that fat content and juiciness were weakly correlated. Further-more, sausages with the highest intensity of juiciness were all high in water content and had high water activity. As salt increased, sausages became drier and harder. Sensory saltiness was highly correlated to salt content.
[1]
S. Wold,et al.
The multivariate calibration problem in chemistry solved by the PLS method
,
1983
.
[2]
Tormod Næs,et al.
Using cumulative ranks to detect individual differences in sensory profiling
,
1994
.
[3]
M. Stone.
Cross‐Validatory Choice and Assessment of Statistical Predictions
,
1976
.
[4]
J. R. Piggott,et al.
Sensory analysis of foods.
,
1988
.
[5]
L. M. Eadie,et al.
Assessment of the Sensory Characteristics of Meat Patties
,
1991
.
[6]
Tormod Naes,et al.
A graphical technique for assessing differences among a set of rankings
,
1994
.
[7]
Peter J. Zemroch,et al.
Cluster Analysis as an Experimental Design Generator, With Application to Gasoline Blend ing Experiments
,
1986
.
[8]
T. Næs.
The design of calibration in near infra‐red reflectance analysis by clustering
,
1987
.