The score sheets required the judges to rank their 4 samples
from least (=1) to most (=4) pleasant for each attribute (Meilgaard
et al., 1999). The rank sum of each sample was used to calculate
Friedman value (F). Comparing it with values given in the Friedman
test (level of significance = 0.05), it can be concluded whether there
is an overall difference among all the Kougelhopf samples.