2.6. Data analysis
The potential growth of Listeria monocytogenes in all products was
predicted using mathematic models developed by the Danish Meat
Research Institute (Mejlholm et al., 2010).
The data for the total count and the lactic acid bacteria growth
were analysed by the following model:
LogðcountÞ ¼ μþsalt level þ day þ salt level day þ ε
The sensory data were analysed using Panelcheck (version 1.3.2).
Analysis of variance was performed using the following model:
Attribute ¼ μ þ salt level þ ASSESSOR þ ASSESSOR SALT LEVEL þ ε
in which the effects written in capital letters were included as
random effects.
Principal Component Analysis (PCA) was performed using original
data.
The sensory shelf life data was analysed using SAS (version 9.2).
Based on a visual evaluation of relevant distribution, it was decided to
analyse the data using a Weibull distribution. The statistical analysis
can be described as a one-way analysis of variance, with Y as the logarithmof
the time of “not acceptable” and ε following aWeibull distribution
(Lawless, 2003).