Sas 9.1 statistical

One way and two way analysis of variance techniques were used to find out the effect of treatment, storage and their interaction. Duncan Multiple Range Test was used to denote the significant difference among the treatments on SAS 9.1 statistical software. Results were declared significant al p-value 0.05 (P ≤ 0.05) [23 ].

The experiment was conducted in a completely randomized design with four treatments and five replicates per treatment. The statistical model is represented by the Eq. (5): $$Yi=\mu +\mathrm{\alpha }\text{i}+\beta \left(X,i,-,\overline{X}\right)+\mathrm{\delta }\text{ij},$$ where Yi was the observed dependent variable, μ was the overall mean, αi was the effect of the diet, β was the regression coefficient or functional relationship with the covariate, Xi the observed value of the covariate applied to the experimental unit, $\overline{\text{X}}$ the mean of the covariate and δij was the random error.
The data were subjected to analysis of variance through the PROC GLM command of the SAS statistical package (SAS version 9.1 2003), and the means were subjected to regression analysis through the PROC REG command of the SAS^® (9.1) statistical package (SAS University Edition). The initial weight was used in the statistical model as a covariate when significant. Analysis of ingestive behaviour also included a t-test to verify the effect of the period (daytime and night-time). Significance was declared when P ≤ 0.05.