2.4. Statistical analyses
All experimental data was analyzed in accordance with the
General Linear Model Procedure established by the Statistics
Analysis Systems Institute (SAS, 1996). The individual pen
served as the experimental unit. The results obtained from pigs
that were fed the CON diet were compared with those that
received the BioPlus 2B® diets via the polynomial regression
(Peterson, 1985) method in order to determine the linear and
quadratic effects on growth performance and CTTAD for DM
and N. The variability of all of the data was expressed as the
standard error (SE) and a probability level of P b 0.05 was
considered to be statistically significant, whereas a P b 0.10
was considered to constitute a tendency. In addition, significant
differences in the slurry noxious gas emission and the pH
among dietary treatments were analyzed by Duncan's test
(Duncan, 1955) using the GLM procedure, with a P b 0.05
being used to indicate significance.