Skip to main content
. 2018 Sep;56(3):381–397. doi: 10.17113/ftb.56.03.18.5285

Table 3. Analysis of variance (ANOVA) of mixture design and regression equations for bacterial change cycles under gastric pH and bile salt conditions.

Bacterial change cycle
Gastric pH Bile salt
Source Sum of squares Df Mean square F-value p-value Sum of squares Df Mean square F-value p-value
Model 1.86 6 0.31 246.54 <0.0001* 2.36 5 0.47 60.03 <0.0001*
Linear mixture 0.27 2 0.14 109.09 <0.0001* 1.86 2 0.93 118.13 <0.0001*
AB 0.34 1 0.34 272.17 <0.0001* 0.01 1 0.01 0.90 0.36
AC 1.35 1 1.35 1075.74 <0.0001* 0.26 1 0.26 33.19 0.0001*
BC 0.00 1 0.00 1.07 0.3245 0.21 1 0.21 26.60 0.0003*
ABC 0.24 1 0.24 186.77 <0.0001*
Residual 0.01 10 0.00 0.09 11 0.01
Lack of fit 0.01 5.00 0.00 1.77 0.27 0.07 6.00 0.01 2.94 0.13
Pure error 0.00 5.00 0.00 0.02 5.00 0.00
Corr. total 1.88 16.00 2.44 16.00
R2 0.99 0.96
Final equation in terms of real components N/(log CFU/mL)=1.052A+0.692B+1.282C–
–2.493AB–3.965AC+12.521ABC
N/(log CFU/mL)=0.006A+0.504B+0.935C–
–1.589AC+1.484BC

A=maltodextrin, B=gum arabic, C=whey; *significant at α=0.05