Design expert software v11

The experimental results were examined methodically using the Design-Expert software V11 (StatEase, Minneapolis, MN, USA). A second-order quadratic model is developed to predict surface roughness. As reported in [35 ], ANOVA has been used to determine the model’s suitability. The ANOVA table for Ra prediction is shown in Table 4.
The F-value of 19.96 suggests that the model is significant, where this large F-value might occur due to noise of 0.01% of the time. p-values less than 0.0500 indicate that the model terms, i.e., the model terms V_c, D_c, V_c, F_z, and V_c² are significant. On the other hand, values over 0.1000 suggest that the model terms are not significant. Model reduction may help for a model with many insignificant terms (except those necessary to enable hierarchy).
The F-value of 0.2783 for the lack of fit suggests that it is minimal compared to the pure error. A substantial lack of fit F-value may occur 95.96% of the time due to noise, and a non-significant lack of fit is preferred. The following Design-Expert software’s (StatEase, Minneapolis, MN, USA) regression equation of the objective factors was obtained upon running the regression.

In this study, the effects of extraction parameters on flavonoid yields were analysed via a single-factor assisted one-way analysis of variance (ANOVA) test (p < 0.05 significance level; SPSS software v. 18 for Windows, IBM, New York, United States) and the response surface method (Design Expert Software, v. 11, Stat-Ease Inc., Minneapolis, MN, USA). All total flavonoid data in the text and tables are presented as the mean ± standard deviation (s.d.); additionally, the s.d. is indicated as error bars in the figures.

The design space was constructed and analyzed using the Design-Expert^® Software, v.11 (Stat-Ease, Inc., Minneapolis, MN, USA). Analysis of variance (ANOVA) was applied to test the chosen model and individual terms in the model. Concerning the in vitro release and ex vivo permeation experiments, data distribution was tested using the Shapiro-Wilk (S-W) normality test. Significance was set at p < 0.05 level and all tests were two-tailed with 95% Confidence Intervals (CI). Results are expressed as mean ± SD for the in vitro release experiments and mean ± standard error (SE) for ex vivo permeation experiments. The permeation values were statistically compared between the different formulations and per timepoint within the formulation. Outlier detection occurred applying the Interquartile Range (IQR) using a step of 1.5 × IQR. No outliers were detected. Normality test indicates a non-Gaussian distribution of the data. Hence, non-parametric tests were applied for the data of both in vitro and ex vivo experiments. Kruskal-Wallis was performed to statistically evaluate the differences between the formulations at every time point of the experiment and post-hoc Mann-Whitney to detect individual differences. Data analysis was performed using SPSS version 26.0 (IBM SPSS Statistics for Windows, Version 26.0, IBM Corporation, Armonk, NY, USA) software package.