Fig. 10
The PLSPM for data from the type (A), size (B), PS-MPs (C), PE-MPs (D), and PE-NPs (E) treatments. Each rectangle represents a latent variable indicated by a set of manifest variables. The arrow widths are proportional to the strength of the path coefficient, with only significant paths marked with a star (*, **, and *** representing significance levels of p < 0.05, p < 0.01, and p < 0.001, respectively). Red arrows indicate positive effects between latent variables, while green arrows indicate negative effects. R2 represents the proportion of variance explained. The goodness of fit values > 0.7 represents a good prediction for the entire model