Statistical Analysis

In this study, the analytical data were manipulated using the Excel 2010 spreadsheet. The SPSS 21.0 software was used for pattern recognition computations. Pattern recognition methods were applied involving the multivariate data analysis. The discrimination and authenticity of the cultivated P. tenuifolia samples were carried out by the following multivariate data analysis (chemometric) techniques: hierarchical cluster analysis (HCA), principle component analysis (PCA), and discriminant analysis (DA). The statistical analysis used in this study is as follows:

The elemental fingerprints and radar plots were analyzed. The statistical analysis was performed using Microsoft Office Excel 2010.

HCA is an unsupervised classification procedure that involves measuring the similarity between samples to be clustered. Hierarchical cluster analysis (HCA) of samples was performed using the selected chemical descriptors as variables, the Ward’s method as the amalgamation rule, and the squared Euclidean distance as the similarity measurement. Samples were grouped in clusters based on their nearness and similarities³⁷. The groups were represented by the branches of the dendrogram. The dendrogram showed the different groups at a normalized or rescaled distance. The between-group linkage method was the clustering method used in this study.

PCA is a projection method that allows for easy visualization of all information contained in the data set. PCA reduces the dimensionality of the data matrix and transforms original variables into principal components (PCs)³⁸. By plotting the PCs, one can view the interrelationships between different samples and examine the grouping of samples. Finally, PCAs quantify the amount of useful information, as opposed to meaningless variations, contained in the data³⁹.