Use of Multidimensional Scaling Analysis together with Multivariate Analysis of Variance in Determining Differences between Groups

Multidimensional Scaling analysis (MDS) and Multivariate analysis of variance (MANOVA) are among the commonly used multivariate statistical methods. While MANOVA is used to evaluate whether there are statistically significant differences between the mean vectors of the experimental groups in terms of more than one independent variable; MDS analysis is used both for dimension reduction and to classify individuals/variables according to their differences. In cases where the relationships between individuals/variables are not known, but the distances between them can be calculated, MDS analysis allows to reveal the relationships between individuals by using these distances, and unlike MANOVA, it does not require any assumptions. In this study, the numerical values produced by the simulation technique were used as the input data, with reference to the real data regarding 5 kinds of pistachios in terms of 13 fatty acids. These data were evaluated with both the MDS analysis and the MANOVA test and the results were interpreted. Considering the convenience in the evaluation of the data, the usability of the MDS analysis as supportive to the MANOVA test and subsequent multiple comparison tests was evaluated.