Discriminant analysis is a statistical model that produces a discriminant function (or a set of discriminant functions if there are more than two groups) capable of classifying new cases for which their membership in a group is unknown. The development of this technique began in 1920 with the work of the English statistician Karl Pearson, but it was in 1930 that R. A. Fisher proposed a methodology to obtain the linear combination of random variables. The original linear discriminant was described for two-class discrimination. However, the multiclass version was later generalized by C.R Rao as Multivariate Discriminant Analysis. In this work the main elements that are related to the procedure to carry out multiple discriminant analysis will be shown and explained. Additionally, the application of this method to a data set called palmerpenguins that includes several physical characteristics of three different species of penguins.