Students gain comprehensive knowledge of the methods of multivariate statistics and know how to use them appropriately.

None

Multivariate methods are characterized by the option of simultaneously analyzing several attributes of individuals. Their advantage over single, univariate analyzes for each attribute is that they can measure how these attributes depend on each other. The dependence between several outcome variables and several predictor variables can be shown by using multivariate regression models. Further methods discussed in this course are cluster analysis and discriminant analysis. The goal of a cluster analysis is to group the objects into categories in such a way that objects from one category are as similar to each other as possible and objects from different categories are as different from each other as possible. A discriminant analysis aims to find a classification rule based on these established categories, so that a new observation can be assigned to one category.

• Bühl A. (2016): SPSS 23: Einführung in die moderne Datenanalyze. 15th rev.ed. Pearson Studium.
  
• Bortz J./ Schuster C. (2010): Statistik für Human- und Sozialwissenschaftler: Limitierte Sonderausgabe. 7th ed. Springer.

Lecture, tutorial

Assessment of tutorial 30%, written final exam 70%