Multivariate statistics

An important characteristic of business processes (think: marketing, sales, service, human resources, new product development, procurement) is that a single outcome is usually influenced by multiple variables. At the same time, one variable may influence multiple outcomes. For instance, customers’ adoption of a new high-tech product may be determined by its ease of use, its usefulness, whether it is enjoyable to use, and whether it is sustainably produced. At the same time, for some customers sustainability may be more important in their decision than for others. To analyze and understand situations like these, we resort to multivariate statistics.

This course is aimed at Master level students who want to conduct research using multivariate statistics. Students will develop an understanding of experimental designs, (big) data sets, statistical assumptions, data design preparation, data screening, multivariate statistical techniques, the multivariate software packages SPSS, PROCESS, and R. In addition, they will learn how to conduct these multivariate analyses, interpret the related statistical software output, draw implications, and how to report their outcomes. We will particularly pay attention to the following topics: data examination, factor analysis, multiple regression analysis, experimental design, (M)ANOVA, mediation and moderation analyses, and multilevel data analysis.

The course features lectures and tutorials. Lectures are plenary, on-campus, and explain the multivariate statistical techniques. Tutorials are on-campus, small(er) scale, and allow groups to work on their group assignments under the guidance of a tutor. Self-study on the materials and working on the group assignments outside the tutorial hours is expected.

The primary objective of this course is to help prepare you to function as an effective analyst of business and consumer data, both in applied (industry) as well as academic settings. To be able to reliably analyze business processes and their outcomes, this course extends prior knowledge on univariate/bivariate analysis techniques to multivariate techniques. Rather than focusing on statistical formulas or algebra, this course provides an applied perspective and employs intuitive, yet state-of-the-art, SPSS and PROCESS software to run a wide variety of multivariate data analyses. You will also be introduced to R. Special attention will be provided to the topic of experimental design and analysis. Extending the applied perspective, each module provides optional, advanced study material for students who wish to further deepen their understanding of a specific topic by getting into statistical and mathematical details underlying the multivariate techniques.
Upon course completion, students should be able to:

• Evaluate a raw multivariate data set with respect to missing values, outliers, and normally distributed data, and prepare it for further analysis • Explain the concept of endogeneity
• Set up an experimental research design, including specifications of factors, manipulations, and control groups Evaluate and address the assumptions for multiple linear regression
• Conduct, and evaluate the outcomes of key multivariate statistical analyses (i.e., exploratory factor analysis, reliability and validity of latent variables, multiple linear regression, ANOVA, MANOVA) using SPSS
• Conduct, and evaluate the outcomes of, mediation, moderation, and moderated mediation analyses, with and without the PROCESS macro • Understand the idea of multilevel analyses and being able to decide whether it is the appropriate analytical strategy (e.g., using ICC)
• Conduct, and evaluate the outcomes of, multilevel analyses using R, including an intercept-only model, a random intercept model, a random slope model, and cross-level interaction

You must meet one of the following collections of requirements

• Collection 1
• Completed Final examination Bsc program
• Collection 2
• Completed Pre-Master

• Hair, Joseph F., Barry J. Babin, Rolph E. Anderson, and William C. Black (2019), Multivariate Data Analysis - 8th Edition, Cengage Learning, ISBN: 978-1-4737-5654-0 (ISBN 978-1-4737-5654-0)
• Other materials (i.e., book chapters and an article) provided on Canvas