Upon successful completion of this course, students are expected to be able to select adequate statistical techniques to answer different types of research questions, to analyse quantitative data, to interpret findings related to the research question/problem statement, and to draw respective conclusions.
Intended Learning Outcomes
Students are expected to be able to:
• Independently conduct multivariate analysis, for instance as part of their master thesis (application level)
• Systematically use a basic methodological toolbox for business analytics (application level)
• Characterise requirements and assumptions of various multivariate statistical techniques (comprehension cognition level)
• Understand the methods and results section of scientific articles
• Better judge the quality of analyses and validity of empirical findings.
• Make well-grounded decisions with regard to the empirical part of the master thesis (integration level)
The course provides all MSc Business Administration students with both active and passive knowledge on multivariate statistical techniques typically used in empirical business research studies. Active knowledge comprises the abilities to choose the adequate multivariate technique for combinations of data and research questions, to autonomously conduct multivariate analyses using statistical software such as SPSS or ADANCO, and to interpret and report the obtained results. Passive knowledge refers to the ability to critically reflect upon the assumptions made as well as the reliability and validity multivariate analyses.|
This course conveys this knowledge by covering five families of multivariate techniques:
1) Cluster analysis as a theoretical technique useful for classification tasks,
2) Factor analysis (more specifically, principal component analysis and common factor analysis) as a technique for data reduction and modelling of latent variables,
3) Multivariate Analysis of Covariance (MANCOVA) as a comprehensive technique to analyse the outcomes of experiments,
4) Multiple linear regression as the cornerstone of econometrics,
5) Structural equation modelling as a technique for theory-testing that combines factor analysis and multiple linear regression.
The multivariate techniques will be presented in different forms. First, each statistical technique will be explained in detail and discussed in the context of a specific business research issue. Subsequently, the technique will be central in a corresponding assignment. For the assignments, the techniques have to be applied using statistical software, e.g. SPSS or ADANCO, to analyse data relevant for the business research issue at hand. The assignments have to be carried out independently by small groups of students. In the corresponding assignment paper, students have to elaborate on the assumptions, choices made that are relevant for the specific statistical technique, to interpret and to report the relevant findings, and to discuss the scientific and/or managerial implications. The actual statistical analyses of the data will be facilitated by computer classes (tutorials).
As a follow-up in the so-called application lectures, findings will be presented by student groups and discussed. Furthermore, additional information on the strengths and limitations of the techniques and analyses will be discussed.
This course is taught in a full time mode for the first 3-4 weeks of the quartile, at the end of this part the exam takes place. A resit for the exam is offered in the last week of the quartile.
Participation in the exam is only possible if all assignments have been submitted within the previous 12 months and in total for these assignments earned 5 points (out of a maximum of 10).
Assumed previous knowledge |Required materials|
Recommended materials-Instructional modes
|Hair et al. (2013), Multivariate Data Analysis, Pearson, 7th edition, ISBN 129202190X (6th or 5th edition are okay as well).|
|Self study with assistance|
|Self study without assistance|