Multivariate statistics in practice for DaSCA

Course goals

 

Content

This course will teach students to use multivariate statistics for the analysis of empirical data. At the end of the course students should be able to determine which statistical model should be used to investigate different types of research questions, e.g. regression, ANOVA, repeated measures and logistic regression analysis. Furthermore, the students should be able to execute the corresponding analyses using SPSS and interpret the results. Students probably have encountered most of these models in the bachelor programmes, but may have to develop confidence in applying these models and procedures while analysing empirical data. Special attention will be given to some practical issues, like checking the model assumptions and dealing with violations. Potentially new topics are power, contrast testing, bootstrapping, missing data and imputation techniques, and Bayesian analysis. 

There are options to follow this course as a non-Research Master student (eg Elective student, PhD student etc). Please contact the teacher about these options on time, ie before July 15 (for courses in semester 1) / December 15 (for courses in semester 2). You will need written approval from the teacher (an email is sufficient) in order to register for this course at the Faculty’s Student Information Point ( Faculty Student Desk ). Note that for external parties, costs for participation may be involved.
 
Relation between tests and goals of course

The following aims of the course will be tested by means of several assignments. The details of these assignments and their weights towards the final grade will be clarified in the course manual.
 
Aims:
1.      Refreshing previously mastered knowledge of multivariate analyses.
2.      Enhancing knowledge about multivariate analyses by studying and/or practicing topics not encountered in previous education (e.g. repeated measure analysis and logistic regression), and/or by studying related topics (e.g., contrast testing and multiple testing issues).
3.      Apply knowledge of General Linear Model to design an appropriate statistical model for a given research question and dataset.
4.      Being able to perform the analyses mentioned in the previous three points using SPSS and being able to interpret the results.
5.      Knowing which statistical model should be used to investigate different types of research questions.
6.      Acquire knowledge on new developments in statistics (e.g. Bootstrapping and the changing ideas on hypothesis testing).
7.      Reporting skills in writing about design choice and choice for statistical test.
8.      Presentation skills in criticizing methodlogy of published article.