Admission requirements
Master’s students Psychology with specialisation Methodology and Statistics, Research Masterstudents.
Description
In this course the general focus is on nonparametric multivariate analysis. In the first part of the course a review on null hypothesis significance testing procedures is given that identifies several statistical and interpretation difficulties concerning these procedures and p-values. Alternatives are presented for alleviating these problems, like cross validation, bootstrapping, permutation testing, and the Bayes factor.
In the second part, the focus is on two major methods for multivariate analysis, multiple regression analysis (MR) and principal component analysis (PCA). Classical MR and PCA are applicable to quantitative data and rely on assumptions of normality and linear relationships. More general methods, CATREG and CATPCA, will be discussed, that can handle both quantitative and qualitative data and do not rely on normality and linearity assumptions.
The third part of the course consists of multidimensional scaling. In the first meeting, some historical aspects concerning classical and least squares scaling, the specific data requirements, and the taxonomy of models are discussed. In the final two meetings, one mode models (PROXSCAL) and two mode models (PREFSCAL) are discussed in detail. The assignments consist of programming simple models in R and analysing own data with complex models using IBM SPSS and specialized software.
Course objectives
Students:
Acquire knowledge of p-values and other model selection statistics
Acquire knowledge of, and insight into, the assumptions of multivariate data analysis techniques
Acquire knowledge of, and insight into, statistical techniques which require a minimal set of assumptions
Learn to use statistical software for nonparametric multivariate analysis
Acquire knowledge of Multidimensional Scaling
Acquire knowledge of Multidimensional Unfolding
Learn how to use SPSS software for Scaling and Unfolding
Learn how to use R for scaling and unfolding
Timetable
Applied Multivariate Analysis (2014-2015):
Registration
Course
Students need to enroll for lectures and work group sessions. Please consult the Instructions registration
Mode of instruction
Seven lectures.
Assessment method
Three graded assignments during course.
The Faculty of Social Sciences has instituted that instructors use a software programme for the systematic detection of plagiarism in students’ written work. In case of fraud disciplinary actions will be taken. Please see the information concerning fraud
Blackboard
Information on blackboard.leidenuniv.nl
Reading list
Cohen, J. (1994). The earth is round (p<.05). American Psychologist, 49, 997-1003. http://www.ics.uci.edu/~sternh/courses/210/cohen94_pval.pdf
Wagenmakers, E.J. (2007). A practical solution to the pervasive problems of p values. Psychonomic Bulletin and Review, 14, 779-804. <http://www.ejwagenmakers.com/2007/pValueProblems.pdf>
Linting, M. (2007). Nonparametric Inference in Nonlinear Principal Components Analysis: Exploration and Beyond. Thesis, Leiden University <https://openaccess.leidenuniv.nl/handle/1887/12386>
Van der Kooij, A.J. (2007). Prediction Accuracy and Stability of Regression with Optimal Scaling Transformations. Thesis, Leiden University. <https://openaccess.leidenuniv.nl/handle/1887/12096>
Kruskal (1964a). <http://dx.doi.org/10.1007/BF02289565>
Kruskal (1964b). <http://dx.doi.org/10.1007/BF02289694>
Y. Takane (2008). <http://takane.brinkster.net/Yoshio/c039.pdf>
Busing, F.M.T.A. & K. van Deun (2009). <https://openaccess.leidenuniv.nl/bitstream/1887/15279/20/02.pdf>
Busing,F.M.T.A., G. Cleaver, & W.J. Heiser (2010). <http://dx.doi.org/10.1016/j.foodqual.2009.08.006>
Contact information
Dr. M. De Rooij
Room 3B21
Tel.: +31 (0)71 527 4102
E-Mail: rooijm@fsw.leidenuniv.nl