SUBJECT

Title

Person Oriented Multivariate Statistical Methods

Code

DPSY16-QNR-101

Type of instruction

practice

Level

Doctoral

Faculty

Faculty of Education and Psychology

Part of degree program

Doctoral School of Psychology

Credits

Recommended in

Semester 1-4

Typically offered in

Autumn/Spring semester

Course description

The aim of the course is to present the most important person-oriented methods applied in empirical psychological research, their theoretical models and their use, primarily in statistical software ROPstat. The topic of the first class is an introduction to ROPstat, and the data import and export in SPSS, ROPstat, and Excel, by means of which data file created in SPSS or Excel can be analyzed in ROPstat. During the second class some rank modules of ROPstat will be presented by means of which person-oriented inferences can also be drawn. The next topic is the screening of missing data, giving descriptive information on the configurations of missing values. It is then followed by the different methods of data imputations and residual analysis, the identification of outlying, extreme cases in a sample. The next topic is configural frequency analysis that can be performed for identifying value patterns of a small set of discrete variables. It is followed by a detailed introduction of several methods of hierarchical cluster analysis, the nonhierarchical k-means cluster analysis (relocation analysis), and dense point analysis. A special class will be devoted to the specific cluster adequacy measures (multivariate etasquared, MORI coefficients, etc.) and procedures by means of which the goodness of a clusterstructure can be evaluated. The course ends with introducing some special methods enabling better interpretation of classification results (such as procedures Centroid and Exacon), and with some complex procedures that integrate classification methods (LICUR, ISOA, TYFO).

Readings

Bergman L. R. Magnusson, D., El-Khouri, B. M. (2003). Studying individual development in an interindividual context: a person-oriented approach. Mahwah, NJ: Lawrence Erlbaum Associates.
Bergman L. R., Nurmi J.-E., & von Eye, A. A. (2012). I-states-as-objects-analysis (ISOA): Extensions of an approach to studying short-term developmental processes by analyzing typical patterns. International Journal of Behavioral Development, 36 (3), 237-246.
Pardo, M. (2010). Clustering. http://lectures.molgen.mpg.de/algsysbio10/clustering.pdf
Vargha, A., Bergman, L. R. & Takács, Sz. (2016). Performing cluster analysis within a personoriented context: Some methods for evaluating the quality of cluster solutions. Journal for Person-Oriented Research, 2 (1-2), 78–86. DOI: 10.17505/jpor.2016.08. http://www.personresearch.ouradmin.se/articles/volume2_1_2/filer/5.pdf
Vargha, A., Torma, B. & Bergman, L. R. (2015). ROPstat: a general statistical package useful for conducting person-oriented analyses. Journal for Person-Oriented Research, 1 (1-2), 87-98.
http://www.person-research.ouradmin.se/articles/volume1_1_2/filer/20.pdf