Course objectives:
- Learn how to conduct exploratory data analysis and find typologies in the data using cluster analysis and latent class modelling.
- Learn how to use obtained typologies in further analysis (e.g., to establish associations between individual- characteristics and cluster membership).
- Focus on cluster analysis in SPSS and latent variable modelling (specifically latent profile analysis – LPA) in R.
Course content:
The course will consist of 10 meetings.
An overview of the topics covered in each meeting is provided below:
- (1st meeting: Introduction to R, if needed)
- 1st meeting: An introduction to cluster analysis (incl. general description, applications, and distinction between different clustering algorithms – i.e. deterministic vs. probabilistic and connectivity-based, centroid-based, distribution-based vs. density-based).
- 2nd meeting: Running k-means clustering (using the PhD’s own data or a dataset available online e.g., World Values Survey)
- 3rd meeting: Running hierarchical clustering (again, using the PhD’s own data or a dataset available online)
- 4th meeting: Running robustness checks and sensitivity analyses for k-means and hierarchical clustering results; addressing issues and errors
- 5th meeting: Running a multinomial logistic regression analysis using cluster memberships as dependent variable and a set of (individual-level) characteristics as independent variables
- 6th meeting: Introduction to latent variable modelling in general and latent profile analysis (LPA) and latent class analysis (LCA) specifically ·
- 7th and 8th meeting: Running LPA and LPCA in R
- 9th/10th meeting: Q&A
