F4SPSSAdvancedAbout 30 min to complete
Cluster Analysis
Segmentation as a research question.
Slides · Dataset
About 30 minutes to complete
Enrol to open the slides and download the data.
What this module covers
segmentation as a research question; hierarchical clustering, the dendrogram and how to read a cut; K-means and how to choose k; two-step clustering and why it handles mixed variable types; standardising variables before clustering and why it is not optional; distance measures; profiling and naming the resulting clusters; validating a solution; the honest caveat that cluster analysis always returns clusters, whether or not they exist.
By the end of it
- Segmentation as a research question
- Hierarchical clustering, the dendrogram and how to read a cut
- K-means and how to choose k
- Two-step clustering and why it handles mixed variable types
- Standardising variables before clustering and why it is not optional
- Profiling and naming the resulting clusters
- Validating a solution
- The honest caveat that cluster analysis always returns clusters, whether or not they exist
The dataset
Health-seeking behaviour survey — Sigma teaching dataset
A teaching dataset of 300 cases, generated from a fixed seed so every figure in the module reproduces exactly.
- CodebookIncluded
- Dataset (CSV)Included
- Dataset (CSV)Included
- SPSS syntaxIncluded