Definition
Cluster analysis divides observations into groups (clusters) so that units within a cluster are as similar as possible and clusters differ from each other — without predefined classes. Hierarchical clustering builds a tree (dendrogram); k-means assigns observations to a given number of centres. Typical uses are customer segmentation, product typologies or pattern discovery in measurements.
Multivariate methods
The result depends on the distance measure (Euclidean, Manhattan, correlation), the linkage method (Ward, average) and on variable scaling — standardize, otherwise wide-range variables dominate. For k-means, choose the number of clusters via the within-cluster variability plot or silhouette.
Clusters are a descriptive construct, not proof: validate them by stability on another sample and by meaningful profiles (variable means within clusters). For large datasets k-means or EM clustering are preferable to hierarchical methods.
In Statistica
Statistics → Multivariate Exploratory Techniques → Cluster Analysis contains hierarchical clustering with a dendrogram, k-means with a plot of means and clustering of variables; for big data and automatic choice of cluster count use Generalized EM & k-Means Clustering in Data Mining with cross-validation.
Related terms
- Principal component analysis (PCA)Principal component analysis (PCA) condenses many correlated variables into a few new uncorrelated components that…
- Discriminant analysisDiscriminant analysis finds combinations of variables that best separate groups known in advance — healthy from ill…
- Decision treesA decision tree splits data by a sequence of simple rules ("age > 45 and income < 30 thousand") into ever more…
- OutliersAn outlier is an observation that differs conspicuously from the rest — a measurement or entry error, a different population, or a genuine rare event.
Knowledgebase guides
FAQ
- How many clusters is right?
- There is no single correct answer — combine statistical criteria (silhouette, the elbow in variability) with whether the clusters are interpretable and usable in practice.
- What is the difference between cluster and discriminant analysis?
- Cluster analysis searches for groups (unsupervised learning); discriminant analysis separates and predicts groups known in advance (supervised learning).
Try it on your own data
Statistica free for 30 days
Full version, no credit card. Or get a pricing estimate in a minute.
Multivariate methods
Updated: September 2026.