This half-day course covers the essentials of cluster analysis - a technique that can be used to explore a dataset and identify groups of cases that exist in the sample.

It is ideal for those working in many areas who either (i) know that groups exist in their data and want to identify which cases belong to which group, (ii) are unsure what groups, if any, exist in their data and want to discover patterns which may exist.

The course will cover both k-means clustering and hierarchical clustering, including issues of choosing distance measures, appropriate numbers of clusters and linkage methods. Microsoft Excel will be used to demonstrate the methods, using a free add-in designed (by the author of this course) to undertake multivariate analyses.

See our full range of courses (and options for bespoke courses) at https://go.herts.ac.uk/sscu or contact the course organiser, Professor Neil Spencer, at statistics@herts.ac.uk.

Only a basic understanding of operating with data is required.