Cluster analysis is a family of techniques that sorts - or more accurately, classifies - cases into groups of similar cases. ‘Data mining’ encompasses a whole host of methodological procedures that are used for cluster analysis while ‘classification’ that is the analytical catalyst to the methodological approach. Thinking about issues of ‘classification’, ‘cluster analysis’ and ‘data mining’ together in this four-volume collection is appropriate, therefore, specifically with regards to developing a case based ‘attitude’ to quantitative analysis.

This collection does not simply focus on a set of methods, but in presenting a range of existing work together, the logic of what is arguably a methodological phase-shift in quantitative research is exposed. In effect, this four-volume collection sets forth an analytical strategy which is increasingly, both implicitly and explicitly, acknowledged across the disciplines as being rooted in the exploratory and descriptive investigation of cases.

Bringing work on classification, cluster analysis and data mining together in a way that is both accessible and timely with respect to the level of ‘activity’ going on in each of these related areas is important to signal a step-change in the kind of data analysis that is currently taking place, nationally and internationally, and to facilitate further research by demarcating the methodological research where the cutting edge approaches to data analysis lie.

Volume One: The Classics

Volume Two: (Useful) Key Texts

Volume Three: Cluster Analysis in Practice

Volume Four: Data Mining with Classification