solution

Segmentation with surgical precision

Cluster analysis was used to classify and segment participants, based upon their preferences for hospitals that provide inpatient care. The clustering was based on the reasons participants gave for preferring a particular hospital. The demographic profiles of the grouped participants were compared to learn whether the segments could be identified more efficiently. The k-means clustering method (SPSS) was used for grouping the participants based on their answers to the hospital preference items. The squared Euclidean distances between all clustering variables were minimised. Because different individuals perceive scales of importance differently, each individual’s ratings were normalised before clustering. The results indicated that the participants could be best classified into four clusters. The cross-validation procedure for cluster analysis was run twice, on halves of the total sample. As expected, the four groups differed substantially by their distributions and average responses to the reasons for their hospital preferences. The names assigned to the four groups reflected the demographic characteristics and reasons for hospital preferences: ‘old-fashioned’, ‘affluent’, ‘value conscious’ and ‘professional want-it-alls’.