POST-HOC SEGMENTATION USING MARKETING RESEARCH

• Main Author: CRISTINEL CONSTANTIN
• Format: Article
• Language: English
• Published: University of Petrosani 2012-10-01
• Series: Annals of the University of Petrosani: Economics
• Subjects: marketing research; market segmentation; multivariate analysis; Kmeans cluster; TwoStep cluster; statistic significance
• Online Access: http://www.upet.ro/annals/economics/pdf/2012/part3/Constantin.pdf

This paper is about an instrumental research conducted in order to compare the information given by two multivariate data analysis used fordividing a population in clusters. These methods are K-means cluster and TwoStep cluster, which are available in SPSS system. Such methods could be used in post-hoc market segmentations, which allow companies to find segments with specific behaviours or attitudes. The research scope is to find which of the two methods is better for market segmentation practice. The outcomes reveal that every method has strong points and weaknesses. These ones are related to the relevance of segments description and the statistic significance of the difference between segments. In this respect, the researchers should compare the results of the named analyses and choose the method which better discriminate between the market segments.