A Classification and Regression Tree (CART) to Analyse Cruisers’ Expenditure Pattern and Perception in a Port of Call

This study analyses cruise visitors’ perceptions in a port-of-call. The empirical data were collected via a survey in Cartagena de Indias (Colombia) during 2009. A high quality and an original database was obtained for such a niche of economy activity. The novelty of this paper consists of a few methodological enhancements. First, a hierarchical cluster analysis, based on the results of a correspondence analysis and two stopping rules, has robustly demonstrated the existence of six different clusters. Second, a CART is used to identify which variables are better predictors to classify individuals in the different clusters. The empirical findings imply several policy directions for each of the clusters. Specifically, institutions inte rvention should enhance security in Cartagena to guarantee inland repeated visits; an ad hoc marketing policy may encourage revisit by young South American customers; destination managers should extend the inland visiting time that is likely to produce local multiplier effects.