Compositional Data Analysis in Tourism: Review and Future Directions

Compositional data analysis (CoDa) is the standard statistical methodology when data contain information about the relative importance of parts of a whole. Many research questions in tourism are either related to distribution of a whole (e.g., distribution, share, allocation, etc.) or relative importance (e.g., dominance, concentration, profile, etc.). Example research questions might be: How does time allocated to different types of activities relate to tourist satisfaction? or Which origins and destinations concentrate the most tourist flows per tourist segment? The first aim of this article is to present the manner in which CoDa solves statistical problems that arise when treating compositional data with classical statistical methods (e.g., spurious correlations, meaningless distances, assumption violation). The second aim is to review all CoDa applications in tourism and hospitality to date. The third is to present CoDa applications in related fields (e.g., finance, sociology, geography, economics, management, ecology, education), which can be translated into future research in tourism. In order to show how to apply the most common CoDa tools (exploratory analysis of compositions, and use of compositions as variables in a model) an example of restaurant menu styles is used.