Domestic and international tourists' behaviour on the length of stay and selection of destination in Australia

Abstract

Although tourism is one of the main reasons for inter-regional travel, but there are relatively little quantitative models of inter-regional and national-level travel demand developed specifically to explain and simulate its pattern. This research project attempts to answer three questions for the different group of tourists: 1) how long will a tourist spend in a journey? 2) what destinations are the tourist going to select and how much time they spend at each selected destination? also, 3) in which order the tourist visit these selected destinations? A sequential modelling framework is proposed and developed in this study to answer the three issues. A Negative Binomial (NB) model is formulated to model tourists’ length of stay in Australia. A Multiple Discrete-Continuous Extreme Value (MDCEV) model is applied to estimate tourists’ selection on destinations and the allocation of time to each selected destination under the constraint of the time budget. A Mixed Logit (ML) model is developed to model the visiting sequence of the selected destinations. The performances of these three models are evaluated by using the root mean square error and correct ratio metrics. These models are calibrated by the data obtained from the National Visitor Survey (2007) and International Visitor Survey (2012) in Australia. The modelling results show that trip duration and destination selection are affected by age group, origin, income, and employment status. The finding of this research is useful for stakeholders in the tourism industry to better understand domestic and international tourists’ behaviour in Australia and make better policies for improving the service level.