Statistical mechanics underlying the geographic distribution of visitors: a power law-inspired explanation

Abstract

Rank-size relationships governed by the power law have been widely observed and studied in various domains including biology, physics, computer science, economics and other social sciences. In the tourism literature, power law has been invoked in the study of the geographic distribution of visitors and tourism. Although the law provides an excellent and simple guide in relating the abundance and size of tourism destinations, to-date, an important limitation in the studies of power law in tourism has been the lack of an explainable mechanism for the emergence of the power law pattern. In this context, the first aim of this thesis is to empirically examine the applicability of power law in explaining the rank-size relationship in the geographic distribution of visitors. The second aim is to investigate the potential mechanisms underlying the distributions by adopting a deductive approach, which is contrary to most previous investigations in tourism. Deriving from an assumption of herd behaviour for visitors, the Polya urn process for Visitor Distribution (PVD) model presented in this thesis yields a “power law-like” pattern at the macro level. The PVD model can serve as an analytical link between micro-level choice behaviours of visitors to the macro-level distributions, such that the complex network of visitor distribution may be associated with the individual’s choice behaviours. The PVD model has an advantage that it generates conveniently interpretable parameters for visitor distributions. Using data on international visitors’ geographic distribution in Australia, this thesis obtains evidence supporting the PVD 2 model as one of the potential mechanisms underlying power law or similar distributions in tourism. Two tourist characteristics - holiday visitors and visitors from Asian countries, were found to fit the PVD model particularly well. An additional finding indicates the potential existence of homogenous destination groups. The emergence of destination groups in the macroscopic tourism distribution may be attributed to the hierarchical destination choice behaviour of tourists. One of the important conclusions is that the choice behaviour and the geographic distribution of visitors are tightly related, and this link may offer insights into why power law-type distributions emerge in social systems.