Travel Reference
In-Depth Information
contain important limitations. To overcome such limitations, conceptual studies have proposed
comprehensive models that approach decision-making from different perspectives. From the
fi eld of psychology to the vast set of economic models, the only consensus is that decision-
making is a complex process that involves a plethora of variables. Such a topic is as complex as it
is challenging. This chapter attempts to discuss the tourist decision-making process in a practical
sense, highlighting how economics, psychology and marketing should contribute to develop a
tourism-based decision model. This chapter, therefore, approaches the heuristics that should be
taken into account when modelling tourism decision-making. These are critically discussed, and
implications for marketing strategies in a tourism context are outlined.
Seminal theories of consumer behaviour
The study of tourist decision-making is grounded mostly in neoclassical economic theory and
discrete choice theory. Neoclassical economic theory (CCB) frames consumption choices as
rational and purposeful. Therefore, individuals tend to maximize the utility of the 'basket of
assets' they can purchase, bearing in mind the need for diversity and budget restrictions. Following
this stream of literature, Lancaster (1966) contends that utility does not derive directly from the
product but from the attributes the product has that enable it to fulfi l the needs of the consumer.
Broadly speaking, a 'technologic constraint' was introduced in terms of optimization problems,
which is commonly referred to as a preference function. In the context of tourism decisions
another constraint was considered, the availability of time for leisure (Bull 1995). From this
perspective (CCB, incorporating Lancaster's extensions [1966]) choice in tourism has been
approached by various scholars (e.g. Rugg 1973; Morley 1992; Papatheodorou 2001), who argue
that travel decision-making is a rational choice process that emerges from the evaluation of
several alternatives constrained by the tourist's pervasive availability of time and money in light
of destination attributes (preference function). Utility functions were estimated but these models
did not consider interpersonal and intrapersonal variables, which led tourism research to consider
discrete choice theory (Jeng and Fesenmaier 1996).
Discrete choice theory arose with contributions from economists and cognitive psychologists.
Discrete choice problems involve choices between two or more discrete alternatives, such as
going or not going on holiday, or choosing between destinations. Such choices contrast with
standard consumption models in which the quantity of each good consumed is assumed to be a
continuous variable. In a continuous case, demand can be modelled using regression models.
Regression models allow us to answer 'how much' type questions. In discrete choice problems
the outcome is discrete and therefore discrete choice models should be applied; hence discrete
choice models help us to answer 'which' type questions.
Two streams of discrete choice models could be considered: revealed and stated preferences
approaches. Revealed preference theory assumes that the preferences of consumers can be
observed, being utility functions derived from their choices, given their budget constraints. For
instance, if a tourist chooses Hawaii instead of Fiji islands, both being affordable, it means that
this tourist prefers Hawaii. Furthermore, this preference is stable and irreversible, over the
observed time period. This theory was widely criticized, since in the real world, when it is
observed that a consumer purchases a certain commodity, it is impossible to say what good or
set of goods was discarded, and so there is not clear evidence that the commodity bought is
necessarily the preferred one. In this sense, preference is not revealed at all in the sense of ordinal
utility. Applications of revealed preference rely mostly on destination attributes (Perdue 1986;
Morey, Shawand and Rowe 1991; Dubin 1998; Colledge and Timmermans 1990; Siderelis and
Moore 1998; Schroeder and Louviere 1999; Haider and Ewing 1990).
