Further, Hoek, Gendall and Esslemont (1996) have argued that despite sophisticated
approaches to market segmentation, one of the key difficulties in undertaking such studies is that
the selection of variables involves significant subjective judgements. For example, researchers
using consumption benefits as a segmentation basis, must determine which benefits to measure and
select appropriate means of assessing their relative importance to respondents. It needs to be
recognised that this process may have a significant impact on the research outcome. Everitt (1974,
p. 48) argues that ‘the initial choice of variables is itself a categorisation of the data which has no
mathematical or statistical guidelines and which reflects the investigator’s judgement of relevance
for the purpose of the classification.’ However, the subjective decisions and assumptions inherent
in segmentation studies do not preclude the studies from being potentially useful to gaining an
improved understanding of the key factors influencing the choice of a tourism destination.
However, such assumptions need to be made explicit and transparent so that users of the models
understand the limitations of any findings. While numerous segmentation studies have been
undertaken within the realm of tourism destination marketing, few of these studies make clear the
subjective elements of the research design