An investigation of measurement error in visitor expenditure surveys

The purpose of this paper is to investigate whether providing additional prompts in a visitor expenditure survey results in higher reported expenditure. Respondents to a self-completion survey of event visitors were randomly allocated either an aggregated or disaggregated expenditure format in a quasi-experimental design. ANOVA is used to identify significant differences in mean reported expenditure to the alternative formats. The research finds that provision of additional prompts in the expenditure module of a visitor survey results in higher reported expenditures in half the expenditure categories and, most importantly, in total expenditure. Collection of accurate visitor expenditure data is critical to estimation of the economic benefits of tourism and special events. Over or under estimation of direct expenditures associated with an event may have implications for future investment in the event by public and/or private agencies. Very few field tests of this fundamental issue in measurement error have been reported in the tourism literature. The few reported examples have tended to report results inconsistent with a priori expectations, although they have been based on very small sample size and therefore are limited by low power. This study is based on a large sample size and produces results consistent with a priori expectations.