The validity of a survey's findings is dependent upon the clarity and lack of ambiguity contained in each individual item yet minimal attention has been directed to this issue in most prevalence studies. Researchers have shown a tendency to accept the uncritical assumption that respondents interpret seemingly simple and straightforward items such as 'how much do you spend gambling?' in a consistent manner. No attempt is made to confirm the uniformity of responses by clarifying the mathematical formulae used by respondents to derive their answers. The purpose of this paper was to examine the consistency shown by a sample of 181 medical undergraduate subjects in estimating the level of gambling expenditure in a series of five case vignettes describing various scenarios of wins and losses during a session of gambling. Results revealed a wide variation in calculated figures depending upon whether or not subjects interpreted the item to mean net expenditure or turnover. Only two thirds or less of subjects calculated the figure to be the difference between the initial amount risked and the residual at the conclusion of the session. It is suggested that more attention be paid in prevalence and clinical studies to providing subjects with clear instructions on how to calculate expenditure estimates.