The purpose of this paper is to suggest a convenient method to measure the amount and direction of bias in survey interviewing. Repeating a central question at the beginning and end of a thematic interview permits one readily to establish the amount of change admitted by the respondents — presumably as a result of the interview experience. In the current study on gambling attitudes and behaviour the following findings are reported:1About one in six respondents changed their attitudes toward the casino in the course of the interview;2Positive and negative change was balanced suggesting a fair interview schedule;3Young respondents were least likely to change their mind;4Respondents with less than 10 years of education were most likely to change;5Gender, political ideology and religious commitment were not related to proclivity to change;6Interviewers' attitudes toward the casino did not influence the respondents' attitudes toward it;We recommend the inclusion of such testing strategies in all thematic interview