Discrepancy, dissonance, and the "dissonant" smoker

This paper reexamines, in the light of social psychological research, the theoretical assumptions made by McKennell and Thomas in their study of smoking habits and attitudes in Britain. One assumption made by them is that antismoking messages which are very discrepant from the opinions held by smokers should be avoided since they may produce attitude change in the opposite direction to that intended. It is argued that there is little empirical basis for this assumption, and that the discrepancy of a communication from the initial attitudes of a target audience is not necessarily the most important determinant of its effectiveness. McKennell and Thomas also distinguish "consonant" from "dissonant" smokers--the latter being those who say that they would like to give up smoking if they could do so easily. It is argued that the term "dissonant" is a misnomer when applied to smokers who see themselves as unable to choose not to smoke. Practical and theoretical implications are discuseed.