Measuring adolescent smoking expectancies by incorporating judgments about the expected time of occurrence of smoking outcomes.

French and English Canadian adolescents completed a smoking expectancy questionnaire and 2 measures of current smoking status. Multiple regression revealed that beliefs about the expected time of occurrence of smoking outcomes explained unique variance in current smoking after controlling for judgments about the probability and desirability of these outcomes. In addition, the relationship between the perceived probability of the general costs of smoking and current smoking was moderated by beliefs about the expected time of occurrence of these costs. There was no relationship between perceived probability of general costs and smoking for adolescents who expected the costs to occur far in the future, whereas there was a significant negative relationship between these 2 variables for adolescents who expected the costs to occur soon after smoking. The authors' results suggest that it may be possible to increase the concurrent validity of traditional smoking expectancy measures by incorporating expected-time-of-occurrence judgments. ((c) 2005 APA, all rights reserved).