Evidence for priming effects of the relationship between smoking perceptions and smoking behaviour among 12–17 year olds
| Type of state campaign | Odds ratio of interaction term between time period and beliefs about smoking | Difference in trends (p value) | ||
|---|---|---|---|---|
| Period 1 | Period 2 | Period 3 | ||
| This table shows the odds ratio between the index of beliefs about smoking and smoking in the past 30 days, controlling for media use and GRPs of exposure to truth® ads in the media market, and demographic and background variables (for example, age, sex, race/ethnicity, living with both parents, religiosity, employment, and weekly earnings) and a control variable for the number of months since baseline. | ||||
| *Odds ratio significantly different from zero at the p<0.05 level. | ||||
| Current smoking | ||||
| Established states | 0.804* | 0.781* | 0.760* | |
| Newer states | 0.724* | 0.796* | 0.677* | |
| All campaign states | 0.825* | 0.793* | 0.748* | 0.051 |
| All other states | 0.756* | 0.747* | 0.723* | |