Objective: Based on a survey of first-year university students, this study examined the effectiveness of antismoking education for first-year university students in light of differences in departments.
Methods: Multiple regression analysis was performed with department as dummy variables to determine whether the coefficient of the dummy variable representing the department was significant after controlling for other factors. The effectiveness of anti-smoking education was analyzed based on the intercept and slope of the regression line.
Results: Concerning the department of social sciences, the coefficient of the department dummy variable was positive and significant (partial regression coefficient 2.367, p < 0.01) even after controlling for other variables indicating the characteristics of the department. On the other hand, the coefficient of the dummy variable for the department of welfare was a negative and significant (-1.067, p < 0.05). In other words, antismoking education was ineffective for social science department compared to welfare department.
Discussion: The social inequality in smoking (the coefficient of the dummy variable for the department of social sciences was positive and significant) was due to norms that cannot be assessed by the Kano Test for Social Nicotine Dependence (KTSND). If the norms not ascertained with the KTSND in this study were included in the multiple regression model as an explanatory variable, then the coefficient for this explanatory variable would be positive and significant, while the coefficient for the dummy variable of the social science department would not be significant.
Conclusion: Adding factors that cannot be ascertained with the KTSND to the multiple regression model used here should yield methods of education that increase the effectiveness of this anti-smoking intervention.
View full abstract