Factors associated with current smoking
A logistic regression model was used to explore which characteristics might be associated with current smoking. This identifies associations, not causes; in other words, factors which identify pupils with an increased or decreased likelihood of being smokers. See Appendix B for more information on the regression model used.
The 8 factors (explanatory variables) shown below had a significant association with current smoking. The size of the circles represents an estimate of the relative contribution to the model (table 1.10 shows the odds ratios for each possible value of each variable in the model).
It was estimated that e-cigarette use had the strongest association, followed by drug use, and having friends who smoke.