Table 3.

Adjusted models estimating main effect of source credibility and interaction effects of source credibility and e-cigarette use on ad and e-cigarette perceptions (N = 497).^a

Variable		b	95% CI	P-value
Perceived ad relevance
Source credibility (stratified by e-cigarette use)	Never	0.22	0.16, 0.28	<0.001
	Ever	−0.02a	−0.31, 0.27	0.91
	Current	0.11b	−0.20, 0.42	0.491
Perceived ad effectiveness
Source credibility (stratified by e-cigarette use)	Never	0.33	0.27, 0.40	<0.001
	Ever	−0.03a	−0.33, 0.28	0.868
	Current	−0.34a	−0.66, −0.02	0.038
Ad liking
Source credibility		0.41	0.36, 0.46	<0.001
Use interest
Source credibility (stratified by e-cigarette use)	Never	0.16	0.08,0.23	<0.001
	Ever	0.17a	−0.18, 0.52	0.335
	Current	0.18b	−0.18, 0.55	0.329
Harm perceptions
Source credibility		−0.04	−0.10, 0.03	0.278

^aLinear mixed effects models with random intercepts were used to estimate associations. We report significant main effects and interaction effects for perceived source credibility and e-cigarette use status on outcome variables. Covariates included age, gender, sexual orientation, race, ethnicity, and cigarette use. P-values were calculated using partial F-tests. Tukey’s tests were used to assess statistical significance of pairwise comparisons. P-values that meet the criteria for statistical significance are bolded. Means without a common superscript letter differ (p < 0.001). Models analyzed the interaction between e-cigarette use status and age group, but results are only stratified by age group status when the interaction was statistically significant. Stratified results are presented from models with statistically significant (p < 0.05) interactions between age group and e-cigarette use status.