CitationMaes, Hermine; Miles, Donna; R; Neale, Michael; & C (2008). Testing for measurement invariance in genetic analyses of smoking and nicotine dependence in adolescence. 2008 Add Health Users Conference. Bethesda, MD.
AbstractUnderstanding the contribution of genetic and environmental influences to the various stages of smoking initiation to more severe forms of smoking behavior (regular smoking, smoking persistence, nicotine dependence) is an important first step in finding specific etiological factors for smoking. Previously, we used an item response theory approach which allows for possible measurement invariance of the scale by sex and age. We examined the relationship between genetic and environmental risk factors for smoking initiation, regular smoking, and nicotine dependence in adult twins from the population based Mid Atlantic Twin Registry. Results showed that the significance of heterogeneity in genetic and environmental parameters changed as a function of accounting for sex differences in the measurement. We attempted to replicate these findings in data from adolescent twins collected as part of Add Health. Smoking behavior was assessed in female, male and opposite sex 18 to 25 year old adult twin pairs by questionnaire. Results indicated significant mean differences by sex (but not age) in the latent ‘nicotine dependence’ factor or thresholds, and sex differences in factor loadings of the nicotine dependence items (FTND) in addition to smoking initiation and regular smoking items. We test whether the heritability of the latent liability to nicotine dependence and the residual heritability on the individual items differs between male and female adolescents, allowing for differences in measurement parameters.
Reference TypeConference proceeding
Book Title2008 Add Health Users Conference