We conducted analyses using Mplus version 7.31 (Muthén & Muthén, 1998 –2012). Clustering within families was accounted for using the CLUSTER command in Mplus. All analyses included offspring’s sex and age at the last interview as covariates. In addition, given the wide range of age in our sample, we controlled for potential birth cohort effects by creating a cohort variable indexed using three dummy-coded variables derived from participant year of birth: [1902–1950], [1951–1970], and [1971–2000] (Grucza et al., 2008). Missing data were accounted for using full information maximum likelihood estimation method.
