The base sample for present analyses was individuals who completed the Wave 4 interview, since it covered lifetime psychiatric and psychosocial information in young adulthood and had the largest number of participants (one dropped due to incomplete data). Over 95% of Wave 4 participants completed at least one additional interview (Wave 1, 3, or 5) that served as a source of data for coding variables in the current analyses. The sample had a mean age of 21.7 (SD=2.8) years at Wave 4, and included 1,038 monozygotic (MZ) twin pairs (911 EA, 127 AA), 734 dizygotic (DZ) twin pairs (606 EA, 128 AA) and 242 twins (199 EA, 43 AA) whose co-twin did not participate. Zygosity was determined using an algorithm based on responses to standard questions regarding similarity (Nichols and Bilbro, 1966). Zygosity for both members of 940 EA and 150 AA twin pairs was confirmed using genotyping, yielding 90.2% and 92.7% accuracy, respectively, which is broadly consistent with reports on other samples of European (Eaves et al., 1989; Reed et al., 2005) and African descent (Whitfield, 2012).
