Two types of reporting inconsistency for sexual initiation were analyzed--event occurrence and its timing--using data from two waves of the National Longitudinal Study of Adolescent Health (Add Health). Overall, 11.1% of those who reported they were sexually active at the time of first interview denied this at the subsequent one. Males of each race/ethnic group had higher percentages of inconsistency than their female counterparts. Being older, not living with parents, or having a highly educated mother was negatively associated with rescinding. Among those reporting sexual experience at both interviews, only 22.2% reported the same date of first sex. On average, teens revised their age at first sex to older ages, and boys, especially African American boys, had large variability in reporting dates, as did teens with lower verbal ability. Seven strategies for resolving inconsistent reports are presented and implications for substantive findings are discussed.