Sexual Initiation Patterns and Subsequent Relationship History: Findings From the National Longitudinal Study of Adolescent to Adult Health


Glick, Gary C. & Halpern, Carolyn Tucker (2017). Sexual Initiation Patterns and Subsequent Relationship History: Findings From the National Longitudinal Study of Adolescent to Adult Health. Society for Adolescent Health and Medicine. New Orleans, LA: Journal of Adolescent Health.


Purpose There is a burgeoning research literature examining how characteristics of initiation into partnered sexual behaviors (e.g., timing, spacing, sequencing) predict adult well-being. The present paper examines associations between five distinct sexual initiation patterns and subsequent adult relationship history. Methods We used data from the National Longitudinal Study of Adolescent to Adult Health (Add Health, N = 11,991), a nationally representative study of U.S. adolescents in grades 7 – 12 (ages 11 to 21) during the 1994 – 1995 school year who have been followed into adulthood. Controlling for race, ethnicity, respondent educational attainment, parental educational attainment, childhood family structure, and chronological age, we tested a series of multinomial logistic regression models to assess associations between sexual initiation patterns (identified using Latent Class Analysis, classes are: Vaginal Initiators/Multiple Behaviors [Referent in regression models; n=5945], Dual Initiators [n=3755], Vaginal Initiators/Single Behavior [n=957], Postponers [n=719], and Early/Atypical Initiators [n=615]) and five indicators of adult relationship history: 1) having ever married, 2) number of lifetime cohabiting partners, 3) having ever had a non-marital, non-cohabiting partner with whom a pregnancy occurred, 4) number of long-term (six months or longer) relationship partners in adulthood, and 5) number of short-term (less than six months) relationship partners in adulthood. All analyses were conducted using survey commands in Stata 13.0 to adjust for Add Health’s complex survey design and apply sampling weights that would yield nationally representative population estimates. Results Both male and female early/atypical initiators (mean age first partnered sexual behavior = 15 years, initiated anal sex before 18) did not differ from the normative Vaginal Initiators/Multiple Behaviors class, except females in the early/atypical class were LESS likely to report nonmarital/noncohabiting pregnancy partners (OR=0.59 [0.36 – 0.96]). On the other hand, Postponers (average age of initiation = 21.7 years) differed from the normative group on multiple indicators, and patterns varied by sex. Among males, Postponers reported significantly fewer Non-Marital Cohabiting Partners, short and long term dating partners, and nonmarital/noncohabiting pregnancy partners. However, they did not differ in likelihood of marriage. Female Postponers were significantly less likely to have married, cohabited or to have had a non-marital/non-cohabitation pregnancy partner. Yet unlike males, female Postponers did not differ from the normative class in number of long-term or short-term dating partners. Conclusions These findings are in concert with a growing literature suggesting that the negative impact of early sexual debut, here measured more comprehensively, on adult relationship outcomes may be overstated. Findings also suggest that postponement of sexual debut until emerging adulthood, well beyond what is statistically normative, may afford these individuals an opportunity to avoid relationship patterns that are more likely to disrupt other life domains (e.g., having a child with a non-residential partner), although partnering patterns differ for males and females.


Reference Type

Conference proceeding

Book Title

Society for Adolescent Health and Medicine


Glick, Gary C.
Halpern, Carolyn Tucker

Year Published


Volume Number





Journal of Adolescent Health

City of Publication

New Orleans, LA





Reference ID