Poster #54 - Differential Effects of Early Maturation on the Trajectories of Nonviolent and Violent Behaviors From Adolescence to Adulthood

• In Event: 3-159 - Poster Session 12
  In Poster Session: PS 12 Section - Developmental Psychopathology

Integrative Statement

It is known that early pubertal timing has deleterious effects on adolescent problem behaviors (Dimler & Natsuaki, 2015) but little is known regarding how long its effect lasts. These mixed findings may be due to externalizing behaviors being too broad of a term to describe a range of behaviors that may be best divided into two domains: nonviolent and violent behaviors. The attenuation hypothesis (Copeland et al., 2010) suggests early maturation’s effect weakens as maturity increases towards adulthood. Based on this hypothesis and by using data from a subset of participants who participated in National Longitudinal Study of Adolescent to Adult Health (Add Health), we asked: (1) Does early maturation have a differential effect on nonviolent and violent externalizing behaviors? (2) If so, does this effect last through adulthood or dissipate after adolescence? We hypothesized that (1) early maturation specifically affects nonviolent behaviors; and (2) effects of early maturation dissipates as adolescents approach adulthood.
METHOD. The sample includes individuals who were 12-16 years old at wave 1 (Mage=14.6 years; 52.8% female; N=4,255). The assessment of pubertal timing, which occurred in waves 1 and 2, combined three sources of information: one interviewer-report item, one self-report perception item on pubertal status compared to peers, and three self-report items about sex-specific secondary sexual characteristics. Across four waves, participants reported the levels of their externalizing behaviors using delinquency, fighting, and violence scales created specifically for Add Health participants. We created three categories of externalizing behaviors: total externalizing, nonviolent, and violent behaviors. Nonviolent and violent items were chosen based on prior research (Beaver et al., 2016) and face validity. Some items changed across waves to keep measures age-appropriate and some items were commonly assessed across all waves. Via a parceling technique for measurement invariance models, we extracted items from each wave, creating total nonviolent behavior scores (αs=.610-.762) and total violent behavior scores (αs=.524-.743). Total externalizing behavior scores for each wave combined nonviolent and violent behaviors. Age was centered at 12-years old. Socioeconomic status and race were included as covariates, and we separately analyzed each sex.
RESULTS AND DISCUSSION. Measurement invariance analyses indicated partial strong factorial invariance, suggesting externalizing behaviors had the same underlying structure over time, despite some items changing to accommodate the participants’ age. Growth curve analyses found early maturers reported significantly elevated total externalizing (γs=.002 and .008, p<.01, for males and females, respectively), nonviolent (γs=.0002 and .0001, p<.01), and male violent (γ =-.002, p<.001) behaviors throughout adolescence but became indistinguishable from on-time and late-maturers after adolescence, supporting the attenuation hypothesis (Figures 1 and 2). The early maturation effect was only significant for males’ violent behaviors, suggesting early maturing males, not females, engaged in more violent behaviors at age 12, but declined in these behaviors more quickly than on-time or late-maturing males. These findings help clarify the differential effect of early maturation on two distinct domains of externalizing behaviors (nonviolent and violent behaviors), especially for females.

Authors

• Laura M Dimler, Regent University
  Presenting Author

• Misaki N Natsuaki, University of California, Riverside
  Non-Presenting Author