Deborah M. Capaldi

Peer ecology of male adolescent drug use

This report represents the perspective that adolescent substance use is best understood as an adaptation to an ecology defined jointly by families and peers. Hypotheses were tested on a sample of 206 boys in the Oregon Youth Study. The analyses proceeded in four steps. First, it was found that the transition from middle to high school was a period of rapid growth in smoking for boys with a prior history of low sociometric status. Second, a structural equation model was tested showing that deviant peer association in early adolescence mediated the relation between peer and family experiences in middle childhood and later substance use. Third, an observational study of the boys with their best friends revealed that active support for rule breaking and substance use was associated with immediate escalation in substance use during the transition to high school. Finally, it was found that ineffective parental monitoring practices were highly associated with the boys involvement in a deviant peer network. In fact, a high degree of similarity was found between boys and their best friends for substance use when parental monitoring was low. These analyses show that substance use in adolescence is embedded within the proximal peer environment, which in turn, emerges and is amplified within a context of low adult involvement and monitoring.

A Systematic Review of Risk Factors for Intimate Partner Violence.

A systematic review of risk factors for intimate partner violence IPV was conducted. Inclusion criteria included publication in a peer-reviewed journal, a representative community sample or a clinical sample with a control group comparison, a response rate of at least 50%, use of a physical or sexual violence outcome measure, and control of confounding factors in the analyses. A total of 228 articles were included (170 articles with adult and 58 with adolescent samples). Organized by levels of a dynamic developmental systems perspective, risk factors included (a) contextual characteristics of partners (demographic, neighborhood, community, and school factors), (b) developmental characteristics and behaviors of the partners (e.g., family, peer, psychological/behavioral, and cognitive factors), and (c) relationship influences and interactional patterns. Comparisons to a prior review highlight developments in the field in the past 10 years. Recommendations for intervention and policy along with future directions for IPV risk factor research are presented.

Co-occurrence of conduct problems and depressive symptoms in early adolescent boys: II. A 2-year follow-up at Grade 8

The current study examines outcomes at Grade 8 for boys who, at Grade 6, displayed elevated, though not necessarily clinical, levels of conduct problems and depressive symptoms. An at-risk community sample of 203 early adolescent boys in the Oregon Youth Study, a multimethod/multiagent study, was divided into the following groups at Grade 6: (a) co-occurring conduct problems and depressed mood, (b) conduct problems only, (c) depressed mood only, and (d) neither problem. The four groups were compared at Grade 8 on family management and boys adjustment, using multivariate analyses of variance. The groups were compared also on relationships with parents, delinquency, and suicidal ideation. As hypothesized, conduct problems showed higher stability than depressive symptoms from Grade 6 to Grade 8. The conduct-problem-only boys and boys with co-occurring conduct problems and depressive symptoms continued to show considerable adjustment deficits at Grade 8, whereas boys with only depressive symptoms showed some improvement. The co-occurring group showed elevated levels of suicidal ideation. As hypothesized, conduct problems at Grade 6 were predictive of increases in depressed mood by Grade 8, but depressed mood was not predictive of an increase in the conduct problems measure. Results are consistent with a failure model whereby lack of skill and noxious behavior lead to pervasive failures and vulnerability to depressed mood.

Development and Validation of an Early Adolescent Temperament Measure

Two studies were employed to develop a self-report temperament measure for the early adolescent period. The measure was based on the work of Rothbart and colleagues with adults and focused on emotionality, reactivity, and self-regulation. In Study 1, 97 middle school students (SO girls and 47 boys), aged 11 to 14 years, completed the Early Adolescent Temperament Questionnaire (EATQ). In addition, 93 parents reported on their adolescents temperament. Analyses assessing scale reliability, dimensionality, and discriminant validity were conducted. Two of the EATQs 14 scales were eliminated, and 2 were collapsed to form 1 scale; 92 of the original 168 items were retained after these analyses. The 11 final scales assessed fear, irritability, shyness, sadness, high-intensity pleasure, low-intensity pleasure, sensitivity, autonomic reactivity, motor activation, activity level, and attention. Alphas for the)) scales were high, and average convergence between parent report and adolescent report for the scales was .29. Three factors were identified in an analysis of the remaining 11 scale scores: negative emotion and somatic arousal, positive emotion and sensitivity, and high intensity pleasure or sensation seeking. In Study 2, participants were 64 boys and 64 girls, aged 11 to 14 years. Scale reliability and the factor structure of the modified adolescent scale were replicated in Study 2. In addition, eight scales from other measures were administered to assess convergent validity of the scales, and retests were conducted. Retest stability was high, and correlations with parallel scales averaged .50, indicating scale validity. Results indicate that the EATQ provides reliable and valid assessment of 1) dimensions of temperament for early adolescents.

Co-occurrence of conduct problems and depressive symptoms in early adolescent boys: III. Prediction to young-adult adjustment

The prediction of young-adult adjustment from early adolescent conduct problems and depressive symptoms was examined for an at-risk sample of approximately 200 males. Conduct problems and depressive symptoms were expected to show stability to young adulthood. It was predicted that early adolescent conduct problems would be associated with a broad range of adjustment problems in young adulthood due to cumulative adjustment failures. Early adolescent depressive symptoms were expected particularly to predict poor relationships with parents and peers. Additive and interactive effects of the two predictors were examined. Conduct problems and depressive symptoms showed significant stability to young adulthood. Conduct problems were associated with a broad range of adjustment problems including continuing problems in peer associations, substance use, self-esteem, relationships with parents, and new problems in noncompletion of education, unemployment, drivers license suspensions, and causing pregnancies. Depressive symptoms predicted particularly to problems in social relationships. Higher levels of both conduct problems and depressive symptoms in early adolescence did not predict to increased difficulties for any one outcome over either problem alone, either due to main or interaction effects. Such co-occurrence, however, did result in problem outcomes in multiple areas, thus, the poorest adjustment overall.

Psychometric properties of fourteen latent constructs from the Oregon Youth Study

1. Introduction.- 2. Academic Skills.- 3. Deviant Peers.- 4. Early Problems with the Target Child.- 5. Parent Depression.- 6. Peer Relations.- 7. Positive Parenting: Parent Inolvement.- 8. Positive Parenting: Positive Reinforcement.- 9. Child Self-Esteem.- 10. Parental Stress.- 11. Monitoring.- 12. Child Depressed Mood.- 13. Antisocial Behavior.- 14. Discipline.- 15. Problem-Solving.- References.

Middle Childhood Antecedents to Progressions in Male Adolescent Substance Use: An Ecological Analysis of Risk and Protection

Initiation of substance use before the age of 15 to 16 is a distinct risk factor for a variety of mental health problems and eventual drug abuse. Using multimethod, multi-agent measures of child, family, and peer antecedents at age 9 to 10, we studied the longitudinal effects in an at-risk sample of 206 boys. Event history analysis was used to examine the antecedents to patterned alcohol and tobacco use as well as experimentation with marijuana between ages 11 and 16. Univariate models revealed that at Grade 4, most constructs were prognostic of boys’ early substance use. Multivariate event history models clarified the risk and protective structure associated with tobacco, alcohol, and marijuana use. The level of intercorrelation among the predictor variables, however, suggested that family, peer, and child characteristics were inextricably connected within an ecology of development. A structural equation prediction model suggested a higher order construct, “childhood risk structure. ”

Can Violent Offenders be Distinguished from Frequent Offenders: Prediction from Childhood to Adolescence:

It was predicted that violent offending at adolescence is part of a general pattern of high-rate antisocial behavior that emerges in childhood. Family characteristics, including family management practices and childhood behavior, were compared for violent adolescent arrestees and nonviolent adolescent arrestees who were matched for arrest frequency. Further, regression analyses were conducted using self-report of violent offending. Also predicted was that multiple arrestees with no arrests for violence would self-report as much violence as multiple arrestees with arrests for violence. Hypotheses were tested on an at-risk sample of males in the Oregon Youth Study (OYS), a multimethod, multiagent longitudinal study. The findings generally supported the contention that violent offenders show the same background as frequent but nonviolent offenders.

Continuity of Parenting Practices Across Generations in an At-Risk Sample: A Prospective Comparison of Direct and Mediated Associations

A prospective model of parenting and externalizing behavior spanning 3 generations (G1, G2, and G3) was examined for young men from an at-risk sample of young adult men (G2) who were in approximately the youngest one third of their cohort to become fathers. It was first predicted that the young men in G2 who had children the earliest would show high levels of antisocial behavior. Second, it was predicted that G1 poor parenting practices would show both a direct association with the G2 sons subsequent parenting and a mediated effect via his development of antisocial and delinquent behavior by adolescence. The young fathers had more arrests and were less likely to have graduated from high school than the other young men in the sample. Findings were most consistent with the interpretation that there was some direct effect of parenting from G1 to G2 and some mediated effect via antisocial behavior in G2.

Heterosexual risk behaviors in at-risk young men from early adolescence to young adulthood: prevalence, prediction, and association with STD contraction.

Health-compromising lifestyles involve stable patterns of behavior and are associated with high-risk social environments and accelerated developmental trajectories. Developmentally, antisocial behavior is associated with such lifestyles. Mediational models predicting a measure of lifetime average sexual risk behavior assessed over a 10-year period (from ages 13-14 to 22-23 years) were examined for a sample of at-risk young men. The measure included years of abstinence from intercourse as well as levels of 3 key heterosexual indicators of risk: frequency of intercourse, number of intercourse partners, and condom use. Predictors included lifetime average measures of contextual, family, and peer process variables and individual behaviors. In addition, similar models for prediction of STD contraction were assessed. A younger age of onset of intercourse was associated with higher numbers of intercourse partners after onset. As hypothesized, findings indicated mediational associations of socioeconomic status, parental monitoring, deviant-peer association, antisocial behavior, and substance use in the prediction of sexual risk behavior. Lower condom use also predicted STD contraction.