A Path to Serious,Violent,Chronic Delinquency: The Harmful Aftermath of Adverse Childhood Experiences

Abstract

Adverse childhood experiences can affect the development of a child in many ways, leading to highly maladaptive behaviors, such as serious, violent, and chronic (SVC) delinquency. This study uses data from 64,329 Florida Department of Juvenile Justice youth, collected from 2007 to 2012, to examine both the direct and indirect effects of adverse childhood experiences (ACEs) on SVC delinquency. Using a generalized structural equation model, the effects of ACE scores are estimated on a youth’s likelihood of engaging in SVC delinquency while considering pathways through maladaptive personality traits (aggression and impulsivity), as well as adolescent problem behaviors (deviant peer imitation, school difficulties, substance abuse problems, and mental illness). The results suggest that a large proportion of the relationship between childhood adversity and SVC delinquency is mediated by maladaptive personality traits and adolescent problem behaviors. Study limitations and directions for future research are also discussed.

Keywords

juveniles violence ACE assessment personality problem behaviors

Introduction

Experiences of childhood adversity are found to be quite common for children around the world. Recent estimates of childhood trauma have indicated that more than half of all children endure at least one type of adverse experience (Anda et al., 2006; Copeland, Keeler, Angold, & Costello, 2007; Felitti et al., 1998). These experiences may include different types of maltreatment, such as abuse or neglect, or dysfunctional household environments, due to witnessing family violence, experiencing parental separation/divorce, household member incarceration, household mental illness, or household substance abuse. Modern-day research in the developmental psychology field has investigated the effects of childhood adversity on the likelihood of a number of undesirable outcomes later in life. These studies have shown that early childhood adversity can lead to a variety of developmental problems occurring over the life course (Cicchetti & Toth, 1995; Lamphear, 1985; Trickett & McBride-Chang, 1995).

A leading theoretical context for research in this area is the developmental psychopathology perspective. This perspective is “the study of the origins and course of patterns of behavioral maladaptation, whatever the age of onset, whatever the causes, whatever the transformations in behavioral manifestation, and however complex the course of the developmental pattern may be” (Sroufe & Rutter, 1984, p.18). Research in this area examines the progression of adaptive as well as maladaptive behaviors, while considering a range of difficulties at each stage of a child’s growth to determine potential antecedents and causes for abnormal development (Causadias, 2013). One of the central considerations in this field examines the effects of maltreatment and other adverse childhood experiences (ACEs). The inclusion of child maltreatment in the developmental psychopathology discipline was principally influenced by the research of developmental and clinical psychologist, Dante Cicchetti, in the 1980s and 1990s (Cicchetti, 1984; Cicchetti & Toth, 1995).

According to this area of investigation, because child maltreatment is such a powerful experience for youth, abused or neglected children may experience developmental risks for inadequate maturation and adaptation later in life (Cicchetti & Toth, 2005). Consequently, these children may instead show symptoms of maladaptive functioning throughout each stage of subsequent development (Cicchetti & Toth, 1995, 2008; Manly, Cicchetti, & Barnett, 1994). Children who experience abuse, neglect, or other adverse experiences are hypothesized to be more likely to display developmental difficulties with regulating their emotions, fostering positive personal relationships, and succeeding in school, and are found to display certain psychopathological traits during later life stages (Cicchetti & Toth, 1995).

As a result of the research examining the harmful effects of childhood adversity, researchers have developed specialized ways to empirically measure it. One such measurement method, developed in the medical field, is the ACE assessment. The ACE assessment was developed in a 1998 study by Felitti and colleagues to examine the relationship between ACEs and the most common causes of death in America. This project surveyed more than 17,000 adults to distinguish the negative childhood experiences that were related to serious health problems in adulthood (Felitti et al., 1998). The initial questionnaire included measures of emotional abuse, physical abuse, sexual abuse, witnessing household violence, household substance abuse, household mental illness, and having an incarcerated member of the family. Ensuing research has included physical neglect, emotional neglect, and parental separation/divorce. Altogether, these forms of adversity comprise a 10-item assessment used to calculate an individual’s ACE score, which represents his or her respective level of childhood adversity.

Since the seminal Felitti and colleagues (1998) study, other researchers have further examined the effects of childhood adversity on other negative outcomes using the ACE assessment. These studies have demonstrated that higher ACE scores may be correlated with smoking (Anda et al., 1999), alcoholism (Dong et al., 2005), obesity (Burke, Hellman, Scott, Weems, & Carrion, 2011), mental illness (Chapman, Dube, & Anda, 2007), depression (Dube et al., 2003), risky sexual behavior (Hillis, Anda, Felitti, & Marchbanks, 2001), adolescent pregnancy (Hillis et al., 2004), homelessness (Herman, Susser, Struening, & Link, 1997), and suicidal behavior (Dube, Anda, Felitti, Edwards, & Croft, 2002; Perez, Jennings, Piquero, & Baglivio, 2016). Research in the criminological field has also linked ACE scores to delinquent behavior (Baglivio et al., 2014; Baglivio, Wolff, Piquero, & Epps, 2015), and even to a youth’s likelihood of serious, violent, and chronic (SVC) delinquency (Fox, Perez, Cass, Baglivio, & Epps, 2015).

The Current Study

Although past studies have shown that higher levels of childhood adversity can predict criminal involvement, and specifically, serious, violent, and chronic delinquency, potential intervening mechanisms between the adverse experiences and SVC delinquency have not been fully examined simultaneously. Consequently, the present study investigates the mediating processes that may explain the relationship between a child’s adverse experiences and SVC behavior using data from 64,239 Florida Department of Juvenile Justice (FDJJ) youth. This research aims to recommend more effective interventions to prevent the progression toward this SVC behavior. This project follows the model of a recent study that examined the mediating role of maladaptive personality traits and problem behaviors (see Jessor, 1987; Jessor & Jessor, 1977) in the relationship between ACEs and suicide (see Perez et al., 2016).

Specifically, the current study seeks to answer three main research questions regarding the role of childhood adversity in the lives of (SVC) juvenile offenders:

Research Question 1: Does the ACE score predict the development of two maladaptive personality traits (aggression and impulsivity)?

Research Question 2: Does the ACE score predict the presence of four adolescent problem behaviors (deviant peer imitation, school difficulties, substance abuse, and mental illness), and is this relationship mediated by aggression and impulsivity? and,

Research Question 3: Does the ACE score predict SVC delinquency, and is this relationship fully (or partially) mediated by the presence of maladaptive personality traits or adolescent problem behaviors?

Method

The data used for the current study were originally collected by the FDJJ. This population consists of all de-identified juveniles who received a delinquency referral in Florida and aged out of the system between January 1, 2007, and December 31, 2012. At the time of each FDJJ referral, each youth was administered a Positive Achievement and Change Tool (PACT) assessment. The PACT is a risk/needs assessment that includes a semi-structured interview with a juvenile probation officer, a case file examination, and an appraisal of the child’s official child abuse records. Through the automated nature of the PACT assessment, the probation officer is given more time to devote to interacting with the youth. For example, the criminal history items on the PACT are fully automated, eliminating the need for self-report or recall of charges, adjudications, or system placements. Based on the PACT assessment, a specialized case plan is created to appropriately assist the youth.

The PACT consists of two versions: a Pre-Screen (46 items) and a Full Assessment (126 items). The FDJJ assesses each juvenile with the Pre-Screen when they first enter the system. Any youth who scores as moderate-high or high risk to re-offend is given the Full PACT Assessment. The Full Assessment contains 12 distinct domains: “criminal history, gender, school, use of free time, employment, relationships, family and living arrangements, alcohol and drugs, mental health, attitudes/behaviors, aggression, and skills” (Baird et al., 2013). The current study’s final sample is comprised of those who were administered the Full PACT Assessment (N = 64,329). Each of the following measures was compiled by the juvenile caseworker based on the semi-structured interview and his or her review of the youth’s official records.

Measures

Adverse Childhood Experiences (ACE) Score

Based on information collected in the PACT, each ACE item was coded dichotomously representing the presence (1) or absence (0) of each ACE. The ACE items included (a) emotional abuse, (b) physical abuse, (c) sexual abuse, (d) emotional neglect, (e) physical neglect, (f) witnessing household violence, (g) household substance abuse, (h) household mental illness, and (i) household member incarceration. An ACE score, ranging from 0 to 9, was created by summing the number of ACE items endorsed (see also Fox et al., 2015; Perez et al., 2016).

Maladaptive personality traits

Aggression

Aggression was measured using a latent construct, comprised of the following six ordinal-level indicator variables: (a) the youth’s level of belief in yelling and verbal aggression to resolve a conflict; (b) the youth’s level of belief in fighting and physical aggression to resolve a conflict, (c) the youth’s tolerance for frustration (d) the youth’s empathy, remorse, sympathy, or feelings for the victim; (e) the youth’s hostile interpretation of actions and intentions of others in non-confrontational settings; and (f) evidence of prior violent or aggressive behavior not included in the juvenile’s criminal record (α = .79).

Impulsivity

Impulsivity was measured from the caseworker’s determination of the juvenile’s impulsivity according to one of the following four categories that best described the youth: (a) usually thinks before acting; (b) sometimes thinks before acting; (c) impulsive, often acts before thinking; or (d) highly impulsive, usually acts before thinking.

Adolescent problem behaviors

Deviant peer imitation

Deviant peer imitation represented a youth’s admiration/imitation of antisocial peers, and the juvenile was coded by the caseworker as being someone who (a) does not admire/imitate antisocial peers, (b) somewhat admires/imitates antisocial peers, or (c) admires/imitates antisocial peers.

School difficulties

School difficulties was assessed using a latent construct, comprised of the following seven ordinal-level variables: (a) youth’s belief in the value of an education, (b) school involvement, (c) school suspensions, (d) school conduct, (e) school attendance (reverse-coded), (f) school performance (reverse-coded), and (g) dropout/expulsion (α = .71).

Substance abuse

Substance abuse was based on the juvenile’s past or current use of a substance in a manner that may disrupt his or her education, cause family conflict, interfere with friends, cause health problems, or contribute to delinquency. The caseworker also looked for signs of tolerance or withdrawal. These criteria resemble those used to diagnose substance abuse in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-V; American Psychiatric Association, 2013). If the youth experienced symptoms of substance abuse, they were coded “1,” while all others were coded “0.”

Mental illness

Juveniles who indicated symptoms of general mental health problems, depression, somatic complaints, or thought disturbances were coded “1,” while all others without any mental illness symptoms were coded “0.”

SVC delinquency

All juveniles who had committed three or more serious felony offenses, with at least one violent, were coded “1” (i.e., engaged in SVC delinquency), while all others were coded “0” (Fox et al. 2015).

Demographics

Demographic variables included the gender of the youth (0 for females and 1 for males), the race/ethnicity of the juvenile (White, African American, Hispanic, or Other) with White being utilized as the reference category, and the socioeconomic status (SES) of the juvenile’s family. SES was represented by four categories of household income: (a) less than US$15,000; (b) from US$15,000 to US$34,999; (c) from US$35,000 to US$49,999; and (d) US$50,000 and more.

Analytic Method

To accommodate the unique features of the PACT data with many dichotomous or ordinal measures, a generalized structural equation model (GSEM) was used, which allows for variables that do not fit the necessary conditions of a normal distribution required for SEM (Muthén, 1984; Skrondal & Rabe-Hesketh, 2004). The structural model for the GSEM is found in Figure 1. This model consisted of the demographic characteristics, the ACE score, the two maladaptive personality traits, the four adolescent problem behaviors, and SVC delinquency (e.g., the outcome). Each stage of the model allows for the mediation of the effects of the previous stage and the estimation of the direct and indirect effects on the outcome at the succeeding stage of analysis. As such, the GSEM model was estimated in an effort to predict the level of mediation between ACEs and SVC delinquency by considering the intervening effects of aggression, impulsivity, deviant peer imitation, school difficulties, substance abuse problems, and mental illness (see also Perez et al., 2016).

Figure 1.

Full GSEM structural model.

Results

Descriptive Statistics

Table 1 presents the demographic background information for the sample. The sample is predominantly male (78.33%) and widely dispersed among racial/ethnic minorities with 42.88% African American (non-Hispanic), 38.23% White (non-Hispanic), 15.37% Hispanic, and 3.52% of the “other” racial/ethnic category. On average, the youth were 17 years old at the time of the last assessment. In addition, the majority of household incomes of the youths’ families fall below US$35,000 per year.

Table 1.

Descriptive Statistics for Demographic Characteristics.

Variable		%
Gender	Male	78.33
Gender	Female	21.67
Race/Ethnicity	White, non-Hispanic	38.23
	African American, non-Hispanic	42.88
	Hispanic	15.37
	Other	3.52
Socioeconomic status	Less than US$15,000	19.46
	US$15,000-US$34,999	53.87
	US$35,000-US$49,999	18.03
	US$50,000 or more	8.64
ACE items	Emotional abuse	32.53
	Physical abuse	26.53
	Sexual abuse	9.22
	Emotional neglect	13.16
	Physical neglect	13.28
	Household violence	33.25
	Household substance abuse	24.37
	Household mental illness	12.28
	Incarcerated household member	65.92
SVC delinquency	SVC delinquent	16.66
SVC delinquency	Non-SVC delinquent	83.34

Note. ACE = adverse childhood experience; SVC = serious, violent, and chronic.

Table 1 also presents the prevalence of each ACE item among youth in the sample. The results demonstrate that each ACE item is experienced by between 9% and 66% of the juveniles. The most prevalent ACE was having an incarcerated family member (65.92%), followed by having experienced emotional abuse (32.53%), having witnessed household violence (33.25%), having experienced physical abuse (26.53%), having grown up with a family member who abused substances (24.37%), having experienced physical neglect (13.28%), having experienced emotional neglect (13.16%), having grown up with a mentally ill family member (12.28%), and having experienced sexual abuse (9.22%).

The ACE score for each juvenile was calculated by adding these nine measures for each juvenile, producing a score ranging from 0 to 9. In this sample, the average ACE score was 2.31 (SD = 1.85). Only 16.74% of youth did not experience any ACEs, indicating that more than 83% experienced at least one of the nine ACEs. In fact, nearly 60% of the sample experienced two or more ACEs. However, less than 2.5% of the sample experienced an ACE score higher than 7. The individual and cumulative prevalence of each ACE score can be found in Table 2.

Table 2.

Percentage of Sample With Each Total ACE Score.

ACE score	%	Cumulative %
0	16.74	16.74
1	24.06	83.26
2	19.53	59.21
3	14.97	39.68
4	10.70	24.71
5	7.30	14.01
6	4.29	6.71
7	1.92	2.42
8	0.45	0.50
9	0.05	0.05

Note. ACE = adverse childhood experience.

The final descriptive statistic integral to this study related to the number of youth who were classified as SVC delinquents. The results indicated that 10,714 juveniles (16.66% of the sample) were classified as SVC delinquents (committed three or more felony offenses, with at least one classified as “against persons”).

The Effect of ACE Scores on SVC Delinquency Without Mediators

To estimate the effect of the ACE score on a juvenile’s likelihood of being an SVC delinquent, a preliminary logistic regression model was run. This model included only the ACE score and the demographic control variables to test the relationship between ACEs and SVC delinquency without any mediators. The results of this regression can be found in Table 3. Overall, the model was significant (p < .001). A higher ACE score significantly increased the odds of a youth being classified as an SVC delinquent (odds ratio [OR] = 1.30, p < .001, 95% confidence interval [CI] = [1.29, 1.32]). The model indicated that males (OR = 4.15, p < .001, 95% CI = [3.87, 4.47]), African Americans (OR = 3.19, p < .001, 95% CI = [3.02, 3.37]), Hispanics (OR = 1.66, p ≤ .001, 95% CI = [1.55, 1.80]), and the “other” racial/ethnic category (OR = 2.89, p < .001, 95% CI = [2.58, 3.25]) each had an increased likelihood of SVC delinquency.

Table 3.

Logistic Regression Predicting SVC Delinquency.

Predictor	b	SE	OR	95% CI
ACE score	0.27	.01	1.30***	[1.29, 1.32]
Gender	1.42	.15	4.15***	[3.87, 4.47]
Race/Ethnicity: African American	1.16	.09	3.19***	[3.02, 3.37]
Race/Ethnicity: Hispanic	0.51	.06	1.67***	[1.55, 1.80]
Race/Ethnicity: Other	1.06	.17	2.89***	[2.57, 3.24]
Socioeconomic status	.02	.01	1.02	[0.99, 1.05]

Note. SVC = serious, violent, and chronic; ACE = adverse childhood experience; OR = odds ratio; CI = confidence interval; Nagelkerke R² = .087, p < .001.

p < .05. **p < .01. ***p < .001.

The Mediation Model of ACE Scores and SVC Delinquency

In light of the logistic regression results, which demonstrated a significant predictive relationship between the ACE score and SVC delinquency, the GSEM model was estimated (Figure 1) to examine potentially intervening mechanisms in this relationship. The results of this model are found in Table 4. In the ensuing sections, the results will be described in sequential order, starting with the demographic control variables on the left of the model/table and progressing to the estimation of all predictor variables on the likelihood of SVC delinquency.

Table 4.

GSEM Coefficient Estimates.

Predictor	Direct effect
	1: ACE		2: Maladaptive personality development				3: Adolescent problem behaviors								4: SVC delinquency
	ACE score		Aggression		Impulsivity		Deviant peer imitation		School difficulties		Substance abuse		Mental illness		SVC Delinquency
	b	95% CI	b	95% CI	b	95% CI	B	95% CI	B	95% CI	Exp(b)	95% CI	Exp(b)	95% CI	Exp(b)	95% CI
Gender	−0.77*** (0.02)	[−0.80, −0.73]	0.03*** (0.00)	[0.02, 0.03]	0.21*** (0.01)	[0.19, 0.23]	0.37*** (0.02)	[0.34, 0.41]	0.05*** (0.00)	[0.05, 0.06]	1.73*** (0.01)	[1.70, 1.85]	0.73*** (0.02)	[0.69, 0.77]	3.86*** (0.04)	[3.60, 4.14]
Race/Ethnicity: African American	−0.52*** (0.02)	[−0.55 −0.49]	0.11*** (0.00)	[0.11, 0.12]	0.11*** (0.02)	[0.09, 0.13]	0.28*** (0.02)	[0.24, 0.31]	0.08*** (0.00)	[0.08, 0.09]	0.57*** (0.01)	[0.55, 0.60]	0.64*** (0.02)	[0.61, 0.67]	2.89*** (0.03)	[2.71, 3.06]
Race/Ethnicity: Hispanic	−0.78*** (0.02)	[−0.83, −0.74]	0.05*** (0.01)	[0.05, 0.06]	−0.01 (0.02)	[−0.04, −0.01]	0.18*** (0.02)	[0.13, 0.23]	0.05*** (0.00)	[0.05, 0.06]	0.90*** (0.02)	[0.87, 0.96]	0.82*** (0.03)	[0.78, 0.87]	1.53*** (0.04)	[1.42, 1.67]
Race/Ethnicity: Other	−0.85*** (0.04)	[−0.93, −0.77]	0.11*** (0.10)	[0.09, 0.12]	0.08** (0.02)	[0.03, 0.12]	0.26*** (0.04)	[0.17, 0.35]	0.06*** (0.00)	[0.05, 0.07]	0.57*** (0.03)	[0.53, 0.63]	0.76*** (0.05)	[0.69, 0.84]	2.59*** (0.06)	[2.29, 2.89]
Socioeconomic status	−0.29*** (0.01)	[−0.30, −0.27]	0.003 (0.002)	[−0.01, 0.01]	0.01* (0.02)	[0.00, 0.02]	0.05*** (0.01)	[0.03, 0.07]	−0.03*** (0.00)	[−0.03, −0.02]	1.20*** (0.01)	[1.19, 1.23]	1.04*** (0.01)	[1.02, 1.07]	1.01 (0.04)	[0.98, 1.04]
ACE score	—	—	0.12*** (0.001)	[0.12, 0.12]	0.23*** (0.02)	[0.22, 0.23]	0.06*** (0.01)	[0.05, 0.07]	0.02*** (0.00)	[0.02, 0.03]	1.14*** (0.01)	[1.12, 1.17]	1.43*** (0.01)	[1.41, 1.45]	1.08*** (0.01)	[1.06, 1.09]
Aggression	—	—	—	—	—	—	1.39*** (0.02)	[1.34, 1.44]	0.27*** (0.01)	[0.27, 0.28]	2.03*** (0.05)	[1.93, 2.14]	3.63*** (0.03)	[3.42, 3.85]	2.61*** (0.04)	[2.43, 2.80]
Impulsivity	—	—	—	—	—	—	0.92*** (0.01)	[0.90, 0.95]	0.07*** (0.00)	[0.07, 0.07]	1.25** (0.02)	[1.21, 1.28]	1.18*** (0.01)	[1.15, 1.21]	1.12*** (0.02)	[1.07, 1.15]
Deviant peer imitation	—	—	—	—	—	—	—	—	—	—	—	—	—	—	1.12*** (0.02)	[1.07, 1.16]
School difficulties	—	—	—	—	—	—	—	—	—	—	—	—	—	—	1.73*** (0.04)	[1.58, 1.89]
Substance abuse	—	—	—	—	—	—	—	—	—	—	—	—	—	—	1.12*** (0.02)	[1.07, 1.17]
Mental illness	—	—	—	—	—	—	—	—	—	—	—	—	—	—	1.36*** (0.03)	[1.28, 1.43]

Note. Standard errors in parentheses; GSEM = generalized structural equation model; ACE = adverse childhood experience; SVC = serious, violent, and chronic; CI = confidence interval; n = 63,400; AIC = 778,183.0 0; BIC = 778,872.20; df = 75. AIC = Akaike Information Criterion; BIC = Bayesian Information Criterion.

p < .05. **p < .01. ***p < .001.

The effects of demographics on the ACE score

Each of the control variables was a significant predictor of the ACE score. Specifically, being male was associated with a significantly lower ACE score (b = −0.77, p < .001, 95% CI = [−0.80, −0.73]). In addition, each of the three racial/ethnic categories, African American (b = −0.52, p < .001, 95% CI = [−0.55, −0.49]), Hispanic (b = −0.78, p < .001, 95% CI = [−0.83, −0.74]), and other (b = −0.85, p < .001, 95% CI = [−0.93, −0.77]), had lower ACE scores than the White reference group. Finally, youth reporting a higher SES scored lower on the ACE assessment (b = −0.29, p < .001, 95% CI = [−0.30, −0.32]).

The effect of the ACE score on maladaptive personality traits

The model revealed that, even when controlling for demographics, higher ACE scores were associated with significant increases in a juvenile’s aggression (b = 0.12, p < .001, 95% CI = [0.12, 0.12]). The youth’s demographics, however, were not fully mediated by the inclusion of the ACE score. For example, aggression was significantly higher in males (b = 0.03, p < .001, 95% CI = [0.02, 0.03]) and in each of the three racial/ethnic groups, African American (b = 0.11, p < .001, 95% CI = [0.11, 0.12]), Hispanic (b = 0.05, p < .001, 95% CI = [0.05, 0.06]), and other (b = 0.11, p < .001, 95% CI = [0.09, 0.12]), relative to Whites.

As with aggression, a higher ACE score was associated with higher impulsivity (b = 0.23, p < .001, 95% CI = [0.22, 0.23]). Impulsivity, though, was also higher for males (b = 0.21, p < .001, 95% CI = [0.19, 0.23]) and African Americans (b = 0.11, p < .001, 95% CI = [0.09, 0.13]). In addition, those of higher socioeconomic status were found to have higher levels of impulsivity (b = 0.01, p = .04, 95% CI = [0.00, 0.02]).

The effect of the ACE score on adolescent problem behaviors

The first adolescent problem behavior examined was deviant peer imitation. Higher ACE scores were directly predictive of a juvenile’s level of admiration and imitation of his or her deviant peers (b = 0.06, p < .001, 95% CI = [0.05, 0.07]). There was also a significant indirect effect found through the mediator variables of aggression (b = 0.17, p < .001) and impulsivity (b = 0.21, p < .001). This increased the total effect of the ACE scores on deviant peer imitation from 0.06 to 0.44 (p < .001).¹ The direct, indirect, and total effects of the ACE score on each outcome in the model are listed in Table 5.

Table 5.

Direct, Indirect, and Total Effects.

Model path	Effects
Model path	Direct	Indirect	Total
ACE Score to Aggression	0.12***	—	0.12***
ACE Score to Impulsivity	0.23***	—	0.23***
ACE Score to Deviant Peer Imitation	0.06***	0.17*(Aggression) 0.21*(Impulsivity)	0.44***
ACE Score to School Difficulties	0.02***	0.03* (Aggression) 0.02* (Impulsivity)	0.07***
ACE Score to Substance Abuse	1.14***	1.09* (Aggression) 1.05* (Impulsivity)	1.31***
ACE Score to Mental Illness	1.43***	1.17* (Aggression) 1.03* (Impulsivity)	1.75***
ACE Score to SVC Delinquency	1.08***	1.12* (Aggression) 1.03 (Impulsivity) 1.05*** (Deviant Peer Imitation) 1.01* (School Difficulties) 1.03* (Substance Abuse) 1.11* (Mental Illness)	1.31***

Note. Unstandardized effect estimates. ACE = adverse childhood experience; SVC = serious, violent, and chronic.

p < .05. **p < .01. ***p < .001.

Demographic characteristics also displayed significant effects on a juvenile’s deviant peer imitation. Males (b = 0.37, p < .001, 95% CI = [0.34, 0.41]), African Americans (b = 0.28, p < .001, 95% CI = [0.24, 0.31]), Hispanics (b = 0.18, p < .001, 95% CI = [0.13, 0.23]), members of the other racial/ethnic category (b = 0.26, p < .001, 95% CI = [0.17, 0.35]), and youth of higher SES (b = 0.05, p < .001, 95% CI = [0.03, 0.07]) displayed an increased level of admiration and imitation of deviant peers. Last, higher levels of both aggression (b = 1.39, p < .001, 95% CI = [1.34, 1.44]) and impulsivity (b = 0.92, p < .001, 95% CI = [0.90, 0.95]) also directly predicted increases in a juvenile’s deviant peer imitation.

The second adolescent problem behavior was school difficulties. Higher ACE scores were directly predictive of a greater level of school difficulties (b = 0.02, p < .001, 95% CI = [0.02, 0.03]). A significant indirect effect also existed through aggression (b = 0.03, p < .001) and impulsivity (b = 0.02, p < .001), bringing the total effect of the ACE score on school difficulties from 0.02 to 0.07 (p < .001).

Again, all demographic control variables also significantly affected the juvenile’s school difficulties. Youth who were males (b = 0.05, p < .001, 95% CI = [0.05, 0.06]), African Americans (b = 0.08, p < .001, 95% CI = [0.08, 0.09]), Hispanics (b = 0.05, p < .001, 95% CI = [0.05, 0.06]), members of the other racial/ethnic category (b = 0.06, p < .001, 95% CI = [0.05, 0.07]), and of lower socioeconomic status (b = −0.03, p < .001, 95% CI = [−0.03, −0.02]) each experienced higher levels of school difficulties. Higher levels of aggression (b = 0.27 p < .001, 95% CI = [0.27, 0.28]) and impulsivity (b = 0.07, p < .01, 95% CI = [0.07, 0.07]) were each predictive of higher school difficulties.

The third adolescent problem behavior included in the model was substance abuse problems. Because this is a dichotomous outcome, these effects are reported as ORs. A higher ACE score directly and significantly predicted the odds that a juvenile would display a substance abuse problem (OR = 1.14, p < .001, 95% CI = [1.12, 1.17]). The results also indicated that a significant indirect effect was found through both aggression (OR = 1.09, p < .001) and impulsivity (OR = 1.05, p < .001), which increased the total effect from 1.14 to 1.31 (p < .001).

Each of the control variables significantly predicted the odds of a youth presenting a substance abuse problem. Males (OR = 1.78, p < .001, 95% CI = [1.70, 1.85]); Whites, as opposed to African Americans (OR = 0.57, p < .001, 95% CI = [0.55, 0.60]); Hispanics (OR = 0.90, p < .001, 95% CI = [0.87, 0.96]); or members of the other racial/ethnic category (OR = 0.57, p < .001, 95% CI = [0.53, 0.63]); and youth of higher SES (OR = 1.20, p < .001, 95% CI = [1.19, 1.23]) each exhibited a higher likelihood of substance abuse problems. Aggression (OR = 2.03, p < .001, 95% CI = [1.93, 2.14]) and impulsivity (OR = 1.25, p < .01, 95% CI = [1.21, 1.28]) each was predictive of a higher likelihood of substance abuse problems.

The fourth and final adolescent problem behavior examined in the GSEM model was mental illness. Again, because this measure is dichotomous, these effects are reported as ORs. The ACE score was directly predictive of a higher likelihood of mental illness (OR = 1.43, p < .001, 95% CI = [1.41, 1.45]), although the results showed a significant indirect effect through aggression (OR = 1.17, p < .001) and impulsivity (OR = 1.03, p < .001). These indirect effects increased the total effect from 1.43 to 1.75 (p < .001).

All demographic control variables also displayed significant relationships with a youth’s likelihood of a mental illness. Females (OR = 0.73, p < .001, 95% CI = [0.69, 0.77]); Whites, as opposed to African Americans (OR = 0.64, p < .001, 95% CI = [0.61, 0.67]); Hispanics (OR = 0.82, p < .001, 95% CI = [0.78, 0.87]); or member of the other racial/ethnic category (OR = 0.76, p < .001, 95% CI = [0.69, 0.84]); and those of higher SES (OR = 1.04, p < .001, 95% CI = [1.02, 1.07]) each displayed an increased likelihood of mental illness. Finally, higher levels of aggression (OR = 3.63, p < .001, 95% CI = [3.42, 3.85]) and impulsivity (OR = 1.18, p < .01, 95% CI = [1.15, 1.21]) also were directly predictive of a greater likelihood of mental illness.

The effect of the ACE score on SVC delinquency

In the final stage of the GSEM model, the ACE score, the two maladaptive personality traits, and the four adolescent problem behaviors were estimated as predictors of SVC delinquency, while controlling for demographics. Nearly every model predictor was associated with an increase in the likelihood of a juvenile being an SVC delinquent. The ACE score was a significant direct predictor of a youth’s higher odds of SVC delinquency (OR = 1.08, p < .001, 95% CI = [1.06, 1.09]). A higher likelihood of becoming an SVC delinquent was also predicted by increases in both maladaptive personality traits—aggression (OR = 2.61, p < .001, 95% CI = [2.43, 2.80]) and impulsivity (OR = 1.12, p < .001, 95% CI = [1.07, 1.15])—and all four adolescent problem behaviors—deviant peer imitation (OR = 1.12, p < .001, 95% CI = [1.07, 1.16]), school difficulties and dropout (OR = 1.73, p < .001, 95% CI = [1.58, 1.89]), substance abuse problems (OR = 1.12, p < .001, 95% CI = [1.07, 1.17]), and mental illness (OR = 1.36, p < .001, 95% CI = [1.28, 1.43]).

When considering the demographic control variables, males (OR = 3.86, p < .001, 95% CI = [3.60, 4.14]), African Americans (OR = 2.89, p < .001, 95% CI = [2.71, 3.06]), Hispanics (OR = 1.53, p < .001, 95% CI = [1.42, 1.67]), and member of the other racial/ethnic category (OR = 2.59, p < .001, 95% CI = [2.29, 2.89]) were significantly more likely to be identified as SVC delinquents.

The ACE score’s indirect effect through the maladaptive personality traits and adolescent problem behaviors was calculated for SVC delinquency. The results indicated that the relationship between the ACE scores was partially mediated by the six mediator variables, as significant indirect effects existed through aggression (OR = 1.12, p < .001) and impulsivity (OR = 1.03, p < .01), as well as through deviant peer imitation (OR = 1.05, p < .001), school difficulties (OR = 1.01, p < .05), substance abuse problems (OR = 1.03, p < .01), and mental illness (OR = 1.11, p < .001). When combining the direct effect and these indirect effects, the total effect of the ACE score on SVC delinquency was increased from 1.08 to 1.31 (p < .001).²

Discussion

Prior research has shown that SVC delinquency may be linked to early experiences of childhood adversity (Fox et al., 2015). Until now, the intervening variables that may explain this relationship, however, have not fully been examined simultaneously. As such, the current study explored this relationship by estimating a GSEM that included a variety of mediating variables related to maladaptive personality development and adolescent problem behaviors, using data from the FDJJ. The estimation of this model allowed for the consideration of a number of developmental factors that may occur during the life of a juvenile and may clarify the nature of this relationship. As a result of this analysis, a number of important findings emerged.

First and foremost, the ACE score was a significant predictor of SVC delinquency. This relationship was found in both a simple logistic regression model and the more complicated GSEM model. This finding supported the work of Fox and colleagues (2015). In addition, the ACE score significantly predicted each of the hypothesized mediator variables (aggression, impulsivity, deviant peer imitation, school difficulties, substance abuse problems, and mental health problems). These relationships were anticipated as a result of the wealth of literature connecting ACEs and many of these maladaptive developmental outcomes (for review, see Perez et al., 2016).

As a result of the relationships with the aforementioned mediator variables, the total effect of the ACE score was partially mediated by the inclusion of the maladaptive personality traits and adolescent problem behaviors. Significant indirect effects existed for each of the mediator variables in the model. The ACE score, however, did remain a significant predictor of SVC delinquency. Despite the importance of these intervening mechanisms, the ACE score’s enduring direct effect suggests that other aspects of development may also be affected by adversity and influence SVC delinquency that were not included in the present analysis.

Implications

A number of implications emerge from this study’s findings. First, this study further validates the use of the ACE assessment as a predictive tool for serious violent behavior and other maladaptive adolescent outcomes. The ACE score significantly predicted SVC delinquency, maladaptive personality traits, and adolescent problem behaviors. The ACE assessment is a 10-item tool that could easily be administered by a number of different individuals, such as pediatricians, counselors, and teachers, among others. The simplicity of administering this assessment may provide practitioners important insights into potential outcomes in the lives of those who experience a variety of adverse experiences early in life. As a result, certain interventions could be implemented for youth with higher ACE scores to prevent their maladaptive development and potential for serious violent delinquency.

In addition, this study suggests the importance of interventions for youth who have experienced multiple adverse experiences that target the specific developmental changes that affect the relationship with SVC delinquency. For example, programs aiming to reduce aggression and impulsivity levels (see Musci et al., 2014; Piquero, Jennings, & Farrington, 2010; Piquero, Jennings, Farrington, Diamond, & Reingle Gonzalez, 2016) in youth who experience adversity may assist in altering the path of development toward SVC delinquency. Furthermore, interventions designed to preclude adolescent problem behaviors may also affect the relationship between childhood adversity and SVC offending. Interventions such as these include (a) “Big Brother/Big Sisters” and peer mediation programs that encourage positive peer associations and reduce involvement with delinquent peers (Grossman & Tierney, 1998; Herrera, Grossman, Kauh, Feldman, & McMaken, 2011; Tierney, Grossman, & Resch, 1995; Wasserman & Miller, 1998), (b) interventions to address school difficulties (see Oyserman, Terry, & Bybee, 2002), (c) substance abuse reduction programs (Gorman, 2014; Sussman et al., 2012), and (d) interventions targeting depression and other mental illnesses (see Clarke et al., 1995; Clarke et al., 2001; Dadds, Spence, Holland, Barrett, & Laurens, 1997).

Finally, this study also highlights the importance of the prevention of ACEs. Because the ACE score was a direct predictor of all maladaptive outcomes in the study, the prevention of such adverse experiences may play a pivotal role in affecting the path toward these outcomes. For example, improved prenatal care and assistance, parent and family training interventions, and home visitation programs have all shown potential in reducing adverse experiences and improving the ensuing development of the child (Cohen, Piquero, & Jennings, 2010; Zigler & Hall, 1989). These programs have consistently been found to meaningfully affect subsequent delinquent and criminal behavior in youth who receive these interventions (see Olds et al., 1997; Olds et al., 1998; Piquero, Farrington, Welsh, Tremblay, & Jennings, 2009; Piquero, Jennings, Diamond, et al., 2016). As such, these programs may prevent youth from experiencing childhood adversity in the first place, further pushing a youth away from a path toward maladaptation and SVC delinquent behavior.

Limitations

The current study has its limitations. For starters, it should be noted that, although the estimated model proposes a sequential developmental path toward an outcome, the cross-sectional nature of the data preclude us from making any definitive statements on the true causal ordering of these outcomes. The PACT data used were compiled at the time of the youth’s last contact with FDJJ, and as a result, we are unable to determine which behaviors occurred during early childhood and which occurred at later times. Instead, the current results only suggest significant associations between the variables of interest. A prospective longitudinal research design that tracks youth development from childhood until adulthood could help address this limitation in future research.

In addition, the FDJJ PACT data were not collected specifically to conduct this analysis or this type of research. Rather, the data were collected in an effort to assess risks and needs of youth who came into contact with the juvenile justice system. As a result, certain measures were not effectively assessed to include in the model (such as biological factors, community- or neighborhood-level factors, other personality measures, and other behavioral variables). Relatedly, all juveniles included in the sample came into contact with the FDJJ during the study time frame. As a result, there were no “comparison” (non-delinquent) youth in the sample. Future research should consider incorporating a comparison sample when data are available.

Finally, although these current results do indicate significant and substantive findings, these relationships may actually simply reflect an underlying inherited or biological influence. For example, Moffitt (1993) suggested that certain neurobiological problems that are inherited by the individual may influence all subsequent stages of their development including temperament, mental illness, level of antisocial behavior, and propensity for violence (see also Grove et al., 1990; Sullivan, Neale, & Kendler, 2000). In addition, those who experience these neurobiological abnormalities may be more likely to be difficult children for their parents to handle and thus may be more likely to experience higher levels of childhood adversity. Because the PACT data have no method of measuring this potentially stable underlying biological source, we cannot say for certain that the genetic problems that Moffitt (1993) described cannot explain the associations between the variables included in the model.

Conclusion

This study’s findings touch on a number of important aspects of development, violence, and the juvenile justice system’s response to the more serious forms of delinquency. Prior research has suggested that SVC delinquency often begins during the phase of adolescence between ages 12 and 20, and as such, its onset may be influenced by a variety of developmental factors during childhood (Elliott, 1994). Accordingly, the best time to prevent a youth from becoming an SVC delinquent is likely during the formative years that precede this stage. A better understanding of development and the path toward SVC delinquency can only help practitioners identify youth at risk of engaging in serious violent and chronic offending later in life and meaningfully address the juvenile’s needs to prevent further progression toward it (Wasserman & Miller, 1998). By considering the predictive role of childhood adversity and other intervening developmental risk factors and processes, more effective assessments and interventions can be implemented to preclude the emergence or persistence of SVC delinquency.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Nicholas M. Perez, PhD, is an assistant professor in the School of Criminology, Criminal Justice, and Emergency Management in the College of Health and Human Services at California State University, Long Beach. His research interests are the development of delinquency, policing, and criminological theory.

Wesley G. Jennings, PhD, is an associate professor, associate chair, and undergraduate director in the Department of Criminology; has a courtesy appointment in the Department of Mental Health Law and Policy; and is a faculty affiliate of the Florida Mental Health Institute in the College of Behavioral and Community Sciences at the University of South Florida. In addition, he also has a courtesy appointment in the Department of Health Outcomes & Policy and is a faculty affiliate of the Institute for Child Health Policy in the College of Medicine at the University of Florida.

Michael T. Baglivio, PhD, is the director of Research and Program Development at G4S Youth Services, LLC in Tampa, Florida.

References

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Washington, DC: Author.

Anda

R. F.

Croft

J. B.

Felitti

V. J.

Nordenberg

Giles

W. H.

Williamson

D. F.

Giovino

G. A.

(1999). Adverse childhood experiences and smoking during adolescence and adulthood. Journal of the American Medical Association, 282, 1652-1658.

Anda

R. F.

Felitti

V. J.

Bremner

J. D.

Walker

J. D.

Whitfield

Perry

B. D.

. . . Giles

W. H.

(2006). The enduring effects of abuse and related adverse experiences in childhood: A convergence of evidence from neurobiology and epidemiology. European Archives of Psychiatry and Clinical Neuroscience, 256, 174-186.

Baglivio

M. T.

Epps

Swartz

Sayedul

Sheer

Hardt

N. S.

(2014). The prevalence of Adverse Childhood Experiences (ACE) in the lives of juvenile offenders. Journal of Juvenile Justice, 3, 1-23.

Baglivio

M. T.

Wolff

K. T.

Piquero

A. R.

Epps

(2015). The relationship between adverse childhood experiences (ACE) and juvenile offending trajectories in a juvenile offender sample. Journal of Criminal Justice, 43, 229-241.

Baird

Healy

Johnson

Bogie

Dankert

E. W.

Scharenbroch

(2013). A comparison of risk assessment instruments in juvenile justice. Madison, WI: National Council on Crime and Delinquency.

Burke

N. J.

Hellman

J. L.

Scott

B. G.

Weems

C. F.

Carrion

V. G.

(2011). The impact of adverse childhood experiences on an urban pediatric population. Child Abuse & Neglect, 35, 408-413.

Causadias

J. M.

(2013). A roadmap for the integration of culture into developmental psychopathology. Development and Psychopathology, 25, 1375-1398.

Chapman

D. P.

Dube

S. R.

Anda

R. F.

(2007). Adverse childhood events as risk factors for negative mental health outcomes. Psychiatric Annals, 37, 359-364.

10.

Cicchetti

(1984). The emergence of developmental psychopathology. Child Development, 55, 1-7.

11.

Cicchetti

Toth

S. L.

(1995). A developmental psychopathology perspective on child abuse and neglect. Journal of the American Academy of Child & Adolescent Psychiatry, 34, 541-565.

12.

Cicchetti

Toth

S. L.

(2005). Child maltreatment. Annual Review of Clinical Psychology, 1, 409-438.

13.

Cicchetti

Toth

S. L.

(2008). The past achievements and future promises of developmental psychopathology: The coming of age of a discipline. Journal of Child Psychology and Psychiatry, 50, 16-25.

14.

Clarke

G. N.

Hawkins

Murphy

Sheeber

L. B.

Lewinsohn

P. M.

Seeley

J. R.

(1995). Targeted prevention of unipolar depressive disorder in an at-risk sample of high school adolescents: A randomized trial of a group cognitive intervention. Journal of the American Academy of Child & Adolescent Psychiatry, 34, 312-321.

15.

Clarke

G. N.

Hornbrook

Lynch

Polen

Gale

Beardslee

Seeley

(2001). A randomized trial of a group cognitive intervention for preventing depression in adolescent offspring of depressed parents. Archives of General Psychiatry, 58, 1127-1134.

16.

Cohen

M. A.

Piquero

A. R.

Jennings

W. G.

(2010). Estimating the costs of bad outcomes for at-risk youth and the benefits of early childhood interventions to reduce them. Criminal Justice Policy Review, 21, 391-434.

17.

Copeland

W. E.

Keeler

Angold

Costello

E. J.

(2007). Traumatic events and posttraumatic stress in childhood. Archives of General Psychiatry, 64, 577-584.

18.

Dadds

M. R.

Spence

S. H.

Holland

D. E.

Barrett

P. M.

Laurens

K. R.

(1997). Prevention and early intervention for anxiety disorders: A controlled trial. Journal of Consulting and Clinical Psychology, 65, 627-635.

19.

Dong

Anda

R. F.

Felitti

V. J.

Williamson

D. F.

Dube

S. R.

Brown

D. W.

Giles

W. H.

(2005). Childhood residential mobility and multiple health risks during adolescence and adulthood: The hidden role of adverse childhood experiences. Archives of Pediatrics & Adolescent Medicine, 159, 1104-1110.

20.

Dube

S. R.

Anda

R. F.

Felitti

V. J.

Edwards

V. J.

Croft

J. B.

(2002). Adverse childhood experiences and personal alcohol abuse as an adult. Addictive Behaviors, 27, 713-725.

21.

Dube

S. R.

Felitti

V. J.

Dong

Chapman

D. P.

Giles

W. H.

Anda

R. F.

(2003). Childhood abuse, neglect, and household dysfunction and the risk of illicit drug use: The adverse childhood experiences study. Pediatrics, 111, 564-572.

22.

Elliott

D. S.

(1994). Serious violent offenders: Onset, developmental course, and termination—The American Society of Criminology 1993 presidential address. Criminology, 32, 1-21.

23.

Felitti

V. J.

Anda

R. F.

Nordenberg

Williamson

D. F.

Spitz

A. M.

Edwards

. . . Marks

J. S.

(1998). Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults. American Journal of Preventative Medicine, 14, 245-258.

24.

Fox

B. H.

Perez

N. M.

Cass

Baglivio

M. T.

Epps

(2015). Trauma changes everything: Examining the relationship between adverse childhood experiences and serious, violent and chronic juvenile offenders. Child Abuse & Neglect, 46, 163-173.

25.

Gorman

D. M.

(2014). Alcohol & drug abuse: The best of practices, the worst of practices: The making of science-based primary prevention programs. Psychiatric Services, 54, 1087-1089.

26.

Grossman

J. B.

Tierney

J. P.

(1998). Does mentoring work? An impact study of the Big Brothers Big Sisters program. Evaluation Review, 22, 403-426.

27.

Grove

W. M.

Eckert

E. D.

Heston

Bouchard

T. J.

Segal

Lykken

D. T.

(1990). Heritability of substance abuse and antisocial behavior: A study of monozygotic twins reared apart. Biological Psychiatry, 27, 1293-1304.

28.

Herman

D. B.

Susser

E. S.

Struening

E. L.

Link

B. L.

(1997). Adverse childhood experiences: Are they risk factors for adult homelessness? American Journal of Public Health, 87, 249-255.

29.

Herrera

Grossman

J. B.

Kauh

T. J.

Feldman

A. F.

McMaken

(2011). Making a difference in schools: The Big Brothers Big Sisters school-based mentoring impact study. Child Development, 82, 346-361.

30.

Hillis

S. D.

Anda

R. F.

Dube

S. R.

Felitti

V. J.

Marchbanks

P. A.

Marks

J. S.

(2004). The association between adverse childhood experiences and adolescent pregnancy, long-term psychosocial consequences, and fetal death. Pediatrics, 113, 320-327.

31.

Hillis

S. D.

Anda

R. F.

Felitti

V. J.

Marchbanks

P. A.

(2001). Adverse childhood experiences and sexual risk behaviors in women: A retrospective cohort study. Family Planning Perspectives, 33, 206-211.

32.

Jessor

(1987). Problem-behavior theory, psychosocial development, and adolescent problem drinking. British Journal of Addiction, 82, 331-342.

33.

Jessor

S. L.

(1977). Problem behavior and psychosocial development: A longitudinal study of youth. New York, NY: Academic Press.

34.

Lamphear

V. S.

(1985). The impact of maltreatment on children’s psychosocial adjustment: A review of the research. Child Abuse & Neglect, 9, 251-263.

35.

Manly

J. T.

Cicchetti

Barnett

(1994). The impact of maltreatment on child outcome: An exploration of dimensions within maltreatment. Developmental Psychopathology, 6, 121-143.

36.

Moffitt

T. E.

(1993). Adolescence-limited and life-course-persistent antisocial behavior: A developmental taxonomy. Psychological Review, 100, 674-701.

37.

Musci

R. J.

Bradshaw

C. P.

Maher

Uhl

G. R.

Kellam

S. G.

Ialongo

N. S.

(2014). Reducing aggression and impulsivity through school-based prevention programs: A gene by intervention interaction. Prevention Science, 15, 831-840.

38.

Muthén

(1984). A general structural equation model with dichotomous, ordered categorical, and continuous latent variable indicators. Psychometrika, 49, 115-132.

39.

Olds

D. L.

Eckenrode

Henderson

C. R.

Kitzman

Powers

Cole

Luckey

(1997). Long-term effects of home visitation on maternal life course and child abuse and neglect: Fifteen-year follow-up of a randomized trial. Journal of the American Medical Association, 278, 637-643.

40.

Olds

D. L.

Henderson

C. R.

Cole

Eckenrode

Kitzman

Luckey

Powers

(1998). Long-term effects of nurse home visitation on children’s criminal and antisocial behavior: 15-year follow-up of a randomized controlled trial. Journal of the American Medical Association, 280, 1238-1244.

41.

Oyserman

Terry

Bybee

(2002). A possible selves intervention to enhance school involvement. Journal of Adolescence, 25, 313-326.

42.

Perez

N. M.

Jennings

W. G.

Piquero

A. R.

Baglivio

M. T.

(2016). Adverse childhood experiences and suicide attempts: The mediating influence of personality development and problem behaviors. Journal of Youth and Adolescence, 45, 1527-1545.

43.

Piquero

A. R.

Farrington

D. P.

Welsh

B. C.

Tremblay

Jennings

W. G.

(2009). Effects of early family/parent training programs on antisocial behavior and delinquency. Journal of Experimental Criminology, 5, 83-120.

44.

Piquero

A. R.

Jennings

W. G.

Diamond

Farrington

D. P.

Tremblay

R. E.

Welsh

Reingle Gonzalez

J. M.

(2016). A meta-analysis update on the effects of early family/parent training programs on antisocial behavior and delinquency. Journal of Experimental Criminology, 12, 229-248.

45.

Piquero

A. R.

Jennings

W. G.

Farrington

(2010). Self-control interventions for children under age 10 for improving self-control and delinquency and problem behaviors: A systematic review. Campbell Systematic Reviews, 6, 1-117.

46.

Piquero

A. R.

Jennings

W. G.

Farrington

D. P.

Diamond

Reingle Gonzalez

J. M.

(2016). A meta-analysis update on the effectiveness of early self-control improvement programs to improve self-control and reduce delinquency. Journal of Experimental Criminology, 12, 249-264.

47.

Skrondal

Rabe-Hesketh

(2004). Generalized latent variable modeling: Multilevel, longitudinal, and structural equation models. New York, NY: Chapman & Hall.

48.

Sobel

M. E.

(1982). Asymptotic confidence intervals for indirect effects in structural equation models. In Leinhardt

(Ed.), Sociological methodology 1982 (pp. 290-312). Washington, DC: American Sociological Association.

49.

Sobel

M. E.

(1986). Some new results on indirect effects and their standard errors in covariance structure models. In Tuma

(Ed.), Sociological methodology 1986 (pp. 159-186). Washington, DC: American Sociological Association.

50.

Sroufe

L. A.

Rutter

(1984). The domain of developmental psychopathology. Child Development, 55, 17-29.

51.

Sullivan

P. F.

Neale

M. C.

Kendler

K. S.

(2000). Genetic epidemiology of major depression: Review and meta-analysis. American Journal of Psychiatry, 157, 1552-1562.

52.

Sussman

Sun

Rohrbach

L. A.

Spruijt-Metz

(2012). One-year outcomes of a drug abuse prevention program for older teens and emerging adults: Evaluating a motivational interviewing booster component. Health Psychology, 31, 476-485.

53.

Tierney

J. P.

Grossman

J. B.

Resch

N. L.

(1995). Making a difference: An impact study of big brothers/big sisters. Philadelphia, PA: Public/Private Ventures.

54.

Trickett

P. K.

McBride-Chang

(1995). The developmental impact of different forms of child abuse and neglect. Developmental Review, 15, 311-337.

55.

Wasserman

G. A.

Miller

L. S.

(1998). The prevention of serious and violent juvenile offending. In Loeber

Farrington

D. P.

(Eds.), Serious & violent juvenile offenders: Risk factors and successful interventions (pp. 197-247). Thousand Oaks, CA: Sage.

56.

Zigler

Hall

N. W.

(1989). Physical child abuse in America: Past, present, and future. In Cicchetti

Carlson

(Eds.), Child maltreatment: Theory and research on the causes and consequences of child abuse and neglect (pp. 38-75). New York, NY: Cambridge University Press.