J Child Fam Stud (2016) 25:2372–2386

DOI 10.1007/s10826-016-0414-9

ORIGINAL PAPER

An Exploratory Study of Neglect and Emotional Abuse

in Adolescents: Classifications of Caregiver Risk Factors
Cassandra Simmel1 • Darcey Merritt2 • Hillary Mi-Sung Kim3 • Soyoun Kim4

Published online: 5 April 2016

Ó Springer Science+Business Media New York 2016

Abstract Recent research has advanced exploration of regarding the duration of time in out-of-home placements
developmental stage and age-related distinctions in between Wave 1 and Wave 3 also emerged significant.
understanding the perpetration of child maltreatment.
Using longitudinal data from the National Survey of Child Keywords Latent class analysis
Emotional abuse

and Adolescent Well-being I (NSCAW-I), this exploratory Neglect
Adolescents
Child welfare
study investigated heterogeneity in caregiver risks in cases
of both neglect and emotional abuse of adolescents
(N = 511). Using MPlus software, a person-centered data Introduction
analytic strategy–latent class analysis was performed to
identify distinct classes of caregiver risk factors that were In recent years, it has been widely determined that both
associated with the reports of maltreatment among ado- child maltreatment referrals and entrants to child welfare
lescents. Subsequently exploring the distinct characteristics system supervision have been declining (DeVooght et al.
of the classes, we examined the related demographic 2014; Institute of Medicine (IOM) and National Research
characteristics, child welfare system variables (e.g., type of Council (NRC) 2014; U.S. Department of Health and
placement status; change in placement type between Human Services 2013). However, an important qualifica-
waves), and youth factors such as the presence of clinical tion to this period is that not all forms of maltreatment are
range problem behaviors (Child Behavior Checklist). indeed declining. Of note, during this seemingly optimistic
Results demonstrate four distinct classes of caregiver risk trend, reports of neglect have not changed substantially,
factors, with a nearly even split between those who have an and reports of emotional or psychological abuse have
absence of risk factors and those who have a whole host of actually increased (IOM and NRC 2014). Moreover,
risk determinants. Differences between the classes although research on the causes, risk factors, and ramifi-
cations of maltreatment (e.g. physical and sexual abuse)
has expanded, research on neglect, and especially emo-
& Cassandra Simmel tional abuse is under developed (Merritt and Snyder 2014;
csimmel@ssw.rutgers.edu Snyder and Merritt 2014). The need for more research on
1
child neglect and emotional abuse also coincides with the
School of Social Work, Rutgers University, 536 George
Street, New Brunswick, NJ 08901, USA
need for a developmentally specific research focus
2
regarding the impact of child maltreatment. Such inquiries
Silver School of Social Work, New York University, 1
Washington Square North, New York, NY 10003, USA
would bolster our understanding of distinct types of mal-
3
treatment based on the developmental capabilities and
Department of Pediatrics, Institute for the Study of Child
vulnerabilities of all age groups of victims (Cicchetti and
Development, Rutgers Robert Wood Johnson Medical
School, 89 French Street, New Brunswick, NJ 08901, USA Rogosch 2002; Jones Harden 2004; Simmel 2010).
4 Historically, research on maltreatment in adolescence
Department of Social Welfare, Ewha Womans University,
52, Ewhayeodae-gil, Seodaemun-gu, Seoul 03760, Republic has primarily focused on sexual abuse, since the incidence
of Korea of this abuse type is highest in this age group (U.S. DHHS

123
J Child Fam Stud (2016) 25:2372–2386 2373

2015). As an unintentional consequence, other abuse per- Recent research and summaries of administrative data
petrated in adolescence receives less focus in research. report the presence of all types of maltreatment as children
Moreover, the characterization of types of neglect (e.g. mature, including neglect and emotional abuse noting that
supervisory neglect) also suggests the victimization of there are multiple types of maltreatment perpetrated against
younger children and toddlers as more pressing in com- adolescents (Simmel 2011; United States Department of
parison to neglect experienced by young adolescents. For Health and Human Services 2013). For instance, Finkelhor
instance, when examining the definitions of neglect—par- et al. (2005) found in their large scale national survey of
tially comprising caregivers’ failure to appropriately children and adolescents that self-reported disclosures of
supervise their children; impairments in protecting children maltreatment by respondents included all forms of mal-
from harm; deficits in caring for children’s basic physical treatment across all age groups. Raissian et al. (2014) used
needs (Erickson and Egeland 2002)—these parental the National Child Abuse and Neglect Data System
behaviors are largely indicative of neglect of infants, tod- (NCANDS) to examine child maltreatment referrals for
dlers, and preschoolers. While almost all incidences of adolescents and younger children. In their study, they
neglect are inherently difficult to discern and recognize, examined how child welfare authorities responded to abuse
developmental theory informs us that some of the previ- allegations and whether age or other demographic charac-
ously described features of salient types of neglect seem teristics emerged as distinct. Similar to Finkelhor et al.
less pertinent to older children or adolescents, thus leaving (2005) study, their findings revealed the presence of all
a definitional vacuum when considering how types of types of abuse among adolescents, though there was vari-
neglect are manifested in adolescence. ation in how child welfare authorities responded to specific
Similar complexities arise with recognizing emotional maltreatment referrals for adolescents.
abuse. Such conduct is partially defined as ‘‘denying Given cursory accounts in the literature regarding
emotional responsiveness’’ and ‘‘hostile rejection/degrad- neglect and emotional abuse among adolescents specifi-
ing’’ (Hart, Brassard, Binggeli, and Davidson 2002). cally, Simmel’s 2011 study of NSCAW data supports the
Consider that one complexity of emotional maltreatment current literature and highlights distinctions according to
may be that it is embedded in the parent–child relation- developmental stage finding neglect and emotional abuse
ship (Hart et al. 2002) and not explicitly evident as abuse. to indeed be prevalent among older youth. In this study,
A long-term dysfunctional pattern of parental behavior latency-aged children and adolescents—both boys and
may be at play, leading into the children’s adolescent girls—had comparable rates of emotional abuse and all
years. With adolescents’ changing sensitivities around types of neglect relative to their younger counterparts.
notions of their burgeoning autonomy vis a vis their Moreover, recent administrative data from the Children’s
caregivers (Parke and Buriel 2008), this suggests emo- Bureau documents that nationwide, despite their relatively
tional abuse may have a differential impact and also older age, many young adolescents face initial involvement
associated with meaningfully differential caregiver char- with the child welfare system (US DHHS 2013). Although
acteristics. Developmentally, adolescents are generally relative to other types of maltreatment, sexual abuse is the
more cognitively and emotionally capable of relating and most frequent type of maltreatment reported for the ado-
interacting within their family systems at a more lescent age group nationwide, all forms of neglect, as well
advanced and interdependent level. They also exhibit as emotional abuse are indeed reported in this age group
growth in self-care skills and emotional regulation, while (U.S. DHHS, 2015). Thus adolescents continue to face
also possessing a keener awareness to parental behavior risks for all types of maltreatment, yet the extent to which
(Parke and Buriel 2008; Villodas et al. 2012). Despite this caregiver characteristics to these risks is not well under-
developmentally appropriate pause from parental reliance, stood. Furthermore, given the developmental uniqueness of
adolescents are still in need of predictable parenting and emotional abuse and neglect in older youth, a detailed
nurturing. As this relationship evolves in developmental descriptive account focused on these forms of abuse solely
context, it is likely that the vulnerabilities to adverse in this age group is needed.
parental conduct—as demonstrated in neglect and emo- Beyond exploring the incidence of specific maltreatment
tional abuse–also transform. This level of cognitive mat- types in adolescence, it is necessary to explore the impact
uration, in turn, may further affect adolescents’ ability to of abuse that adolescents experience. For example, in a
recognize parental conduct as abusive. Hence, new chal- recent study of the propensity toward intergenerational
lenges in the caregiving relationship may arise due to abuse, the history of child maltreatment in caregivers’
child developmental maturation, thus creating stress on backgrounds was found to contribute to subsequent per-
the parent–child relationship. However, this line of petration of abuse by these individuals as they parented
inquiry has not been examined with respect to emotional their own children (Thornberry and Henry 2013). Yet, this
abuse or neglect among this age group. subsequent perpetration was associated only with

123
2374 J Child Fam Stud (2016) 25:2372–2386

maltreatment that persisted into the adolescent stage of forms of child maltreatment (discrete categories of neglect
development and not maltreatment that was ‘‘childhood- and emotional abuse) among adolescents, with a particular
limited’’ and solely occurred prior to their children reach- focus on how the caregiving context is associated with
ing adolescence. Hence, the impact of having experienced reports of such abuse. This is an important step in
maltreatment as an adolescent has long-term effects on advancing prevention, detection, and ultimately protection
future parenting across developmental stages, insofar as efforts for adolescents, who may potentially be under-
this study demonstrates. served by child welfare systems and agencies (Raissian
Understanding the context of caregiving is important. In et al. 2014).
the recent Institute of Medicine and National Research This study is considered exploratory in nature and
Council (2014) report, the authors emphasize a broader and therefore guided by the following research aims: (1) mea-
more comprehensive framework for understanding multi- suring the extent to which adolescents in the NSCAW
ple aspects of the etiology of maltreatment. For instance, dataset are involved with the child welfare system due to
the report authors posit that due to social and economic emotional abuse and neglect as the primary maltreatment
inequality, individuals are ‘‘stratified’’ in society, resulting report type; (2) exploring how the caregiver risk factors,
in negative influences on their capacity for caregiving (p. both historical and current conditions are related to these
27). This framework provides a useful perspective in which maltreatment reports and cluster together; and (3) exam-
to examine the confluence of caregiver risk factors asso- ining differences between these classes in terms of ado-
ciated with the onset of maltreatment and they identify lescents’ subsequent child welfare involvement and
several categories of risk factors spanning micro, meso, behavioral outcomes.
and macro elements: (1) individual parent or caregiver risk
attributes (e.g., history of childhood abuse and neglect;
early childbearing; and parental psychopathology); (2) Method
individual child factors (e.g., developmental disabilities;
behavioral health difficulties); (3) family characteristics Participants
(e.g., family structure; deficient parenting skills; intimate
partner violence; and social isolation); and, (4) community This study used data from the National Survey of Child and
and environmental ‘‘contextual factors’’ (e.g., poverty; Adolescent Well-being (NSCAW I), a nationally repre-
unemployment; low socioeconomic status; neighborhood sentative longitudinal study that investigated the compre-
characteristics) (IOM and NRC 2014, pp. 24–26). hensive functioning of children and families involved with
The influence of these factors, as well as how they the child welfare system. NSCAW I uses a stratified cluster
conjointly operate on manifestations of distinct forms of sampling design resulting in a final cohort of 6228 children.
maltreatment in exclusive age groups of children and youth Further, this sample comprises two sub-samples aged birth
is an underexplored topic. Moreover, the IOM and NRC to 16 years: the Child Protection Service sample (CPS:
framework is largely represented in the NSCAW study (the 5,501 children who had been investigated by Child Pro-
basis for our current study), which includes comprehensive tection Service agencies for child abuse or neglect) and the
data on many of these caregiver risk conditions. To what Long-Term Foster Care sample (LTFC: 727 children who
extent are these risk factors apparent in caregivers of had stayed in out-of-home placement for approximately
adolescents? The limited research available does reveal 1 year at the time of sampling). Data were collected at
intriguing comparisons. Contextual factors such as poverty, multiple time points from children, caregivers, child wel-
unemployment, and inadequate housing have been clearly fare caseworkers, and teachers, inclusive of administrative
tied with the potential for abuse generally (Merritt 2009); records. The baseline interviews were conducted at
impairing reunification efforts (Fowler et al. 2013); neglect 2–6 months after the initial contact in 1999–2000 (Wave
in early childhood (Fallon et al. 2011; Moore et al. 2002); 1). Subsequently, data were collected at 12 months (Wave
and to a lesser measureable degree, abuse among adoles- 2), 18 months (Wave 3), and 36 months (Wave 4) after the
cents as well (Simmel 2011). Other risk factors that have initial investigation. Wave 5 data were collected at
been gleaned from recent studies include that adolescents 59–96 months after the close of investigation (Dowd et al.
may be vulnerable to maltreatment due to their own indi- 2006). Our study used the Wave 1 to 3 data from the CPS
vidual level risk factors (e.g., socio-emotional challenges, sample only.
behavioral health difficulties) perhaps in conjunction with The primary focus of our study is on the measurement of
impaired parental response to these factors (Simmel 2010). neglect and emotional abuse in adolescence and therefore
Taken together, our research builds on empirical work had the following inclusionary criteria: (1) children enter-
investigating developmental stage-specific indices of mal- ing the child welfare system due to neglect or emotional
treatment. Here, we explore two related but individual abuse; and (2) the age of children at Wave 1 was between

123
J Child Fam Stud (2016) 25:2372–2386 2375

11 and 15 years. Because we wanted to examine certain We measured OOH placement status according to the
system factors at Wave 3 (e.g., placement type), we capped following three metrics. The first metric defines the in-
the maximum age at Wave 1 to 15 years so the youth home or OOH care group at each wave (i.e., Wave 1 or
would not be 18 years or older as we followed-up at Wave Wave 3). The second metric classified the youth into four
3. This resulted in a final sample size of 511 youth groups regarding placement changes from Wave 1 to Wave
(weighted N = 268,594). The bulk of the sample consisted 3: (1) remained in-home, (2) moved from in-home to OOH,
of those who had experienced a combination of neglect (3) moved from OOH to in-home, and (4) remained in
types (e.g. failure to supervise and moral/legal/educational OOH care. Finally, we calculated the third metric by
neglect) (59.43 %) followed by those who experienced determining the proportion of OOH care days from Wave 2
physical neglect (24.52 %) This youth sample had an to Wave 3. We used this method because baseline Wave 1
average age of 12.6 years (linearized SE = .09) at Wave 1, data did not provide the length of previous time spent in
and was composed of 53 % girls and 47 % boys. Close to OOH care, and thus the proportions from Wave 1 to Wave
half of the sample was Caucasian (43 %), 30 % were 3 could not be generated. Further, the interval between
African American, 17 % were Hispanic, and 10 % were Wave 2 and Wave 3 varied largely across the sample, so
from other ethnicities (e.g., Asian, Hawaiian, Pacific the proportion of OOH care days were calculated with the
Islanders). The majority (77 %) of the children lived in interval as a denominator.
urban areas and the remaining 23 % in non-urban areas. In Table 3, we provide information on caregiver and
Detailed demographics of the youth and their primary youth risk factors. Caregiver and environmental risks. This
caregivers are presented in Table 1. information was gathered at Wave 1 by caseworkers at the
time of the case investigation, with each item recorded as
Measures presence versus no-presence on risk assessment instru-
ments. NSCAW utilized established risk assessment
In Table 2 we provide information on maltreatment and information yielded from the respective states’ child wel-
child welfare involvement. Types of maltreatment. The fare investigation techniques. We therefore had 14 risk
alleged primary maltreatment types were recorded by the assessment items for the caregivers who were associated
caseworkers (indicated as the most serious type of mal- with referrals to the state child welfare systems. The 14
treatment—notwithstanding co-occurrence among types) items for caregivers include measures of: previous reports
at Wave 1, providing the basis of the CPS reports. For the of child maltreatment (58.4 %), high stress in family
present study we solely focused on neglect and emotional (49.4 %), poor parenting skills (39.4 %), low social sup-
abuse. Four types of primary maltreatment were assessed: port (32 %), economic difficulties (26.8 %), mental health
(1) emotional abuse (14.1 %), (2) physical neglect (failure difficulties (18.7 %), recent history of arrest (13.3 %),
to provide) (24.5 %), (3) supervisory/other types of substance abuse (alcohol = 11.1 %; drug use = 9.93 %),
neglect (consisting of failure to supervise and moral/legal/ ongoing intimate partner violence (9.6 %), caregivers’ own
educational neglect) (59.4 %), and (4) abandonment childhood history of child abuse (9.6 %), cognitive
(1.9 %). While it is more than likely that these youth impairments (9.2 %), inappropriate parenting (7.8 %), and
endured other types of abuse as well, we opted to focus physical impairment (6.6 %).
on the primary allegation only, as a means for investi- Youths’ special needs. Presence of special needs was
gating collective parental risks apparent among these gathered on risk assessment tools at Wave 1 by case-
relatively understudied forms of maltreatment among workers and indicated as presence vs. no-presence. This
adolescents. risk assessment variable reflects the general category of
children’s developmental disabilities or special socio-
Out of Home Placement Status emotional needs.
Youths’ behavioral/emotional functioning. The Child
The caseworkers recorded whether the youth stayed at Behavior Checklist (CBCL: Achenbach 1991) was used to
home or were placed in out-of-home (OOH) care at each measure behavioral health of the youth and examine
wave (commonly referred to as ‘‘foster care’’). These OOH associations over time with the perpetration of emotional
placement types included non-relative foster care, kin care, abuse and neglect. In this study we used the subscales on
group homes, and other placements. However, we exam- the CBCL: externalizing, internalizing, as well as the
ined OOH care placements collectively and not by each cumulative total problem behavior scale. Caregiver report
placement type. The length of OOH placement through the of the CBCL was the source of the behavioral rating at the
youths’ life was reported as well. Unfortunately, we do not Wave 3. T-scores standardized by age and gender were
have data on the youths’ child welfare involvement prior to used, with higher scores indicating more problem behav-
Wave 1 of NSCAW. iors. The youth were also grouped into the clinical range

123
2376 J Child Fam Stud (2016) 25:2372–2386

Table 1 Types of Maltreatment

Variable Min. Max. Weighted % Weighted Linearized
and OOH status (unweighted
mean SE
n = 511, weighted
N = 268,594) Type of maltreatment
Emotional abuse 14.11
Physical neglect 24.52
Other forms of neglect 59.43
Abandonment 1.94
OOH status
At W1
In-home 77.98
OOH 22.02
Foster care 5.49
Kin care 4.94
Group home 2.29
Other OOH 9.30
At W3
In-home 83.41
OOH 16.59
Foster care 6.11
Kin care 4.93
Group home 3.08
Other OOH 2.47
From W1 to W3
Stay in-home 71.71
In-home to OOH 5.85
OOH to in-home 4.03
Stay OOH 9.22
Not ascertained 9.20
Proportion of OOH days bw 0.00 100.00 6.58 1.60
W2 and W3
Categories of Neglect include: physical neglect (failure to provide); supervisory (failure to supervise); other
(moral/legal/educational neglect); abandonment

and the non-clinical range (e.g. normal, borderline) group completely at random (MCAR) we used Little’s MCAR
for externalizing, internalizing, and total problem behav- test (Little 1988), which indicated that the data were
iors, respectively. The clinical range was defined as a missing completely at random, v2 (22) = 23.06, p = .40.
T-score of 64 or above on the respective scale. The CBCL Thus, latent class analysis (LCA) using caregiver/envi-
is norm-referenced for large populations, and therefore ronmental risk items were conducted with the entire sample
socioeconomic status and race have little effect on the (n = 511), utilizing the Full Information Maximum Like-
scores. lihood (FIML) method (Muthén and Muthén 2010).

Data Analyses Primary Analysis

Missing Value Analysis To test if there were discrete classes identified by different
constellations of caregiver risks, latent class analysis
Before conducting our primary analysis we examined the (LCA) was conducted using Mplus 5. Next, to explore
data for missing values, which were less than 10 % across distinguishing characteristics among the identified classes,
all variables, except for the four caregiver/environmental a series of bivariate analyses, using uncorrected v2 for
risk items (10 % for alcohol and drug abuse, respectively; categorical variables or adjusted Wald F test for continuous
and 26 % for one’s own history of abuse) and the CBCL variables with STATA 11, examined whether the identified
measures (13 %). To test whether data were missing classes were related to abuse type, OOH placement status,

123
J Child Fam Stud (2016) 25:2372–2386 2377

Table 2 Descriptives of
Variable Min. Max. Weighted % Weighted mean Linearized SE
demographics (unweighted
n = 511, weighted Characteristics of caregiver
N = 268,594)
Country of birth
U.S. 92.12
Non-U.S. 7.85
Not ascertained 0.03
Gender
Male 14.05
Female 84.02
Not ascertained 1.03
Age 19.00 79.00 38.44 .60
Education
[HS 30.26
=HS 34.99
\HS 31.68
Not ascertained 3.07
Romantic relation
Married 27.96
Separated 13.88
Divorced 14.39
Widowed 3.86
Never married 21.11
Live-in partner 15.27
Not ascertained 3.52
Relation to the child
Bio-parent 70.66
Other relatives 20.79
Non-relatives 8.55
Urbanicity
Urban 77.42
Non-urban 22.58
Characteristics of child
Race
White 43.33
Black 29.74
Hispanic 16.79
Other 9.64
Not ascertained 0.39
Gender
Boy 46.99
Girl 53.01
Age at W1 (years) 11.00 15.00 12.60 .09
11 23.33
12 23.82
13 27.53
14 20.37
15 4.96
Special needs
Yes 24.84
No 72.81
Not ascertained 3.35

123
2378 J Child Fam Stud (2016) 25:2372–2386

Table 2 continued
Variable Min. Max. Weighted % Weighted mean Linearized SE

Behavioral/emotional outcomea
Internalizing 31.00 86.00 55.60 .99
Clinical rangeb 25.28
No 61.60
Not ascertained 13.11
Externalizing 32.00 88.00 59.86 .92
Clinical rangeb 37.51
No 49.38
Not ascertained 13.11
Total problems 23.00 86.00 59.48 1.03
Clinical rangeb 36.88
No 50.00
Not ascertained 13.11
a
T scores of the Achenbach‘s Child Behavior Check List (CBCL)
b
T scores C 64

and characteristics of the youth and their primary care- social support along with history of abuse. We thus refer to
givers, as well the youth’s behavioral/emotional function- this group as ‘‘Relational Risk.’’ Finally, Class 4 consisted
ing. Subsequent post hoc analyses were conducted, using of 3.5 % of the caregivers, who manifested no salient risk
multinomial logistic regression or Wald F test. To address factors. We labeled Class 4 as the ‘‘No Risk’’ group. The
the complex sampling design attributes, Taylor Series lin- four-class model showed a clear distinction between clas-
earization methods were applied in all analyses. ses, resulting in only one class with a high probability of
membership in that class and the other classes having a low
probability. For example, primary caregivers in Class 1
Results (Historical Risk) had an average probability of .935 of
classifying in that group, and probabilities of grouping in
Results of the LCA resulted in four classes of caregivers’ Class 2 (Comprehensive Risk), Class 3 (Relational Risk),
risks, among those who were involved with the child and Class 4 (No Risk) was .065, .000, and .000,
welfare system due to the following types of maltreatment: respectively.
emotional abuse, physical neglect, supervisory/other The conditional probabilities endorsing each risk item
neglect, or abandonment. Several models were compared within each group contributed to the categorization of the
using a different number of classes in order to determine four classes. Those risk items with high probabilities (that
the overall fit of the models and arrive at substantive is, probabilities greater than .40) were considered highly
meanings of the classes (Muthén and Muthén 2009). Ulti- endorsed by individuals in that particular class.
mately, a four-class solution was determined most appro- In this analysis, we investigated whether the four pre-
priate, resulting in four subtypes of caregiver risks among viously identified latent classes of caregiver risks were
this population of maltreated youth (See Table 4). related to a host of child welfare case characteristics,
Profiles of the four classes are displayed in Fig. 1. The caregiver attributes, and child factors both at baseline and
majority of the primary caregivers belonged to either Class subsequently at Wave 3. These results are displayed in
1 (49.0 % of the caregivers) or Class 2 (42.6 %). Class 1 Table 5.
caregivers were more likely to have no other risk but
personal history of abuse and we therefore call this group Types of Abuse
‘‘Historical Risk’’. The group of Class 2 caregivers was
differentiated by the following indicators: recent history of There were no significant differences between the latent
arrest, serious mental health problems, inappropriate classes regarding maltreatment reports (emotional; physi-
parenting, low social support, and problems in paying cal neglect; supervisory/other neglect; abandonment). This
necessities, as well as history of abuse. As such, we refer to may be due in part to the co-occurrence of these types of
this group as ‘‘Comprehensive Risk.’’ maltreatment inherent in the NSCAW derived variable
Class 3, consisting of 4.9 % of the caregivers was pri- (‘‘most serious form of abuse’’ notwithstanding co-
marily characterized by the following two risk factors: low occurrence).

123
J Child Fam Stud (2016) 25:2372–2386 2379

Table 3 Caregiver/environmental risks (unweighted n = 511, Table 3 continued

weighted N = 268,594)
Variable Weighted %
Variable Weighted %
Not ascertained 6.46
Prior report of abuse Inappropriate parenting
Yes 58.44 Yes 7.84
No 37.35 No 87.66
Not ascertained 4.11 Not ascertained 4.50
High stress in family Physical impairment
Yes 49.37 Yes 6.62
No 47.89 No 88.48
Not ascertained 2.74 Not ascertained 4.89
Poor parenting skills
Yes 39.44
No 57.63
Changes in Placement Status
Not ascertained 2.94
Low social support
There were, however, significant differences regarding
Yes 32.04
OOH placement between the four classes at Wave 3
No 64.82
(v2 = 43.29, df = 3, p = .012), whereas such differences
Not ascertained 3.14
were not significant at Wave 1. Post-hoc comparisons
Trouble to pay necessities
indicated that at Wave 3, youth from the Comprehensive
Yes 26.82 Risk class were more likely to be placed in OOH care than
No 69.27 those from either the Historical Risk class (Post-hoc: log-
Not ascertained 3.91 odds = 1.75, p = .002, 95 % CI .69, 2.81, relative risk
Mental health problems ratio = 5.76) or No Risk classes (Post-hoc: log-
Yes 18.66 odds = 21.20, p = .000, 95 % CI 19.61, 22.78, relative
No 74.29 risk ratio = 1.61e?09); and youth from Historical Risk
Not ascertained 7.05 were more likely to be placed in OOH care than those from
Recent arrest No Risk (Post-hoc: log-odds = 19.45, p = .000, 95 % CI
Yes 13.31 17.67, 21.22, relative risk ratio = 2.79e?08). Further, the
No 77.10 proportion of days in OOH care between Wave 2 and 3 was
Not ascertained 9.59 different across the classes (Wald F = 6.67, df = 1, 80,
Active drinking p = .012). The children from the Comprehensive Risk class
Yes 11.14 were more likely to stay longer in OOH care than those
No 78.49 from all other classes (post hoc: t = 2.61, p = .011;
Not ascertained 10.37 t = 4.16, p = .000; and t = 2.36, p = .021, respectively).
Active drug use
Yes 9.93 Demographic Characteristics of the Caregiver
No 79.70
Not ascertained 10.37 The latent classes differed significantly by sociodemo-
Active domestic violence graphic characteristics of primary caregivers with regard to
Yes 9.85 gender and age. However, country of birth, levels of edu-
No 83.89 cation, marital status, and relation to the child were not
Not ascertained 6.26 related to class membership.
History of abuse Caregivers’ gender was different by the classes
Yes 9.55 (v2 = 55.98, df = 3, 81, p = .009). Compared to the
No 64.03 Comprehensive Risk class, caregivers in the No Risk and
Not ascertained 26.42 Relational Risk classes tended to consist of male caregivers
Cognitive impairment rather than female caregivers (post hoc: log-odds = 3.35,
Yes 9.22
p = .023, 95 % CI, relative risk ratio = 28.62: .47, 6.24;
No 84.32
log-odds = 1.65, p = .014, 95 % CI .35, 2.96, relative risk
ratio = 5.23, respectively).

123
2380 J Child Fam Stud (2016) 25:2372–2386

Table 4 Latent Class Analysis Fit Indices by Number of Classes

Fit index Number of classes
1 2 3 4 5

Log likelihood -4868.92 -4385.49 -4052.24 -3948.64 -3838.63

AIC 9793.82 8884.97 8276.47 8127.28 7965.26
Adj-BIC 9823.57 8945.52 8367.82 8249.43 8118.22
Entropy N/A .985 .892 .916 .895
Vuong-Lo-Mendel Rubin LRT N/A 960.70 (p = .68) 674.13 (p = .65) 114.54 (p = .69) 203.46 (p = .77)
AIC Akaike information criteria, Adj-BIC adjusted Bayesian information criteria, LRT likelihood ratio test

Fig. 1 Profiles of parents’ risks Historical 49.0% Comprehensive 42.6% Relaonal 4.9% No Risk 3.5%
for each of the latent classes. 1.00
Note: Class 1 = historical risk,
0.90
Class 2 = comprehensive risk,
CLASS 3 = relational risk, 0.80
Conditional Probability

Class 4 = no risk 0.70

0.60
0.50
0.40
0.30
0.20
0.10
0.00

Caregivers’ age was also significantly related to the 6.23, relative risk ratio = 53.35; log-odds = 3. 39,
classes (Wald F = 7.42, df = 1, 81, p = .008). As such, p = .001, 95 % CI 1.39, 5.39, relative risk ratio = 29.99,
No Risk caregivers were more likely to be younger than respectively). Alternatively, girls were more likely to be
those from the other classes (post hoc: t = -4.93, children of caregivers associated with this No Risk class.
p = .000; t = -3.19, p = .002; and t = -9.44, p = .000, Documentation of youths’ special needs at Wave 1 was
respectively). And, Comprehensive Risk caregivers were also associated with the classes (v2 = 78.46, df = 3, 81,
more likely to be older than caregivers in the Historical p = .000). As such, those with special needs were more
Risk class (post hoc: t = 2.60, p = .011). likely to come from the Relational and the Comprehensive
Risk classes, compared to the Historical Risk class (post
Characteristics of the Youth hoc: log-odds = 1.76, p = .000, 95 % CI 1.05, 2.47, rel-
ative risk ratio = 7.25; log-odds = 1.68, p = .001, 95 %
In the analysis of Wave 1 youth related factors that could CI 1.07, 4.29, relative risk ratio = 4.69, respectively).
be associated with the classes, we found only two that were Compared to the No Risk class, caregivers in the Relational
significant: gender and presence of special needs. Youths’ and the Comprehensive classes were also more likely to be
gender was associated with the classes (v2 = 17.65, related to raising youth with special needs (post hoc: log-
df = 3, 81, p = .023). Three of the classes (Historical, odds = 19.31, p = .000, 95 % CI 18.60, 20.02, relative
Relational, and Comprehensive Risk) were more likely to risk ratio = 2.44e?09; log-odds = 20.23, p = .000, 95 %
contain boys, compared to the No Risk class (post hoc: log- CI 18.62, 21.84, relative risk ratio = 2.44e?09, respec-
odds = 3.23, p = .004, 95 % CI 1.06, 5.37, relative risk tively). Finally, the No Risk caregivers relative to the
ratio = 24.93; log-odds = 3.98, p = .001, 95 % CI 1.72, Historical Risk caregivers were less likely to be associated

123
 Table 5 Relations between variables and the latent classes of primary caregivers’ risks
Variable Percentage (Mean)
Class1 Class2 Class3 Class4 v2 Wald F df p
Historical risk Comprehensive risk Relational risk No risk

Type of abuse 25.69 9 .483

Emotional abuse 12.9 15.4 27.1 0.0
Physical neglect 22.3 23.5 26.0 67.3
Other neglect 63.4 58.3 47.0 32.7
Abandonment 1.5 2.9 0.0 0.0
OOH care status
J Child Fam Stud (2016) 25:2372–2386

At W1 31.29 3 .064
OOH 6.4 22.1 28.0 0.0
In-home 93.6 77.9 72.0 100.0
At W3 43.29 3 .012
OOH 7.1 30.3 12.4 0.0
In-home 93.0 69.7 87.6 100.0
From W1 to W3 44.03 9 .103
Stay in home 87.1 66.2 69.2 100.0
OOH to in-home 5.0 4.0 19.1 0.0
In-home to OOH 3.8 6.9 0.0 0.0
Stay OOH 4.1 22.8 11.7 0.0
Proportion of OOH days between W2 and W3 6.67 1, 80 .012
(3.4) (11.5) (3.4) (0.0)
Characteristics of caregiver
Country of birth 22.84 3 .064
U.S. 88.7 96.8 99.8 70.6
Non-U.S. 11.3 3.2 0.2 29.4
Gender 55.98 3 .009
Male 13.7 6.7 27.4 67.3
Female 86.3 93.3 72.6 32.7
Age (years) 7.42 1, 81 .008
(37.4) (40.1) (38.1) (31.7)
Education 41.32 6 .096
[HS 31.3 33.9 30.3 0.0
=HS 32.0 32.9 62.8 96.7
\HS 36.8 41.4 7.0 3.3
Romantic relationship 63.51 15 .088
Married 22.1 29.8 55.5 67.3
2381

123
 Table 5 continued
2382

Variable Percentage (Mean)

123
Class1 Class2 Class3 Class4 v2 Wald F df p
Historical risk Comprehensive risk Relational risk No risk

Separated 14.9 16.3 3.4 0.0

Divorced 9.8 20.9 7.9 29.4
Widowed 2.7 6.3 1.5 0.0
Never married 28.6 16.4 10.7 3.3
Live-in partner 20.9 10.3 21.0 0.0
Urbanicity 33.62 3 .065
Urban 85.8 65.2 77.0 100.0
Non-urban 14.2 34.8 23.0 0.0
Relation to the child 50.67 6 .067
Bio-parent 75.8 70.1 49.1 32.7
Other relatives 20.7 15.2 35.5 67.3
Non-relatives 3.5 14.7 15.4 0.0
Characteristics of child
Race 65.53 81 .064
White 41.48 49.74 45.57 0.0
Black 27.9 33.99 34.68 3.27
Hispanic 21.5 7.93 7.08 67.33
Other 9.11 8.33 12.67 29.4
Gender 17.65 3, 81 .023
Male 46.0 50.0 64.3 3.3
Female 54.0 50.0 35.7 96.7
37.56 12, 81 .322
Age (years) (12.4) (12.8) (12.6) (12.3)
11 27.1 19.1 36.9 0.0
12 25.5 20.2 3.0 70.6
13 27.9 26.9 27.3 29.4
14 16.8 25.8 27.0 0.0
15 2.7 8.1 5.7 0.0
Special needs 78.46 3, 81 .000
Yes 12.2 44.5 66.8 0.0
No 87.8 55.5 33.2 100.0
Child behavioral/emotional outcomea
Internalizing problems 2.07 3 .615
(55.9) (55.3) (57.4) (46.1)
J Child Fam Stud (2016) 25:2372–2386
J Child Fam Stud (2016) 25:2372–2386 2383

.442

.452
with youths’ special needs (post hoc: log-odds = -31.55,

p
p = .000, 95 % CI -33.64, -29.46, relative risk
ratio = 1.83e-14).
The second longitudinal outcome we explored (in
addition to placement status across Wave 1 and Wave 3)

3
df

was the presence of behavior problems at this subsequent

time point. Although there were high rates of total
behavioral problems (36.9 %) overall (externaliz-
Wald F

ing = 37.5 % and internalizing = 25.3 %), these difficul-

ties were not significantly related to the classes.
3.73

3.57

Discussion
v2

This exploratory study builds on the current research lit-

erature in a number of ways. First, it extends on the recent
No risk
Class4

(42.7)

(39.9)
8.7

8.7

8.7
91.3

91.3

91.3

progress in examining specific categories of child mal-

treatment (Merritt and Snyder 2014; Simmel and Shpiegel
For categorical variables, uncorrected Chi square tests were used. For continuous variables, adjusted Wald F tests were used

2013; Snyder and Merritt 2014; Snyder and Merritt 2015),

especially neglect and emotional abuse. Given the hetero-
Relational risk

geneity in characteristics associated with types of neglect

among Western definitions of child rearing, understanding
Class3

(61.6)

(61.1)
25.0
75.0

52.9
47.1

51.8
48.2

each type of neglect and associated parental characteristics

is a complex task. Current research lacks consensus on
refined definitions of the meaning of specific types of child
maltreatment (e.g. emotional neglect), based on child and
Comprehensive risk

parental characteristics. There are many complexities to

consider when assessing what parental behaviors are
neglectful and for whom (e.g. child’s developmental stage).
Second, our study offers new directions for exploring risk
Class2

(60.3)

(60.1)
31.6
68.4

43.3
56.7

42.6
57.5

factors that culminate in the perpetration of maltreatment

T scores in the Achenbach‘s Child Behavior Check List (CBCL) were used

of adolescents (e.g., Trickett et al. 2011; Villodas et al.

Percentage (Mean)

2012). Third, this study advances the methodological

approaches utilized in recent research on child maltreat-
Historical risk

ment and child welfare (Fowler et al. 2013; Li and Godinet

2014) by using the person-centered method of Latent Class
Class1

(59.9)

(59.4)
28.1
71.9

43.1
56.7

42.4
57.6

Analysis (LCA) in a large, nationally representative sample

of adolescents involved with the child welfare system. It is
important to underscore the uniqueness of this study sam-
ple as a channel for exploring risks associated with neglect
and emotional abuse. By using a broad population involved
T scores in the CBCL scale of 64 or more

with child welfare systems, but not necessarily placed in

out of home care (OOH), we thus have ample heterogeneity
in the families’ presentation of risk conditions and out-
comes in children. Whereas some of our results are curious
and suited for speculative interpretation, our discussion of
Externalizing problems

the overall story and implications are framed around the

most salient and meaningful results.
Table 5 continued

Clinical rangeb

Clinical rangeb

Clinical rangeb
Total problems

The identification of four distinct classes of caregiver

Normative

Normative

Normative

risk (Historical Risk, No Risk; Relational Risk; Compre-

hensive Risk) among child neglect categories provides the
Variable

template for an assessment of covariates at the child, par-

ental, and situational level. Caregiver risk among parents in
b
a

123
2384 J Child Fam Stud (2016) 25:2372–2386

this sample is nearly evenly split between those who have a development suitable to address the co-occurrence of
personal history of abuse or neglect (‘Historical Risk,’ multiple types of neglect in a broad manner.
Class 1) and those who have a whole host of risk deter- Additionally, given risk factors are more salient once
minants (‘Comprehensive Risk,’ Class 4). Fewer parents families are subjected to the oversight of the child welfare
proportionally suffer from both a history of abuse and lack system (post baseline data collection), perhaps the risk
of social support or those with no risk factors than those determinant’s of parents is less so linked to initial child
with a comprehensive set of risk factors or only a history of welfare involvement (i.e. precipitating factors inciting an
abuse. This supports initiatives that intentionally maximize investigation and caseworker determination) than the
services with particular attention to the intergenerational trauma triggered when parents are placed under the
transmission component of learned behavior, as well as supervision of the child welfare system. Such that it fol-
cases wherein there are a multitude of co-occurring risk lows, caregiver risk factors play a larger role once children
factors. are placed in OOH care, which may play a role in children
This study supports the work of Thornberry and Henry being ultimately placed in care and the increased length of
(2013), in that parents who have suffered previous abuse time in placement between Waves 2 and 3. Policy and
have a propensity for maltreatment of their children into service enhancements would be best designed foremost in
the adolescent stage of development of their children. consideration of the placement outcomes, both based on
Despite the dearth of research demonstrating a salient link maltreatment type and resulting from involvement with the
between the history of abuse as a sole predictive factor child welfare system. Secondarily with respect to the
related to types of neglect, a cycle seems to be at play, caregiver traits most closely associated with the type of
worthy of future research and accompanying appropriate maltreatment in addition to parental report of a history of
service design and delivery. Above and beyond any expe- maltreatment.
riential or environmental risk factors manifesting in Moreover, given there were no significant associations
adulthood, the sole experience of having a history of with the specific types of maltreatment and caregiver risk
childhood abuse should be considered a risk to address in factors at baseline data collection, these results are most
preventative services. Services should begin with the useful for understanding how caregiver risk factors are
identification of prior exposure to child maltreatment fol- related to specific child/caregiver demographic character-
lowed by efforts to encourage positive parenting tech- istics. Consistent with previous research (Simmel 2010),
niques in an attempt to minimize the impact of negative our study finds types of child maltreatment are best
learned behavior. Moreover, parents with only a history of assessed through the identification of associations between
abuse and those with a comprehensive set of risks would specific child and caregiver demographic characteristics.
benefit from tailored services to address their individual Not surprisingly and perhaps as validation of cautious
and environmental challenges and encourage positive par- caseworker decision making, youth under the care of par-
enting behaviors. ents plagued by multiple maltreatment risk factors are
Our results did not provide insight as to how caregiver placed in out of home care and remain for longer periods of
risk factors manifest among the assessed maltreatment time (between the last two waves of data collection) under
types. This is an interesting finding that child neglect types the purview of child welfare oversight. Further, youth
(e.g. emotional abuse; physical neglect; supervisory/moral/ cared for by parents indicating a history of child abuse
legal/educational neglect; abandonment) do not signifi- were deemed more likely in need of OOH placements than
cantly differ across caregiver risk classes. This is quite parents without salient risk factors for maltreatment.
likely due to the co-occurrence of maltreatment types Whereas, we cannot speculate as to why these youth were
inherently captured in the NSCAW derived variable, thus flagged as a more pressing consideration for services, there
muddying potential associations. An additional potential is clearly a prevalence of maltreated youth in out of home
limitation here is that our sample size was relatively small care among caregivers who endured child maltreatment as
for detecting significant differences between the abuse well.
types. Future assessments regarding the correlates of Of note, more men classified by either the absence of
specific types of maltreatment would be beneficial pro- risk factors, or a history of maltreatment coupled with low
viding the sample size was ample to make reliable infer- social support, compared to women were more likely to
ences. Our sample did not allow for an inquiry specific to display multiple risk factors (Comprehensive Risk). These
the maltreatment types comprised in the ‘other forms of findings suggest at the very minimum, the need for services
neglect’ category. We entered into this research without to provide basic parenting support for men, while
specific expectations regarding how classes of caregiver addressing any indication of a history of maltreatment.
risk might associate differentially with maltreatment type, Additionally, women whom are considered primary to the
however, these findings suggest a need for service rearing of children are especially vulnerable to co-

123
J Child Fam Stud (2016) 25:2372–2386 2385

occurring parental challenges, indicating a need for com- approaches that focus attention on specific caregiver traits
prehensive services in an effort to prevent child neglect according to age related levels of risk, prevention efforts
based on their particular individual and environmental should be driven by early assessments of these risk factors
contexts. prior to children and adolescents being placed and spending
Whereas, one might assume younger parents would have an inordinate amount of time in OOH care.
a more difficult time with fulfilling a new parental role, it Our findings should be interpreted with caution due to
seems younger parents in this sample are less likely to certain inherent limitations in this secondary dataset. For
display noticeable caregiver risk factors. Similarly and instance, our analyses were limited in that our research
perhaps due in part to the trials that come along with life questions were confined to the parameters of the dataset. Of
experience, older parents are more likely to classify in the note is that the measures of caregiver risk factors, though
group enduring multiple challenges (Comprehensive Risk) culled from protocols used by state child welfare profes-
related to parenting. Older parents, who have endured a sionals, are only dichotomous indices of such risks and do
longer time in the more challenged strata of our society, not reflect the degree to which these difficulties manifest in
have likely experienced myriad parental challenges. Fur- families’ lives. Missing data on certain variables across
ther, by the sheer virtue of having experienced more life, waves may have hindered our sample size. Although we
one might be understandably more prone to a compre- were able to include a comprehensive set of covariates due
hensive set of risks. to the richness of the dataset, we cannot be certain that we
In addition to the gender differences among caregivers, have captured all the possible explanatory associations
our results indicate some striking youth gender differences with each of the classes. For instance, the out of home
as associated with caregiver risk factors. As such, the placement (OOH) variable is comprised of quite distinct
findings suggest that support for parents of boys may need caregiver settings, and thus doesn’t allow us to comment on
heightened attention based on the finding that boys are differences between these types of settings, such as foster
more likely cared for by parents classifying in the three care versus kinship care or group homes. Further, the
classes indicating a range of risk factors from solely a maltreatment type variable is limiting in that it assesses the
history of abuse to multiple challenges. Comparatively, most serious type of maltreatment in the context of likely
girls are significantly more likely to be cared for by those co-occurrence among types.
parents without any identified risk factors. In sum, this exploratory research indicates parents with
The challenges of parenting are even more heightened a history of childhood abuse and those with multiple
when caring for a youth with special needs (i.e., emo- combined risk factors are clearly vulnerable to poor par-
tional and behavioral health difficulties). Not surprisingly, enting practices. Although our results confirm reasonable
youth with special needs are less likely associated with expectations, our study further supports the notion that
parents without salient risk factors than those parents with intergenerational transmission of abusive behavior is a
a history of personal experiences of abuse and neglect. salient characteristic to address in efforts to prevent the
Similarly, special needs children are more likely cared for perpetuation of maltreatment (Pears and Capaldi 2001;
by parents with low social support and multiple risk Thornberry and Henry 2013; Thornberry et al. 2012).
factors, including a history of abuse and neglect than Further, addressing comprehensive risk factors, especially
parents with an absence of risk determinants. Good for female caregivers, is paramount in supporting families
practice would support flagging parents displaying mul- with adolescents at risk for maltreatment, as well as
tiple risks and also caring for youth with special needs in decreasing the time spent in OOH care. Ideally, com-
an effort to buffer the extraneous parenting challenges of prehensive services are needed to address the complexi-
caring for a special needs youth. ties of co-occurring risk factors at multiple levels (i.e.
As child behavior differs according to age group, so fol- endogenous, micro level, as well as environmental) with
lows parental behavior as a response to these developmental specific attention to understanding risk factors affecting
stages, and based on the parent–child relationship. In order to caregivers of adolescents. We would suggest future
address distinct forms of maltreatment, particularly neglect research continue to tease out ways in which parents can
and emotional abuse, one must understand how develop- be supported (based on their specific circumstances and
mental stages are associated with each of these forms of accompanying challenges) and youth (based on their
abuse. Because our study emphasized caregivers risk factors endogenous issues) can be protected from harm, particu-
associated with collective types of emotional abuse and larly nebulous forms of maltreatment, such as neglect and
neglect—as opposed to between these types of maltreat- emotional abuse. Our research findings reinforce the
ment–we were able to highlight demographic characteristics existing literature suggesting that the child welfare system
distinctly associated with certain classes of caregiver risk lacks a comprehensive prevention focus for families with
factors. In addition to a need for useful intervention adolescents.

123
2386 J Child Fam Stud (2016) 25:2372–2386

References Muthén, L. K., & Muthén, B. O. (2010). Mplus user’s guide (6th ed.).
Los Angeles, CA: Muthen & Muthen.
Cicchetti, D., & Rogosch, F. A. (2002). A developmental psy- Parke, R. D., & Buriel, R. (2008). Socialization in the family: Ethnic
chopathology perspective on adolescence. Journal of Consulting and ecological perspectives. In W. Damon & R. M. Lerner
and Clinical Psychology, 70(1), 6–20. (Eds.), Child and adolescent development: An advanced course
DeVooght, K., Fletcher, M., & Cooper, H. (2014). Federal, state, and (pp. 95–128). Hoboken, NJ: Wiley.
local spending to address child abuse and neglect in SFY 2012. Pears, K. C., & Capaldi, D. M. (2001). Intergenerational transmission
Washington, DC: Child Trends. of abuse: A two generational prospective study of an at-risk
Dowd, K. L., Kinsey, S., Wheeless, S., Thissen, R., Richardson, J., sample. Child Abuse and Neglect, 25(11), 1439–1461.
Suresh, R., et al. (2006). National Survey of Child and Raissian, K. M., Dierkhising, C. B., Geiger, J. M., & Schelbe, L.
Adolescent Well-Being: NSCAW, Combined waves 1-4 data file (2014). Child maltreatment reporting patterns and predictors of
user’s manual, restricted release version. Ithaca, NY: National substantiation: Comparing adolescents and younger children.
Data Archive on Child Abuse and Neglect (dataset archivists). Child Maltreatment, 19(1), 3–16.
Erickson, M. F., & Egeland, B. (2002). Child neglect. In J. B. Myers, Simmel, C. (2010). Why do adolescents become involved with the
L. Berliner, J. Briere, C. T. Hendrix, C. Jenny, & C. T. A. Reid child welfare system? Exploring risk factors that affect young
(Eds.), The APSAC handbook on child maltreatment (2nd ed., adolescents. Children and Youth Services Review, 32(12),
pp. 3–20). Thousand Oaks, CA: Sage Publications. 1831–1836.
Fallon, B., Trocmè, N., MacLaurin, B., Sinha, V., & Black, T. (2011). Simmel, C. (2011). Demographic profiles of children reported to the
Untangling risk of maltreatment from events of maltreatment: child welfare system. Journal of Public Child Welfare, 5(1),
An analysis of the 2008 Canadian incidence study of reported 87–110.
child abuse and neglect. International Journal of Mental Health Simmel, C., & Shpiegel, S. (2013). Describing the context and nature
and Addiction, 9, 460–479. of emotional maltreatment reports in children. Children and
Finkelhor, D., Ormrod, R., Turner, H., & Hamby, S. L. (2005). The Youth Services Review, 35(4), 626–633.
victimization of children and youth: A comprehensive, national Snyder, S., & Merritt, D. H. (2014). Do childhood experiences of
survey. Child Maltreatment, 10, 5–25. neglect affect delinquency among child welfare involved youth?
Fowler, P. J., Henry, D. B., Schoeny, M., Landsverk, J., Chavira, D., Children and Youth Services Review, 46, 64–71.
& Taylor, J. J. (2013). Inadequate housing among families under Snyder, S., & Merritt, D. H. (2015). The influence of supervisory
investigation for child abuse and neglect: Prevalence from a neglect on subtypes of emerging adult substance use after
national probability sample. American Journal of Community controlling for familial factors, relationship status, and individ-
Psychology, 52, 106–114. ual traits. Substance Abuse, 36(4), 507–514.
Hart, S. N., Brassard, M. R., Binggeli, N. J., & Davidson, H. A. Thornberry, T. P., & Henry, K. L. (2013). Intergenerational continuity
(2002). Psychological maltreatment. In J. B. Myers, L. Berliner, in maltreatment. Journal of Abnormal Child Psychology, 41,
J. Briere, C. T. Hendrix, C. Jenny, & T. A. Reid (Eds.), The 555–569.
APSAC handbook on child maltreatment (2nd ed., pp. 79–104). Thornberry, T., Knight, K. E., & Lovegrove, P. J. (2012). Does
Thousand Oaks, CA: Sage Publications. maltreatment beget maltreatment? A systematic review of the
Institute of Medicine and National Research Council. (2014). New intergenerational literature. Trauma Violence and Abuse, 13(3),
directions in child abuse and neglect research. Washington, DC: 135–152.
The National Academics Press. Trickett, P. K., Negriff, S., Ji, J., & Peckins, M. (2011). Child
Jones Harden, B. (2004). Safety and stability for foster children: A maltreatment and adolescent development. Journal of Research
developmental perspective. The Future of Children, 14(1), on Adolescence, 21(1), 3–20.
31–47. United States Department of Health and Human Services. (2013).
Li, F., & Godinet, M. T. (2014). The impact of repeated maltreatment Adoption Foster Care Analysis and Reporting System
on behavioral trajectories: From early childhood to early (AFCARS), Volume 20 (2012). Retrieved from: http://www.acf.
adolescence. Children and Youth Services Review, 36, 22–29. hhs.gov/programs/cb/resource/afcars-report-20
Merritt, D. H. (2009). Child abuse potential: Correlates with child U.S. Department of Health and Human Services, Administration for
maltreatment rates and structural measures of neighborhoods. Children and Families, Administration on Children, Youth and
Children and Youth Services Review, 31(8), 927–934. Families, Children’s Bureau. (2015). Child maltreatment 2013.
Merritt, D. H., & Snyder, S. M. (2014). Maltreatment type and Available from http://www.acf.hhs.gov/programs/cb/research-
behaviors: Does listening matter? Child Abuse and Neglect, data-technology/statistics-research/child-maltreatment.
38(12), 2062–2071. Villodas, M. T., Litrownik, A. J., Thompson, R., Roesch, S. C.,
Moore, K. A., Redd, Z., Burkhauser, M., Mbwana, K., & Collins, A. English, D. J., Dubowitz, H., et al. (2012). Changes in youth’s
(2002). Children in poverty: Trends, consequences, and policy experiences of child maltreatment across developmental periods
options. Child Trends Research Brief, #2009-11. Child Trends, in the LONGSCAN consortium. Psychology of Violence, 2(4),
Washington, DC. 325–338.
Muthén, L. K., & Muthén, B. O. (2009). Categorical latent variable
modeling using Mplus: Cross-sectional data. Retrieved from
http://www.statmodel.com/download/Topic%205.pdf

123