Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Emotion regulation as a mechanism linking parents’ marital aggression to adolescent behavioral problems: a longitudinal analysis
(USC Thesis Other)
Emotion regulation as a mechanism linking parents’ marital aggression to adolescent behavioral problems: a longitudinal analysis
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
EMOTION REGULATION AS A MECHANISM LINKING PARENTS’ MARITAL
AGGRESSION TO ADOLESCENT BEHAVIORAL PROBLEMS:
A LONGITUDINAL ANALYSIS
by
Estibaliz Maitena Iturralde
A Thesis Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF ARTS
(PSYCHOLOGY)
December 2009
Copyright 2009 Estibaliz Maitena Iturralde
ii
Table of Contents
List of Tables iii
List of Figures iv
Abstract v
Chapter 1: Background and Significance 1
Marital Aggression and Youth Outcomes 2
Emotion Regulation 4
Emotion Regulation as Mediator 7
The Present Study and Hypotheses 9
Chapter 2: Experimental Design and Methods 11
Participants 11
Procedures 13
Measures 14
Chapter 3: Results 18
Missing Data 18
Descriptive Statistics: Marital Aggression 20
Descriptive Statistics: Relations Between Constructs Over Time 22
Structural Equation Models 25
Chapter 4: Discussion 39
References 45
Appendices 51
Appendix A: Calculation of Indirect Effect Sizes in Final Model 51
Appendix B: Domestic Conflict Index 52
Appendix C: Conflict Tactics Scale-Revised 60
Appendix D: Emotion Regulation Checklist 61
Appendix E: Child Behavior Checklist 62
Appendix F: Youth Self Report 64
iii
List of Tables
Table 1: Characteristics of Families in Wave 1 (N=119) 13
Table 2: Descriptive Statistics 17
Table 3: Bivariate Correlations Among Marital Aggression Scores in the First 21
and Last Waves
Table 4: Bivariate Correlations Between Marital Aggression and Later Youth 23
Emotion Dysregulation and Behavioral Outcomes
iv
List of Figures
Figure 1: Model 1: Direct and Indirect Effects 28
Figure 2: Model 4: Indirect Effects 32
Figure 3: Model 4d: Final Model 35
v
Abstract
Youths’ exposure to marital aggression is a widespread problem. Many studies
point to adverse behavioral consequences for young children, but more examination is
needed of factors that might influence a relation between exposure to marital aggression
and adolescent behavioral outcomes. The present study focused on emotion regulation as
one possible mechanism. Data were analyzed from four waves of a prospective
longitudinal study on two-parent families with a 14- to 16-year-old in Wave 4 (N=83).
Questionnaires from parents and youth measured frequency of marital physical and
psychological aggression, and youth emotion regulation and behavioral outcomes.
Structural Equation Models revealed that emotion dysregulation provided an indirect link
between marital aggression and adolescents’ internalizing and externalizing problems.
Effects were observed cross-sectionally and longitudinally. A reciprocal relation was also
found between emotion dysregulation and behavioral outcomes such that the two had
independent and bidirectional influences on each other across time.
1
Chapter 1: Background and Significance
Unfortunately, the experience of witnessing aggression between parents is
common in childhood. A nationally representative sample of 1,615 families estimated
that 15.5 million American children live in households that have had physical aggression
between marital partners in the past year (McDonald, Jouriles, Ramisetty-Mikler, &
Caetano, 2006). This estimate equates to 29.4% of U.S. children living with cohabiting
parents. Youths’ exposure to this aggression is a widespread problem that affects
approximately 30% of those youth living in two-parent households (McDonald et al.,
2006). Psychological aggression is even more common, has been found to be related to
physical aggression, and is reported by children as highly upsetting (El-Sheikh,
Cummings, Kouros, Elmore-Staton, & Buckhalt, 2008; Margolin, Oliver, & Medina,
2001).
Much data exist to link aggression between parents to adverse outcomes in
children. Perhaps the two most examined outcomes are behavioral problems, particularly
externalizing and internalizing problems (Kitzmann, Gaylord, Holt, & Kenny, 2003;
Margolin & Gordis, 2000). However, mixed findings and modest effect sizes would
suggest that some exposed youth do not develop behavioral problems at all. Moreover,
there is insufficient information about how marital aggression affects adolescents, as
contrasted with younger children (Wolfe, Crooks, Lee, Mclntyre-Smith, & Jaffe, 2003).
Considering that adolescents are beginning to assume more adult responsibilities and
enter romantic relationships of their own, it is of interest to know how they may be
influenced by aggression between their most immediate adult role models, their parents.
The present study examined one possible mediator of this relation, emotion regulation,
2
which was conceptualized as a salient developmental task that is important to healthy
adolescent adjustment.
Marital Aggression and Youth Outcomes
More than 100 studies have linked negative outcomes in children to exposure to
marital aggression (Evans, Davies, & DiLillo, 2008; Kitzmann et al., 2003). Although the
present study focused on two of the most frequently studied problems (internalizing and
externalizing behavior), it should be noted that marital aggression has been linked to
many other deleterious outcomes as well. In addition to depression and conduct
problems, these include academic difficulties, impaired social competence, post-
traumatic symptoms, altered physiological reactivity, and greater utilization of health
services (Cummings & Davies, 1994; Gottman & Katz, 1989; Grych & Fincham, 1990;
Margolin & Gordis, 2000; Saltzman, Holden, & Holahan, 2005). These and many other
studies support the conclusion that exposure to aggression between parents is a
significant risk factor for youth that threatens adjustment in a variety of interpersonal and
intrapersonal domains.
Much of the literature has examined behavioral problems as primary outcomes of
exposure to marital aggression. Two of the most frequently investigated are externalizing
problems (i.e. aggression and rule-breaking) and internalizing problems (depression,
anxiety, and somatic symptoms). These are typically measured using the mother’s report
on the Child Behavior Checklist (Achenbach, 1991). One meta-analysis of 41 studies
(Wolfe et al., 2003) found an effect size of 0.28 for total externalizing and internalizing
behavior problems. Although genetic factors may help to explain part of this relation
3
between parents’ and children’s behavior, evidence supports the existence of
environmental influences as well. One study of 1,103 five-year-old twins found a direct
effect of marital aggression on internalizing and externalizing problems; 29.2% of the
variance was explained by environmental factors and specifically 2–8% by exposure to
marital aggression in particular (Jaffee, Moffitt, Caspi, Taylor, & Arseneault, 2002). A
moderate effect size of 0.28 (Wolfe et al., 2003) suggests that many exposed children do
not develop behavioral problems at all. In particular, the mixed findings on the
association between marital aggression and children’s aggressive behavior suggest a
complex picture (Margolin & Gordis, 2000).
One factor that may complicate this association is age. When comparing effects
across developmental periods, a conclusive pattern of age interactions has not always
emerged (Evans et al., 2008; Kitzmann et al., 2003; Wolfe et al., 2003), but it has been
suggested that a transition from childhood to adolescence reduces the impact of exposure
to marital aggression (Levendosky et al., 2002; Morris, Silk, Steinberg, Myers, &
Robinson, 2007; Schulz, Waldinger, Hauser, & Allen, 2005). It may be that adolescents,
due to their greater cognitive maturity, autonomy, and social involvements outside the
home, are more able to escape household conflict and evade some of its repercussions.
However, it is also possible that adolescents’ maturity may allow them to become
more involved in parents’ disputes, which could be to their detriment. The literature on
marital discord has found that adolescents are more likely than younger children to try to
intervene in their parents’ conflict (Davies, Myers, & Cummings, 1996), although it is
unclear if this pattern holds in more aggressive situations. Furthermore, marital
aggression may interfere with important developmental processes, whether in
4
adolescence or earlier in childhood, and this may in turn disrupt the mastery of tasks that
are fundamental to adolescent adjustment.
Until recently, there have been few studies of how adolescents react to aggression
between their parents. In their meta-analysis, Wolfe and colleagues (2003) found similar
effect sizes when comparing total problems across adolescent, school-aged, and
preschool-aged youth. However, of the 41 studies they included, only seven employed
specifically adolescent samples and their effect sizes ranged from 0.07 to 0.40. These
early studies were difficult to interpret because they were typically correlational and often
drew samples from populations with a high base rate for other problems, such as child
protection agencies, low-income housing projects, and psychiatric services (DuRant,
Cadenhead, Pendergrast, Slavens, & Linder, 1994; Kempton, Thomas, & Forehand,
1989; McGee, Wolfe, & Wilson, 1997; Muller, Goebel-Fabbri, Diamond, & Dinklage,
2000; Rogers & Holmbeck, 1997). Given the current state of the literature, more research
is warranted to examine the effects of marital aggression on the adolescent population
specifically. Two issues that remain to be understood are by what mechanisms marital
aggression puts youth at risk and how these evolve developmentally over time.
Emotion Regulation
The term emotion regulation (ER) has been used quite ambitiously (and often
inconsistently) to encompass a wide array of phenomena through which individuals
experience emotion. Much research has converged around a definition of ER as the
diverse set of processes by which we, both effortfully and automatically, monitor,
interpret, maintain, and adjust our emotions, often in response to the demands of the
5
social context (Eisenberg & Spinrad, 2004; Gross & Thompson, 2007; Zeman, Cassano,
Perry-Parrish, & Stegall, 2006). The slipperiness of this definition has introduced
numerous challenges of measurement. Competent emotion regulation is understood to
result in appropriate behaviors that help the individual adapt to environmental demands;
however, the nature of these responses is highly dependent on the situation (Campos,
Frankel, & Camras, 2004). As a result, measurement of ER often involves assessing how
well an individual appears to bounce back from affective challenges and achieve
objectives that are socially valued.
The development of emotion regulation is a critical task for children to master as
they grow up (Cole, Michel, & Teti, 1994). It may consist of many subtasks, including
impulse control, use of reflection, frustration tolerance, delay of gratification, and control
of excitement (Katz & Gottman, 1991). When children are small, they depend on the
involvement of others for help with ER, whether it is to soothe a baby’s cries or distract a
toddler from a desired but forbidden object. As children age, they develop more
experience and more cognitive resources that allow for greater intra-individual and
effortful forms of emotion regulation.
Researchers have become increasingly interested in the role of ER in adolescence.
As youth age, their maturation process allows them a greater facility with managing their
emotions (Zeman et al., 2006). Adolescent ER is often characterized by a more
sophisticated set of cognitive strategies with which difficult feelings are handled
(Underwood, 1997). Adolescents are able to reflect on emotions before acting; they can
imagine others’ perspectives and develop alternative explanations; they are better able to
plan ahead and consider the consequences of their responses (Garnefski, Legerstee,
6
Kraaij, van den Kommer, & Teerd, 2002). They may also develop a greater awareness of
emotion itself, learning for example that they can consciously regulate their affect, that
emotional responses are and sometimes should be used deceptively, and that emotions
can often be mixed and even contradictory (Harris, 1989). These regulatory processes
succeed in managing unpleasant affect, but they also serve a vital social function.
Adolescents depend less on their parents to aid them in ER and instead look to their peers
for help in regulating their emotions (Morris et al., 2007; Silk, Steinberg, & Morris,
2003).
Numerous studies have found associations between poor emotion regulation and
psychosocial maladjustment in both children and adults (Cicchetti, Ackerman, & Izard,
1995; Davies, Harold, Goeke-Morey, & Cummings, 2002; Eisenberg, Shepard, Fabes,
Murphy, & Guthrie, 1998; Silk et al., 2003; Walden, 1997). Emotion regulation is often
an implicit goal of clinical interventions (Cole et al., 1994). It seems to play a central role
in understanding adolescent maladjustment considering the many psychological disorders
for which adolescence is a typical period of onset (Costello et al., 2003). In particular,
Cluster B personality disorders are often characterized by difficulties with moderating
extreme affective states or generating more sophisticated emotional responses, such as
empathy (American Psychiatric Association, 2000).
Something that this maladjustment may have in common is a frequent mismatch
between the types of environmental demands made on the individual and the individual’s
range of emotional responses used to adapt to these demands. Consistent with a
developmental psychopathology framework, a pattern of failure moderating one’s
internal affective states in a way that meets environmental demands can precipitate
7
failures in social and functional domains (Cicchetti, Ganiban, & Barnett, 1991; Dodge &
Garber, 1991; Zeman et al., 2006). Because adolescents are expected to fulfill greater
responsibilities in their families, schools, and personal lives, there is pressure put on them
to use ever more sophisticated strategies of emotion regulation to navigate these
increasingly challenging social expectations.
Emotion Regulation as Mediator
Exposure to family violence is thought to interfere with critical tasks such as self-
regulation, which may then lead to failure on critical tasks in later developmental stages
(Cicchetti, Toth, Bush, & Gillespie, 1988). It is possible that exposure to marital
aggression disrupts the important developmental task of mastering emotion regulation,
and that this contributes to negative outcomes. Children exposed to marital aggression
appear to become emotionally dysregulated; they often show more sensitivity to signs of
impending conflict and are likely to overreact due to this hypervigilance (Katz &
Gottman, 1991). Davies and colleagues (2002) have found that greater reactivity is
related to problems of internalizing and externalizing behavior in children exposed to
marital aggression. In a study of adolescents, difficulty managing anger mediated in boys
the relation between exposure to marital aggression and verbal and aggressive treatment
of a romantic partner (Kinsfogel & Grych, 2004).
The role of emotion regulation as a mediator is consistent with Davies and
Cummings’ (1994) emotional security hypothesis, which holds that children exposed to
marital conflict feel less secure and engage in behaviors that allow them to feel more
control in a situation that is inherently out of their control. Some children may react by
8
withdrawing and avoiding conflict, while others may mimic their parents and engage in
more conflictual behavior themselves. Emotion regulation is one facet of emotional
security, as are components such as emotional reactivity and internal representations of
parent figures. Davies and Cummings (1998) found a mediating role for emotional
security in explaining an association between marital dysfunction and children’s
maladjustment.
As Morris and colleagues (2007) have described in a tripartite model of familial
influence, parents are able to affect youths’ emotion regulation through a number of
mechanisms. Consistent with Social Learning Theory (Bandura, 1977), parents act as
models to youth on how emotions should be properly regulated; they may also validate or
invalidate the experience of emotion by how they react to children’s emotional
experiences; and, they may create an emotionally secure or insecure climate. The third of
these mechanisms most resembles the emotional security hypothesis (Cummings &
Davies, 1994) and offers an entryway for marital aggression to make lasting impacts on
youths’ emotion regulation. Social Learning Theory has also been invoked as a possible
explanation for how children exposed to marital aggression go on to develop aggressive
behavior; however, in keeping with Morris et al.’s tripartite model, it may be that
observational learning actually affects how youth develop emotion regulation, which in
turn affects their psychosocial adjustment.
There has been little investigation of this association across time or
developmental stage. In a rare example, Schulz, Waldinger, Hauser, and Allen (2005)
found that emotion regulation mediated the effects of inter-parental hostility on
adolescents, as measured during conflictual family discussions; youth increased their ER
9
and decreased their own hostility over time. In the community violence literature, Kelly
and colleagues (2008) showed that, among children personally victimized by community
violence, emotion regulation mediated the relation between exposure and peer rejection
across measurements taken one year apart.
Studies that use a cross-sectional design to test mediation are limited in their
theoretical implications. By definition, mediation happens when the rise or fall of one
process brings about the rise or fall of another process via a third intervening construct;
this kind of mechanism necessitates a temporal ordering of variables, which is not
possible in a study that collects all measures at the same time point. As Cole and
Maxwell (2003) have illustrated, the processes that comprise a true mediation effect must
be measured at distinct time points. Furthermore, the absence of an apparent mediation
effect when examining variables cross-sectionally does not signify a lack of mediation. If
the causal mechanism is unfolding over an extended period of time, then a snapshot of
this process at a particular time point may not reveal mediation at all.
The Present Study and Hypotheses
The present study utilized four waves of longitudinal data on community families
to test a mediating role for poor emotion regulation in the association between parents’
marital aggression and youths’ behavioral difficulties. Beginning when youth were 9-10
years old, cohabiting parents and their child completed questionnaires on family violence
and children’s adjustment. Parents and youth reported on the frequency of physical and
psychological aggression between parents and the extent of internalizing and
externalizing behavior in the child. Parents also rated the degree to which their child
10
exhibited normative affective responses that were consistent with good emotion
regulation. The input of multiple reporters per construct was intended to reduce the
impact of reporter bias.
Structural equation modeling (SEM) was used to test the associations between
variables during Waves 3 and 4, when youth were aged 11-12 and 14-16, respectively,
while taking into account marital aggression dating back to Wave 1. One benefit of this
approach was that effects could be estimated while adjusting for initial levels of variables
and high correlations among outcomes, which otherwise can confound results and make
interpretation more challenging.
This design allowed mediation effects to be tested cross-sectionally, i.e. among variables
measured within the same wave, as well as longitudinally among variables measured at
different time points. The study’s hypotheses were as follows:
1. Marital aggression measured in Waves 1, 2, and 3 would be associated with an
increase in youths’ internalizing and externalizing behavior measured in Wave 4
(longitudinal direct effects model).
2. Emotion dysregulation would mediate this relation over time, meaning that earlier
marital aggression would be associated with later emotion dysregulation and that
inclusion of this mediator would account for most of the association previously
attributed to the direct effects (longitudinal mediation effects model).
3. A mediation effect would be significant when measured across multiple waves
and when tested cross-sectionally.
11
Chapter 2: Experimental Design and Methods
Participants
The present study was conducted using data from the University of Southern
California’s Family Studies Project. This ongoing prospective longitudinal investigation
contained four assessments of an ethnically diverse sample of two-parent families from
the Los Angeles area. The 83 participating families all had a child aged 9 or 10 at the
study’s inception in 2000–2001. The study utilized data collected during these four
assessments. The project used measures gathered from the mother, father, and youth in
each family.
The participating families consisted, in Wave 1, of a cohabiting mother and father
and one participating child aged 9 or 10. The two parents were either biologically related
to the child or had been living with him or her for at least the past three years. Legal
marriage was not required. All participants were asked to complete data collection
procedures in English.
Families were recruited throughout the Los Angeles area, including the ethnically
diverse neighborhoods near the USC campus. Through flyers placed at sites where
parents and children frequently congregate, such as local schools and community
organizations, as well as newspaper advertisements, families with a child aged 9 or 10
were encouraged to call the lab to participate in a study of how families cope with stress.
In each of Waves 1 through 4, families took part in a four-hour session of data
collection in the project’s lab at the USC campus. Compensation ranged from $100 in
Year 1 to $150 in Year 4. Most families agreed to return annually to the lab as needed
until the target child reached senior year of high school.
12
The child’s mean age at first assessment, i.e. “Wave 1,” was 10.0 years (SD =
0.6). In Wave 2, it was 11.1 years (SD = 0.7), in Wave 3, 12.4 years (SD = 0.7), and in
Wave 4, 15.1 years (SD=1.5). The mean amount of time that elapsed between a family’s
visit in Wave 1 and in Wave 2 was 1.1 years (SD = 0.2; 90% were seen within 12-16
months). Between Wave 2 and Wave 3 it was 1.3 years (SD = 0.28; 90% were seen
within 14-19 months), and between Wave 3 and Wave 4 it was 2.7 years (SD = 0.48;
90% were seen within 25-41 months).
In Wave 1, 119 families participated. In Wave 2, 104 families returned. In Wave
3, 104 families returned, although five of these had participated in Wave 1 but not in
Wave 2. In Wave 4, which was conducted after a greater time lapse, 83 families from
Wave 1 returned, all of whom had participated in Waves 1-3, except for one family that
had skipped Wave 2 and another that had skipped Wave 3. In cases where parents
separated or divorced during the course of the study, the family remained part of the
sample and every effort was made to continue evaluating all three family members even
if they participated in separate appointments. New parental figures (e.g. mother’s new
boyfriend) were also included in data collection as appropriate.
In Wave 1, the ethnic breakdown of participating children was 24% non-Hispanic
white, 26% multiethnic/other, 23% Hispanic/Latino, 22% Black/African-American, and
8% Asian/Pacific Islander. Note that these percentages do not add to 100% due to overlap
between categories. Compared to 11% in the Los Angeles population, this sample’s over-
representative 22% of Black/African-American children likely reflects recruitment efforts
near campus in the central area of Los Angeles. In contrast, the sample under-represents
Hispanic/Latino children (23% versus 47% in the Los Angeles population), which may
13
be due to this study’s administration of all measures in English. Demographic
characteristics of the families in Wave 1 are summarized in Table 1.
Table 1.
Characteristics of Families in Wave 1 (N=119)
________________________________________________________________________
Mothers Fathers Youths
Mean Age 38.6 41.1 9.9
Age Range 25.6–53.5 24.3–55.5 9.0-10.99
Female – – 42.9%
Male – – 57.1%
Mean Education 14.3 yrs. 14.1 yrs. –
Education Range 7–doctoral 8–doctoral –
Modal Education 12 yrs. (34.4%) 12 yrs. (28.6%) –
High School or Less 33% 39% –
American-born 77% 70% –
Ethnicity
Hispanic/Latino 29% 26.9% 23%
Asian/Pac. Islander 10.1% 8.4% 8%
African-American 24.4% 24.4% 22%
Caucasian 30% 31.9% 24%
Other/Mixed 5.9% 21.0% 26%
Parents
Combined Income Mean=$67,810; Range=$8,700-$165,000
Less than $49,000 34%
$50,000 - $99,000 42.9%
Above $100,000 21%
________________________________________________________________________
Procedures
In each wave, each family had its own four-hour appointment time and was
scheduled when both the parents and the youth could attend. Two experimenters worked
with each family, one primarily with the parents and one with the youth. Initially all three
family members met with both experimenters for an introduction to the project, a
description of the protocols, and completion of the consent procedures. Youth were
administered an assent form. It was explained that, in addition to their responses being
14
kept confidential, these answers would also not be shared with other members of the
family. If each family member agreed to participate, parents and youth then went to
separate rooms to complete measures. All measures described below were administered
in all four waves.
Measures
Marital Physical and Psychological Aggression. Marital aggression was defined
as acts of physical or psychological aggression between partners with the intent to harm.
To assess levels of marital aggression, parents completed the Domestic Conflict Index
(DCI; Margolin et al., 1990; see Appendix B for items). This questionnaire’s 56
aggression items included 18 items from the Conflict Tactics Scale (Straus, 1979) and
more recent additions from the Conflict Tactics Scale-Revised (Straus, Hamby, Boney-
McCoy, & Sugarman, 1996). The DCI prompts for incidents of physical abuse,
psychological aggression, and general conflict between spouses, e.g. “done or said
something to spite your spouse,” “screamed or yelled at your spouse,” and “choked or
strangled your spouse.” Respondents rated if the behavior had ever happened, and if so,
how frequently in the last year. Answer choices were “never,” “once,” “2-5 times,” “6-12
times,” “2-4 times per month,” and “more than once a week.” Parents reported on their
own and their spouse’s behavior. Test-retest reliabilities for the DCI fall into a range of
0.63–0.70 (Margolin John, & Foo, 1998). As previously described (Margolin et al.,
2009), in Wave 1, 52% of families reported father-to-mother or mother-to-father physical
aggression, which included acts such as “pushed, grabbed, shoved,” “slapped,” and “
kicked, bit, or hit with fist.”
15
Meanwhile, children reported on their parents’ physical and psychological
aggression in the past year by completing 12 items of the Conflict Tactics Scale-Revised
(CTS2; Straus et al., 1996; see Appendix C for items). For youth, the answer choices
were relatively collapsed as compared to the DCI, consisting of “none,” “once,” “a few
times,” and “a lot.”
Marital aggression scores were calculated per wave by comparing family
members’ answers on each item and summing the most severe responses across items.
Mean substitution was performed for individual items for which there was no response
from any of the three family members, provided that at least one family member had
completed the scale in the given wave. For the purposes of comparison, youths’ answers
of “a few times” and “a lot” on the CTS-2 were treated as equivalent to “2-5 times” and
“6-12 times,” respectively, on the DCI. This approach of taking the maximally severe
response per item was intended to reduce the effects of intra-family disagreement and
under-reporting typically found in studies of family violence (Augustyn, Frank, Posner,
& Zuckerman, 2002; McCabe, Lucchini, Hough, Yeh, & Hazen, 2005; O’Brien, John,
Margolin & Erel, 1994).
Emotion Regulation. Youths’ ability to regulate their emotions was assessed using
a modified 10-item version of the Emotion Regulation Checklist (ERC; Shields &
Cicchetti, 1995; see Appendix D for items), which was completed by each parent
reporting on the youth. The ERC presents 24 statements and uses a four-point Likert
scale (“never true,” “sometimes true,” “often true, and “almost always true”) to gauge
youths’ emotional responses. A principle components analysis (Shields & Cicchetti,
1997) has revealed two factors (correlated at r = -0.5), lability/negativity and emotion
16
regulation. The lability/negativity items are suggestive of dysregulated negative affect
and lack of flexibility, for example, “overreacts to minor frustration” and “goes to pieces
under stress.” The emotion regulation scale items represent appropriateness of affect and
empathy, for example, “I show recognition of others’ feelings,” and “I develop genuine
and close relationships.” The measure has been shown to display convergent validity with
the Emotion Regulation Q-Scale (Shields & Cicchetti, 1997).
The 10 items that were ultimately administered in this study were recommended
by the instrument’s authors as the most representative (Cicchetti, personal
communication to Margolin, 2000). An overall ERC sum score was calculated for each
parent after reverse-scoring the five negatively worded items. The overall scale showed
adequate internal consistency for both parents (Wave 1, Cronbach’s alpha = 0.72 and
0.75 for father and mothers). Scores were multiplied by -1 and used in analyses as
Emotion Dysregulation (rather than Regulation).
Internalizing and Externalizing Behavior. Two questionnaires from the
Achenbach System of Empirically Based Assessment (ASEBA) were used, the Child
Behavior Checklist (CBCL), which was completed by parents with regards to youth, and
the Youth Self-Report (YSR), which was completed by adolescents about themselves
(Achenbach, 1991; see Appendices E and F for items). The CBCL and the YSR list
similar sets of 118 difficulties (112 on the YSR), which respondents rate as 0 (“not true”),
1 (“somewhat or sometimes true”), or 2 (“very true or often true”). These items fall into
emotional, behavioral, academic, and somatic categories. Both questionnaires are widely
used and demonstrate high test-retest reliability and internal consistency (Achenbach,
1991). Both questionnaires yield two broadband factors that were used in the present
17
study, Internalizing Problems (depression, anxiety, and somatic complaints) and
Externalizing Problems (aggression and delinquency). T-scores were computed based on
age- and sex-adjusted norms provided by the questionnaires’ authors (Achenbach, 1991).
A T-score of 65 or more is considered to signify a clinically significant problem.
Descriptive statistics for these measures in select waves are displayed in Table 2.
Table 2.
Descriptive Statistics
Variable N Mean SD Min Max
Wave 1, 2, 3
Parents’ Marital Aggression 83 68.60 42.33 8.00 205.67
Wave 3
Youths’ Emotion Regulation 83 21.57 3.64 12.50 28.50
Youths’ Internalizing Behavior 83 50.49 7.97 35.00 72.92
Youths’ Externalizing Behavior 83 49.00 7.82 33.33 74.58
Wave 4
Parents’ Marital Aggression 83 61.45 38.75 1.00 184.00
Youths’ Emotion Regulation 83 20.93 4.03 12.50 29.00
Youths’ Internalizing Behavior 83 48.92 8.84 30.00 76.58
Youths’ Externalizing Behavior 83 49.08 8.86 29.67 80.92
Note. Marital Aggression scores were the sum of the maximally severe response on each
item comparing across family members (“max report”) on the Domestic Conflict Index
(parents) and the Conflict Tactics Scale-Revised (youth). Wave 1, 2, 3 Marital
Aggression was the mean of max reports from Waves 1-3. Emotion Regulation was the
mean of mothers’ and fathers’ sum scores on the Emotion Regulation Checklist, later
multiplied by -1 and analyzed as Emotion Dysregulation. Internalizing/Externalizing
Behavior were the means of Internalizing/Externalizing T scores for mothers’ and
fathers’ reports on the Child Behavior Checklist and youths’ reports on the Youth Self
Report.
18
Chapter 3: Results
In preparation for a longitudinal analysis, the data were examined in an effort to
maximize statistical power and make best use of the study’s strengths, namely, its use of
repeated measures and multiple reporters. Decisions were guided by a desire to include as
much adolescent data (i.e. data collected in Wave 4) as possible. First, it was important to
evaluate the extent of missing data and develop a solution that would prevent the loss of
incomplete cases. Second, descriptive statistics were computed to guide the construction
of a measurement model that would best capture longitudinal changes as well as the
perspectives of multiple family member reporters. Third, a series of structural equation
models were used to model relations between constructs over time while controlling for
cross-sectional relations among variables. In particular, emotion dysregulation was
modeled at two time points as a possible mediator of a relation between past exposure to
marital aggression and adolescents’ difficulties with internalizing and externalizing
behavior.
Missing Data
A total of 83 families met basic criteria for use in the analyses. Families were
enrolled at Wave 1 and at least one member had participated in Wave 4. Most families
among the 83 participated in all four waves, but one family skipped Wave 2 and one
skipped Wave 3. Analysis of variance (ANOVA) demonstrated that, of the 119 families
who participated in the study in Wave 1, the 36 who did not complete the study did not
differ from the rest on the following characteristics: marital aggression (max report),
youths’ emotion dysregulation, youths’ internalizing or externalizing behavior, total
19
spousal income, and highest year of mothers’ education (all measured in Wave 1). An
almost significant difference existed between groups on father’s highest year of education
(t(117) = 1.98, p = 0.05); fathers in non-completing families had on average one more
year of education.
Within each wave, some scores for some participants were missing, typically
because the family member was not available to complete the measure. Based on the 11
individual reporter scores multiplied by the 83 families per wave that were used in later
analyses, the small number of scores that were missing (3.6% in Wave 1; 3.9% in Wave
2; 4.2% in Wave 3; and, 8.1% in Wave 4) were substituted with the mean of that score
from the waves when it was available from that family member. This completed the
dataset for all but two cases. In one family the youth participated minimally in the study
due to his confinement in an in-patient facility; his missing Internalizing/Externalizing
Behavior scores were computed from the mean of his parents’ reports. In the other
family, the father participated only by mail and did not report on youths’ emotion
regulation in any wave; this score was substituted with the mother’s report.
In the case of three families, due to divorce or separation, parents or youth in
Wave 4 reported marital aggression for an additional partner besides the original mother
or father. In these cases, the reporter’s score was calculated by summing the maximally
severe responses per item comparing across the two partners. As described earlier, a
“max report” of marital aggression was calculated per wave by taking the maximum
response per reporter on each item of the aggression measure. In a few cases, when one
family member did not contribute that measure in that wave, the max report was
computed using the responses from the remaining family members. In computation of
20
individual sum scores, mean substitution was used to impute values for any missing
items.
Descriptive Statistics: Marital Aggression
Table 3 presents Pearson correlations among Wave 1 and Wave 4 marital
aggression variables. When taking the max report, marital aggression showed significant
stability from the first to last wave of the study (r = 0.54, p < 0.001). Mothers’ reports of
their own and their partners’ aggression were highly correlated with fathers’ reports in
both Wave 1 and Wave 4 (r = 0.55, 0.45; p < 0.001). The max report of mother-to-father
and father-to-mother aggression were highly intercorrelated in each wave (r = 0.74, 0.78;
p < 0.001), suggesting that marital conflict in these families was bidirectional among
partners.
Mother-reported (but not father-reported) marital aggression in Wave 4 was
related to youths’ reports in the same wave (r = 0.28; p < 0.05) but parents’ reports back
in Wave 1 did not correlate with youths’ in that wave. Developmental stage may play a
role in this discrepancy. Interestingly, youths’ reports in Wave 4 were related to mothers’
and fathers’ reports from back in Wave 1 (r = 0.23, 0.20; p < 0.05, 0.08), but not
correlated with their own reports in Wave 1. This indicates that there was a delay in
youths’ ability to report aggression between their parents in a way that was concordant
with their parents’ views. Children aged 9 or 10 may lack the necessary awareness or
may be more reluctant to disclose negative information about their parents than would an
adolescent.
21
Table 3.
Bivariate Correlations Among Marital Aggression Scores in the First and Last Waves
Variables 1 2 3 4 5 6 7 8 9 10 11 12
Marital Aggression (W1)
1. Total (M) –
2. Total (F) 0.55
***
–
3. Total (Y) 0.02 0.13 –
4. F-to-M (M, F, Y) 0.85
***
0.76
***
0.12 –
5. M-to-F (M, F, Y) 0.73
***
0.87
***
0.20
b
0.74
***
–
6. Total (M, F, Y) 0.84
***
0.88
***
0.17 0.92
***
0.94
***
–
Marital Aggression (W4)
7. Total (M) 0.63
***
0.35
**
<0.01 0.54
***
0.43
***
0.52
***
–
8. Total (F) 0.50
***
0.61
***
0.03 0.50
***
0.59
***
0.59
***
0.45
***
–
9. Total (Y) 0.23
*
0.20
b
0.07 0.24
*
0.22
a
0.24
*
0.28
*
0.08 –
10. F-to-M (M, F, Y) 0.51
***
0.44
***
-0.01 0.52
***
0.42
***
0.52
***
0.75
***
0.72
***
0.32
***
–
11. M-to-F (M, F, Y) 0.52
***
0.47
***
-0.03 0.47
***
0.50
***
0.49
***
0.75
***
0.76
***
0.29
**
0.78
***
–
12. Total (M, F, Y) 0.55
***
0.48
***
-0.02 0.50
***
0.52
***
0.54
***
0.80
***
0.79
***
0.33
**
0.94
***
0.95
***
–
*
p < 0.05,
**
p < 0.01,
***
p < 0.001,
a
p < 0.07,
b
p < 0.08.
Note. W1, W4 = Wave 1, Wave 4. F-to-M, M-to-F = Father-to-Mother, Mother-to-Father Aggression. Total is the sum of Father-
to-Mother and Mother-to-Father Aggression. Marital Aggression for: M, F, Y is the sum of maximally severe responses per item
per mother (M), father (F), and youth (Y); M or F alone is the sum of responses on the DCI for that parent; Y alone is the sum of
youths’ responses on the CTS-2.
22
The max report captured a significant amount of the variance in fathers’ and
mothers’ reports in both waves (r
2
= 62–77%), whereas it was uncorrelated with youths’
reports in Wave 1 and correlated with them at r = 0.33 (p < 0.01) in Wave 4. That
mothers and fathers contributed disproportionately to the max report is not surprising
since the youth measure consisted of a small subset of the questions given in the adult
measure and its four-point scale was truncated in comparison. Nevertheless, it appears
that youth showed more concordance with their parents on the issue of marital aggression
as they became older. These results also supported the use of a max report as a way to
include aggression that parents but not youth were under-reporting without allowing the
smaller scope of the youth measure to artificially suppress parents’ reports. It was also
apparent that combining mother-to-father and father-to-mother aggression as a unitary
construct was supported by the strong statistical association between the two.
Descriptive Statistics: Relations Between Constructs Over Time
Table 2 presents means and standard deviations for the eight main study variables
that were used in later structural equation analyses. Pearson correlations were also
calculated for Wave 1-4 marital aggression variables and for variables of emotion
dysregulation and internalizing/externalizing behavior from the latter two waves of the
study. These are shown in Tables 3 and 4. (To conserve space, some Wave 4 variables
are summarized as mean scores in Table 4.)
23
Table 4.
Bivariate Correlations Between Marital Aggression and Later Youth Emotion Dysregulation and Behavioral Outcomes
Variables 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Marital Aggression
1. Wave 1 (M, F, Y) –
2. Wave 2 (M, F, Y) 0.72
***
–
3. Wave 3 (M, F, Y) 0.67
***
0.76
***
–
4. Wave 4 (M, F, Y) 0.54
***
0.53
***
0.63
***
–
Emotion Dysregulation
5. Wave 3 (M) 0.01 0.11 0.09 0.07 –
6. Wave 3 (F) 0.13 0.20
a
0.11 0.22
*
0.44
***
–
7. Wave 4 (M, F) 0.05 0.09 -0.02 0.18 0.57
***
0.71
***
–
Internalizing Behavior
8. Wave 3 (M) 0.15 0.25
*
0.09 0.11 0.51
***
0.31
**
0.50
***
–
9. Wave 3 (F) 0.16 0.14 0.07 0.20
b
0.23
*
0.51
***
0.43
***
0.52
***
–
10. Wave 3 (Y) 0.01 -0.09 -0.05 -0.07 0.08 -0.13 0.15 0.13 0.11 –
11. Wave 4 (M, F, Y) 0.16 0.21
a
0.06 0.14 0.29
***
0.39
***
0.62
***
0.63
***
0.51
***
0.35
**
–
Externalizing Behavior
12. Wave 3 (M) 0.11 0.15 0.05 0.11 0.46
***
0.29
**
0.54
***
0.67
***
0.36
**
0.24
*
0.61
***
–
13. Wave 3 (F) 0.07 0.12 0.07 0.15 0.28
*
0.41
***
0.47
***
0.43
***
0.66
***
0.12 0.47
***
0.59
***
–
14. Wave 3 (Y) 0.09 <0.01 -0.02 <0.01 0.18 0.05 0.25
*
0.10 0.05 0.61
***
0.31
**
0.37
**
0.26
*
–
15. Wave 4 (M, F, Y) 0.12 0.13 0.07 0.20
a
0.30
**
0.33
**
0.59
***
0.47
***
0.37
***
0.41
***
0.77
***
0.68
***
0.60
***
0.58
***
–
*
p < 0.05,
**
p < 0.01,
***
p < 0.001,
a
p < 0.07,
b
p < 0.08.
Note. M, F, Y were the sum of maximally severe responses per item per mother (M), father (F), and youth (Y); M or F alone were the sum of
responses on the DCI for that parent; Y alone were the sum of youths’ responses on the CTS-2. Emotion Dysregulation scores denoted with: M, F
were the mean of mothers’ and fathers’ scores on the ERC with sign reversed; M or F alone were the sum score for that parent.
Internalizing/Externalizing Behavior scores denoted with: M, F, Y were the mean of parents’ (and youths’) CBCL (and YSR) T scores for those
constructs; M, F, or Y alone were the T score for that reporter.
24
Wave 3 and 4 Emotion Dysregulation were not related to Marital Aggression
within the same wave, but fathers’ Wave 3 report of Emotion Dysregulation was related
to Marital Aggression in Wave 4 and had a nearly significant relation to Marital
Aggression in Wave 2. These coefficients suggested a possible longitudinal relation
between these constructs, although the directionality of influence was unclear.
There was no correlation between Marital Aggression and Externalizing
Behavior, except for a nearly significant association between the two in Wave 4 (r = 0.20,
p < 0.06). Wave 2 Marital Aggression was associated with Wave 3 Internalizing
Behavior (mothers’ report) and nearly significantly with it in Wave 4 Internalizing
Behavior (mean report); Wave 4 Marital Aggression was correlated at borderline
significance with fathers’ report of Internalizing Behavior in Wave 3. This overall pattern
of coefficients suggests a more consistent relation between Marital Aggression and
youths’ Internalizing Behavior throughout childhood, with Externalizing Behavior
becoming more associated with Marital Aggression later in adolescence.
Mothers’ reports of their children’s Emotion Dysregulation were highly correlated
with fathers’, as indicated in Wave 3 (r = 0.44, p < 0.001), and each was highly
correlated with the mean of parents’ reports in Wave 4 (r = 0.57, 0.71; p < 0.001),
showing a high degree of stability. Emotion Dysregulation was related to parent report of
Internalizing and Externalizing Behavior in Waves 3 and 4, but not with youth self-
report, except for a significant correlation between youth-reported Wave 3 Externalizing
Behavior and mean Emotion Dysregulation scores in Wave 4. These results lent some
limited support to the validity of the Emotion Dysregulation measure in capturing
25
emotional disruption in youth that may underlie problematic behaviors observed by
parents and sometimes youth themselves.
Parents’ individual reports on their children’s behavioral problems were highly
intercorrelated for Internalizing and Externalizing Behavior, as shown in Wave 3 (r =
0.52, 0.59). In Wave 3, mothers’ and fathers’ Internalizing reports did not significantly
correlate with youths’ self-report; their Externalizing reports did correlate, however,
particularly mothers’ (r = 0.37, p < 0.01).
These analyses offered some indication that longitudinal relations among
variables might exist, but that these were inconsistent and difficult to interpret without
taking into account correlations between measures taken at different time points and
correlations between outcomes. In addition, the strong correlations between reporters on
measures of emotion dysregulation and behavioral outcomes justified the use of mean
scores in later analyses to capture the perspective of multiple family members while also
reducing reporter bias. The next set of analyses would build upon these findings by using
these multiply informed variables in models that also took into account the longitudinal
characteristics of the study and the intercorrelations of exogenous and endogenous
variables.
Structural Equation Models
Structural Equation Models (SEMs) use the covariance of variables to estimate
the strength of statistical relations within an a priori model of interest and based on
principles of multiple linear regression. Normality in the distribution of endogenous
variables is an assumption, the violation of which can introduce bias into parameter
26
estimates (Kline, 2005). According to visual inspection of normal probability plots and
the Shapiro-Wilks W test of normality, Emotion Dysregulation sums and Internalizing
Behavior T scores met this requirement. T scores of Externalizing Behavior had slight
positive skew in Waves 3 and 4.
A series of Structural Equation Models was used to test emotion dysregulation as
a mediator of the association between exposure to marital aggression and youths’
externalizing and internalizing behavior. Full mediation was understood to occur if the
following requirements were met: (a) a direct relation existed between the predictor and
outcomes if indirect paths via the mediator were forced to be zero, (b) a path that was
allowed to freely estimate from the predictor to the mediator was significant, and, (c)
inclusion of indirect paths between the predictor and outcomes rendered the direct paths
in a no longer significant (Baron & Kenny, 1986). Conventional fit guidelines were used
to judge model fit; good fit was considered to occur with a ratio of χ
2
to df of less than 3,
RMSEA (Root Mean Square Residual Error of Approximation) of 0.05 or less, and CFI
(Comparative Fit Index; Bentler 1990) of greater than 0.95. The chi-square difference test
(Kline, 2005) was used to determine the significance of individual path coefficients after
constraints were added or removed.
Although four waves of repeated measures were available, inconsistent time lags
between measurement periods discouraged the use of a four-wave model of the kind
described by Cole and Maxwell (2003). Specifically, the longer time lag between Waves
3 and 4 meant that it was less likely that cross-time path coefficients would be equivalent
across all waves, which would reduce the model’s power to be fit and estimated
successfully. Therefore, to make use of the adolescent data in Wave 4, a two-wave model
27
was specified that would allow for the examination of effects in two developmental
stages, pre-adolescent middle childhood and early adolescence, while also estimating
relations between constructs longitudinally. According to Cole and Maxwell (2003),
when a developmental process seems to involve ongoing influences among variables, it is
appropriate to estimate these effects using two time points.
1
Baseline Model (Model 1). To make use of the longitudinal nature of the data, a
model was fit using two variables for marital aggression as predictors: an accumulation
variable that was the mean of the Marital Aggression scores from Waves 1, 2, and 3
(represented as MA
123
); and, the Marital Aggression score in Wave 4 only (MA
4
).
Outcomes were Externalizing Behavior and Internalizing Behavior for each of Waves 3
and 4 (INT
3
, INT
4
, EXT
3
, and EXT
4
). Each of these four outcome variables was
calculated by taking the mean of parents’ CBCL scores and youths’ YSR scores for that
construct in that wave. Proposed mediators were Emotion Dysregulation in Waves 3 and
4 (EDR
3
and EDR
4
), which were computed by taking the mean of mothers’ and fathers’
ERC scores in each wave and reversing the sign. Although latent variables were not used
in this model, each manifest variable was estimated by using scores from multiple
sources to reduce reporter bias.
1
In a two-wave model, the relation between mediator and outcome from Time 1 to Time 2 can be used to
estimate what the relation would have been between Time 2 and Time 3, if stationarity is assumed.
Stationarity is a stable relation between variables based on cross-lagged paths over time. While not testing
stationarity directly, a two-wave design does allow for a direct test of equilibrium, which is the stability of
relations between variables on a cross-sectional basis. Equilibrium is not required to have stationarity, but
stationarity can often foster equilibrium. (Cole & Maxwell, 2003)
28
Figure 1.
Model 1: Direct and Indirect Effects
*
p < 0.05,
**
p < 0.01,
***
p < 0.001.
Note. MA
123
is the mean of Wave 1-3 “max reports” of marital aggression; MA
4
is the
max report from Wave 4 only. EDR
3
and EDR
4
are the means of parents’ emotion
regulation scores with sign reversed. INT
3
, EXT
3
, INT
4
, EXT
4
are the mean of family
members’ reports of youths’ internalizing and externalizing behaviors (T scores) for
Waves 3 and 4.
Model 1 (see Figure 1) was a mediation model in which direct and indirect paths
were specified between the Marital Aggression variables and the youth outcomes of
Internalizing and Externalizing Behavior. Specifically, six direct paths were specified
between Wave 1-3 Marital Aggression (MA
123
) and each outcome in Waves 3 and 4
(INT
3
, EXT
3
, INT
4
, and EXT
4
), as well as between Wave 4 Marital Aggression (MA
4
)
and Wave 4 outcomes (INT
4
, and EXT
4
). Mediation in Wave 3 was examined by
specifying a path between Wave 1-3 Marital Aggression (MA
123
) and Wave 3 Emotion
29
Dysregulation (EDR
3
), as well as between Wave 3 Emotion Dysregulation (EDR
3
) and
each of the Wave 3 outcomes (INT
3
and EXT
3
). Similar paths were also specified
between Wave 4 Marital Aggression (MA
4
), Emotion Dysregulation (EDR
4
), and
outcomes (INT
4
and EXT
4
).
To examine mediation longitudinally, paths were specified between Wave 1-3
Marital Aggression (MA
123
) and Wave 4 Emotion Dysregulation (EDR
4
), as well as
between Wave 3 Emotion Dysregulation (EDR
3
) and Wave 4 outcomes (INT
4
and
EXT
4
). Additional paths were also specified between each Wave 3 outcome (INT
3
and
EXT
3
) and Wave 4 Emotion Dysregulation (EDR
4
) to test the alternative explanation that
Emotion Dysregulation is predicted by Internalizing/Externalizing Behavior rather than
vice versa. This model also allowed investigation of a possible reciprocal relation
between Emotion Dysregulation and behavioral problems over time.
To compensate for the small sample size (N=83) and allow for enough power to
test model fit, a number of paths were forced to equal one another during the estimation
process based on theoretical considerations. Specifically, it was assumed that if two paths
had parallel purposes in two different waves, these should have coefficients of similar
magnitude. For example, the correlation between Internalizing and Externalizing
Behavior in Wave 3 was assumed to be roughly equal to the same correlation in Wave 4.
This and other equality constraints are represented in Figure 1 with the following: a
(MA
123
/MA
4
EDR
3
/EDR
4
), b (EDR
3
/EDR
4
INT
3
/INT
4
), c (EDR
3
/EDR
4
EXT
3
EXT
4
), d (MA
123
/MA
4
INT
3
/INT
4
), e (MA
123
/MA
4
EXT
3
/EXT
4
), and f
(INT
3
/INT
4
EXT
3
/EXT
4
). Later, the correctness of some equality constraints was
tested formally (as explained later). Model 1 met criteria for good fit (χ
2
= 11.57, 11 df,
30
RMSEA = 0.03, CFI = 1.00). Using this model as a baseline, other constraints were
imposed to test hypotheses and facilitate the choice of a model that offered the most
parsimonious explanation of the data.
Tests of Path Significance and Mediation. In Model 2, the two paths that would be
used to indicate a reciprocal relation over time between behavioral outcomes and
Emotion Dysregulation were constrained to zero (i.e. INT
3
EDR
4
and EXT
3
EDR
4
).
The resulting model had poor fit according to some fit indices (χ
2
= 26.67, 13 df,
RMSEA = 0.11, CFI = 0.96) and represented a significant degradation in fit compared to
Model 1 (Δχ
2
= 15.10, 2 df, p < 0.001). These paths were therefore maintained in the
model. The possible implications of these paths are discussed later in this section.
Next, a series of models were tested to examine the hypothesized mediating role
of Emotion Dysregulation. In Model 3a, indirect paths for Internalizing but not
Externalizing Behavior were constrained to zero. (These were EDR
3
INT
3
, EDR
3
INT
4
, and EDR
4
INT
4
.) In Model 3b, indirect paths for Externalizing but not
Internalizing Behavior were constrained to zero (EDR
3
EXT
3
, EDR
3
EXT
4
, and
EDR
4
EXT
4
). Both Model 3a and 3b exhibited poor fit (χ
2
= 51.31, 13 df, RMSEA =
0.19, CFI = 0.90; χ
2
= 40.45, 13 df, RMSEA = 0.16, CFI = 0.93) and resulted in
considerable worsening of fit compared to Model 1 (Δχ
2
= 39.74, 2 df, p < 0.001; Δχ
2
=
28.88, 2 df, p < 0.001). These tests supported the hypothesis that Emotion Dysregulation
was necessary to explain a link between Marital Aggression and
Internalizing/Externalizing Behavior. However, these results did not support a process of
mediation because the remaining direct effect paths did not provide a good fit to the data.
Emotion Dysregulation may be better described as explaining an indirect association
31
between Marital Aggression and behavioral outcomes. Tests described below would
further elucidate this indirect link.
It was still necessary to clarify if all of the indirect pathways were necessary to
explain the data. The reader is reminded that some indirect paths in Model 1 connected
variables within the same wave, as in MA
4
EDR
4
INT
4
, whereas others connected
variables across multiple waves, as in MA
123
EDR
3
INT
4
. Therefore, the next
models assessed whether one set of paths could be discarded without degrading model fit.
In Model 3c, indirect paths within but not across waves were constrained to zero (MA
123
EDR
3
, MA
4
EDR
4
, EDR
3
INT
3
/EXT
3
, and EDR
4
INT
4
/EXT
4
). The resulting
model had poor fit (χ
2
= 56.66, 13 df, RMSEA = 0.20, CFI = 0.89) and was significantly
worse fitting than Model 1 (Δχ
2
= 45.09, 2 df, p < 0.001). In Model 3d, indirect paths
across waves but not within waves were constrained to zero (MA
123
EDR
4
and EDR
3
INT
4
/EXT
4
). The resulting model had mediocre fit (χ
2
= 21.29, 14 df, RMSEA = 0.08,
CFI = 0.89) but had significantly worse fit than Model 1 (Δχ
2
= 9.72, 3 df, p < 0.05). As a
result, all of the across-wave indirect paths were carried over to subsequent models to
allow the coefficients to be evaluated for significance, as they seemed to contribute at
least marginally to model fit.
These last two sets of analyses suggested that Emotion Dysregulation had some
role in explaining an indirect effect of Marital Aggression on youths’ behavioral
outcomes, whether Externalizing Behavior or Internalizing Behavior, and whether within
the same time point or across multiple time points.
Model 4 tested the necessity of the direct effect paths. All six were constrained to
zero (MA
123
INT
3
/EXT
3
, MA
123
INT
4
/EXT
4
, and MA
4
INT
4
/EXT
4
). Given the
32
results of the mediation tests above, which suggested that direct effects alone were not
sufficient to explain the data, this model had predictably good fit (χ
2
= 15.54, 15 df,
RMSEA = 0.02, CFI = 1.00). Although the fit was not significantly different from that of
Figure 2.
Model 4: Indirect Effects
*
p < 0.05,
**
p < 0.01,
***
p < 0.001.
Note. MA
123
is the mean of Wave 1-3 “max reports” of marital aggression; MA
4
is the
max report from Wave 4. EDR
3
and EDR
4
are the means of parents’ emotion regulation
scores with sign reversed. INT
3
, EXT
3
, INT
4
, EXT
4
are the mean of family members’
reports of youths’ internalizing and externalizing behaviors (T scores) for Waves 3 and 4.
33
Model 1 (Δχ
2
= 3.97, 4 df, p = 0.41), Model 4 was kept because it offered a more
parsimonious explanation of the data and is shown in Figure 2.
Tests of Individual Path Coefficients. In both Models 1 and 4, Emotion
Dysregulation was associated with recent Marital Aggression in both Waves 3 and 4.
Although not strong in magnitude (β = 0.16 and 0.14, respectively), these paths were
significant at the 0.05 alpha level (t = 2.19, 2.14). These paths had been constrained to
equal each other, thereby not allowing for independent estimation of each parameter;
however, the good fit of the model suggested that the coefficients of these paths should
be considered similar in magnitude. To test this assumption, an alternative model was
specified in which these two paths were allowed to freely estimate (Model 4a). The
resulting model did not significantly improve fit (Δχ
2
= 0.09, 1 df, p = 0.76). As a result,
these more parsimonious equality constraints were maintained in the model. These results
suggest that recent marital aggression had a similarly negative effect on emotion
regulation whether in middle childhood or in adolescence.
Consistent with an indirect effects model, significant and moderately sized
associations were found between Emotion Dysregulation and each of the behavioral
outcomes in each wave. Despite the equality restrictions placed on these paths (EDR
3
INT
3
set to equal EDR
4
INT
4
, and EDR
3
EXT
3
set to equal EDR
4
EXT
4
), the
good fit of the model suggests that these constraints were justified. To test this
assumption, an alternative model was specified in which these two paths were allowed to
freely estimate (Model 4b). The resulting model did not significantly change fit (Δχ
2
=
0.87, 2 df, p = 0.65), which suggests that the associations between Emotion
34
Dysregulation and behavioral outcomes were roughly similar for these youth at ages 11-
12 as they were at ages 14-16.
The within-wave effect of Emotion Dysregulation on Internalizing Behavior was
slightly higher than its effect on Externalizing Behavior (β = 0.49 versus 0.43 in Wave 3
and β = 0.47 versus 0.41 in Wave 4). To test if these differences between Internalizing
and Externalizing Behavior were significant, another chi-square difference test was used
(Model 4c). All pairs of paths from Emotion Dysregulation to Internalizing and
Externalizing Behavior were forced to equal each other (i.e. EDR
3
INT
3
to equal
EDR
3
EXT
3
, EDR
4
INT
4
to equal EXT
4
, and EDR
3
INT
4
to equal EDR
3
EXT
4
). If the resulting model yielded a significantly worse fit, then this would suggest
that the differences between the Internalizing and Externalizing path coefficients were
significant. In fact, the opposite occurred (Δχ
2
= 1.93, 2 df, p = 0.38). This more
parsimonious model was therefore retained and suggests that the indirect effect of Marital
Aggression via Emotion Dysregulation was significant and similar in magnitude for both
Internalizing and Externalizing Behavior.
A similar procedure was used to test the significance of a difference found
between paths from Wave 4 Internalizing/Externalizing Behavior to Wave 4 Emotion
Dysregulation, which were consistent with a reciprocal association. In Model 4, a
significant relation was found for Externalizing Behavior (β = 0.19, t = 2.34 p < 0.05),
but not for Internalizing Behavior (β = 0.12, t = 1.45, p = 0.15). To test this difference,
the two paths were constrained to equal each other in an alternative model (Model 4d).
As seen in the previous test, fit was not worsened (Δχ
2
= 0.27, 1 df, p = 0.60), suggesting
that these relations were not significantly different from each other. These constraints
35
resulted in equal and significant path coefficients for both INT
3
EDR
4
and EXT
3
EDR
4
(β = 0.16, t = 3.66, p < 0.001). These results signify that, when holding constant
the previous wave’s Emotion Dysregulation scores, the previous wave’s Externalizing
and Internalizing Behavior each made a significant contribution to adolescents’ Emotion
Dysregulation at the next time point two-and-a-half years later.
Figure 3.
Model 4d: Final Model
*
p < 0.05,
**
p < 0.01,
***
p < 0.001.
Note. MA
123
is the mean of Wave 1-3 “max reports” of marital aggression; MA
4
is the
max report from Wave 4 only. EDR
3
and EDR
4
are the means of parents’ emotion
regulation scores with sign reversed. INT
3
, EXT
3
, INT
4
, EXT
4
are the mean of family
members’ reports of youths’ internalizing and externalizing behaviors (T scores) for
Waves 3 and 4.
36
These equality constraints were carried over into the final model, which is shown
in Figure 3 and had a good fit (χ
2
= 17.74, 18 df, RMSEA = 0, CFI = 1.00). The variance
explained in the final model for Waves 3 and 4, respectively, was 3% and 64% for
Emotion Dysregulation, 21% and 51% for Internalizing Behavior, and 21% and 60% for
Externalizing Behavior.
Power Analysis. The final model provided enough statistical power to ensure that
it did not fit poorly, which is an important consideration in light of the small sample size
in this study (N = 83). As shown in Figure 3, the final model contained multiple paths
that were constrained to equal each other. For example, the four paths marked b were
treated as a single parameter and estimated using 332 data points (rather than the
estimation of four parameters with 83 data points each). All together, these constraints
resulted in 17 parameters being estimated with an average of 117 data points each. Based
on procedures described by MacCallum, Browne, and Sugawara (1996), with N = 117, 18
degrees of freedom, and an alpha level of 0.05, the final model had 80% power to detect a
model misfit of RMSEA = 0.10 (compared to perfect fit, i.e. RMSEA = 0).
Magnitude of Indirect Effects. Path coefficients in the final model were inspected
to assess whether indirect effects occurred predominantly within wave or if they also
extended across time to multiple waves. As described earlier, when cross-sectional
indirect effect paths were constrained to zero, significant model misfit was the result; this
was true for longitudinal indirect effect paths, though to a less severe extent. To
understand these relations between variables more deeply, the magnitude of these path
coefficients was assessed and overall effect sizes were calculated.
37
According to the final model, the path from earlier Marital Aggression (MA
123
) to
later Emotion Dysregulation (EDR
4
) and the paths from earlier Emotion Dysregulation
(EDR
3
) to later behavioral outcomes (INT
4
/EXT
4
) all had negative coefficient values.
While these were the opposite sign of what one might expect, these needed to be
interpreted in light of the suppression of these paths by the inclusion of control variables
MA
123
and EDR
3
. Therefore, the overall longitudinal indirect effects were calculated by
using path tracing rules (Kline, 2005) and simultaneously taking into account all paths
that led from Wave 1-3 Marital Aggression to Wave 4 Internalizing/Externalizing
Behavior via Emotion Dysregulation in Wave 3 and/or 4. Seven paths were traced for
each outcome and summed to produce small indirect effects of 0.03 and 0.04 for
Internalizing and Externalizing Behavior (see calculations in Appendix A). These effects
signified that a one standard deviation unit increase of Wave 1-3 Marital Aggression
produced a 0.03 and 0.04 standard unit of increase in Internalizing and Externalizing
Behavior in Wave 4, as explained by indirect paths via Emotion Dysregulation. A similar
calculation using only cross-sectional indirect paths in Waves 3 and 4 yielded effect sizes
of 0.06–0.07, also small, but twice the size of the longitudinal indirect effects. Altogether,
cross-sectional and longitudinal indirect effects accounted for a 0.09–0.10 effect of
marital aggression on behavioral outcomes when youth were aged 14-16.
Summary. The SEM analysis did not find a direct association between marital
aggression and youths’ behavioral problems. Models in which direct paths between the
two were constrained to zero demonstrated the best fit. However, a link between the two
was demonstrated through an intervening construct, emotion dysregulation. The absence
of direct effects violated the conventional criteria for mediation (Baron & Kenny, 1986),
38
however, the pattern of coefficients estimated in the final model indicated an indirect
association between constructs. Marital aggression did predict youths’ internalizing and
externalizing behaviors though its association with emotion dysregulation.
Marital aggression reported by the family when youth were 9-12 was a significant
predictor of youths’ emotion dysregulation at age 11-12; in turn, emotion dysregulation at
age 11-12 was a significant predictor of internalizing and externalizing behavior at that
age. Associations of similar size were found among these variables when youth were
aged 14-16. While these effects were strongest cross-sectionally, they also existed
longitudinally. The best fitting model included paths that predicted later behavioral
problems with earlier marital aggression linked through emotion dysregulation, in
addition to the cross-sectional paths.
There was also indication of a reciprocal association between emotion
dysregulation and behavioral outcomes over time. Although this was not tested through
use of a nonrecursive model, the pattern of relations observed in the final model indicated
that both internalizing and externalizing behavior when youth were aged 11-12 was
predictive of emotion dysregulation when youth were 14-16. These effects were
significant even after taking into account emotion dysregulation at age 11-12 and the
other paths described above between emotion dysregulation and behavioral problems.
39
Chapter 4: Discussion
Contrary to expectations, direct effects between marital aggression and youths’
internalizing and externalizing behavior were weak and not statistically significant; thus
the data did not support the hypothesized mediation model. However, an indirect effects
model offered a good explanation for how exposure to marital aggression was related to
behavioral outcomes in this sample of youth. Emotion dysregulation played a significant
intervening role in connecting predictor with outcome. Youth exposed to higher
frequencies of marital physical and psychological aggression were more likely to show
difficulty regulating their emotional states. This, in turn, made them more prone to
internalizing behaviors (indicative of anxiety and depression) as well as externalizing
behavior (aggressive and delinquent acts). These effects were strongest when assessed
cross-sectionally, but were also perceptible longitudinally. The indirect effects were
roughly the same for youth aged 10-11 and 14-16, which suggests that the associations
between these variables are similar for youth of both age groups. Although not
hypothesized, a significant longitudinal relation was also found between Wave 3
behavioral outcomes and Wave 4 emotion dysregulation. Both externalizing and
internalizing behavior independently predicted emotion dysregulation at the later time
point.
This reciprocal relation between constructs—in which emotion dysregulation
worsens behavioral problems and behavioral problems then feed back to worsen emotion
dysregulation—may act as a self-maintaining process that amplifies the original risks
introduced by exposure to marital aggression. This reciprocal process may have special
importance for adolescents, a developmental stage when externalizing and internalizing
40
behavior may become increasingly serious in nature and may pose greater risks to
youths’ success academically and socially.
The results of this study highlight the impact of marital aggression on
adolescents’ development of problem behaviors. The longitudinal nature of the data,
spanning six years of youths’ lives, supports the conclusion that exposure to marital
aggression has lingering consequences into adolescence. These consequences may arise
due to disruption of salient developmental tasks such as emotion regulation, thus making
youth vulnerable to difficulties in a variety of interpersonal domains. The development of
internalizing or externalizing behaviors may decrease a youth’s ability to succeed
academically or socially. Failures in these domains may then reduce the number of
opportunities a youth has to develop and practice the self-regulatory strategies needed to
fit in with others that age. Thus a reciprocal feedback process may occur such that the
behavioral problems that arose from a lack of emotion regulation ultimately deprive the
youth of critical opportunities to catch up with peers, thus resulting in even more
dysregulation.
The definition of emotion regulation has lately received considerable scrutiny
(e.g. Eisenberg & Spinrad, 2004; Morris et al., 2007; Southam-Gerow & Kendall, 2002),
with much debate focused on which constructs of emotion do or do not belong under the
wide umbrella of ER. The results of this study are suggestive of further research
directions to be pursued. Although this study measured emotion regulation through a
widely used instrument in the literature, the measurement may not capture all of what is
thought to pertain to emotion regulation (for example, reactivity) and did not distinguish
between extrinsic and intrinsic processes of regulation. Future studies may be able to
41
strengthen findings by using an array of ER measures, including psychophysiology and
behavioral observation. Moreover, there may be elements of emotion regulation that are
particularly salient to the adolescent experience.
One area that could be assessed more directly is the construct of emotion
regulation strategies (Morris et al., 2007). Youths’ ability to call on effective strategies
may be indicative of good adjustment and it is possible that different kinds of
maladjustment are related to different kinds of ER strategies. Adolescents seem less
skilled than adults in utilizing emotion-related cognitive coping strategies (Garnefski et
al., 2002). Boys and girls seem to use different strategies as dictated by gender norms,
and youth with externalizing and internalizing problems both seem to use ineffective
means to manage anger (Zeman et al., 2006).
An additional way that emotion regulation should be examined with regards to
adolescence is by expanding the construct beyond intra-individual processes to
encompass the influences of family members and peers. The social world is important in
teaching children acceptable models for how to express emotion in specific contexts
(Cole et al., 1994). And, as proposed by Morris and colleagues (2007), relationships with
parents are likely to influence the success of youth in learning adaptive ER. Some
research has suggested that too much dependence on parents for ER may signal a risk for
depression in adolescents, while too little may introduce risk for conduct problems
(Allen, Hauser, Eickholt, & Bell, 1994). Exposure to marital aggression, in particular,
may model certain ineffective strategies that, if adopted by youth, introduce risk for
behavioral difficulties.
42
To understand how behavioral difficulties arise from exposure to marital
aggression, it is important to understand how such experiences may interfere with
ordinary developmental processes. The small sizes of the effects observed here related to
emotion dysregulation underline the likelihood that a multitude of factors influence the
relation between exposure to marital aggression and negative outcomes. These may
involve psychobiological, interpersonal, cognitive, and affective processes, among others.
Whenever possible, these factors should be examined within the context of
developmental change. As described by Cole and Maxwell (2003), cross-sectional
investigations are likely to not detect mediating mechanisms. When they are detected,
they are difficult to interpret in the absence of time spacing between measures.
There are several limitations to this study. Its use of a non-randomized
community sample limits the degree to which findings may be generalized to the wider
population. Still, this study stands in contrast to other adolescent studies of marital
aggression that relied on clinical samples (e.g. McGee et al., 1997; Muller et al., 2000).
The small size of this study’s sample (by SEM standards) did not allow for the analysis of
covariates that may have been interesting to consider. For example, gender may moderate
some of the associations discussed here, perhaps establishing direct effects or increasing
the influence of a mediator for one sex but not the other. It was also not possible to
incorporate the influence of pubertal stage in these analyses. Adolescence is a time of
rapid change with considerable variability among individuals in the time course of these
changes. More complex designs may be able to interpret the development of emotion
regulation while taking into account the psychobiological changes that accompany
puberty.
43
This study also does not capture the entirety of factors that may contribute to
adolescent risk and resilience in the face of exposure to marital aggression. Other factors
may include social support, educational and health resources, and socio-cognitive
variables such as attitudes towards aggression. Also, this study did not adjust effects for
the influence of other types of violence exposure such as physically harsh parenting or
community violence, which are known to co-occur at a high rate with marital aggression
and contribute to externalizing and internalizing difficulties (Margolin & Gordis, 2000;
Margolin et al., 2009). Others studies have suggested that the co-occurrence of multiple
types of violence amplifies the adverse effects of one alone (e.g. Finkelhor, Ormrod, &
Turner, 2007), and it is possible that emotion dysregulation mediates this process.
Adolescence is a developmental stage marked by two simultaneous and opposing
impulses (Schulz et al., 2005). On one hand, youth wish to achieve greater independence
from their parents and more affiliation with their peers; but at the same time, they look to
their parents for approval and for a model of how to behave in the world. The present
study illustrates how dysregulation of emotion can throw off this balance. A family
stressor such as marital aggression has the capacity to disrupt youths’ emotion regulation,
which may then place youth at greater risk for internalizing and externalizing difficulties.
Youth with behavioral problems are then less able to succeed in the social realm.
Emotion regulation is an important developmental task that not only helps youth manage
strong emotions, but also allows them to adapt within complex interpersonal situations.
Thus, an initial vulnerability introduced by a hostile home environment may have lasting
consequences for youths’ adjustment within a variety of domains. The feedback process
observed in the current study—that emotion dysregulation and behavioral problems
44
exhibit a reciprocal relation over time—underscores a danger therein. Once emotion
regulation becomes disrupted, a process may be set in motion that is self-maintaining and
can put youth at risk even after the original risk factor of inter-parental aggression has
been resolved.
45
References
Achenbach, T. M. (1991). Integrative guide for the 1991 CBCL/4-18, YSR, and TRF
profiles. Burlington, VT: University of Vermont.
Allen, J. P., Hauser, S. T., Eickholt, C., Bell, K. L., & O’Conno, T. G. (1994). Autonomy
and relatedness in family interactions as predictors of expressions of negative adolescent
affect. Journal of Research on Adolescence, 4, 535–552.
American Psychiatric Association. (2000). Diagnostic and statistical manual of mental
disorders (Revised 4th ed.). Washington, DC: Author.
Augustyn, M., Frank, D. A., Posner, M., & Zuckerman, B. (2002) Children who witness
violence, and parent report of children’s behavior. Archives of Pediatric and Adolescent
Medicine, 156, 800-803.
Bandura, A. (1977). Social Learning Theory. Englewood Cliffs, NJ: Prentice Hall.
Baron, R. M. & Kenny, D. A. (1986). The moderator-mediator variable distinction in
social psychological research: Conceptual, strategic, and statistical considerations.
Journal of Personality and Social Psychology, 51, 1173–1182.
Bentler, P. M. (1990). Comparative fit indexes in structural models. Psychological
Bulletin, 107, 238–246.
Campos, J. J., Frankel, C. B., & Camras, L. (2004). On the nature of emotion regulation.
Child Development, 75, 377-394.
Cicchetti, D., Ackerman, B. P., & Izard, C. E. (1995). Emotions and emotion regulation
in developmental psychopathology. Development and Psychopathology, 7, 1–10.
Cicchetti, D., Ganiban, J., & Barnett, D. (1991). Contributions from the study of high risk
populations to understanding the development of emotion regulation. In J. Garber & K.
Dodge (Eds.), The development of emotion regulation and dysregulation (pp. 15-49).
New York: Cambridge University Press.
Cicchetti, D., Toth, S., Bush, M. & Gillespie, J. (1988). Stage-salient issues: A
transactional model of intervention. New Directions for Child Development, 39, 123-145.
Cole, D. A., & Maxwell, S. E. (2003). Testing mediational models with longitudinal data:
Questions and tips in the use of structural equation modeling. Journal of Abnormal
Psychology, 112, 558–577.
46
Cole, P. M., Michel, M. K., & Teti, L. O. (1994). The development of emotion regulation
and dysregulation: A clinical perspective. Monographs of the Society for Research in
Child Development, 59, 73-100.
Costello, E.J., Mustillo, S., Erkanli, A., Keeler, G., Angold, A. (2003). Prevalence and
development of psychiatric disorders in childhood and adolescence. Archives of General
Psychiatry, 60, 837-844.
Cummings, E . M., & Davies, P.T . (1994). Children and marital conflict. New York:
Guilford.
Davies, P. T. & Cummings, E. M. (1994). Marital conflict and child adjustment: an
emotional security hypothesis. Psychological Bulletin, 116, 387-411.
Davies, P. T., & Cummings, E. M. (1998). Exploring children’s emotional security as a
mediator of the link between marital relations and child adjustment. Child Development,
69, 124–139.
Davies, P. T., Harold, G. T., Goeke-Morey, M. C. & Cummings, E. M. (2002). Child
emotional security and interparental conflict. Monographs of the Society for Research in
Child Development, 67, vii–viii.
Davies, P. T., Myers, R. L., & Cummings, E. M. (1996). Responses of children and
adolescents to marital conflict scenario as a function of the emotionality of conflict
endings. Merrill-Palmer Quarterly, 42, 1-21.
Dodge, K. A., & Garber, J. (1991). Emotion and social information processing. In J.
Garber & K. A. Dodge (Eds.), The development of emotion regulation and dysregulation
(pp. 159-181). Cambridge, England: Cambridge University Press.
DuRant, R. H., Cadenhead, C., Pendergrast, R. A., Slavens, G., Linder, C. W. (1994).
Factors associated with the use of violence among urban black adolescents. American
Journal of Public Health, 84, 612-617.
Eisenberg, N., Shepard, S. A., Fabes, R. A., Murphy, B. C, & Guthrie, I. K. (1998).
Shyness and children's emotionality, regulation, and coping: Contemporaneous,
longitudinal, and across context relations. Child Development, 69, 767-790.
Eisenberg, N., & Spinrad, T. L. (2004). Emotion-related regulation: Sharpening the
definition. Child Development, 75, 334-339.
El-Sheikh, M., Cummings, E. M., Kouros, C. D., Elmore-Staton, L. & Buckhalt, J.
(2008). Marital psychological and physical aggression and children's mental and physical
health: Direct, mediated, and moderated effects. Journal of Consulting and Clinical
Psychology, 76, 138-148.
47
Evans, S. E., Davies, C. & DiLillo, D. (2008). Exposure to domestic violence: A meta-
analysis of child and adolescent outcomes. Aggression and Violent Behavior, 13, 131-
140.
Finkelhor, D., Ormrod, R. K., & Turner, H. A. (2007). Poly-victimization: A neglected
component in child victimization. Child Abuse & Neglect, 31, 7-16.
Garnefski, N., Legerstee, J., Kraaij, V., van den Kommer, T., & Teerds, J. (2002).
Cognitive coping strategies and symptoms of depression and anxiety: A comparison
between adolescents and adults. Journal of Adolescence, 25, 603–611.
Gottman, J. M. & Katz, L F (1989). Effects of marital discord on young children’s peer
interaction and health. Developmental Psychology, 25, 373-381.
Gross, J. J. & Thompson, R. (2007). Emotion regulation: Conceptual foundations. In J. J.
Gross (Ed.), Handbook of emotion regulation (pp. 3-24). New York: Guilford Press
Grych, J. H., & Fincham, F. D. (1990). Marital conflict and children’s adjustment: A
cognitive-contextual frame-work. Psychological Bulletin,1 08, 267-290.
Harris, P. L. (1989). Children and Emotion: The Development of Psychological
Understanding. Cambridge, MA: Basil Blackwell; 243.
Hennessy, K. D., Rabideau, G. J., Cicchetti, D. & Cummings, E. M. (1994). Responses of
physically abused and nonabused children to different forms of interadult anger. Child
Development, 65, 815-828.
Jaffe, P., Wolfe, D., Wilson, S. & Zak, L. (1986). Similarities in behavioral and social
maladjustment among child victims and witnesses to family violence. American Journal
of Orthopsychiatry, 56, 142-146.
Jaffee, S. R., Moffitt, T. E., Caspi, A., Taylor, A. & Arseneault, L. (2002). Influence of
adult domestic violence on children's internalizing and externalizing problems: an
environmentally informative twin study. Journal of the American Academy of Child and
Adolescent Psychiatry, 41, 1095-1103.
Katz, L. F., & Gottman, J. M. (1991). Marital discord and child outcomes: A social
psychophysiological approach. In J. Garber & K. A. Dodge (Eds.), The development of
emotion regulation and dysregulation (pp. 129-158). New York: Cambridge University
Press.
Kelly, B. M., Schwartz, D., Hopmeyer Gorman, A., & Nakamoto, J. (2008). Violent
Victimization in the Community and Children’s Subsequent Peer Rejection: The
Mediating Role of Emotion Dysregulation. Journal of Abnormal Child Psychology, 36,
175-185.
48
Kempton, T., McCombs-Thomas, A., & Forehand, R. (1989). Dimensions of
interparental conflict and adolescent functioning. Journal of Family Violence, 4, 297–
307.
Kinsfogel, K. M. & Grych, J. H. (2004). Interparental conflict and adolescent dating
relationships: integrating cognitive, emotional, and peer influences. Journal of Family
Psychology, 18, 505-15.
Kitzmann, K. M., Gaylord, N. K., Holt, A. R. & Kenny, E. D. (2003). Child witnesses to
domestic violence: A meta-analytic review. Journal of Consulting and Clinical
Psychology, 71, 339-352.
Kline, R. B. (2005). Principles and practice of structural equation modeling (2nd ed.).
New York: Guilford.
Levendosky, A. A., Huth-Bocks, A. & Semel, M. A. (2002). Adolescent peer
relationships and mental health functioning in families with domestic violence. Journal of
Clinical Child and Adolescent Psychology, 31, 206-218.
MacCallum, R. C., Browne, M. W., and Sugawara, H. M. (1996). Power analysis and
determination of sample size for covariance structure modeling. Psychological Methods,
1, 130-149.
Margolin, G. & Gordis, E. B. (2000). The effects of family and community violence on
children. Annual Review of Psychology, 51, 445-479.
Margolin, G., Burman, B., John, R. S., & O’Brien, M. (1990). The Domestic Conflict
Index. Unpublished instrument, University of Southern California, Los Angeles.
Margolin, G., John, R. S., & Foo, L. (1998). Interactive and unique risk factors for
husbands’ emotional and physical abuse of their wives. Journal of Family Violence, 13,
315-344.
Margolin, G., Oliver, P., & Medina, A. (2001). Conceptual issues in understanding the
relation between interparental conflict and child adjustment: Integrating developmental
psychopathology and risk/resilience perspectives. In J. Grych & F. Fincham (Eds.), Child
development and interparental conflict (pp. 9-38). New York: Cambridge University
Press.
Margolin, G., Vickerman, K. A., Ramos, M. C., Duman Serrano, S., Gordis, E. B.,
Iturralde, E., Oliver, P. H., & Spies, L. A. (2009). Youth exposed to violence: Stability,
co-occurrence, and context. Clinical Child and Family Psychology Review, 12, 39-54.
49
McCabe, K. M., Lucchini, S. E., Hough, R. L., Yeh, M., & Hazen, A. (2005). The
relation between violence exposure and conduct problems among adolescents: A
prospective study. American Journal of Orthopsychiatry, 75, 575-584.
McGee, R. A., Wolfe, D. A., & Wilson, S. K. (1997). Multiple maltreatment experiences
and adolescent behavior problems: Adolescents’ perspectives. Development and
Psychopathology, 9, 131-149.
McDonald, R., Jouriles, E. N., Ramisetty-Mikler, S. & Caetano, R. (2006). Estimating the
number of American children living in partner-violent families. Journal of Family
Psychology, 20, 137-142.
Morris, A. S., Silk, J. S., Steinberg, L., Myers, S. S., & Robinson, L. R. (2007) The role
of the family context in the development of emotion regulation. Social Development, 16,
361-388.
Muller, R. T., Goebel-Fabbri, A. E., Diamond, T., & Dinklage, D. (2000). Social support
and the relationship between family and community violence exposure and
psychopathology among high risk adolescents. Child Abuse and Neglect, 24, 449-464.
O’Brien, M., John, R. S., Margolin, G. & Erel, O. (1994). Reliability and diagnostic
efficacy of parent's reports regarding children's exposure to marital aggression. Violence
& Victims, 9, 45-62.
Rogers, M. J., & Holmbeck, G. N. (1997). Effects of interparental aggression on
children’s adjustment: The moderating role of cognitive appraisal and coping. Journal of
Family Psychology, 11, 125–130.
Saltzman, K. M., Holden, G. W., & Holahan, C. J. (2005). The psychobiology of children
exposed to marital violence. Journal of Clinical Child and Adolescent Psychology, 34,
129-139.
Shields, A. M. & Cicchetti, D. (1995, March). The development of an emotion regulation
assessment battery: Reliability and validity among at-risk grade-school children. Poster
session presented at the biennial meeting of the Society for Research in Child
Development, Indianapolis, IN.
Shields, A. M. & Cicchetti, D. (1997). Emotion regulation among school-age children:
The development and validation of a new criterion Q-sort scale. Developmental
Psychology, 33, 906-916.
Shields, A., Ryan, R. M. & Cicchetti, D. (2001). Narrative representations of caregivers
and emotion dysregulation as predictors of maltreated children's rejection by peers.
Developmental Psychology, 37, 321-337.
50
Schulz, M. S.,Waldinger, R. J., Hauser, S. T., & Allen, J. P. (2005). Adolescents’
behavior in the presence of interparental hostility: Developmental and emotion regulatory
influences. Development and Psychopathology, 17, 489–507.
Silk, J. S., Steinberg, L., & Morris, A. S. (2003). Adolescents’ emotion regulation in
daily life: Links to depressive symptoms and problem behavior. Child Development, 74,
1869-1880.
Southam-Gerow, M. A., & Kendall, P. C. (2002). Emotion regulation and understanding
implications for child psychopathology and therapy. Clinical Psychology Review, 22,
189-222.
Straus, M. A. (1979). Measuring intrafamily conflict and violence: the Conflict Tactics
(CT) Scales. Journal of Marriage & Family, 41, 75-78.
Straus, M. A., Hamby, S. L., Boney-McCoy, S. & Sugarman., D. B. (1996). The Revised
Conflict Tactics Scales (CTS2). Journal of Family Issues, 17, 283-316.
Underwood, M. K. (1997) Top ten pressing questions about the development of emotion
regulation. Motivation and Emotion, 21, 127-146.
Walden, T. A., & Smith, M. C. (1997). Emotion regulation. Motivation and Emotion, 21,
7–25.
Wolfe, D. A., Crooks, C. V., Lee, V., Mclntyre-Smith, A. & Jaffe, P. G. (2003). The
effects of children’s exposure to domestic violence: A meta-analysis and critique.
Clinical Child and Family Psychology Review, 6, 171-187.
Zeman, J., Cassano, M., Perry-Parrish, C., Stegall, S. (2006). Emotion regulation in
children and adolescents. Developmental and Behavioral Pediatrics, 27, 155-168.
51
Appendix A
Calculation of Indirect Effect Sizes in Final Model
Longitudinal Indirect Effects (Internalizing Behavior)
MA
123
MA
4
EDR
4
INT
4
(0.63)(0.14)(0.44)
MA
123
EDR
4
INT
4
(-0.17)(0.44)
MA
123
EDR
3
INT
4
(0.16)(-0.18)
MA
123
EDR
3
EDR
4
INT
4
(0.16)(0.62)(0.44)
MA
123
EDR
3
INT
3
INT
4
(0.16)(0.45)(0.51)
MA
123
EDR
3
INT
3
EDR
4
INT
4
(0.16)(0.45)(0.16)(0.44)
MA
123
EDR
3
EXT
3
EDR
4
INT
4
+ (0.16)(0.46)(0.16)(0.44)
0.03
Longitudinal Indirect Effects (Externalizing Behavior)
MA
123
MA
4
EDR
4
EXT
4
(0.63)(0.14)(0.44)
MA
123
EDR
4
EXT
4
(-0.17)(0.44)
MA
123
EDR
3
EXT
4
(0.16)(-0.18)
MA
123
EDR
3
EDR
4
EXT
4
(0.16)(0.62)(0.44)
MA
123
EDR
3
EXT
3
EXT
4
(0.16)(0.46)(0.51)
MA
123
EDR
3
INT
3
EDR
4
EXT
4
(0.16)(0.45)(0.16)(0.44)
MA
123
EDR
3
EXT
3
EDR
4
EXT
4
+ (0.16)(0.46)(0.16)(0.44)
0.04
Cross-sectional Indirect Effects (Internalizing Behavior)
MA
123
EDR
3
INT
3
(0.16)(0.45) = 0.07
MA
4
EDR
4
INT
4
(0.14)(0.44) = 0.06
Cross-sectional Indirect Effects (Internalizing Behavior)
MA
123
EDR
3
EXT
3
(0.16)(0.46) = 0.07
MA
4
EDR
4
EXT
4
(0.14)(0.44) = 0.06
52
Appendix B
Domestic Conflict Index
No matter how well a couple gets along, there are times when they disagree on major decisions,
get annoyed about something the other person does, or just have spats or fights because they’re in
a bad mood, or tired, or for some other reason. People have many different ways of expressing
frustration, annoyance, or hostility with one another. Attached you will find a list of some things
that you and your partner may have done. You will find that some of these items apply, while
others do not. Please be sure to consider all items, even if they seem extreme.
First, decide if this behavior has ever occurred (Section A). If the behavior has never occurred in
the history of your relationship, circle “No” under “Ever” and go on to the next question.
If the behavior has occured, indicate whether or not it happened in front of your child (Section B).
Next, indicate how frequently this behavior occurred within the last year (regardless of child’s
presence) (Section C).
Section A Section
B
Section
C
From one year ago until today…
Have you: Ever?
In front of your
child?
0 per
year
1 per
year
2-5 per
year
6-12
per
year
2-4 per
month
>1 per
week
1. screamed or
yelled at your
spouse
No Yes No Maybe Yes
2. insulted or swore
at your spouse
No Yes No Maybe Yes
3. damaged a
household item, or
some part of your
home, out of anger
towards your
spouse
No Yes No Maybe Yes
4. withheld
affection from your
spouse
No Yes No Maybe Yes
5. deliberately
disposed of or hid
an important item
of your spouse’s
No Yes No Maybe Yes
6. sulked or refused
to talk about an
issue
No Yes No Maybe Yes
7. monitored your
spouse’s time and
made him or her
account for where
he/she was
No Yes No Maybe Yes
8. made plans that
left your spouse
feeling excluded
No Yes No Maybe Yes
53
Appendix B (continued)
9. left your spouse
and were unsure
whether you were
going to return
No Yes No Maybe Yes
10. been angry if
your spouse told
you that you were
using too much
alcohol or drugs
No Yes No Maybe Yes
11. been very upset
if dinner,
housework, or
home repair work
was not done when
you thought it
should be
No Yes No Maybe Yes
12. done or said
something to spite
your spouse
No Yes No Maybe Yes
13. been jealous
and suspicious of
your spouse’s
friends
No Yes No Maybe Yes
14. purposely hurt
your spouse’s pet
No Yes No Maybe Yes
15. purposely
damaged or
destroyed your
spouse’s clothes,
car, and/or other
personal
possessions
No Yes No Maybe Yes
16. insulted or
shamed your spouse
in front of others
No Yes No Maybe Yes
17. locked your
spouse out of the
house
No Yes No Maybe Yes
18. told your spouse
that he/she could
not work, go to
school, or go to
other self-
improvement
activities
No Yes No Maybe Yes
19. tried to prevent
your spouse from
seeing/talking to
family or friends
No Yes No Maybe Yes
20. had an
extramarital affair
No Yes No Maybe Yes
21. restricted your
spouse’s use of the
car or telephone
No Yes No Maybe Yes
54
Appendix B (continued)
22. made threats to
leave the
relationship
No Yes No Maybe Yes
23. blamed your
spouse for your
problems
No Yes No Maybe Yes
24. tried to turn
family, friends, or
children against
your spouse
No Yes No Maybe Yes
25. ordered your
spouse around
No Yes No Maybe Yes
26. been insensitive
to your spouse’s
feelings
No Yes No Maybe Yes
27. frightened your
spouse
No Yes No Maybe Yes
28. treated your
spouse like he/she
was stupid
No Yes No Maybe Yes
29. given your
spouse the silent
treatment/cold
shoulder
No Yes No Maybe Yes
30. criticized your
spouse
No Yes No Maybe Yes
31. called your
spouse names
No Yes No Maybe Yes
32. stomped out of
the room, house, or
yard
No Yes No Maybe Yes
33. stayed away
from the house
No Yes No Maybe Yes
34. ridiculed your
spouse
No Yes No Maybe Yes
35. physically
twisted your
spouse’s arm
No Yes No Maybe Yes
36. threatened to hit
your spouse or
throw something at
him/her in anger
No Yes No Maybe Yes
37. pushed,
grabbed, or shoved
your spouse
No Yes No Maybe Yes
38. slapped your
spouse
No Yes No Maybe Yes
39. physically
forced sex on your
spouse
No Yes No Maybe Yes
40. burned your
spouse
No Yes No Maybe Yes
41. shaken your
spouse
No Yes No Maybe Yes
55
Appendix B (continued)
42. thrown,
smashed, hit, or
kicked something
No Yes No Maybe Yes
43. prevented your
spouse from getting
medical care that
he/she needed
No Yes No Maybe Yes
44. thrown or tried
to throw your
spouse bodily
No Yes No Maybe Yes
45. thrown an
object at your
spouse
No Yes No Maybe Yes
46. choked or
strangled your
spouse
No Yes No Maybe Yes
47. kicked, bit or hit
your spouse with a
fist
No Yes No Maybe Yes
48. hit your spouse,
or tried to hit your
spouse, with
something
No Yes No Maybe Yes
49. beat up your
spouse (multiple
blows)
No Yes No Maybe Yes
50. threatened your
spouse with a knife
or gun
No Yes No Maybe Yes
51. used a knife or a
gun on your spouse
No Yes No Maybe Yes
52. used
humiliation to make
your spouse have
sex
No Yes No Maybe Yes
53. used threats to
make your spouse
have sex
No Yes No Maybe Yes
54. coerced your
spouse to engage in
sexual practices
he/she did not want
No Yes No Maybe Yes
55. slammed your
spouse against the
wall
No Yes No Maybe Yes
56. physically
prevented your
spouse from leaving
an argument or
blocked his/her exit
No Yes No Maybe Yes
57. showed your
spouse that you
cared even though
the two of you
disagreed
No Yes No Maybe Yes
56
Appendix B (continued)
58. showed respect
for your partner’s
feelings about an
issue
No Yes No Maybe Yes
59. suggested a
compromise to a
disagreement
No Yes No Maybe Yes
60. agreed to a
solution your
partner suggested
No Yes No Maybe Yes
61. took
responsibility for
your part in a
problem
No Yes No Maybe Yes
In this section, you will answer the same types of questions about your spouse. Again, you
will find that some of these items apply, while others do not. Please be sure to consider all
items, even if they seem extreme.
Section
A
Section
B
Section
C
From one year ago until today…
Has your spouse: Ever?
In front of your
child?
0 per
year
1 per
year
2-5 per
year
6-12
per
year
2-4 per
month
>1 per
week
1. screamed or
yelled at you
No Yes No Maybe Yes
2. insulted or swore
at you
No Yes No Maybe Yes
3. damaged a
household item, or
some part of your
home, out of anger
towards you
No Yes No Maybe Yes
4. withheld
affection from you
No Yes No Maybe Yes
5. deliberately
disposed of or hid
an important item
of yours
No Yes No Maybe Yes
6. sulked or refused
to talk about an
issue
No Yes No Maybe Yes
7. monitored your
time and made you
account for where
you were
No Yes No Maybe Yes
8. made plans that
left you feeling
excluded
No Yes No Maybe Yes
9. left you and
(you) were unsure
whether he/she was
going to return
No Yes No Maybe Yes
57
Appendix B (continued)
10. been angry
when you told
him/her that he/she
was using too much
alcohol or drugs
No Yes No Maybe Yes
11. been very upset
if dinner,
housework, or
home repair work
was not done when
he/she thought it
should be
No Yes No Maybe Yes
12. done or said
something to you
No Yes No Maybe Yes
13. been jealous
and suspicious of
your friends
No Yes No Maybe Yes
14. purposely hurt
your pet
No Yes No Maybe Yes
15. purposely
damaged or
destroyed your
clothes, car, and/or
other personal
possessions
No Yes No Maybe Yes
16. insulted or
shamed you in front
of others
No Yes No Maybe Yes
17. locked you out
of the house
No Yes No Maybe Yes
18. told you that
you could not work,
go to school, or go
to other self-
improvement
activities
No Yes No Maybe Yes
19. tried to prevent
you from
seeing/talking to
family or
friends
No Yes No Maybe Yes
20. had an
extramarital affair
No Yes No Maybe Yes
21. restricted your
use of the car or
telephone
No Yes No Maybe Yes
22. made threats to
leave the
relationship
No Yes No Maybe Yes
23. blamed you for
his/her problems
No Yes No Maybe Yes
24. tried to turn
family, friends, or
children against you
No Yes No Maybe Yes
25. ordered you
around
No Yes No Maybe Yes
58
Appendix B (continued)
26. been insensitive
to your feelings
No Yes No Maybe Yes
27. frightened you No Yes No Maybe Yes
28. treated you like
you were stupid
No Yes No Maybe Yes
29. given you the
silent
treatment/cold
shoulder
No Yes No Maybe Yes
30. criticized you No Yes No Maybe Yes
31. called you
names
No Yes No Maybe Yes
32. stomped out of
the room, house, or
yard
No Yes No Maybe Yes
33. stayed away
from the house
No Yes No Maybe Yes
34. ridiculed you No Yes No Maybe Yes
35. physically
twisted your arm
No Yes No Maybe Yes
36. threatened to hit
you, or throw
something at you,
in anger
No Yes No Maybe Yes
37. pushed,
grabbed, or shoved
you
No Yes No Maybe Yes
38. slapped you No Yes No Maybe Yes
39. physically
forced sex on you
No Yes No Maybe Yes
40. burned you No Yes No Maybe Yes
41. shaken you No Yes No Maybe Yes
42. thrown,
smashed, hit, or
kicked something
No Yes No Maybe Yes
43. prevented you
from getting
medical care that
you needed
No Yes No Maybe Yes
44. thrown, or tried
to throw you,
bodily
No Yes No Maybe Yes
45. thrown an
object at you
No Yes No Maybe Yes
46. choked or
strangled you
No Yes No Maybe Yes
47. kicked, bit or hit
you with a fist
No Yes No Maybe Yes
48. hit you, or tried
to hit you, with
something
No Yes No Maybe Yes
49. beat you up
(multiple blows)
No Yes No Maybe Yes
50. threatened you
with a knife or gun
No Yes No Maybe Yes
59
Appendix B (continued)
51. used a knife or a
gun on you
No Yes No Maybe Yes
52. used
humiliation to make
you have sex
No Yes No Maybe Yes
53. used threats to
make you have sex
No Yes No Maybe Yes
54. coerced you to
engage in sexual
practices you did
not want
No Yes No Maybe Yes
55. slammed you
against the wall
No Yes No Maybe Yes
56. physically
prevented you from
leaving an
argument or
blocked your exit
No Yes No Maybe Yes
57. showed you that
he/she cared even
though the two of
you disagreed
No Yes No Maybe Yes
58. showed respect
for your feelings
about an issue
No Yes No Maybe Yes
59. suggested a
compromise to a
disagreement
No Yes No Maybe Yes
60. agreed to a
solution you
suggested
No Yes No Maybe Yes
61. took
responsibility for
his/her part in a
problem
No Yes No Maybe Yes
60
Appendix C
Conflict Tactics Scale-Revised
61
Appendix D
Emotion Regulation Checklist
Child Behavior Checklist
Appendix E
hild Behavior Checklist
62
63
Appendix E (continued)
64
Appendix F
Youth Self Report
Below is a list of items that describe kids. For each item that describes you now or within the
past 6 months, please circle the 2 if the item is very true or often true of you. Circle 1 if the item
is somewhat or sometimes true of you. If the item is not true of you, circle 0.
0= Not True 1= Somewhat or Sometimes True 2= Very True of Often True
0 1 2 1. I act too young for my age 0 1 2 40. I hear sounds or voices that other people
think aren’t there (describe):
_______________________________
0 1 2 2. I have an allergy:
Describe:________________
0 1 2 41. I act without stopping to think
0 1 2 3. I argue a lot 0 1 2 42. I would rather be alone than with others
0 1 2 4. I have asthma 0 1 2 43. I lie or cheat
0 1 2 5. I act like the opposite sex 0 1 2 44. I bite my fingernails
0 1 2 6. I like animals 0 1 2 45. I am nervous or tense
0 1 2 7. I brag a lot 0 1 2 46. Parts of my body twitch or make nervous
movements (describe): _________________
0 1 2 8. I have trouble concentrating or paying
attention
0 1 2 47. I have nightmares
0 1 2 9. I can’t get my mind off certain
thoughts (describe):
_________________________
0 1 2 48. I am not liked by other kids
0 1 2 10. I have trouble sitting still 0 1 2 49. I can do certain things better than most kids
0 1 2 11. I’m too dependent on adults 0 1 2 50. I am too fearful or anxious
0 1 2 12. I feel lonely 0 1 2 51. I feel dizzy
0 1 2 13. I feel confused or in a fog 0 1 2 52. I feel too guilty
0 1 2 14. I cry a lot 0 1 2 53. I eat too much
0 1 2 15. I am pretty honest 0 1 2 54. I feel overtired
0 1 2 16. I am mean to others 0 1 2 55. I am overweight
0 1 2 17. I daydream a lot 56. Physical problems without known medical
cause
0 1 2 19. I try to get a lot of attention 0 1 2 a. Aches or pains (not stomach or headache)
0 1 2 20. I destroy my own things 0 1 2 b. Headaches
0 1 2 21. I destroy things belonging to others 0 1 2 c. Nausea, feel sick
0 1 2 22. I disobey my parents 0 1 2 d. Problems with eyes (not if corrected by
glasses) (describe): ____________________
0 1 2 23. I disobey at school 0 1 2 e. Rashes or other skin problems
0 1 2 24. I don’t eat as well as I should 0 1 2 f. Stomachaches or cramps
0 1 2 25. I don’t get along with other kids 0 1 2 g. Vomiting, throwing up
0 1 2 26. I don’t feel guilty after doing
something I shouldn’t
0 1 2 h. Other (describe): ____________________
0 1 2 27. I am jealous of others 0 1 2 57. I physically attack people
0 1 2 28. I am willing to help others when
they need help
0 1 2 58. I pick my skin or other parts of my body
(describe): ___________________________
0 1 2 29. I am afraid of certain animals,
situations, or places, other than school
(describe):____________
0 1 2 59. I can be pretty friendly
0 1 2 30. I am afraid of going to school 0 1 2 60. I like to try new things
0 1 2 31. I am afraid I might think or do
something bad
0 1 2 61. My school work is poor
0 1 2 32. I feel that I have to be perfect 0 1 2 62. I am poorly coordinated or clumsy
65
Appendix F (continued)
0 1 2 33. I feel that no one loves me 0 1 2 63. I would rather be with older kids than with
kids my own age
0 1 2 34. I feel that others are out to get me 0 1 2 64. I would rather be with younger kids than
with kids my own age
0 1 2 35. I feel worthless or inferior 0 1 2 65. I refuse to talk
0 1 2 36. I accidentally get hurt a lot 0 1 2 66. I repeat certain acts over and over
(describe):__________________________
0 1 2 37. I get in many fights 0 1 2 67. I run away from home
0 1 2 38. I get teased a lot 0 1 2 68. I scream a lot
0 1 2 39. I hang around with kids who get in
trouble
0 1 2 69. I am secretive or keep things to myself
0 1 2 70. I see things that other people think
aren’t there
(describe):_______________________
0 1 2 92. I like to make others laugh
0 1 2 71. I am self conscious or easily
embarrassed
0 1 2 93. I talk too much
0 1 2 72. I set fires 0 1 2 94. I tease others a lot
0 1 2 73. I can work well with my hands 0 1 2 95. I have a hot temper
0 1 2 74. I show off or clown 0 1 2 96. I think about sex too much
0 1 2 75. I am shy 0 1 2 97. I threaten to hurt people
0 1 2 76. I sleep less than most kids 0 1 2 98. I like to help others
0 1 2 77. I sleep more than most kids during
day and or night (describe):
_________________
0 1 2 99. I am too concerned about being neat or
clean
0 1 2 78. I have a good imagination 0 1 2 100. I have trouble sleeping (describe):
____________________________________
0 1 2 79. I have a speech problem (describe):
________________________________
____
0 1 2 101. I cut classes or skip school
0 1 2 80. I stand up for my rights 0 1 2 102. I don’t have much energy
0 1 2 81. I steal at home 0 1 2 103. I am unhappy, sad, or depressed
0 1 2 82. I steal from places other than home 0 1 2 104. I am louder than other kids
0 1 2 83. I store up things I don’t need
(describe):
________________________________
____
0 1 2 105. I use alcohol or drugs for nonmedical
purposes (describe): ____________________
0 1 2 84. I do things other people thing are
strange
(describe):
___________________________
0 1 2 106. I try to be fair to others
0 1 2 85. I have thoughts that other people
would think are strange (describe):
_____________
________________________________
____
0 1 2 107. I enjoy a good joke
0 1 2 86. I am stubborn 0 1 2 108. I like to take life easy
0 1 2 87. My moods or feelings change
suddenly
0 1 2 109. I try to help other people when I can
0 1 2 88. I enjoy being with other people 0 1 2 110. I wish I were of the opposite sex
0 1 2 89. I am suspicious 0 1 2 111. I keep from getting involved with others
0 1 2 90. I swear or use dirty language 0 1 2 112. I worry a lot
Abstract (if available)
Abstract
Youths’ exposure to marital aggression is a widespread problem. Many studies point to adverse behavioral consequences for young children, but more examination is needed of factors that might influence a relation between exposure to marital aggression and adolescent behavioral outcomes. The present study focused on emotion regulation as one possible mechanism. Data were analyzed from four waves of a prospective longitudinal study on two-parent families with a 14- to 16-year-old in Wave 4 (N=83). Questionnaires from parents and youth measured frequency of marital physical and psychological aggression, and youth emotion regulation and behavioral outcomes. Structural Equation Models revealed that emotion dysregulation provided an indirect link between marital aggression and adolescents’ internalizing and externalizing problems. Effects were observed cross-sectionally and longitudinally. A reciprocal relation was also found between emotion dysregulation and behavioral outcomes such that the two had independent and bidirectional influences on each other across time.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Examining the longitudinal relationships between community violence exposure and aggressive behavior among a sample of maltreated and non-maltreated adolescents
PDF
Parents' attunement: relation to adolescent problem behavior and the moderating role of marital conflict
PDF
Not just talk: observed communication in adolescent friendship and its implications for health risk behavior
PDF
Marital conflict and parenting: moderating effects of ethnicity and parent sex
PDF
Biological and behavioral correlates of emotional flexibility and associations with exposure to family aggression
PDF
Spouse aggression, depression, and physical health: a multivariate longitudinal study of midlife couples
PDF
Family aggression exposure and community violence exposure associated with brain volume in late adolescence: a comparison of automated versus manual segmentation
PDF
How parental psychological control is related to adolescent anger and aggression
PDF
Neighborhood context and adolescent mental health: development and mechanisms
PDF
Using observed peer discussions to understand adolescent depressive symptoms and interpersonal interactions
PDF
Community violence exposure and children's subsequent rejection within the school peer group: the mediating roles of emotion dysregulation and depressive symptoms
PDF
Family aggression, prosocial friends, and the risk of dating and friend victimization in late adolescence
PDF
Examining the associations of respiratory problems with psychological functioning and the moderating role of engagement in pleasurable activities during late adolescence
PDF
Actual and perceived social reinforcements of weight-related cognitions and behaviors in adolescent peer groups
PDF
Adolescent life stress and the cortisol awakening response: the moderating roles of emotion regulation, attachment, and gender
PDF
Reframing parental psychological control: the role of insecure attachment
PDF
The link between maternal depression and adolescent daughters' risk behavior: the mediating and moderating role of family
PDF
Popularity as a predictor of friendship affiliation in adolescence
PDF
Dynamics of victimization, aggression, and popularity in adolescence
PDF
A multiple systems approach to examining physiological stress and its association with internalizing disorders in adolescence
Asset Metadata
Creator
Iturralde, Estibaliz Maitena
(author)
Core Title
Emotion regulation as a mechanism linking parents’ marital aggression to adolescent behavioral problems: a longitudinal analysis
School
College of Letters, Arts and Sciences
Degree
Master of Arts
Degree Program
Psychology
Publication Date
09/04/2009
Defense Date
05/18/2009
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
adolescent development,domestic violence,emotion regulation,externalizing behavior,internalizing behavior,marital aggression,OAI-PMH Harvest
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Margolin, Gayla (
committee chair
), Manis, Franklin R. (
committee member
), Schwartz, David (
committee member
)
Creator Email
esti_iturralde@hotmail.com,iturrald@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-m2586
Unique identifier
UC1209157
Identifier
etd-Iturralde-3176 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-252052 (legacy record id),usctheses-m2586 (legacy record id)
Legacy Identifier
etd-Iturralde-3176.pdf
Dmrecord
252052
Document Type
Thesis
Rights
Iturralde, Estibaliz Maitena
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Repository Name
Libraries, University of Southern California
Repository Location
Los Angeles, California
Repository Email
cisadmin@lib.usc.edu
Tags
adolescent development
domestic violence
emotion regulation
externalizing behavior
internalizing behavior
marital aggression