Nihms 779728

HHS Public Access
Author manuscript
J Abnorm Psychol. Author manuscript; available in PMC 2017 July 01.
Author Manuscript
Published in final edited form as:

J Abnorm Psychol. 2016 July ; 125(5): 664–678. doi:10.1037/abn0000170.
The Multiple Faces of Interparental Conflict: Implications for

Cascades of Children’s Insecurity and Externalizing Problems
Patrick T. Davies,
University of Rochester
Rochelle F. Hentges,
Author Manuscript
Jesse L. Coe,
Meredith J. Martin,
Melissa L. Sturge-Apple, and

E. Mark Cummings
University of Notre Dame
Abstract
Author Manuscript
This multi-study paper examined the relative strength of mediational pathways involving hostile,
disengaged, and uncooperative forms of interparental conflict, children’s emotional insecurity, and
their externalizing problems across two longitudinal studies. Participants in Study 1 consisted of
243 preschool children (M age = 4.60 years) and their parents, whereas Study 2 consisted of 263
adolescents (M age = 12.62 years) and their parents. Both studies utilized multi-method, multi-
informant assessment batteries within a longitudinal design with three measurement occasions.
Across both studies, lagged, autoregressive tests of the mediational paths revealed that
interparental hostility was a significantly stronger predictor of the prospective cascade of
children’s insecurity and externalizing problems than interparental disengagement and low levels
of interparental cooperation. Findings further indicated that interparental disengagement was a
stronger predictor of the insecurity pathway than was low interparental cooperation for the sample
of adolescents in Study 2. Results are discussed in relation to how they inform and advance
Author Manuscript
developmental models of family risk.
Correspondence concerning this article should be addressed to Patrick Davies, Department of Clinical and Social Sciences in
Psychology, University of Rochester, Rochester, New York, 14627. patrick.davies@rochester.edu.
Study 1 was conducted at Mt. Hope Family Center (University of Rochester) and was supported by the Eunice Shriver Kennedy
National Institute of Child Health and Human Development (R01 HD065425) awarded to Patrick T. Davies and Melissa L. Sturge-
Apple. Study 2 was supported by the National Institute of Mental Health (2R01 MH57318) awarded to Patrick T. Davies and E. Mark
Cummings.
Davies et al. Page 2
Keywords
Author Manuscript
interparental conflict; child insecurity; child reactivity to conflict; child externalizing problems
Children with externalizing difficulties characterized by conduct problems, aggression, and

attention deficit-hyperactivity symptoms experience significant psychological burdens that
carry substantial costs to society (Doshi et al., 2012; Foster, Jones, & the Conduct Problems
Prevention Group, 2005). Research focused on identifying the family antecedents of these
problems has shown that children’s exposure to interparental conflict is a consistent
precursor of their externalizing symptoms in childhood and adolescence (e.g., Buehler et al.,
1997; 1998; Jouriles, Rosenfield, McDonald, & Mueller, 2014). According to family process
models, children’s vulnerability to interparental discord may be rooted in their exposure to
three distinctive forms of conflict behaviors (e.g., emotional expressions, verbal content):
Author Manuscript
hostility, disengagement, and uncooperativeness (Harold & Leve, 2012; Morris, Silk,
Steinberg, Myers, & Robinson, 2007; Repetti, Robles, & Reynolds, 2011). Interparental
hostility specifically consists of parental expressions of anger, frustration, and aggression
during conflicts, whereas disengagement between parents is comprised of parental
withdrawal, detachment, and avoidance behaviors. In contrast, low levels of warmth,
support, and collaborative problem solving define uncooperative interparental conflicts.
These family process models further posit that children’s difficulties preserving their
emotional security in the face of interparental conflict mediate associations between hostile,
disengaged, and uncooperative forms of interparental conflict and their behavior problems.
The widespread conceptual emphasis placed on children’s emotional insecurity as a

mediator in the pathway between interparental conflict and children’s behavior problems is
originally rooted in emotional security theory (EST; Davies & Cummings, 1994). According
Author Manuscript
to EST, destructive interparental conflict undermines children’s ability to preserve their

emotional security across a wide range of developmental periods spanning infancy through
late adolescence. Difficulties achieving emotional security in the face of interparental
conflict manifest in elevated fearful distress, avoidance, and negative representations of the
consequences conflict has for family life. Children’s insecurity, in turn, is proposed to lay
the foundation for later behavior problems. In support of this hypothesis, studies have
repeatedly identified children’s insecurity as a mediator in prospective associations between
interparental conflict and child externalizing problems (e.g., Buehler, Lange, & Franck,
2007; Cummings, Schermerhorn, Davies, Goeke-Morey, & Cummings, 2006; Kelly & El-
Sheikh, 2013).
In an effort to distinguish between the specific forms of conflict outlined in the family
Author Manuscript
process models, the reformulation of emotional security theory posits that interparental
hostility, disengagement, and poor cooperation will vary systematically in their strength as
predictors of children’s insecurity (EST-R; Davies & Sturge-Apple, 2007). According to
EST-R, the emotional security system developed over our evolutionary history to selectively
respond to cues of interpersonal danger. As a result, interpersonal threat cues (e.g., angry
faces, loud voices, aggression) assume primacy in organizing children’s fearful responses in
close-knit social contexts such as the family (Davies & Sturge-Apple, 2007; Öhman &

Mineka, 2001). Because diminished happiness, support, and problem solving are far less
Author Manuscript
reliable as signals of danger in the absence of overt hostile threat cues, these parameters of
uncooperative conflict are hypothesized to carry minimal weight in signifying threat and
organizing children’s insecure responses (Öhman & Mineka, 2001). Therefore, EST-R
hypothesizes that hostile interparental conflict is a more consistent and powerful predictor of
children’s insecurity and psychological problems than are variations of interparental
cooperation (Davies & Sturge-Apple, 2007).
As a further distinction, EST-R posits that interparental disengagement will be a modest

predictor of children’s insecurity, falling in between the hypothesized minimal impact of
poor interparental cooperation and the potent risk of interparental hostility. On the one hand,
disengagement may be a marker of interpersonal discord, struggle, and rejection within the
social hierarchy (Gilbert, 2001). Therefore, by signalling the potential for threat, repeated
exposure to disengagement may be more likely to undercut children’s security in the
Author Manuscript
interparental relationship than variations in uncooperative conflict (Dixon, 1998). On the

other hand, withdrawal and disengagement are less reliable prognosticators of imminent
social threat. Thus, hostile interparental conflict is theorized to represent more immediate
signals of threat than disengaged conflict.
Research has yet to test whether unfolding mediational cascades involving interparental
conflict, children’s insecurity, and their behavioral problems vary depending on whether
conflict is expressed through hostility, disengagement, or poor cooperation. Addressing this
gap requires simultaneously examining interparental hostility, disengagement, and poor
cooperation as predictors of children’s insecurity. In contrast, most studies exploring
multiple dimensions of interparental conflict have isolated the predictive power of forms of
conflict in separate analytic models of emotional insecurity in samples of children ranging
Author Manuscript
from 2- to 16-years-old (i.e., Cummings, Goeke-Morey, & Papp, 2004; Davies, Martin, &
Cicchetti, 2012; Goeke-Morey, Cummings, & Papp, 2007; McCoy, Cummings, & Davies,
2009). Only one study has simultaneously examined the predictive roles of interparental
conflict dimensions in models of security and adjustment that roughly correspond with the
three conflict parameters outlined in EST-R (Du Rocher, Schudlich & Cummings, 2007). In
a sample of 8- to 16-year-old children who differ from the children in the samples of the
present paper, the study results indicated that depressive conflict tactics (i.e., sadness,
withdrawal, fear, physical distress) were significantly related to children’s insecurity. In
contrast with EST-R’s hypotheses, hostile and uncooperative conflict failed to explain any
unique variance in children’s insecurity.
However, these findings relied on cross-sectional data, and only longitudinal designs can
Author Manuscript
directly test the hypothesized mediational paths. For example, family process models regard
interparental conflict as a risk factor that gradually erodes children’s adjustment by
progressively setting in motion their pathogenic responses to family stress (Cummings &
Davies, 2011; Morris et al., 2007; Repetti et al., 2011). Thus, security is proposed to develop
over periods of months following exposure to interparental conflict and, in turn, broaden into
psychological problems over an extended time period (Cummings & Miller-Graff, 2015;
Davies, Martin, & Sturge-Apple 2016). Likewise, quantitative psychologists have
demonstrated that cross-sectional tests of mediation produce substantially biased findings

(Cole & Maxwell, 2003; Maxwell & Cole, 2007; Preacher, 2015). The resulting conclusion
Author Manuscript
is that “the bias inherent in cross-sectional designs has disastrous consequences for
hypothesis tests” of mediation (Maxwell & Cole, 2007, p. 39).
The aim of this paper is to provide the first prospective test of the mediational pathways
involving the three forms of interparental conflict, children’s insecurity, and their
externalizing problems. We report on the results from two data sets that are independent of
the previously published paper on mediational links among forms of interparental conflict,
children’s insecurity, and their psychological adjustment (i.e., Du Rocher Schudlich &
Cummings, 2007). Each study employed a longitudinal design with three annual
measurement occasions. Thus, in accord with quantitative recommendations for testing
mediation (e.g., Cole & Maxwell, 2003; Maxwell & Cole, 2007), we used repeated measures
of insecurity and child adjustment problems to permit a full prospective analysis of change
at each link in the mediational chain. In addition, the studies in this paper each used a multi-
Author Manuscript
method, multi-informant measurement approach. Finally, as a step toward examining the

generalizability and specificity of our hypotheses, we tested the mediational role of
insecurity across two studies that varied in their developmental (i.e., preschool versus
adolescent children) and socioeconomic (i.e., economically impoverished versus middle
class sample) backgrounds.
Study 1
To test the EST-R hypotheses, we examined mediational pathways among the three forms of
interparental conflict, children’s insecurity, and their externalizing problems during the
transition from the preschool to early school years. This developmental window has been
posited to be a sensitive period for the operation of emotional security processes. For
Author Manuscript
example, relative to their older counterparts, children in the early school years respond to
adult conflicts with more fear, poorer perceived competence, and more limited coping
strategies (e.g., Cummings, Vogel, Cummings, & El-Sheikh, 1989; El-Sheikh & Cummings,
1995; Grych, 1998). Developmental models have further postulated that advances in social
perspective taking trigger an increase in children’s concerns about the welfare of their
parents and themselves within the broader family unit (Cicchetti, Cummings, Greenberg, &
Marvin, 1990; Davies et al., 2016). As a first step toward understanding the hypothesized
cascade of insecurity, we examined our research questions within a sample of families with
preschool children who are at elevated risk for exhibiting psychopathology. Finally, we
included family income, child gender, and maternal and paternal parenting quality as
covariates in the analyses. We selected these factors based on empirical identification of: (a)
parenting difficulties and economic impoverishment as predictors of children’s insecurity
Author Manuscript
and externalizing problems (e.g., Webster-Statton & Hammond, 1999; Sturge-Apple,

Davies, Winter, Cummings, & Schermerhorn, 2008) and (b) child gender (i.e., boys) as a
risk factor for externalizing difficulties (e.g., Campbell, Shaw, & Gilliom, 2000).
Methods
Participants—Participants included 243 families (i.e., mother, father, and preschool
child) from a moderately sized metropolitan area who were recruited through multiple

agencies including local preschools, Head Start programs, Women, Infants, and Children
Author Manuscript
(WIC) programs, and public and private daycare providers. The average age of children at
Wave 1 was 4.60 years (SD = .44; range = 4 to 5 years old), with 56% of the sample
consisting of girls. Almost half of the families were Black or African American (48%),
followed by families who identified as White (43%), multi-racial (6%), or another race
(3%). Approximately 16% of the family members identified as Latino. At Wave 1, 99% of
the mothers and 74% of the fathers were biological parents. Median household income was
$33,900 per year (range = $1,100 – $121,000), with most families (69%) receiving public
assistance. Moreover, the median education for the sample consisted of a GED or high
school diploma, and 19% of the parents did not earn a high school diploma or GED. Parents
had lived together an average of 3.36 years. Approximately half of the adults (47%) were
married. The longitudinal design consisted of three annual measurement occasions
beginning when children were in their last year of preschool. Retention rates across
Author Manuscript
contiguous waves of data collection were 97% and 94%. Comparisons of families lost to
attrition and those who participated in all three waves along the thirteen family (e.g., forms
of interparental conflict, parenting), child (e.g., insecurity responses, externalizing
problems), and demographic (e.g., family income) variables included in the primary
analyses only yielded one significant difference. Families participating in all waves
evidenced significantly lower incomes than those who dropped from the study, d = .64.
Procedures—Families visited our research laboratory at each measurement occasion.

All research procedures were approved by the Institutional Review Board prior to
conducting the study. Families and teachers were compensated monetarily for their
participation.
Interparental conflict task: Parents participated in an interparental conflict task in which

Author Manuscript
they discussed common, problematic disagreements in their relationship for 10 minutes

(Gordis, Margolin, & John, 2001). While the child was in a separate room, parents selected
multiple issues to discuss so they could move on to another topic if they finished discussing
a previous one during the task. Parents were free to discuss any problematic disagreement
topic as long as they were both comfortable discussing it in front of their child. After parents
selected the topic, an experimenter escorted the child into the room to play with a set of toys.
Parents then engaged in the video-recorded interparental interaction after the experimenter
left the room.
Interparental disagreement interview: At Wave 1, mothers completed the Interparental

Disagreement Interview, a semi-structured, narrative interview designed to assess the
frequency, nature, course, and aftermath of interparental conflicts witnessed by their children
Author Manuscript
(e.g., Davies, Sturge-Apple, Cicchetti, Manning, & Vonhold, 2012). After selecting an
interparental conflict topic that commonly takes place in front of the child, a trained
interviewer presented mothers with a series of open-ended questions (e.g., “How would you
describe your disagreements over [topic]?”; “How do you typically feel during these
disagreements?”; “How does the disagreement end?”). Additional probes were used to
maximize the richness of maternal narratives. Interviews were videotaped for later coding.

Questionnaire assessments: Mothers and fathers also completed survey measures of

Author Manuscript
interparental conflict and demographic characteristics at Wave 1. During Waves 1 and 3,

mothers and teachers reported on children’s disruptive behavior problems.
Family interaction task: At Wave 1, mothers, fathers, and children participated in a 10-
minute task in which they were asked to work together to build a model house using LEGO
blocks (e.g., Schoppe, Mangelsdorf, & Frosch, 2001). Because the objective was to create a
context that elicits child bids for parental support and assistance (McHale, Kuersten-Hogan,
& Lauretti, 2001), the model house was selected so that children could not successfully build
the house without parental assistance. Video records of the task were later coded for
parenting.
Measures—We used a multi-method, multi-informant approach to obtain

comprehensive measures of each form of interparental conflict. For comparability in
Author Manuscript
assessment, interparental hostility, disengagement, and poor cooperation variables each

consisted of mean composites of eight standardized assessments. Each form of conflict was
specifically indexed by three observer ratings from the interparental conflict task, three
parent report measures of interparental conflict, and two coder ratings from the interparental
disagreement interview.
Hostile interparental conflict (Wave 1): For the observational component of the
measurement battery, trained coders rated interparental hostility during the interparental
conflict task using the Negative Escalation code from the System for Coding Interactions in
Dyads (SCID; Malik & Lindahl, 2004) and the Anger code from the Interparental Conflict
Expressions (ICE; e.g., Davies, Coe, Martin, Sturge-Apple, & Cummings, 2015; Davies &
Sturge-Apple, 2013) Coding System. Rated along a five-point scale (1 = Very low; 5 = High)
Author Manuscript
based on dyadic behavior, the SCID Negative Escalation code reflects the degree to which
the couple reciprocates or escalates displays of anger, hostility, and negativity. The ICE
Anger code indexes the intensity and frequency of facial expressions, verbalizations, and
postural displays of anger by mothers and fathers separately in the interaction on a nine-
point scale (1 = Not at all characteristic; 9 = Mainly characteristic). Interrater reliabilities,
based on ICCs of independent ratings on 30% of the interactions, ranged from .80 to .84
across codes. The SCID is a well-established system for coding interparental conflict with
strong psychometric properties (for details, see Lindahl, Malik, Kaczynski, & Simons,
2001). As an adaptation of the SCID, the ICE codes have evidenced predictive validity, as
shown by their significant associations with children’s emotional reactivity to interparental
conflict, internalizing symptoms, and externalizing problems (Davies et al., 2015).
Author Manuscript
In the interview portion of the assessment, coders rated mother and father anger during
conflicts from the interparental disagreement interview. Anger ratings, which ranged from 0
(None) to 6 (High), reflected the degree to which mothers and fathers exhibited signs of
hostility, anger, and irritation. ICCs, indexing reliability between two raters who
independently overlapped on 30% of the videos, were .87 for maternal Anger and .86 for
partner Anger.

For the questionnaires, mothers and fathers both completed the Psychological Aggression
Author Manuscript
Scale from the Conflict Tactics Scale (CTS-2; Straus, Hamby, Boney-McCoy, & Sugarman,
1996), while mothers also reported on the Negative Escalation scale from the Managing
Affect and Disagreements Scale (MADS; Arellano & Markman, 1995). The CTS-2
Psychological Aggression scale is designed to assess verbal and psychological forms of
interparental hostility, whereas the MADS Negative Escalation scale measures reciprocation
of anger between parents. Internal consistencies ranged from .80 to .90 across the three
questionnaire measures. The resulting eight measures were standardized and averaged into a
composite of interparental hostility (α = .80).
Disengaged interparental conflict (Wave 1): In the observational part of the assessment,
trained coders rated the interparental conflict task using the SCID Pursuit-Withdrawal Scale
(Malik & Lindahl, 2004) and the Disengagement scale from the ICE (Davies & Sturge-
Apple, 2013). The Disengagement code assesses maternal and paternal detachment, flat
Author Manuscript
affect, unresponsiveness, and avoidance of conflict topics along a nine-point scale (1 = Not
at all Characteristic; 9 =Mainly Characteristic). The five-point Pursuit-Withdrawal scale (1 =
Very low; 5 = High) is a dyadic code characterized by one partner persistently responding
with detachment to the consistent demands of the other partner to engage in the conflict.
ICCs assessing reliability based on coders’ independent ratings of over 20% of the videos
were .77 for maternal Disengagement, .72 for paternal Disengagement, and .56 for Pursuit-
Withdrawal.
For the two interview assessments of the construct, trained coders rated maternal responses
to the interparental disagreement interview for levels of maternal and paternal
disengagement along seven-point scales (0= None; 6 = High). The Disengagement code
reflects the extent to which each parent is detached during the conflicts (e.g., avoidance,
Author Manuscript
leaves the room or house, sulks, becomes quiet). ICCs indexing interrater reliability among
trained coders who overlapped on over 30% of the interviews were .85 and .88 for maternal
and paternal Disengagement codes, respectively.
As the set of three survey assessments, mothers and fathers each completed the Stalemate
scale from the Conflict and Problem-Solving Scales, and mothers also reported on the CPS
Avoidance scale (CPS; Kerig, 1996). The 14-item Stalemate scale indexes interparental
detachment (e.g., “Sulk, refuse to talk, give the silent treatment”), while the 16-item
Avoidance scale assesses efforts to avoid or withdraw from interparental problems (e.g.,
“Leave the room”). Alpha coefficients ranged from .78 to .84 across the three measures. As
with the interparental hostility composite, the multi-method, multi-informant scales were
standardized and aggregated to form a single index of interparental disengagement (scale-
Author Manuscript
level α = .68).
Uncooperative interparental conflict (Wave 1): The observational assessment from the
interparental conflict task consisted of trained coder ratings of the interparental conflict task
using the ICE Positive Affect scale for mothers and fathers separately and the SCID
Cohesion code at the couple level (Davies & Sturge-Apple 2013; Malik & Lindahl, 2004).
Rated on a nine-point scale (1 = Not at all characteristic; 9 = Mainly characteristic), the ICE
Positive Affect code assesses maternal and paternal expressions of warmth through

verbalizations (e.g., cheerful tone of voice), facial expressions (e.g., smiling), and gestures
Author Manuscript
(e.g., physical affection). The SCID Cohesion scale, which utilizes a five-point scale (e.g., 1
= Very low; 5 = High), assesses the degree of closeness, support, and connectedness between
the parents. ICCs for interrater reliability based on over 20% of the interactions ranged
from .80 to .87 across the three codes.
For the interview portion of the measurement battery, trained coders rated maternal and
paternal contentment from the maternal interparental disagreement interview descriptions
immediately following disagreements. Rated on a seven-point scale (0 = None; 6 = High),
the Contentment code captures the extent to which each parent is happy, relaxed, and
comfortable in the aftermath of the conflict. Thus, at high levels, the narrative portrays
parents as experiencing intense positive affect and satisfaction. Interrater reliabilities, based
on the two coders’ ratings of over 30% of the interactions, were excellent (ICC = .92 for
maternal and paternal contentment).
Author Manuscript
The three survey measures utilized to assess interparental uncooperativeness, included: (a)
mother and father reports on the 12-item CPS Cooperation Scale (Kerig, 1996) and (b)
mother reports on the MADS Editing Scale (Arellano & Markman, 1995). The CPS
Cooperation scale assesses the degree to which parents work collaboratively to solve
conflicts in mutually respectful ways (e.g., “try to find a solution that meets both needs
equally”). The MADS Editing scale measures parental tendencies to respond positively to
each other even during stressful times through listening skills and constructive framing of
interactions (e.g., “I express appreciation for my partner’s help despite his unsuccess”).
Internal consistency coefficients were satisfactory for the three scales (i.e., each α = .89). To
obtain an assessment of poor cooperation, all eight cooperation measures from the
interparental conflict task, interparental disagreement interview, and questionnaires were
Author Manuscript
reverse scored so that higher scores reflect greater uncooperative conflict. Each of these
measures were standardized and averaged together into a single composite of uncooperative
interparental conflict (α = .80).
Children’s emotional insecurity (Waves 1 and 2): At the first two waves, trained raters
coded children’s behavioral reactivity to interparental conflict from the interparental conflict
task along three established assessments of insecurity (Davies, Sturge-Apple, Winter,
Cummings, & Farrell, 2006). Child reactivity ratings at Wave 2 were only included in
analyses when the same parents from Wave 1 participated in the task at Wave 2. As a result,
20 children had missing data for these ratings. The three measures were specified as
manifested indicators of a latent construct of insecurity at each wave. To assess the first
indicator of emotional reactivity, coders rated children along two dimensions: Vigilance and
Author Manuscript
Fearful Distress. Coded along a five-point scale (1 = No vigilance; 5 = Intense vigilance),

signs of vigilance included carefully watching and listening to parents, substantially
decreasing play activities due to preoccupation with the conflict, and verbal concern about
the conflict. The second code consisted of a nine-point molar rating of Fearful Distress (1 =
Not at all characteristic; 9 = Mainly characteristic), assessing the extent to which children
displayed anxiety, tension, fear, worry, or emotional upset in facial (e.g., open mouth, staring
wide-eyed), postural (e.g., freezing), or gestural (e.g., wringing hands) expressions. ICCs,
indexing two independent coders’ ratings of over 20% of the interactions at each wave,

ranged from .78 to .86 for the two codes across the waves. Ratings for each code were
Author Manuscript
standardized and averaged together to create a composite of emotional reactivity at each

wave. Internal consistencies for the composites were .84 at Wave 1 and .72 at Wave 2.
As the second indicator of insecurity, coders rated children’s avoidance of interparental

conflict along two scales: Avoidance and Submissive Disengagement. The five-point scale
for Avoidance (1 = No avoidance; 5 = Intense avoidance) indexes the chronicity, quality, and
intensity of child withdrawal behaviors during the interparental interaction (e.g., leaving the
room, hiding one’s face, making noise to tune out parents, minimizing verbal interactions
with parents, turning away from parents). Submissive Disengagement, which was coded
along a nine-point scale (1 = Not at all characteristic; 9 = Mainly characteristic), was defined
as subtle, passive, anxious, and restrained forms of social distancing and withdrawal from
the interaction (e.g., becoming quiet or withdrawn, avoiding eye contact, gingerly moving
away from the interaction). As indices of inter-rater reliability, ICCs of the ratings of over
Author Manuscript
20% of the videos by two independent coders ranged from .76 to .93 across the two codes at
each wave. In accord with the emotional reactivity measure, ratings for each code were
standardized and averaged to create a composite of avoidance at each wave (composites αs
were .79 at Wave 1 and .81 at Wave 2).
As the third indicator, coders assessed children along a molar scale of Security, ranging from
1 (Not at all characteristic) to 9 (Mainly characteristic). The Security code was defined by
behaviors that are theorized to reflect children’s confidence in parents to manage disputes in
a way that maintains family harmony, including: negligible or mild levels of fearful distress
and involvement in conflicts that are well-managed and followed by quick resumption of
normal activities in the aftermath of parental anger. Trained coders independently rated over
20% of the videos to assess interrater reliability at each wave. ICCs were .89 at Wave 1 and .
Author Manuscript
85 at Wave 2.
Child externalizing problems: At Waves 1 and 3, teachers and parents completed the
respective Teacher and Parent versions of the MacArthur Health and Behavior Conduct
Problems Subscale (HBQ; Ablow et al., 1999; 11 items; e.g., “Physically attacks people”).
In addition, teachers completed the HBQ ADHD Symptoms scale to assess impulsivity,
hyperactivity, and inattention (15 items; e.g., “Can’t stay seated when required to do so”).
Internal consistencies for the measures across the waves ranged from .89 to .95. The three
scales were specified as manifest indicators of the latent construct of child externalizing
problems at each wave.
Covariates: Maternal and paternal parenting quality: Quality of maternal and paternal
Author Manuscript
parenting was assessed during the family interaction task using the Sensitivity and Warmth
codes from the well-established Iowa Family Interaction Rating Scales (IFIRS; Melby &
Conger, 2001). The IFIRS assesses the frequency and intensity of parent caregiving
behaviors on a nine-point scale (1 = Not at all characteristic; 9 = Mainly characteristic). The
Sensitivity scale assesses individual differences in parental awareness of their children’s
needs, emotional states, and abilities; the Warmth Scale indexes parental support and
affection toward the child. To evaluate interrater reliabilities, a second coder independently
rated 21% of the parent-child interactions. ICCs ranged from .90 to .95 across the four

codes. Because the ratings were assessed on the same scale, composites of maternal and
Author Manuscript
paternal parenting quality were created by averaging Warmth and Sensitivity ratings for each
parent (αs = .84 for mom; .92 for dad).
Covariates: Sociodemographic characteristics: Two demographic covariates derived

from a Wave 1 maternal interview included (a) child gender and (b) annual household
income.
Results and Discussion

Table 1 provides the means, standard deviations, and correlations among the primary
variables and covariates. Comparisons of teacher reports of behavioral problems on the HBQ
with comparable reports from a community sample of children from two US regions (i.e.,
Ablow et al., 1999) indicated that our sample evidenced conduct problems and ADHD
Author Manuscript
symptoms that were, on average, 81% higher across the two waves. Percentages of children
in our sample who exceeded the mean level of behavioral difficulties in a clinic-referred
sample of young children in the Ablow et al. (1999) study ranged from 19% to 22% across
the two teacher measures.
Structural equation modeling (SEM) with the Amos 22.0 statistical software program was
used to examine children’s emotional insecurity as a mediator of prospective associations
among interparental hostility, disengagement, and poor cooperation and their externalizing
problems. As shown in Figure 1, structural paths were specified between each predictor (i.e.,
interparental conflict forms and covariates) and the two endogenous variables of Wave 2
child insecurity and Wave 3 child externalizing problems. We also estimated a structural
path between Wave 1 externalizing problems and Wave 2 emotional insecurity (see Figure
1). As a test of the second link in the mediational model, Wave 2 insecurity, in turn, was
Author Manuscript
specified as a predictor of Wave 3 externalizing problems. Autoregressive paths using Wave

1 insecurity and externalizing problems as predictors were included in the analysis. In
accord with strong factorial approaches for maximizing measurement equivalence for
repeated measures of child functioning (Widaman, Ferrer, & Conger, 2010), the following
constraints were placed on the latent endogenous constructs: (1) factor loadings of the
indicators of insecurity and externalizing difficulties were constrained to be equal across
time and (2) intercepts of the same indicators were fixed to be invariant across time. Finally,
correlations were specified among all Wave 1 (i.e., covariates and predictors) variables in the
model and between error terms of comparable manifest indicators of insecurity and
externalizing problems across the waves. SEM analyses used full-information maximum
likelihood (FIML) to estimate missing data (i.e., data were missing for 10.9% of the values)
and retain the full sample for primary analyses (Enders, 2001).
Author Manuscript
The resulting model depicted in Figure 1 provided a good representation of the data, χ2 (109,
N = 243) = 163.52, p = .001, RMSEA = .05, CFI = 0.95, and χ2/df ratio = 1.50. For clarity,
the correlations among covariates and predictors (see Table 1 for correlations among the
measures) are not displayed in the Figure. In support of the measurement model, the
standardized loadings of the manifest indicators onto their latent constructs were all
significant (p < .001) and generally moderate to high in magnitude (mean absolute value for

loadings = .65). Moderate autoregressive paths were identified for children’s insecurity over
Author Manuscript
a one-year period, β = .27, p < .01, and their externalizing difficulties across a two-year span,
β = .47, p < .001. With the inclusion of the autoregressive paths, none of the covariates were
significant predictors of W2 child insecurity or W3 externalizing problems. Consistent with
hypotheses, Wave 1 interparental hostility significantly predicted children’s greater
emotional insecurity over a one-year period, β = .34, p < .01. Children’s insecurity at Wave
2, in turn, predicted child behavior problems at Wave 3, β = .21, p < .05. As a further test of
mediation, we conducted bootstrapping tests for the indirect effect (Preacher & Hayes,
2008). In support of mediation, the results indicated that the indirect path involving Wave 1
interparental hostility, Wave 2 emotional insecurity, and Wave 3 child externalizing
problems was significantly different from zero, with the unstandardized coefficient for the
indirect path = .36, 95% CI [.05, .80].
In contrast to the significant findings for interparental hostility, uncooperative and

Author Manuscript
disengaged forms of conflict were unrelated to child insecurity at Wave 2. As a direct test of
the salience of interparental hostility in predicting insecurity, we conducted pairwise
parameter comparisons comparing the magnitude of associations between the three forms of
interparental conflict and child insecurity. One-tailed tests were used to compare the relative
strength of the links based on the EST-R prediction that interparental hostility would be the
strongest predictor of insecurity. Supporting the hypotheses, results indicated that Wave 1
interparental hostility was a significantly stronger predictor of Wave 2 child insecurity than
interparental disengagement, z = 2.05, p < .05, and poor interparental cooperation, z = 1.91,
p < .05.
Study 2
Author Manuscript
In Study 2, our goal was to further test interparental hostility, disengagement, and
uncooperativeness as unique predictors of children’s insecurity and externalizing problems
in a sample of young adolescents. Based on the Study 1 results and EST-R predictions, we
hypothesized that interparental hostility would more strongly predict insecurity than
interparental disengagement and poor cooperation. The null associations between
disengaged and uncooperative forms of conflict and children’s insecurity in Study 1 did not
support EST-R’s hypothesis that disengagement between parents would fall in between
hostile and uncooperative interparental conflict in its strength as a precursor of security.
However, from a developmental perspective, it is possible that disengaged interparental
conflict only emerges as a threat to children’s insecurity as they make further gains in
processing interpersonal emotions and interactions. In this regard, research has shown that
young adolescents exhibit greater acuity than younger children in discriminating between
Author Manuscript
interparental conflict tactics and draw more systematic causal connections between social
events and their impact on future interpersonal relations (e.g., Cummings, Ballard, El-
Sheikh, & Lake, 1991; El-Sheikh & Cummings, 1995). Therefore, due to their greater
sensitivity to subtle forms of interpersonal threat, we hypothesized that interparental
disengagement would emerge as a stronger predictor of insecurity than poor interparental
cooperation during early adolescence.

In addition, developmental psychopathology models underscore the importance of

Author Manuscript
identifying risk processes across normative as well as atypical conditions (e.g., Cicchetti &
Toth, 2009). Therefore, our aim in Study 2 was to test whether the pathogenic cascade of
insecurity identified in the at-risk sample in Study 1 was similar in a more advantaged
sample. We specifically sought to examine mediational pathways involving the three forms
of interparental conflict, child insecurity, and their externalizing difficulties in a community
sample of adolescents who, on average, experienced relatively low levels of adversity. To
maximize the comparability of the assessment batteries, design, and analysis of the two
studies, we made efforts to maintain some correspondence in measures across the studies.
Therefore, Study 2 made use of similar: (a) interparental hostility, disengagement, and poor
cooperation composites derived from observational ratings, a semi-structured interview, and
parental questionnaire ratings, (b) teacher and maternal assessments of disruptive behavior,
(c) signs of emotional insecurity (e.g., emotional reactivity, avoidance), and (d) covariates
Author Manuscript
(i.e., maternal and paternal parenting quality, household income, and child gender). As in
Study 1, we employed the same lagged autoregressive approach to testing the mediational
pathways over time.
Methods
Participants—Data for this paper were drawn from a study with 280 mothers, fathers,
and adolescents. Participants were recruited through local school districts and community
centers in a Northeastern metropolitan area and a small Midwestern city. Due to our focus
on examining children’s responses to interparental conflict, families were only included in
this study if the mothers, fathers, and adolescents had regular contact with each other (i.e.,
contact as a triad for an average of 2 to 3 days per week during the year). As a result, 17
families were excluded from this paper, yielding a sample of 263 mothers, fathers, and
Author Manuscript
adolescents. Most adolescents were White (74%), followed by smaller percentages of

African-American (17%), multi-racial (8%), and other racial (1%) backgrounds.
Approximately 4% of the adolescents identified as Latino. Most adolescents lived with their
biological mothers (93%) and fathers (79%), with the remainder living with adoptive
parents, stepparents, or adult guardians. The longitudinal design of the study consisted of
three annual measurement occasions. Retention rates across contiguous waves were 92%
and 94%. The average age of adolescents at Wave 1 was 12.62 years (SD = .57; age range =
11 to 14 years old), with 50% of the sample consisting of girls. Median household income of
the families was between $55,000 and $74,999 per year. Mothers and fathers reported
median education levels of some college education. Most parents (i.e., 89%) were married at
the outset of the study. Comparisons of families lost to attrition with those who remained in
the study at the third wave yielded no statistical differences across the thirteen family (e.g.,
Author Manuscript
interparental conflict, parenting), child (e.g., insecurity, externalizing problems), and

demographic variables at Wave 1.
Procedures—Families visited the laboratory at one of two sites at each time point. The
Institutional Review Boards at each research site approved all research procedures. Families
and teachers were compensated monetarily for their participation.

Interparental conflict task: During Wave 1, mothers and fathers participated in an

Author Manuscript
interparental conflict task in which they were asked to engage in a disagreement (Du Rocher
Schudlich, & Cummings, 2007). Parents conferred to select two problematic topics for their
relationship that they were comfortable discussing. The couples then discussed the two
topics for a total of fourteen minutes. The task was video recorded for subsequent coding.
Interparental disagreement interview: At Wave 1, mothers completed the same

interview used in Study 1. The interview was video recorded for later coding.
Questionnaire assessments: Mothers and fathers filled out surveys of interparental

conflict and demographic characteristics at Wave 1. During Waves 1 and 3, teachers and
mothers completed questionnaires to assess adolescent problem behaviors. Finally, teens
completed a survey to assess their emotional insecurity in the interparental relationship at
Waves 1 and 2.
Author Manuscript
Family problem-solving task: At Wave 1, mothers, fathers, and adolescents participated

in a family problem-solving task in which they were asked to select and discuss a topic of
frequent family disagreement for seven minutes with the goal of coming to a resolution
(Gordis, Margolin, & John, 2001). Families were instructed to choose a topic that they felt
comfortable discussing and to talk about the topic as they normally would at home. The
interactions were video recorded for subsequent coding of maternal and paternal parenting
behaviors.
Measures—Consistent with Study 1, we used a multi-method, multi-informant approach

to assess interparental hostility, disengagement, and poor cooperation. Thus, for each form
of conflict, we created a composite of six standardized assessments that consisted of two
Author Manuscript
observer ratings from the interparental conflict task, two parent (i.e., mother and father)
report measures of interparental conflict, and two coder ratings from the interparental
disagreement interview.
Hostile interparental conflict: For the observational part of the interparental hostility
measurement battery, trained coders separately rated maternal and paternal hostility in the
interparental interaction task using the Negativity and Conflict code from the SCID (Malik
& Lindahl, 2004). Negativity and Conflict reflected the degree to which each individual in
the dyad displayed anger, frustration, and tension, as indexed by a five-point rating scale (1
= Very Low; 5 = High). Interrater reliabilities, based on ICCs of coders’ independent ratings
on at least 20% of the interactions, were .87 for mothers and .86 for fathers. For the
interparental disagreement interview part of the measurement, another set of trained
Author Manuscript
observers rated the levels of mother and father anger using the same Anger rating scale from
Study 1. ICCs, assessing reliability between two raters who independently overlapped on
25% of the videos, were .81 and .88 for maternal and partner Anger, respectively. Finally,
mothers and fathers each completed the 10-item O’Leary Porter Scale to assess children’s
exposure to interparental hostility (OPS; Porter, & O’Leary, 1980; e.g., “How often do you
and/or your partner display verbal hostility [raised voices, etc.] in front of your child?”).
Internal consistencies for the maternal and paternal OPS assessments were .78 and .80,

respectively. The resulting six scales were standardized and averaged together into a
Author Manuscript
composite of hostile interparental conflict (α = .68).
Disengaged interparental conflict: For the observational assessment of the interparental

conflict task, trained coders rated mothers and fathers separately along the SCID Withdrawal
scale. The Withdrawal code is characterized by displays of detachment, avoidance of
conflict topics, unresponsiveness, and flat affect, with ratings ranging from 1 (Very low) to 5
(High). ICCs assessing reliability based on coders’ independent ratings of 20% of the
interactions were .85 and .86 for mother and partner withdrawal, respectively. For maternal
responses on the interparental disagreement interview, trained coders rated maternal and
paternal conflict responses separately using the same Disengagement scale from Study 1.
ICCs indexing interrater reliability of trained coders on over 25% of the videos were .75
and .74 for maternal and paternal Disengagement ratings. For the final measures, mothers
and fathers completed the CPS Stalemate scale (αs = .80 and .81 for mom and dad reports)
Author Manuscript
used in Study 1 (Kerig, 1996). As with the interparental hostility measure, the six multi-
method scales were standardized and aggregated into an interparental disengagement
composite (α = .65).
Uncooperative interparental conflict: The uncooperative conflict assessment followed

the same multi-method and multi-informant approach as the other forms of interparental
conflict. First, the observational assessment consisted of trained coder ratings on the SCID
Positive Affect scale during the interparental conflict task. Rated on a five-point scale (e.g.,
1 = Very Low; 5 = High), the SCID Positive Affect code is defined by positive, warm
displays reflected in tone of voice (e.g., cheerful, satisfied), behaviors (e.g., affection,
laughter), and facial expressions (e.g., genuine smiles). Independent coder ratings of 20% of
the interactions yielded adequate interrater reliabilities for maternal (ICC = .90) and paternal
Author Manuscript
(ICC = .91) positive affect. Second, coders rated maternal and paternal behavior from the
interparental disagreement interview using the Contentment code from Study 1. ICCs of
independent coder ratings on over 25% of the interviews were .92 for maternal contentment
and .87 for paternal contentment. Third, mothers and fathers also independently completed
the CPS Cooperation subscale used in Study 1 (Kerig, 1996). Internal consistency
coefficients for the subscales were .87 for father reports and .86 for mother reports.
Consistent with data reduction for the other conflict dimensions, the six assessments of
cooperation were reserve-scored, standardized, and averaged together into a composite of
poor interparental cooperation (α = .70).
Adolescent insecurity in the interparental relationship: In contrast to the younger

children who participated Study 1, studies have shown that adolescents provide reliable and
Author Manuscript
valid reports of their own emotional insecurity, including indices of fearful distress (e.g.,
emotional reactivity, avoidance) as well as appraisals of the meaning interparental conflict
has for their families and themselves (Davies et al., 2002). As a result, adolescents
completed three scales derived from the Security in Interparental Subsystem (SIS) Scales to
assess their emotional insecurity at Waves 1 and 2 (Davies et al., 2002): (1) the Emotional
Reactivity scale, which assesses multiple prolonged fearful distress reactions to conflict
(e.g., nine items; “When my parents argue, I feel scared”), (2) the Avoidance scale, indexing

children’s efforts to reduce their exposure to interparental conflicts (e.g., seven items;
Author Manuscript
“When my parents have an argument, I try to be very quiet”), and (3) the Destructive Family
Representations scale, which assesses negative appraisals of the impact of interparental
conflict for the family (e.g., four items; “When my parents have an argument, I wonder if
they will divorce or separate”). Alpha coefficients ranged from .83 to .89 for the three scales
across Waves 1 and 2. The validity of the SIS scales is supported by previous research (e.g.,
Davies et al., 2002, 2014).
Adolescent externalizing problems: At Waves 1 and 3, teachers completed the Conduct

Problems (e.g., “often fights with other children”) and Hyperactivity/Inattention (“restless,
overactive, cannot stay still for long”) Scales from the Strength and Difficulties
Questionnaire (SDQ; Goodman, 2001). To obtain multiple informant reports, mothers also
completed the Delinquent Behavior Scale from the Child Behavior Checklist (Achenbach,
1991). Alpha coefficients for the three scales at each wave ranged from .67 to .82, and their
Author Manuscript
psychometric properties are well established (Achenbach et al., 2003; Goodman & Scott,
1999).
Covariates: Maternal and paternal parenting quality: Maternal and paternal parenting
quality during the family problem-solving task at Wave 1 was rated using the IFIRS
Relationship Quality and Inductive Reasoning scales (Melby & Conger, 2001). Each code is
rated on a nine-point scale (1 = Not at all characteristic; 9 = Mainly characteristic). Whereas
high scores on the Relationship Quality code reflect warm and comfortable interactions
between parents and children, low relationship quality reflects unhappy, distressing, and
dissatisfying interactions. The Inductive Reasoning code assesses the degree to which the
parent sensitively structures interactions and explanations in ways that facilitates teens to
consider the consequences of their own behavior and the feelings and perspectives of others.
Author Manuscript
To assess interrater reliability, coders independently rated over 20% of the interactions. ICCs
ranged from .77 to .92 for maternal and paternal ratings of Relationship Quality and
Inductive Reasoning. Because the codes were assessed on the same scales, the two parenting
measures were averaged together for each parent to form composites of maternal and
paternal parenting quality (α = .64 and .63, respectively).
Covariates: Sociodemographic characteristics: Two covariates were derived from

parent reports of demographic characteristics: (1) children’s gender (1 = boys; 2 = girls) and
(2) annual family income based on a 13-point ordinal scale ranging from 1 (< $6,000) to 13
(≥ $125,000).
Results and Discussion

Author Manuscript
Table 2 provides the means, standard deviations, and correlations among the covariates,
interparental conflict characteristics, adolescents’ insecurity, and externalizing difficulties.
Consistent with Study 1, SEM analyses were conducted with the Amos 22.0 statistical
software program to examine the mediational role of children’s emotional insecurity in the
prospective pathways between the three forms of interparental conflict and externalizing
difficulties. Analyses utilized FIML to estimate missing data (i.e., data were missing for
13.0% of the values) and retain the full sample for primary analyses (Enders, 2001).

Following the same analytic approach as Study 1, the three interparental conflict
Author Manuscript
characteristics and four covariates were specified as primary predictors of teens’ emotional
insecurity at Wave 2 and their externalizing difficulties at Wave 3 (see Figure 2). An
additional structural path was estimated between Wave 2 insecurity and Wave 3
externalizing difficulties to test the second link in the mediational chain. Autoregressive
paths at Wave 1 were also included for adolescent insecurity and externalizing problems to
control for stability in the proposed mediator and outcome. Figure 2 also shows that we
included predictive paths among all four Wave 1 covariates and adolescent insecurity at
Wave 2 and their externalizing problems at Wave 3. As an additional covariate, a structural
path was also estimated between adolescent externalizing problems at Wave 1 and their
insecurity at Wave 2. As with Study 1, we implemented strong factorial invariance
constraints for emotional security and externalizing problem constructs to attain
correspondence in the measurement of constructs across time (Widaman et al., 2010).
Author Manuscript
Correlations were further specified among all Wave 1 variables (i.e., covariates and
predictors) in the model and between error terms of the manifest indicators for insecurity
and externalizing problems across the assessment waves.
The model, which is depicted in Figure 2, fit the data well, χ2 (109, N = 263) = 154.96, p < .
01, RMSEA = .04, CFI = .97, and χ2/df ratio = 1.42. For clarity, correlations are not
displayed in the figure. Standardized loadings of the manifest indicators onto their latent
constructs were all significant (p < .001) and generally high in magnitude (mean loading = .
73). Autoregressive paths were significant for teen insecurity, β = .47, p < .001, and their
externalizing problems, β = .79, p < .001. None of the Wave 1 covariates were significant
predictors of adolescent insecurity or their externalizing problems. In testing the first link in
the proposed mediational model, adolescent insecurity at Wave 2 was predicted by higher
levels of interparental hostility, β = .38, p < .001, and disengagement, β = .16, p = .05, at
Author Manuscript
Wave 1. Higher insecurity, in turn, was associated with more externalizing difficulties at
Wave 3, β = .21, p = .01. In further support of mediation, bootstrapping analyses of the
indirect links involving the two forms of interparental conflict, insecurity, and externalizing
problems were significantly different from 0 for (a) hostile conflict: unstandardized
coefficient = .13, 95% CI [.03, .25] and (b) disengaged conflict: unstandardized coefficient
= .05, 95% CI [.001, .13] (Preacher & Hayes, 2008).
Although Wave 1 poor interparental cooperation predicted greater teen externalizing

problems at Wave 3, β = .19, p < .05, it also was related to lower rather than higher
insecurity at Wave 2, β = −.21, p < .05. Given that the correlations between poor
interparental cooperation and the Wave 2 insecurity indicators were all significant and in the
expected (i.e., negative) direction (see Table 2), we examined whether the counterintuitive
Author Manuscript
finding resulted from multicollinearity. However, variance inflation factor (VIF) values for
poor interparental cooperation with each of the covariates and other predictors in the model
ranged from 1.07 to 2.42, all falling below even conservative standards (i.e., 2.50) for
problematic levels of multicollinearity (Allison, 1999). Next, we examined if the negative
links between uncooperative conflict and insecurity were the result of the variance it shared
with the other forms of conflict. Therefore, we re-ran the model in Figure 1 after eliminating
interparental hostility and disengagement from the analysis. The results indicated that the
sign of the structural path between Wave 1 uncooperative conflict and Wave 2 insecurity

changed to the expected direction (β = .11, p = .15) after excluding the other conflict forms.
Author Manuscript
Thus, the unexpected finding could not be reproduced in follow-up analyses.
Consistent with Study 1, we calculated pairwise parameter comparisons to test whether the
prospective association between interparental conflict and teen insecurity varied
significantly as a function of the form of conflict. One-tailed tests were used in the statistical
comparisons. As hypothesized, interparental hostility predicted insecurity more strongly
than interparental cooperation, z = 4.18, p < .001, and disengagement, z = 1.74, p < .05. In
further supporting predictions, interparental disengagement was a stronger predictor of
insecurity than uncooperative interparental conflict, z = 2.67, p < .01.
General Discussion
Multiple models of risky family processes have highlighted the importance of examining
Author Manuscript
children’s emotional insecurity as a mediating mechanism in associations among

interparental hostility, disengagement, and poor cooperation and children’s psychopathology
(Harold & Leve, 2012; Macfie, Brumariu, & Lyons-Ruth, 2015; Morris et al., 2007; Repetti
et al., 2011). However, research has yet to examine the relative roles of the three forms of
interparental conflict as predictors of mediational paths involving children’s insecurity and
psychopathology. Simultaneous inclusion of interparental hostility, disengagement, and poor
cooperation in the analyses of both of our studies consistently indicated that interparental
hostility was a significantly stronger predictor of the cascade of children’s insecurity and
behavior problems in comparison with interparental disengagement and uncooperativeness.
Thus, these findings are consistent with the EST-R hypothesis that children’s insecurity is
more likely to develop from earlier histories of interparental hostility than from prior
experiences with disengaged or uncooperative conflict (Davies et al., 2016). According to
Author Manuscript
EST-R, the emotional security system is selectively designed to contend with threat in
interpersonal contexts. As a result, the security system itself should be most sensitive to
imminent danger cues characterized by anger, yelling, dominant posturing, and aggression in
the interparental relationship. At a broader level, our empirical findings also correspond with
evidence that children are biologically prone to prioritize detecting and responding to anger
displays over other emotions (e.g., LoBue, 2009; Öhman & Mineka, 2001). For example,
LoBue (2009) showed that young children and adults are biased toward rapid identification
and processing of angry facial cues over happy or sad facial cues.
According to EST-R, interparental disengagement should also be a stronger predictor of

children’s insecurity and, in turn, their adjustment than variations in exposure to
interparental cooperation. Results from our two studies yielded partial support for this
hypothesis. Study 1 results indicated that mediational paths involving interparental
Author Manuscript
uncooperativeness and disengagement, child insecurity, and externalizing problems were

negligible in the early school years. However, Study 2 findings revealed that interparental
disengagement more strongly predicted mediational pathways involving adolescent
insecurity and behavior problems than did poor interparental cooperation. Thus, both studies
indicated that variability in cooperation, in and of itself, is unlikely to be a dependable sign
of threat. Moreover, the more modest and inconsistent role of interparental disengagement as
a predictor of insecurity across the two studies is broadly consistent with the hypothesis that

it is a less reliable prognosticator of imminent interpersonal threat than interparental hostility

Author Manuscript
(Davies et al., 2016; Dixon, 1998; Gilbert, 2001; LoBue, 2009).
Although definitive explanations for our findings will require additional research, the results
beg the question of why the potency of interparental disengagement as a precursor of the
insecurity cascades differs across the studies. From a developmental perspective, it is
possible that age moderates the mediational pathway between interparental disengagement,
children’s insecurity, and their externalizing problems. Early adolescence may usher in a
number of developmental processes that heighten children’s sensitivity to interparental
disengagement (Cummings, Schermerhorn, Davies, Goeke-Morey, & Cummings, 2006).
Some evidence suggests that adolescents may be more proficient in identifying subtle
expressions of interparental discord (e.g., disengagement, withdrawal) than their younger
counterparts (e.g., Cummings, Ballard, El-Sheikh, & Lake, 1991; El-Sheikh & Cummings,
1995). Research has also highlighted the disproportionate risk for relationship instability and
Author Manuscript
dissolution following histories of apathy, helplessness, and detachment between parents

(e.g., Christensen & Heavey, 1990; Gottman, 1993). Thus, by virtue of their increasing
sensitivity to interparental interactions, adolescents may be better able to recognize the
insidious repercussions of disengagement for the family and themselves. Furthermore,
adolescents’ greater tendencies to become involved in interparental disagreements may
further sensitize them to subtler forms (i.e., disengagement) of interparental discord (e.g.,
Buchanan, Maccoby, & Dornbusch, 1991; Cummings et al., 1991). Because both parents and
children report that the effectiveness of children’s efforts to mediate interparental conflicts
diminishes substantially from childhood into adolescence (Covell & Miles, 1992),
heightened involvement in conflicts may do little more than embroil adolescents in cold,
detached interactions between parents and, in the process, increase their insecurity.
Author Manuscript
However, study differences in interparental disengagement as a precursor of insecurity may

also be due to methodological variations across the studies. Although research has shown
that emotional insecurity operates similarly as a mediator of destructive interparental
conflict for children from a wide array of demographic backgrounds (Davies et al., 2016), it
is still possible that the lower SES levels and proportions of married families in Study 1
diluted the risk associated with interparental disengagement. Likewise, although we
attempted to maximize consistency in the measurement of many of the primary constructs
(e.g., interparental conflict), the design (e.g., three annual waves of data), and the analyses
(e.g., autoregressive design, inclusion of similar covariates), assessments of insecurity
differed across the studies due to the age differences of the children. Following previous
recommendations for assessing security at different ages (Davies et al., 2006), we relied on
behavioral indicators (e.g., emotional reactivity, avoidance) to assess young children’s
Author Manuscript
insecurity in Study 1 and adolescent reports of their insecurity (e.g., emotional reactivity,
avoidance) in Study 2.
Given our empirical identification of children’s emotional insecurity as a consistent

predictor of their externalizing symptoms in both childhood and adolescence, it is also
important to consider the question of how this pathogenic process unfolds. More
specifically, why does insecurity, which is rooted in fear and anxiety, increase children’s
vulnerability to disruptive behavior problems? One possibility is that children high in

emotional insecurity may adopt their highly defensive ways of defending against
Author Manuscript
interparental conflict as blueprints for interpreting and responding to new or challenging

social contexts outside of the interparental relationship (Johnston, Roseby, & Kuehnle,
2009). For example, hostile processing of stressful peer events has been shown to mediate
the association between children’s insecurity in the interparental subsystem and school
maladjustment (e.g., uncooperative behavior, attention problems) over a one-year period
(Bascoe, Davies, Sturge-Apple, & Cummings, 2009). It is also possible that prolonged
concerns about insecurity in the interparental relationship may increase children’s behavior
problems by undermining their abilities to master approach-oriented goals (Davies et al.,
2016). Supporting this hypothesis, insecurity has been shown to increase children’s
vulnerability to externalizing problems through its association with impairments in
exploration and problem solving during play and cognitive tasks (e.g., Davies, Manning, &
Cicchetti, 2013).
Author Manuscript
Further discussion of the limitations of our paper is also necessary to fully interpret the
findings. First, more research is needed to identify why the studies yielded different results
for interparental disengagement as a predictor. Second, despite fairly wide diversity in the
racial and demographic backgrounds of families in the paper, caution should be exercised in
generalizing the findings to high-risk or clinical samples of children. Third, even with the
inclusion of a number of predictors and covariates in the analyses, our longitudinal designs
do not rule out all potential confounding variables (e.g., genetic mechanisms, temperament).
Fourth, based on prior work, we did not expect that interparental uncooperativeness in Study
2 would predict lower child insecurity in the context of interparental hostility and
disengagement. Given that this finding was not reproduced in any of our other analyses (i.e.,
Study 1 results, Study 2 correlations and models excluding the other forms of conflict as
predictors), any definitive conclusions about this result will require further research. Finally,
Author Manuscript
our identification of the modest to negligible roles of interparental disengagement and

uncooperativeness as sources of children’s insecurity should not be misinterpreted as
evidence that they have no significant health implications for children or families. Although
EST-R hypothesizes that child concerns about security hinge heavily on experiential
histories with interparental hostility, it also proposes that other forms of interparental
conflict (e.g., disengagement, minimal cooperation) increase children’s risk for
psychological difficulties through other mechanisms. For example, interparental
disengagement and poor cooperation may increase psychopathology by undermining
parenting practices and children’s approach-oriented (e.g., social affiliation, exploration)
goals (Davies et al., 2013; McCoy, George, Cummings, & Davies, 2013). Supporting this
possibility, uncooperative interparental conflict in Study 2 did significantly predict higher
levels of teen externalizing symptoms over time.
Author Manuscript
In closing, this paper was designed to break new ground by testing the relative strength of
mediational pathways involving hostile, disengaged, and uncooperative forms of
interparental conflict, children’s emotional insecurity, and their externalizing problems
across two multi-method longitudinal studies. The results highlight the value of
distinguishing between different forms of interparental conflict in understanding differences
in developmental pathways of children’s coping and psychopathology. From a clinical
perspective, the findings may also have important translational implications for alleviating

the burden of child psychopathology. Children’s concerns about safety are commonly
Author Manuscript
viewed as targets of change allaying their anxiety and depressive symptoms. However, our
mediational findings suggest that adding security modules to prevailing behavioral
approaches for treating externalizing problems may also reduce children’s disruptive
behaviors (e.g., parenting training or cognitive-behavioral programs; Lochman, Powell,
Boxmeyer, & Carnargo, 2011; Menting, de Castro, & Matthys, 2013). Given the growing
number of treatment programs that emphasize promoting child security in the family (e.g.,
Coatsworth, 2013; Cummings & Schatz, 2012; Johnston et al., 2009; Lieberman, Van Horn,
& Ippen, 2005), the present results also underscore the potential merits of prioritizing
clinical changes in specific forms of interparental conflict as a way of stopping the
pathogenic cascade of insecurity in the development of behavior problems. Thus, reducing
interparental hostility and, to a lesser degree, interparental disengagement may be regarded
as a stronger clinical priority than enhancing interparental cooperation for alleviating
Author Manuscript
children’s safety concerns in the family and their disruptive problems (Coatsworth, 2013).
References
Ablow JC, Measelle JR, Kraemer HC, Harrington R, Luby J, Smider N, … Kupfer DJ. The MacArthur
Three-City Outcome Study: Evaluating multi-informant measures of young children’s
symptomatology. Journal of the American Academy of Child & Adolescent Psychiatry. 1999;
38:1580–1590. DOI: 10.1097/00004583-199912000-00020 [PubMed: 10596259]
Achenbach, TM. Manual for the Child Behavior Checklist/ 4–18 and 1991 Profile. Burlington, VT:
University of Vermont, Department of Psychiatry; 1991.
Achenbach TM, Dumenci L, Rescorla LA. DSM-oriented and empirically based approaches to
constructing scales from the same item pools. Journal of Clinical Child and Adolescent Psychology.
2003; 32(3):328–340. DOI: 10.1207/S15374424JCCP3203_02 [PubMed: 12881022]
Allsion, PD. Multiple regression: A primer. Thousand Oaks, CA: Pine Forge Press; 1999.
Arellano CM, Markman HJ. The managing affect and differences scale (MADS): A self-report
Author Manuscript
measure assessing conflict management in couples. Journal of Family Psychology. 1995; 9:319–
334. DOI: 10.1037/0893-3200.9.3.319
Bascoe SM, Davies PT, Sturge-Apple ML, Cummings EM. Children’s representations of family
relationships, peer information processing, and school adjustment. Developmental Psychology.
2009; 45:1740–1751. DOI: 10.1037/a0016688 [PubMed: 19899928]
Buchanan CM, Maccoby EE, Dornbusch SM. Caught between parents: Adolescents’ experience in
divorced homes. Child Development. 1991; 62:1008–1029. DOI: 10.1111/j.
1467-8624.1991.tb01586.x [PubMed: 1756653]
Buehler C, Anthony C, Krishnakumar MS, Stone G, Gerard J, Pemberton S. Interparental conflict and
youth problem behaviors: A meta-analysis. Journal of Child and Family Studies. 1997; 2:233–247.
doi: http://dx.doi.org/10.1023/A:1025006909538.
Buehler C, Krishnakumar A, Stone G, Anthony C, Pemberton S, Gerard J, Barber BK. Interparental
Conflict Styles and Youth Problem Behaviors: A Two-Sample Replication Study. Journal of
Marriage and Family. 1998; 60:119–132. DOI: 10.2307/353446
Author Manuscript
Buehler C, Lange G, Franck KL. Adolescents’ cognitive and emotional responses to marital hostility.
Child Development. 2007; 78:775–789. DOI: 10.1111/j.14678624.2007.01032.x [PubMed:
17517004]
Campbell SB, Shaw DS, Gilliom M. Early externalizing behavior problems: Toddlers and preschoolers
at risk for later maladjustment. Development and psychopathology. 2000; 12:467–488. [PubMed:
11014748]
Christensen A, Heavey CL. Gender and social structure in the demand/withdraw pattern of marital
conflict. Journal of Personality and Social Psychology. 1990; 59:73–81. DOI:
10.1037/0022-3514.59.1.73 [PubMed: 2213491]

Cicchetti, D.; Cummings, EM.; Greenberg, MT.; Marvin, RS. An organizational perspective on
attachment beyond infancy. In: Greenberg, M.; Cicchetti, D.; Cummings, EM., editors. Attachment
Author Manuscript
in the preschool years: Theory, research, and intervention. Chicago: The University of Chicago
Press; 1990. p. 3-49.
Cicchetti D, Toth SL. The past achievements and future promises of developmental psychopathology:
The coming of age of a discipline. Journal of Child Psychology and Psychiatry. 2009; 50:16–25.
DOI: 10.1111/j.1469-7610.2008.01979.x [PubMed: 19175810]
Coatsworth. Could emotional security theory help advance family-focused prevenetative
interventions?. In: Lansdale, NS.; McHale, SM.; Booth, A., editors. Families and child health.
New York: Springer; 2013. p. 95-104.
Cole DA, Maxwell SE. Testing mediational models with longitudinal data: Questions and tips in the
use of structural equation modeling. Journal of Abnormal Psychology. 2003; 112:558–577. DOI:
10.1037/0021-843X.112.4.558 [PubMed: 14674869]
Covell K, Miles B. Children’s beliefs about strategies to reduce parental anger. Child Development.
1992; 63:381–390. DOI: 10.1111/j.1467-8624.1992.tb01634.x [PubMed: 1611941]
Cummings EM, Ballard M, El-Sheikh M, Lake M. Resolution and children’s responses to interadult
Author Manuscript
anger. Developmental Psychology. 1991; 27:462–470. DOI: 10.1037/0012-1649.27.3.462

Cummings, EM.; Davies, PT. Marital conflict and children: An emotional security perspective. New
York: Guilford Press; 2011.
Cummings EM, Miller-Graff LE. Emotional Security Theory An Emerging Theoretical Model for
Youths’ Psychological and Physiological Responses Across Multiple Developmental Contexts.
Current Directions in Psychological Science. 2015; 24:208–213. DOI:
10.1177/0963721414561510
Cummings EM, Goeke-Morey MC, Papp LM. Everyday marital conflict and child aggression. Journal
of Abnormal Child Psychology. 2004; 32:191–202. DOI: 10.1023/B:JACP.0000019770.13216.be
[PubMed: 15164860]
Cummings EM, Schatz JN. Family conflict, emotional security, and child development: Translating
research findings into a prevention program for community families. Clinical Child and Family
Psychology Review. 2012; 15:14–27. DOI: 10.1007/s10567-012-0112-0 [PubMed: 22311087]
Cummings EM, Schermerhorn AC, Davies PT, Goekey-Morey MC, Cummings JS. Interparental
discord and child adjustment: Prospective investigations of emotional security as an explanatory
Author Manuscript
mechanism. Child Development. 2006; 77:132–152. DOI: 10.1111/j.1467-8624.2006.00861.x

[PubMed: 16460530]
Cummings EM, Vogel D, Cummings JS, El-Sheik M. Children’s responses to different forms of
expression of anger between adults. Child Development. 1989; 60:1392–1404. DOI:
10.2307/1130929 [PubMed: 2612248]
Davies PT, Coe JL, Martin MJ, Sturge-Apple ML, Cummings EM. The developmental costs and
benefits of children’s involvement in interparental conflict. Developmental Psychology. 2015;
51:1026–1047. DOI: 10.1037/a0016688 [PubMed: 26053147]
Davies PT, Cummings EM. Marital conflict and child adjustment: An emotional security hypothesis.
Psychological Bulletin. 1994; 116:387–411. DOI: 10.1037/0033-2909.116.3.387 [PubMed:
7809306]
Davies PT, Forman EM, Rasi JA, Stevens KI. Assessing children’s emotional security in the
interparental subsystem: The Security in the Interparental Subsystem (SIS) Scales. Child
Development. 2002; 73:544–562. DOI: 10.1111/1467-8624.00423 [PubMed: 11949908]
Author Manuscript
Davies PT, Manning LG, Cicchetti D. Tracing the cascade of children’s insecurity in the interparental
relationship: The role of stage-salient tasks. Child Development. 2013; 84:297–312. DOI:
10.1111/j.1467-8624.2012.01844.x [PubMed: 22925122]
Davies PT, Martin MJ, Cicchetti D. Delineating the sequelae of destructive and constructive
inteparental conflict for children within an evolutionary framework. Developmental Psychology.
2012; 48:939–955. DOI: 10.1037/a0025899 [PubMed: 22004336]
Davies, PT.; Martin, MJ.; Sturge-Apple, ML. Emotional security theory and developmental
psychopathology. In: Cicchetti, D.; Cohen, DJ., editors. Developmental Psychopathology: Vol. 1:
Theory and Method. 3. New York: Wiley; 2016. p. 199-264.

Davies PT, Sturge-Apple ML. Advances in the formulation of emotional security theory: An
ethologically-based perspective. Advances in Child Development and Behavior. 2007; 35:87–137.
Author Manuscript
DOI: 10.1016/B978-0-12-009735-7.50008-6 [PubMed: 17682324]

Davies, PT.; Sturge-Apple, ML. Interparental conflict and emotion (ICE) coding system. University of
Rochester; Rochester, NY: 2013. Unpublished manual
Davies PT, Sturge-Apple ML, Cicchetti D, Manning LG, Vonhold SE. Pathways and processes of risk
in associations among maternal antisocial personality symptoms, interparental aggression, and
preschooler’s psychopathology. Development and Psychopathology. 2012; 24:807–832. DOI:
10.1017/S0954579412000387 [PubMed: 22781856]
Davies PT, Sturge-Apple ML, Winter MA, Cummings EM, Farrel D. Child adaptational development
in contexts of interparental conflict over time. Child Development. 2006; 77:218–233. DOI:
10.1111/j.1467-8624.2006.00866.x [PubMed: 16460535]
Dixon AK. Ethological strategies for defence in animals and humans: Their role in some psychiatric
disorders. British Journal of Medical Psychology. 1998; 71:417–445. DOI: 10.1111/j.
2044-8341.1998.tb01001.x [PubMed: 9875954]
Doshi JA, Hodgkins P, Kahle J, Sikrica V, Cangelosi J, … Neumann PJ. Economic impact of childhood
Author Manuscript
and adult attention deficit/hyperactivity disorder in the United States. Journal of the American
Academy of Child and Adolescent Psychiatry. 2012; 51:990–1002. DOI: http://dx.doi.org/10.1016/
j.jaac.2012.07.008. [PubMed: 23021476]
Du Rocher Schudlich T, Cummings EM. Parental dysphoria and children’s adjustment: Marital
conflict styles, children’s emotional security, and parenting as mediators of risk. Journal of
Abnormal Child Psychology. 2007; 35:627–639. DOI: 10.1007/s10802-007-9118-3 [PubMed:
17390219]
El-Sheikh M, Cummings EM. Children’s responses to angry adult behavior as a function of
experimentally manipulated exposure to resolved and unresolved conflict. Social Development.
1995; 4:75–91. DOI: 10.1111/j.1467-9507.1995.tb00052.x
El-Sheikh M, Cummings EM, Kouros CD, Elmore-Staton L, Buckhalt J. Marital psychological and
physical aggression and children’s mental and physical health: Direct, mediated, and moderated
effects. Journal of Consulting and Clinical Psychology. 2008; 76:138–148. DOI:
10.1037/0022-006X.76.1.138 [PubMed: 18229991]
Enders CK. A primer on maximum likelihood algorithms for use with missing data. Structural
Author Manuscript
Equation Modeling. 2001; 8:128–141. DOI: 10.1207/S15328007SEM0801_7

Foster EM, Jones DE, Bierman KL, Coie JD. the Conduct Problems Prevention Research Group. The
high costs of aggression: Public expenditures resulting from conduct disorder. American Journal of
Public Health. 2005; 95:1767–72. DOI: 10.2105/AJPH.2004.061424 [PubMed: 16131639]
Gilbert P. Depression and stress: a biopsychosocial exploration of evolved functions and mechanisms.
Stress: The International Journal of the Biology of Stress. 2001; 4:121–135.
Goeke-Morey MC, Cummings EM, Papp LM. Children and marital conflict resolution: Implications
for emotional security and adjustment. Journal of Family Psychology. 2007; 21:744–753. DOI:
10.1037/0893-3200.21.4.744 [PubMed: 18179346]
Goodman R. Psychometric properties of the strengths and difficulties questionnaire. Journal of the
American Academy of Child & Adolescent Psychiatry. 2001; 40:1337–1345. DOI:
10.1097/00004583-200111000-00015 [PubMed: 11699809]
Goodman R, Scott S. Comparing the Strengths and Difficulties Questionnaire and the Child Behavior
Checklist: is small beautiful? Journal of abnormal child psychology. 1999; 27:17–24. DOI:
Author Manuscript
10.1023/A:1022658222914 [PubMed: 10197403]

Gordis EB, Margolin G, John RS. Parents’ hostility in dyadic marital and triadic family settings and
children’s behavior problems. Journal of Consulting and Clinical Psychology. 2001; 69:727–734.
DOI: 10.1037/0022-006X.69.4.727 [PubMed: 11550741]
Gottman JM. A theory of marital dissolution and stability. Journal of Family Psychology. 1993; 7:57–
75. DOI: 10.1037/0893-3200.7.1.57
Grych JH, Fincham FD. Marital conflict and children’s adjustment: A cognitive-contextual framework.
Psychological Bulletin. 1990; 108:267–290. DOI: 10.1037/0033-2909.108.2.267 [PubMed:
2236384]

Grych JH. Children’s appraisals of interparental conflict: Situational and contextual influences. Journal
of Family Psychology. 1998; 12:437–453. DOI: 10.1037/0893-3200.12.3.437
Author Manuscript
Grych JH, Seid M, Fincham FD. Assessing marital conflict from the child’s perspective: The
Children’s Perception of Interparental Conflict Scale. Child Development. 1992; 63:558–572.
DOI: 10.2307/1131346 [PubMed: 1600822]
Harold, GT.; Leve, LD. Parents as partners: How the parental relationship affects children’s
psychological development. In: Balfour, A.; Morgan, M.; Vincent, C., editors. How Couples
Relationships Shape Our World: Clinical Practice, Research, and Policy Perspectives. London:
Karnac Books; 2012. p. 25-55.
Johnston, J.; Roseby, V.; Kuehnle, K. In the name of the child: A developmental approach to
understanding and helping children of conflict and violent divorce. New York, NY: Springer
Publishing; 2009.
Jouriles EN, Rosenfield D, Mcdonald R, Mueller V. Child involvement in interparental conflict and
child adjustment problems: A longitudinal study of violent families. Journal of Abnormal Child
Psychology. 2014; 42:693–704. DOI: 10.1007/s10802-013-9821-1 [PubMed: 24249486]
Kelly RJ, El-Sheikh M. Longitudinal relations between marital aggression and children’s sleep: The
Author Manuscript
role of emotional insecurity. Journal of Family Psychology. 2013; 27:282–292. DOI: 10.1037/
a0031896 [PubMed: 23458697]
Kerig P. Assessing the links between interparental conflict and child adjustment: The Conflict and
Problem-Solving Scales. Journal of Family Psychology. 1996; 10:454–473. DOI:
10.1037/0893-3200.10.4.454
Lieberman AF, Van Horn P, Ippen CG. Toward evidence-based treatment: Child-parent psychotherapy
with preschoolers exposed to marital violence. Journal of the Academy of Child and Adolescent
Psychiatry. 2005; 44:1241–1248. DOI: 10.1097/01.chi.0000181047.59702.58
Lindahl KM, Malik NM, Kaczynski K, Simons JS. Couple power dynamics, systematic family
functioning, and child adjustment: A test of a mediational model in a multiethnic sample.
Development and Psychopathology. 2004; 16:609–630. doi:10.10170S0954579404004699.
[PubMed: 15605628]
LoBue V. More than just another face in the crowd: Superior detection of threatening facial
expressions in children and adults. Developmental Science. 2009; 12:305–313. DOI: 10.1111/j.
1467-7687.2008.00767.x [PubMed: 19143803]
Author Manuscript
Lochman JE, Powell NP, Boymeyer CL, Camargo LJ. Cognitive-behavioral therapy for externalizing
disorders in children and adolescents. Child and Adolescent Psychiatric Clinics of North America.
2011; 20:305–318. DOI: 10.1016/j.chc.2011.01.005 [PubMed: 21440857]
Macfie J, Brumariu LE, Lyons-Ruth K. Parent-child role-confusion: A critical review of an emerging
concept. Developmental Review. 2015; 36:34–57. DOI: 10.1016/j.dr.2015.01.002
Malik, NM.; Lindahl, KM. System for Coding Interactions in Dyads (SCID). In: Kerig, PK.; Baucom,
DH., editors. Couple observational coding systems. Mahwah, NJ: Erlbaum; 2004. p. 173-188.
Maxwell SE, Cole DA. Bias in cross-sectional analyses of longitudinal mediation. Psychological
Methods. 2007; 12:23–44. DOI: 10.1037/1082-989X.12.1.23 [PubMed: 17402810]
McCoy K, Cummings EM, Davies PT. Constructive and destructive marital conflict, emotional
security and children’s prosocial behavior. Journal of Child Psychology and Psychiatry. 2009;
50:270–279. DOI: 10.1111/j.1469-7610.2008.01945.x [PubMed: 18673403]
McCoy KP, George MR, Cummings EM, Davies PT. Constructive and destructive marital conflict,
parenting, and children’s school and social adjustment. Social Development. 2013; 22:641–662.
Author Manuscript
DOI: 10.1111/sode.12015
McHale, J.; Kuersten-Hogan, R.; Lauretti, A. Evaluating coparenting and family-level dynamics during
infancy and early childhood: The coparenting and family rating system. In: Kerig, PK.; Lindahl,
KM., editors. Family observational coding systems: Resources for systemic research. Hillsdale NJ:
Lawrence Erlbaum Associates; 2001. p. 147-166.
Melby, JN.; Conger, RD. The Iowa Family Interaction Rating Scales: Instrument summary. In: Kerig,
PK.; Lindahl, KM., editors. Family observational coding systems: Resources for systemic
research. Mahwah, NJ: Erlbaum; 2001. p. 33-58.

Menting ATA, de Castro BO, Matthys W. Effectiveness of the Incredible Years parent training to
modify disruptive and prosocial child behavior: A meta-analytic review. Clinical Psychology
Author Manuscript
Review. 2013; 33:901–913. DOI: 10.1016/j.cpr.2013.07.006 [PubMed: 23994367]

Morris AS, Silk JS, Steinberg L, Myers SS, Robinson LR. The role of the family context in the
development of emotion regulation. Social Development. 2007; 16:361–388. DOI: 10.1111/j.
1467-9507.2007.00389.x [PubMed: 19756175]
Ohman A, Mineka S. Fears, phobias, and preparedness: Toward an evolved module of fear and fear
learning. Psychological Review. 2001; 108:483–522. DOI: 10.1037/0033-295X.108.3.483
[PubMed: 11488376]
Porter B, O’Leary KD. Marital discord and childhood behavior problems. Journal of Abnormal Child
Psychology. 1980; 8:287–295. [PubMed: 7410730]
Preacher KJ, Hayes AF. Asymptotic and resampling strategies for assessing and comparing indirect
effects in multiple mediator models. Behavior Research Methods. 2008; 40:879–891. DOI:
10.3758/BRM.40.3.879 [PubMed: 18697684]
Preacher KJ. Advances in mediation analysis: a survey and synthesis of new developments. Annual
Review of Psychology. 2015; 66:825–852. DOI: 10.1146/annurev-psych-010814-015258
Author Manuscript
Repetti RL, Robles TF, Reynolds B. Allostatic processes in the family. Development and
Psychopathology. 2011; 23:921–938. DOI: 10.1017/S095457941100040X [PubMed: 21756442]
Schoppe SJ, Mangelsdorf SC, Frosch CA. Coparenting, family process, and family structure:
implications for preschoolers’ externalizing behavior problems. Journal of Family Psychology.
2001; 15:526–545. DOI: 10.1037/0893-3200.15.3.526 [PubMed: 11584800]
Straus MA, Hamby SL, Boney-McCoy S, Sugarman DB. The revised conflict tactics scales (CTS2):
Development and preliminary psychometric data. Journal of Family Issues. 1996; 17:283–316.
DOI: 10.1177/019251396017003001
Sturge-Apple ML, Davies PT, Winter MA, Cummings EM, Schermerhorn A. Interparental conflict and
children’s school adjustment: The explanatory role of children’s internal representations of
interparental and parent-child relationships. Developmental Psychology. 2008; 44:1678–1690.
DOI: 10.1037/a0013857 [PubMed: 18999330]
Webster-Stratton C, Hammond M. Marital conflict management skills, parenting style, and early-onset
conduct problems: Processes and pathways. Journal of Child Psychology and Psychiatry. 1999;
40:917–927. [PubMed: 10509886]
Author Manuscript
Widaman KF, Ferrer E, Conger RD. Factorial invariance within longitudinal structural equation
models: Measuring the same construct across time. Child Development Perspectives. 2010; 4:10–
18. DOI: 10.1111/j.1750-8606.2009.00110.x [PubMed: 20369028]
Author Manuscript

General Scientific Summary

Author Manuscript
Although there has been a longstanding interest in understanding how children’s sense of
insecurity in the family plays a role in explaining why exposure to interparental conflict
increases their vulnerability to behavior problems, little is known about how specific
forms of conflict between parents serve as unique antecedents of their psychological
difficulties. In two studies, we found that interparental hostility more strongly predicted
children’s insecurity and, in turn, their behavior problems than interparental
disengagement or poor cooperation. For the sample of adolescents in Study 2,
interparental disengagement was also a stronger precursor of their insecurity and
ultimately their externalizing problems than was poor cooperation between parents.
Author Manuscript
Author Manuscript
Author Manuscript

Author Manuscript
Author Manuscript
Figure 1.
An autoregressive structural equation model examining children’s emotional insecurity as a
mediator in prospective pathways between interparental hostility, disengagement, and poor
Author Manuscript
cooperation and their externalizing difficulties in Study 1. Tch. = Teacher; Cond. = Conduct
Problems Sale; Em. React. = Emotional Reactivity. * p < .05 for structural paths in the
figure.
Author Manuscript

Author Manuscript
Author Manuscript
Author Manuscript
Figure 2.
An autoregressive structural equation model examining adolescent emotional insecurity as a
mediator in prospective pathways between interparental hostility, disengagement, and poor
cooperation and their externalizing difficulties in Study 2. Tch. = teacher; Conduct =
Conduct Problems; Delinq = Delinquency. * p < .05 for structural paths in the figure.
Author Manuscript

Author Manuscript Author Manuscript Author Manuscript Author Manuscript
Table 1
Means, Standard deviations, and Correlations for the Primary Variables in the Study 1 Analyses.
M SD 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Davies et al.
Wave 1 Covariates
1. Child Gender 1.44 0.50 --
2. Family Income 41.51 23.54 .07 --
(thousands)
3. Maternal Parenting 3.84 1.78 −.06 .38* --
Quality
4. Paternal Parenting 3.57 1.89 −.01 .39* .38* --
Quality
Wave 1 Interparental Conflict
5. Hostile 0.00 0.65 .01 −.17* −.14* −.12 --
6. Disengaged 0.00 0.56 −.04 −.30* −.27* −.18* .70* --
7. Uncooperative 0.00 0.59 .01 −.09 −.16* −.19* .62* .60* --
Wave 1 Child Emotional Security

8. Emotional Reactivity 0.00 0.93 −.09 −.09 −.14* −.02 .16* .14* .06 --
9. Avoidance 0.00 0.91 .02 −.15* −.25* −.13* .22* .25* .10 .20* --
10. Overall Security 5.31 1.96 .03 .05 .03 .06 −.12 −.09 −.10 −.45* −.29* --

11. Emotional Reactivity 0.00 0.89 −.03 −.07 −.17* −.14* .28* .15* .15* .28* .11 −.15* --
12. Avoidance 0.00 0.92 .08 −.03 −.24* −.13 .19* .22* .17* .07 .39* −.11 .28* --
13. Overall Security 4.96 2.05 .01 .15* .15* .17* −.33* −23* −.20* −.21* −.13 .27* −.68* −.30* --

Wave 1 Child Externalizing Problems
14. Conduct Problems 1.54 3.40 .12 −.07 −.09 −.08 .04 .03 .11 −.05 −.01 .01 .02 .07 −.03 --
(Teacher)
15. ADHD Symptoms 6.46 6.66 .08 −.12 −.09 −.22* .03 −.06 .09 −.10 −.09 −.02 .02 .10 .00 .63* --
(Teacher)
16. Conduct Problems 1.77 2.26 .10 −.11 −.08 −.04 .19* .18* .15* .11 .13* −.12 .14 .16* −.16* .31* .25* --
(Mother)
17. Conduct Problems 1.75 3.31 −.02 −.27* −.21* −.23* .12 .15* .09 .10 .13 −.18* .28* .02 −.22* .38* .25* .36* --
(Teacher)
Page 28
M SD 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
18. ADHD Symptoms 7.51 7.76 .05 −.27* −.18* −.21* .04 .11 .03 .01 .01 −.11 .10 −.08 −.14 .38* .44* .26* .72* --
(Teacher)
19. Conduct Problems 1.67 2.64 .06 −.13* −.09 −.06 .22* .19* .11 .10 .14* −.11 .17* .10 −.21* .30* .18* .65* .52* .39*
(Mother)
Davies et al.
Note. Child gender: 1= girls; 2 = boys.

*
p ≤ .05.

Page 29
Table 2
Means, Standard deviations, and Correlations for the Primary Variables in the Study 2 Analyses.
M SD 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Davies et al.
Wave 1 Covariates
1. Child Gender 0.50 0.50 --
2. Family Income 8.47 3.15 −.06 --
3. Maternal Parenting Quality 4.36 1.63 −.10 .34* --
4. Paternal Parenting Quality 3.86 1.61 −.09 .30* .31* --
Wave 1 Interparental Conflict

5. Hostile 0.02 0.66 −.02 −.17* −.30* −.22* --
6. Disengaged 0.02 0.64 .04 −.31* −.23* −.18* .49* --
7. Uncooperative −0.01 0.67 .11 −.16* −.31* −.20* .60* .55* --

8. Emotional Reactivity 14.65 5.62 .09 −.15* −.12 −.01 .18* .13* .22* --
9. Avoidance 15.40 5.32 .01 −.13* −.04 −.01 .15* .08 .15* .61* --
10. Insecure Representations 5.74 2.58 .00 −.28* −.19* −.02 .22* .23* .21* .73* .49* --

11. Emotional Reactivity 14.13 5.57 .03 −.13 −.15* −.03 .42* .32* .21* .48* .38* .41* --
12. Avoidance 15.11 5.48 −.01 .04 −.09 −.03 .34* .16* .17* .43* .47* .31* .69* --
13. Negative Representations 5.58 2.44 .04 −.29* −.21* −.14* .31* .30* .20* .33* .22* .43* .70* .54* --

14. Conduct Problems (Teacher) 0.56 1.37 −.12 −.31* −.25* −.22* .27* .20* .22* .08 .08 .20* .09 .03 .12 --
15. SDQ Hyperactive (Teacher) 2.27 2.37 −.27* −.19* −.11 −.11 .18* .16* .13* .11 .07 .22* .05 .00 .08 .54* --
16. Delinquency (Mother) 1.28 1.77 −.18* −.22* −.18* −.13* .18* .15* .04 .04 .02 .16* .13 −.03 .19* .33* .34* --

17. Conduct Problems (Teacher) 0.50 1.20 −.07 −.28* −.06 −.21* .11 .09 .10 .17* .08 .26* .22* .10 .26* .47* .35* .28* --
18. Hyperactive (Teacher) 2.26 2.34 −.19* −.17* −.06 −.12 .09 .13 .08 .14* .08 .24* .10 .08 .21* .38* .48* .17* .58* --
19. Delinquency (Mother) 1.37 2.12 −.08 −.26* −.17* −.13 .20* .06 .18* .19* .07 .25* .21* .03 .29* .42* .29* .62* .37* .22*
Note. Child gender: 0= boys; 1 = girls.

Page 30
p ≤ .05.
* Davies et al. Page 31
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript

Nihms 779728

Uploaded by

Copyright:

Available Formats

Nihms 779728

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Nihms 779728

Uploaded by

Copyright:

Available Formats

HHS Public Access

Published in final edited form as:

The Multiple Faces of Interparental Conflict: Implications for

Melissa L. Sturge-Apple, and

developmental models of family risk.

Children with externalizing difficulties characterized by conduct problems, aggression, and

The widespread conceptual emphasis placed on children’s emotional insecurity as a

to EST, destructive interparental conflict undermines children’s ability to preserve their

J Abnorm Psychol. Author manuscript; available in PMC 2017 July 01.

As a further distinction, EST-R posits that interparental disengagement will be a modest

interparental relationship than variations in uncooperative conflict (Dixon, 1998). On the

J Abnorm Psychol. Author manuscript; available in PMC 2017 July 01.

method, multi-informant measurement approach. Finally, as a step toward examining the

and externalizing problems (e.g., Webster-Statton & Hammond, 1999; Sturge-Apple,

J Abnorm Psychol. Author manuscript; available in PMC 2017 July 01.

Procedures—Families visited our research laboratory at each measurement occasion.

Interparental conflict task: Parents participated in an interparental conflict task in which

they discussed common, problematic disagreements in their relationship for 10 minutes

Interparental disagreement interview: At Wave 1, mothers completed the Interparental

J Abnorm Psychol. Author manuscript; available in PMC 2017 July 01.

Questionnaire assessments: Mothers and fathers also completed survey measures of

interparental conflict and demographic characteristics at Wave 1. During Waves 1 and 3,

Measures—We used a multi-method, multi-informant approach to obtain

assessment, interparental hostility, disengagement, and poor cooperation variables each

J Abnorm Psychol. Author manuscript; available in PMC 2017 July 01.

J Abnorm Psychol. Author manuscript; available in PMC 2017 July 01.

Fearful Distress. Coded along a five-point scale (1 = No vigilance; 5 = Intense vigilance),

J Abnorm Psychol. Author manuscript; available in PMC 2017 July 01.

standardized and averaged together to create a composite of emotional reactivity at each

As the second indicator of insecurity, coders rated children’s avoidance of interparental

J Abnorm Psychol. Author manuscript; available in PMC 2017 July 01.

Covariates: Sociodemographic characteristics: Two demographic covariates derived

Results and Discussion

specified as a predictor of Wave 3 externalizing problems. Autoregressive paths using Wave

J Abnorm Psychol. Author manuscript; available in PMC 2017 July 01.

In contrast to the significant findings for interparental hostility, uncooperative and

J Abnorm Psychol. Author manuscript; available in PMC 2017 July 01.

In addition, developmental psychopathology models underscore the importance of

adolescents. Most adolescents were White (74%), followed by smaller percentages of

interparental conflict, parenting), child (e.g., insecurity, externalizing problems), and

J Abnorm Psychol. Author manuscript; available in PMC 2017 July 01.

Interparental conflict task: During Wave 1, mothers and fathers participated in an

Interparental disagreement interview: At Wave 1, mothers completed the same

Questionnaire assessments: Mothers and fathers filled out surveys of interparental

Family problem-solving task: At Wave 1, mothers, fathers, and adolescents participated

Measures—Consistent with Study 1, we used a multi-method, multi-informant approach

J Abnorm Psychol. Author manuscript; available in PMC 2017 July 01.

composite of hostile interparental conflict (α = .68).

Disengaged interparental conflict: For the observational assessment of the interparental

Uncooperative interparental conflict: The uncooperative conflict assessment followed

Adolescent insecurity in the interparental relationship: In contrast to the younger

J Abnorm Psychol. Author manuscript; available in PMC 2017 July 01.

Adolescent externalizing problems: At Waves 1 and 3, teachers completed the Conduct

Covariates: Sociodemographic characteristics: Two covariates were derived from

Results and Discussion

J Abnorm Psychol. Author manuscript; available in PMC 2017 July 01.

Although Wave 1 poor interparental cooperation predicted greater teen externalizing

J Abnorm Psychol. Author manuscript; available in PMC 2017 July 01.

Thus, the unexpected finding could not be reproduced in follow-up analyses.