Play Therapy Interventions for Family Separation
Play Therapy Interventions for Family Separation
Order Description
1.Locate at least three articles that involve play therapy interventions for divorced families or families that have experienced a separation. share important information from those articles.
2. Discuss at least two evidence-based play therapy interventions for use with children who have experienced divorce and why are they effective (Support with peer-reviewed research)
Linking Marital Conflict and Children’s Adjustment:
The Role of Young Children’s Perceptions
Jennifer C. Ablow and Jeffrey R. Measelle
University of Oregon
Philip A. Cowan and Carolyn P. Cowan
University of California at Berkeley
Young children’s (n 96) perceptions and appraisals of their parents’ marital conflict were
evaluated at age 5 and again at age 6. Concurrent reports of marital conflict by each parent
and teachers’ reports of children’s classroom adjustment served as criteria against which to
evaluate the validity of young children’s perceptions. Children’s perceptions of their parents’
marital relationship were significantly correlated with spouses’ reports at ages 5 and 6, as well
as correlated with teacher reports of internalizing and externalizing problems. Consistent with
the cognitive– contextual theory, children’s tendency to blame themselves for their parents’
conflict partially mediated the link between marital conflict and children’s internalizing
symptoms. In contrast, children’s reports that they become involved in their parents’ conflict
partially mediated the effect of marital conflict on externalizing problems.
Keywords: marital conflict, children’s perceptions, Berkeley Puppet Interview, children’s
adjustment
Over the past 15 years, many studies have found links
between parents’ marital conflict and their children’s behavior
problems (Cummings & Davies, 1994; Erel & Burman,
1995; Fosco & Grych, 2007; Grych & Fincham, 1990;
Jouriles, Spiller, Stephens, McDonald, & Swank, 2000;
McDonald & Grych, 2006). Until recently, there has been
little progress in identifying the mechanisms by which marital
conflict comes to affect children’s adaptation. One
promising theory—the cognitive– contextual framework
first proposed by Grych and Fincham (1990)—suggests that
children’s perceptions and interpretations of conflict in their
parents’ relationship play a central role in determining the
effect that marital conflict has on children’s emotional and
behavioral adjustment. At present, studies of children’s perceptions
as a mediating mechanism have been limited to
children who are 7 years or older. Evidence suggests, however,
that it may be the cognitive limitations of younger
children that make them particularly vulnerable to the effects
of interparental conflict, the content of which frequently
lies beyond their comprehension (Jouriles et al.,
2000; McDonald & Grych, 2006; Turner & Cole, 1994).
Recognizing the difficulty of assessing perceptions in very
young children, we created the Berkeley Puppet Interview
(BPI; Ablow & Measelle, 1993), which is designed to elicit
4 1/2- to 7 1/2-year-old children’s perceptions and interpretations
of different aspects of their family environment. The
present study reports on the psychometric properties of the
BPI scales designed to assess children’s perceptions and
appraisals of their parents’ conflict and tests the hypothesis
that individual differences in young children’s perceptions
of their parents’ conflict mediate the effect of conflict on
children’s adjustment. Central to the present study is the
contention that young children are reasonably astute observers
of their parents’ marital interaction and that, when
exposed to conflict, they try to determine the extent to which
they themselves are implicated, either as causes or as potential
sources of resolution.
Previous Studies of Children’s Perception
of Marital Conflict
Research on the role of children’s appraisals of their
parents’ conflict has considered a number of different dimensions
of children’s social cognitive processes. For example,
Cummings and Davies (1994) have demonstrated
that children’s perceptions of the degree to which marital
conflict is resolved serves as the salient mechanism linking
parents’ conflict and their children’s adjustment. Buchanan,
Maccoby, and Dornbusch (1991) and Kerig (1995, 1996)
reported that the degree to which children perceive being
“pulled into†or triangulated within their parents’ conflict is
Jennifer C. Ablow and Jeffrey R. Measelle, Department of
Psychology, University of Oregon; Philip A. Cowan and Carolyn
P. Cowan, Department of Psychology, University of California at
Berkeley.
Work on this project was supported by National Institute of
Mental Health Grants to Jennifer C. Ablow (1 F3 MH11038), to
Jeffrey R. Measelle (1 F3 MH10816), and to Philip A. Cowan and
Carolyn P. Cowan (MH-31109). Work on the development of
training materials for the Berkeley Puppet Interview was supported
by Grant PR-10004 from the MacArthur Foundation Research
Network on Psychopathology and Development to Jennifer C.
Ablow and Jeffrey R. Measelle. We would like to express our
gratitude to Joan Kaplan and the research assistants who worked
on this project, as well as to the teachers and families who
participated in the study.
Correspondence concerning this article should be addressed to
Jennifer C. Ablow, Department of Psychology, University of Oregon,
Eugene, OR 97403-1227. E-mail: jcablow@uoregon.edu
Journal of Family Psychology © 2009 American Psychological Association
2009, Vol. 23, No. 4, 485–499 0893-3200/09/$12.00 DOI: 10.1037/a0015894
485
associated with negative effects of marital conflict on the
child’s adaptation, a finding consistent with family systems
therapists’ accounts of family factors in children’s psychopathology
(e.g., Minuchin & Fishman, 1981). Children’s
perceptions of the content, intensity, and frequency of their
parents’ conflict have been related to children’s selfreported
distress and adjustment as well—in particular, internalizing
symptomatology (Grych & Fincham, 1993). In
summary, older children’s perceptions of the content, frequency,
intensity, and resolution of their parents’ conflict
have all been implicated in children’s response to interparental
conflict.
At the outset, we want to clarify two aspects of our
terminology. In the present report, we refer to children’s
perceptions to describe how children report on the properties
or characteristics of their parents’ marital conflict; for
example, children’s perceptions of their parents’ level of
conflict, conflict resolution, and affection. We use conflict
appraisals to refer to children’s cognitive processing of the
cause or the meaning that they derive from their perceptions.
Specifically, we use appraisals to refer to children’s
reports of self-blame for their parents’ conflict, the distress
they experience as a result of their parents’ conflict, and
their perceived involvement in their parents’ conflict.
Although the association between interparental conflict
and children’s socioemotional adjustment has been demonstrated
(see review by Fincham, 1998), insights into the
processes that give rise to these associations are more recent.
Models designed to account for such associations have
increasingly emphasized children’s appraisals of threat and
attributions of blame for the conflict (Grych & Fincham,
1990; Grych, Fincham, Jouriles, & McDonald, 2000). Using
the Children’s Perceptions of Interparental Conflict Scale
(CPIC; Grych, Seid, & Fincham, 1992) to test the
cognitive– contextual model in a study of 10- to 14-year-old
children from the community sample and from a battered
women’s shelter, Grych and colleagues (2000) found that
children’s perceived threat mediated the link between interparental
conflict and internalizing for all children. Selfblame,
however, mediated the same link for all boys in the
study, but only for the girls in the shelter sample. In the
same study, children’s sense of threat and self-blame did not
act as mediators of the link between interparental conflict
and externalizing problems (Grych et al., 2000).
More recently, McDonald and Grych (2006) adapted the
CPIC for 7- to 9-year-olds to see whether the threat and
self-blame experienced by younger children might mediate
the effect of interparental conflict on adjustment (Grych,
1998; Jouriles et al., 2000). As in the study with older
children, McDonald and Grych (2006) found that perceptions
of threat and, to a lesser extent, self-blame mediated
the association between interparental conflict and children’s
internalizing but not externalizing behaviors. This study
raised interesting, yet mostly unexplored, questions about
the earliest ages at which children’s conflict perceptions and
appraisals assume a central role in shaping their emotional
reactions to interparental discord. Research already has begun
to show that interparental conflict during the first years
of life is associated with infants’ fear of novel stimuli,
emotion regulation, psychophysiology (Crockenberg,
Leerkes, & Lekka, 2007; Porter, Wouden-Miller, Silva, &
Porter, 2003), and subsequently, their emotional security
(Cummings & Davies, 2002; Davies, Myers, & Cummings,
1996). Because children’s perceptions of interpersonal
events are thought to have their roots in the early experiences
of emotional security (Cummings & Davies, 2002), it
would be prudent to extend tests of the cognitive–
contextual framework to earlier developmental periods (see
Grych, Wachsmuth-Schlaefer, & Klockow, 2002, for similar
reasoning). From a theoretical point of view, the emotional
security theory provides insight into how early interpersonal
perceptions are shaped, whereas the cognitive–contextual
framework guides our understanding of how, specifically, such
interpersonal perceptions and appraisals might contribute to
children’s emotional and behavioral adjustment.
Assessing Younger Children’s Perceptions and
Appraisals of Marital Conflict
Despite data suggesting that children’s perceptions and
appraisals of their parents’ conflict are related to their socioemotional
adjustment, there have been few investigations
of young children’s perceptions of their parents’ marital
conflict. At least two factors may be responsible for this
gap.
First, although young children are able to make causal
attributions about events, the sophistication of their causal
reasoning is limited (Miller & Aloise, 1989; Turner & Cole,
1994). Older children are likely to understand that varied
factors can lead to conflict between their parents and, therefore,
are more likely to make more appropriate causal attributions
to marital conflict. (Covell & Abromovitch, 1987;
Kurdek, 1986; Turner & Cole, 1994). By contrast, children
who are at an egocentric level of thought may not understand
that their parents’ marital conflict could have little to
do with them. Cummings (1987; Cummings, Ballard,
El-Sheikh, & Lake, 1991) found that preschoolers revealed
more distress and reported feeling more frightened than
both younger (1- to 3-year-old) and older (9- to 19-year-old)
children in response to anger expressed between adults.
Although 4- to 6-year-old children appear capable of expressing
distress about anger between adults, their emotional
and cognitive immaturity might support unreasonable
attributions about the causes of the conflict. In one of the
few studies investigating young children’s attributions concerning
their parents’ behavior, Covell and Abramovich
(1987) found that 5- and 6-year-old children believed that
they were the sole cause of their mothers’ anger, whereas 8-
and 9-year-old children were able to identify factors other
than themselves that could cause anger. Similarly, research
on children’s reasoning about their parents’ divorce
(Kurdek, 1986; Neal, 1983; Wallerstein & Kelly, 1980)
showed that 5- to 8-year-old children tended to believe that
they were the primary cause of their parents’ divorce,
whereas older children realized that factors such as parental
incompatibility could lead to divorce (Kurdek, 1986). Although
most children do not see themselves as the primary
cause of their parents’ divorce, this tendency occurs more
486 ABLOW, MEASELLE, COWAN, AND COWAN
frequently in younger than in older children (Ablow, 2005).
These findings suggest that, because of their limited understanding
of causality, young children may be at heightened
risk for blaming themselves, even when this is not justified
by actual family interaction patterns.
Unlike perceptions of blame and threat or distress, children’s
appraisals that they should become involved in their
parents’ conflict to aid in its resolution have received less
attention, in particular among younger children (Ablow,
2005). From the perspective of the emotional security theory,
Cummings and colleagues (Cummings, Schermerhorn,
Davies, Goek-Morey, & Cummings, 2006) have reasoned
that interparental conflict can evoke response or regulatory
tendencies in children that are designed to return oneself
and others within a family system to optimal states of
security. Behavioral responses such as venting distress,
avoiding, or seeking to become involved (e.g., taking sides,
helping with tasks that may have caused the conflict, attempting
resolution) may serve the goal of preserving and
restoring a child’s emotional security within the context of
interparental conflict (Cummings et al., 2006). Although
each of these responses can be described as different forms
of coping with interparental conflict (Shelton & Harold,
2007), children’s inclination to become involved may carry
particular significance for their adjustment (Davies & Forman,
2002). In their study of 7- to 9 year-olds and 13- to
15-year-olds, Davies et al. (1996) found that the older
children reported greater motivation to intervene in adults’
unresolved hostile conflict. Adolescents, therefore, may be
more inclined to become involved if they can accurately
perceive an opportunity to be helpful, whereas children in
their middle childhood years may appraise few coping options
(Shelton & Harold, 2007). Given that younger children
are developmentally prone to egocentric or grandiose thinking
(Flavell & Green, 1999), their tendency may be to
become involved or perceive themselves as able to stop
their parents’ arguments. Studies addressing this question
remain to be conducted.
Second, the absence of data on young children’s experiences
of their family relationships may be due to the dearth
of developmentally appropriate self-report methodologies
for this age range (Measelle, Ablow, Cowan, & Cowan,
1998). When young children’s perceptions of family and
self are assessed, researchers tend to rely on instruments
with constructions that are structurally less appropriate for
younger children. By using pictures, icons that represent
quantity (e.g., larger and smaller circles), and emotionally
expressive faces, researchers have made efforts to adapt
existing measures for younger research participants. Although
theoretically consistent with the constructs they are
meant to assess, at the core, each of these paper-and-pencil
approaches are often too long and complex and, ultimately,
a foreign medium for young children, most of who are just
learning to read. As a result, children’s boredom and lack of
attention may contribute to the frequently reported lower
reliability in studies of younger children (Hughes, 1984).
The BPI (Ablow & Measelle, 1993) was created to provide
a more age-appropriate methodology for assessing
children’s perceptions of their family environment. In contrast
to the few paper-and-pencil questionnaires available
for young children, the BPI is designed to ask 4 1/2- to-7
1/2-year-old children how they perceive and respond to
different aspects of their family environment. With an interactive
approach to interviewing children adapted from
Eder (1990) and drawing on puppetry’s long history of
success in the therapeutic context (Irwin, 1985; Schaefer &
O’Conner, 1983; Woltmann, 1952), two identical puppets
“volunteer opposing information about themselves†and
then ask children to say which of two statements is most like
them. To date, work in our lab (Ablow, 2005; Ablow et al.,
1999; Measelle, 2005; Measelle et al., 1998, Measelle, John,
Ablow, Cowan, & Cowan, 2005) and in others (Arseneault
et al., 2003; Atzaba-Poria & Pike, 2008; Luby et al., 2002;
Pike, Coldwell, & Dunn, 2005) provide clear support for the
BPI method as a source of self-report data, as well as a way
to evaluate how children think and feel about relationships
in which they are participants (parent– child, peers, teacher–
child, sibling) or observers (marital, coparenting).
Aims of the Present Study
Internal Consistency, Differentiation,
and Factor Structure
Previous research has not addressed the age at which
children might be expected to differentiate among various
dimensions of interparental conflict. As such, it was not
clear whether all of the BPI’s marital conflict scales would
emerge as distinct factors in young children’s reports. Thus,
in the present research, in addition to estimating standard
measures of internal consistency and intercorrelation, we
used confirmatory factor analysis (CFA) to test whether
young children could differentiate between the properties of
their parents’ conflict and self-related appraisals of this
conflict. We predicted that young children would hold internally
consistent and meaningfully distinct perceptions of
their parents’ conflict (conflict, resolution, and affection), as
well as internally consistent and distinct appraisals of this
conflict (self-blame, distress, and involvement).
Temporal Stability
Interparental difficulties tend to demonstrate stability during
children’s early years (Fincham, 1998). Nevertheless,
the stability of young children’s conflict perceptions and
appraisals have not been assessed directly. Consequently,
we do not yet know whether the perceptions and appraisals
of young children demonstrate comparable stability across
time. Given previous evidence of self-report stability using
the BPI (Measelle et al., 2005), we predicted that children’s
perception and appraisals of their parents’ conflict would demonstrate
significant stability between the ages of 5 and 6.
Criterion Validity
Third, we sought to evaluate the criterion validity of
young children’s reports of their parents’ conflict in three
ways. First, we compared children’s reports of their parents’
CHILDREN’S PERCEPTIONS AND MARITAL CONFLICT 487
conflict with comparable marital ratings provided by mothers
and fathers. Second, because young children’s reports of
marital conflict are rarely accorded the same consideration
in developmental research when adults’ reports are available
(Measelle et al., 2005), we compared child–parent
associations with the level of agreement between spouses’
reports of their own marital conflict dynamics. Third, we
examined the real-world implications of young children’s
perceptions of their parents’ conflict for their classroom
adjustment, as reported by teachers. In each of these instances,
we predicted that young children’s perceptions of
marital conflict would demonstrate significant validity.
Testing the Meditational Role of Young
Children’s Appraisals
Finally, we tested the mediation hypothesis proposed by
the cognitive– contextual framework (Cummings & Davies,
1994; Grych & Fincham, 1990). Specifically, we hypothesized
that children’s conflict appraisals would mediate the
link between interparental conflict and children’s internalizing
and externalizing symptoms as rated by teachers. To
date, children’s reports of threat and self-blame have been
found to mediate the link between interparental conflict and
internalizing problems in children as young as age 7
(McDonald & Grych, 2006), but not younger. Social–
cognitive mechanisms linking marital conflict and externalizing
problems have not been identified clearly among children.
Our goal was to extend the empirical assessment of the
cognitive– contextual framework to younger ages by demonstrating
that 5- and 6-year-olds who experience their
parents’ conflict as threatening and as their fault would be at
risk for adjustment difficulties, given that the problem is
likely outside of their influence (Shelton & Harold, 2007).
Because children’s reports of higher involvement in their
parents’ conflict would be inconsistent with attributions of
helplessness, we did not expect that the BPI Involvement
scale would operate as a mediator in a model designed to
explain internalizing. Alternatively, we conjectured that
children’s reported involvement might be a source of frustration
that would mediate the link between marital conflict
and externalizing problems.
Method
Participants
Families were participants in a longitudinal investigation
of the transition to school (see Cowan, Cowan, Ablow,
Kahen-Johnson, & Measelle, 2005) and were assessed annually
when children were 4, 5, and 6 years of age. Approximately
110 two-parent families were followed prospectively
as their oldest child made the transition from
preschool to kindergarten and first grade. Families had been
recruited to join the study through preschools, day care
programs, and local media, and they were predominantly
middle-class residents of the greater San Francisco Bay
area. In each recruitment context, the project was described
as a study of family factors related to children’s successful
entry into school. Of the families, 21% were of African
American, Hispanic American, or Asian American ethnicity,
and the remaining 79% were European American. The
median income of the sample was $71,000 (SD $17,000).
Because the BPI marital conflict questions were not administered
during the first wave of data collection when
children were 4 years of age, in the present investigation,
only data from the second and third waves were used. For
the present study, the sample consists of 53 boys and 43
girls (n 96) with a mean age of 5.6 years (SD 0.37) at
the Time 2 assessment and a mean age of 6.4 years (SD
0.35 years) at the Time 3 assessment. There were no significant
differences on any of the central variables between
families in the present investigation and the families (n
14) who did not continue beyond the first year of the study.
Finally, although the present sample was an unselected
community sample, approximately 20% to 30% of the
mothers or fathers reported significant clinical levels of
distress on entering the study. Specifically, 18% of the
adults reported clinically elevated depressive symptoms
(16) on the Center for Epidemiological Studies—
Depression scale (CES–D; Radloff, 1977), and 21% reported
clinical levels of marital distress (100) on the
Dyadic Adjustment Scale (DAS; Spanier, 1979).
Procedure
The second wave of data (age 5 assessment) occurred
during the spring or summer of children’s kindergarten year,
and the third wave of data (age 6 assessment) was collected
during the spring or summer of their first grade year. In
conjunction with both waves, families visited a university
laboratory in the San Francisco Bay Area. At this visit, each
parent completed questionnaire packets addressing a range
of issues related to family life, including their marital relationship
(e.g., conflict, communication, problem solving
efficacy, and satisfaction with partner). Children’s perceptions
of marital conflict were assessed during two separate
summer assessments, the first right after kindergarten and
the second just after first grade. Target children were visited
in their homes, where they were interviewed with the BPI
about their perceptions of themselves and the major relationships
in their lives, including their views of their parents’
marital conflict. Each target child’s kindergarten and
first-grade teachers were contacted by mail and then by
telephone and were asked to complete behavioral adjustment
assessments of all of the children in their classrooms,
while remaining unaware of the target project participant.
Teachers were asked to complete the ratings during the
spring semester of the school year. In summary, for each
assessment period (assessment at age 5 or 6), parental
reports of their conflict, children’s reports of their parents’
conflict, and teacher reports of the children’s classroom
behavior were typically collected during the spring and
summer, or within 2 to 4 months.
Measures
Children’s perceptions of marital conflict. Children’s
perceptions of their parents’ conflict were assessed using the
488 ABLOW, MEASELLE, COWAN, AND COWAN
BPI, an interviewing and coding method (Ablow & Measelle,
1993) designed to assess young children’s perceptions
of their family environment (interparental conflict, parent–
child relationships, sibling relationships, and parents’ work
lives) and themselves (academically, socially, and psychological
symptomatology). Extensive work with the BPI has
demonstrated its utility as a psychometrically sound measure
of 4- to 8-year-old children’s self-perceptions and
perceptions of interpersonal relationships in and out of the
family (e.g., Ablow, 2005; Ablow et al., 1999; Luby et al.,
2002; Measelle et al., 1998; Measelle et al., 2005; Pike et
al., 2005).
The BPI methodology builds on Harter and Pike’s (1994)
and Eder’s (1990) work on assessing self-perceptions and
Grych and Fincham’s (1993) work in assessing older children’s
perceptions of marital conflict. Using an interactive
approach, two identical puppets volunteer opposing information
about themselves and then ask the children how the
issue pertains to them. For example, one puppet says,
“When my parents fight, I think I did something wrong,â€
whereas the second puppet says, “When my parents fight, I
don’t think I did something wrong . . . How about you?â€
Like Harter and Pike’s self-concept measure (Harter & Pike,
1994), the technique is designed to offer either alternative as
possible and acceptable in an attempt to help the child to
feel free to report their own perception about a particular
issue.
Children’s verbal responses to the interview were videotaped
and later coded on a scale ranging from 1 (never) to
7 (always) depending on which statement children say is
most like them. If, for example, a child responds negatively
by endorsing the item “When my parents have a fight, I
think I did something wrong,†her response is coded 1, 2, or
3, depending on the degree to which she endorses the
negative alternative. Endorsements of positive alternatives
are coded 5, 6, or 7. For example, children who essentially
repeat a puppet’s statement receive a score of 2 or 6,
depending on the valence of their response. A child who
amplifies a puppet’s statement (e.g., “I always think I did
something wrong,†or “I never think I did something
wrongâ€) is scored a 1 or 7, respectively. A child who
indicates that one of the puppet statements pertains to them,
but to a lesser degree (e.g., “I usually think I did something
wrong,†or “Most of the time I don’t think I did something
wrongâ€), is scored either a 3 or 5, respectively. If the child
indicates that “both†options pertain to her, this response is
coded 4. Two additional codes (8 and 9) are reserved for
unable-to-code responses. Finally, although most children
respond verbally to the puppets by making statements that
clearly correspond to either puppet’s statement, the BPI’s
interviewing and coding system provides procedures by
which to ensure that nonverbal responses are codable. For
example, with prohibitively shy children, the puppets would
encourage the child to point at the desired puppet (and in
turn, the response is coded either a 2 or 6 depending on the
valence).
For the purpose of this study, only the items designed to
assess children’s perceptions and processing of marital conflict
were used. Development of these items was shaped by
the theoretical arguments and empirical results of studies
exploring the relation between marital conflict and children’s
adaptation, in particular, work conducted on and with
the CPIC (Grych, Seid, & Fincham, 1992). Review of this
literature resulted in the development of five theoretical
scales for the BPI: Conflict, Conflict Resolution, Self-
Blame, Distress About Conflict, and Perceived Involvement
of Self in the Conflict. In addition to these five scales, a
sixth scale was created based on Gottman’s identification of
volatile yet stable couples (Gottman, 1993). Gottman describes
these couples as having stable relationships because
affection between parents buffers their relatively volatile
expressions of marital conflict. Thus a sixth BPI scale,
Parental Affection, was created to reflect the fact that positive
and negative marital behaviors commonly co-occur.
The six scales developed for the BPI are designed to assess
dimensions of marital conflict that have been shown to be
particularly salient in influencing (a) the effect of marital
conflict on children, (b) children’s response to the conflict,
and (c) children’s adjustment (Fosco & Grych, 2007; Grych
& Fincham, 1990; McDonald & Grych, 2006). A description
of each of the scales follows.
Perceptions of marital conflict properties. The Marital
Conflict scale reflects the degree of conflict that children
perceive between their parents. The scale includes four
items (e.g., “My parents have fights/My parents don’t have
fights,†“My parents fight a lot/My parents don’t fight a
lotâ€). The Conflict Resolution scale is designed to assess the
degree to which children perceive that their parents are able
to resolve marital conflict. There are four items in this scale
(e.g., “After my parents have a fight, they say they are sorry
to each otherâ€/“After my parents have a fight, they don’t say
they are sorry to each other,†“After my parents have a fight,
they stop talking to each other for a whileâ€/“After my
parents have a fight, they don’t stop talking to each otherâ€).
Finally, the Spousal Affection scale reflects the degree to
which children perceive their parents as being affectionate
with one another and engaging in warm exchanges. This
scale comprises five items (e.g., “My parents hug each
other/My parents don’t hug each other,†“My parents smile
at each other/My parents don’t smile at each otherâ€).
Appraisals of marital conflict. The Self-Blame scale is
designed to assess children’s tendency to blame themselves
for the conflict they perceive between their parents. The
scale includes five items (e.g., “It’s my fault when my
parents have a fightâ€/“It isn’t my fault when my parents
have a fight,†“When my parents have a fight, I think that
they are mad at meâ€/“When my parents have a fight, I don’t
think that they are mad at meâ€). The Distress About Conflict
scale reflects the level of emotional distress that a child
reports because of his or her parents’ conflict. The distress
scale includes four items (e.g., “When my parents have a
fight I get scaredâ€/“When my parents have a fight I don’t get
scared,†“When my parents have a fight I think something
bad will happenâ€/“When my parents have a fight I don’t
think something bad will happenâ€). Finally, the Involvement
of Self scale assesses whether the child perceives that
he or she becomes involved when marital conflict occurs.
The scale includes four items (e.g., “When my parents fight,
CHILDREN’S PERCEPTIONS AND MARITAL CONFLICT 489
I yellâ€/ “When my parents fight I get quiet,†“When my
parents fight, I can make them stopâ€/“When my parents
fight, I can’t make them stopâ€).
Four trained research assistants coded the videotaped
interviews. Each child’s interview was coded by at least two
trained observers. Interrater reliability was calculated using
intraclass correlations and averaged to .83 for all six scales
(conflict .94, resolution .72, affection .90; selfblame
.87; distress .82; involvement .89). To create
one score for each item, we averaged the ratings made by
independent coders.
Parents’ Self-Reports of their Marital Relationship
Husbands and wives independently completed the Couple
Communication Questionnaire (CCQ; Cowan & Cowan,
1990), which is a 41-item questionnaire designed to assess
partners’ level of comfort with and ability to communicate
about relationship issues, such as intimacy, conflict, and
problem solving. Responses were made on a scale ranging
from 1 (a lot) to 7 (none). For this study, the following three
theoretically derived scales were used (for both spouses’
average internal consistency, 0.80, range 0.72– 0.85;
for 1-year test–retest, 0.83): (a) Positive Emotion and
Intimacy (five items; e.g., “We enjoy one another,†“We
have a warm relationshipâ€), (b) Expression of Conflict (five
items; e.g., “We argue,†“We yell or insult one anotherâ€),
and (c) Effective Problem-Solving Strategies (five items;
e.g., “We talk about it to clarify the problem,†“We discuss
both our points of viewâ€). Scales from the CCQ are significantly
correlated with marital satisfaction for both spouses
(rs .56 to .64).
Measures of Children’s Adjustment
Measures of children’s adjustment were obtained from
children’s kindergarten teacher and a year later from their
first-grade teachers. Unaware as to which child in their
classrooms was a participant in the study, teachers completed
the Child Adaptive Behavior Inventory (CABI) for
each student in the classroom during the spring semester of
the academic year. The CABI is a 106-item questionnaire,
adapted by Cowan and Cowan (Cowan et al., 1994) from
Schaefer and Hunter’s (1983) 60-item scale, the Adaptive
Behavior Inventory, with an additional 17 items adapted
from the downward extension of the Quay–Peterson Behavior
Problem Checklist (O’Donnell & VanTuinen, 1979), and
23 items from Achenbach and Edelbrock’s Child Behavior
Checklist (CBCL; Achenbach & Edelbrock, 1983). The
remaining items were developed specifically for the School
Children and Their Families Project (SAF) to describe the
child’s behavior with peers.
CABI ratings are made on a 4-point scale ranging from 1
(not at all) to 4 (very much like). Scores on each scale are
then converted to z scores to represent the behavior of the
target child relative to the other children of the same gender
in his or her classroom, thus controlling for teachers’ general
negative or positive bias and classroom atmosphere
effects and facilitating comparison of ratings between teachers.
Finally, the CABI has good internal consistency (average
0.81; range 0.66–0.90) and validity (Cowan
& Cowan, 1987; Gottman, 1994). Recent data demonstrate
the CABI’s sensitivity in assessing both externalizing and
internalizing behaviors in children from ages 4 to 8 (Cowan
et al., 2005). Katz and Gottman (1993) reported significant
correlations between the Internalizing and Externalizing
scales from the CABI and the Teacher Report Form of the
CBCL, as reported by teachers who used both instruments
to rate their 7- to 8-year-old students’ behaviors (Externalizing,
r .63; Internalizing, r .49). In the present study,
teachers’ ratings of children’s internalizing symptoms
(combination of depression and anxiety symptoms) and
externalizing symptoms (combination of antisocial and oppositional
behavior) were used.
Results
Psychometric Properties of BPI Marital
Relationship Scales
Initially, we sought to determine whether there were
gender differences in how boys and girls responded to the
BPI marital relationship scales by testing separate multivariate
analyses of variance (MANOVAs; Gender Scale) at
ages 5 and 6. At both ages, there were no significant mean
differences between boys’ and girls’ responses on any of the
BPI’s marital scales (ps .10). Additionally, we found no
significant differences in the average levels or patterns of
intercorrelations (using Fisher’s r-to-z transformation)
among the BPI marital relationship scales for boys and girls.
Accordingly, subsequent analyses were conducted on the
entire sample.
The alphas for each BPI scale are presented in Table 1.
With the exception of children’s reports of marital conflict
at age 5 (Cronbach’s 0.55), the internal consistency of
each was within an acceptable range (Cronbach’s
0.65), especially in light of the fact that these scales comprised
either four or five items each. In addition, the mean
interitem correlation coefficients, which are independent of
scale length, all exceeded r .28 (range .28 to .39). This
latter estimate indicated further that young children could
describe interparental conflict with reasonable consistency.
The scale means and standard deviations for each of the
marital relationship scales for both years are also shown in
Table 1. Children generally perceived their parents as low in
conflict and as high in affection and conflict resolution.
Correspondingly, children tended to report low levels of
conflict related self-blame. Relative to their reports of marital
conflict, children reported moderate, although significantly
higher levels of distress and involvement at age 5,
ts(97) 2.98, ps .01; and at age 6, ts(94) 3.42, ps
.001, respectively. Children’s mean scores on all of the
scales did not change significantly. Despite low means, the
standard deviations for most of the scales were fairly large,
indicating a good deal of variability in children’s reports.
The intercorrelations among the BPI marital relationship
scales are also presented in Table 1, both within and across
years. At the age 5 and age 6 assessments, the BPI marital
490 ABLOW, MEASELLE, COWAN, AND COWAN
relationship scales were only modestly correlated on average,
mean rs .16 and .18, respectively, suggesting that the
scales represented fairly distinct dimensions of children’s
perceptions of their parents’ marital relationship.
The highlighted correlation coefficients in Table 1 provide
a measure of the stability of children’s conflict perceptions
and appraisals across a 1-year period of time. With the
exception of children’s reports of their distress, which
showed little to no stability across 1 year (r .08, ns), these
data indicate that children’s reports on the BPI marital
relationship scales showed significant stability.
Factor Structure
We used confirmatory factor analysis (CFA) to test the
dimensionality of young children’s reports of interparental
conflict at both ages. We were particularly interested in
testing whether the encouraging finding regarding internal
consistency and the low scale intercorrelations reported
earlier would translate into distinct factors. We tested four
different a priori factor structures, the results of which are
presented in Table 2. The first was a single-factor model,
which suggested that young children were not distinguishing
between the BPI’s conflict property scales and the
conflict attribution scales. A two-factor model was tested to
determine whether children differentiated between the three
conflict property scales and the three conflict appraisal
scales. A five-factor model was tested in which children’s
self-blame and distress were loaded on to the same factor to
reflect ideas of learned helplessness. Finally, a six-factor
model was tested to determine whether children differentiated
between all of the BPI marital scales; support for this
model would be consonant with similar tests of the CPIC
with older children (McDonald & Grych, 2006). We tested
the CFA models using maximum likelihood estimation procedures
(Mplus 5.1, Muthe´n & Muthe´n, 2007), with children’s
continuously measured responses to each scale item
used as factor indicators. Factors were allowed to correlate
in each model. Because the key interest in these CFAs was
the relative fit of the alternative models, all items were
allowed to load on only one factor, and no changes were
made on the basis of modification indices. For the purposes
of model comparison, our six-factor model was our base
model.
As shown in Table 2, at the age 5 assessment, the onefactor
model showed the worst relative fit, followed by the
two- and five-factor models. Because the first three models
did not differ in their degrees of freedom, we used Akaike’s
information criteria (AIC; Akaike, 1987) to estimate differences
between AIC values (AIC). AIC values that are
less than 2 indicated that a particular model is essentially
identical to the best fitting model within a set of related
models (Burnham & Anderson, 2002); AIC values that
exceed 10 suggest no support for the present model relative
to the best fitting model. At the age 5 assessment, the
Table 1
Internal Consistencies, Means, Standard Deviations, and Intercorrelations of BPI Marital Scales at Ages 5 and 6
BPI scales
Age 5 Age 6 BPI scale intercorrelations at ages 5 and 6
M (SD) M (SD) Conflict Resolution Affection Self-blame Distress Involvement
Conflict 0.55 2.84 (1.08) 0.77 2.93 (1.24) .38 .01 .14 .16 .01 .08
Resolution 0.67 4.83 (1.26) 0.68 4.87 (1.19) .15 .38 .27 .14 .19 .05
Affection 0.62 5.32 (0.79) 0.74 5.34 (0.95) .23 .51 .28 .04 .08 .06
Self-Blame 0.69 2.62 (0.81) 0.67 2.44 (0.68) .39 .26 .23 .39 .22 .32
Distress 0.80 3.72 (1.27) 0.78 3.49 (1.25) .26 .03 .04 .11 .08 .12
Involvement 0.67 3.58 (1.25) 0.60 3.39 (1.01) .11 .04 .02 .02 .24 .52
Note. n 98 at age 5; n 96 at age 6. Age 5 intercorrelations are presented above the bold diagonal; age 6 intercorrelations are printed
below the diagonal. Correlations in bold on the diagonal represent 1-year stability correlations between age 5 and 6. BP Berkeley Puppet
Interview.
p .05. p .01. p .001.
Table 2
Summary of Confirmatory Factor Analyses Comparing Competing Models of Children’s Reports of Interparental
Conflict at Ages 5 and 6
Model tested
Age 5 Age 6
2 df 2 AIC AIC 2 df 2 AIC AIC
One factor 554.8 275 79 8040 50 435.6 275 71 7449 130
Two factors 547.7 275 72 8035 45 421.8 275 57 7403 84
Five factors 538.3 275 63 8022 32 393.0 275 28 7387 68
Six correlated factors 475.5 260 — 7990 0 364.6 260 — 7319 0
Note. n 96 at ages 5 and 6; 2 Chi-square difference indicating the difference in fit compared to the predicted model, six correlated
factors model; AIC Akaike’s information criterion; AIC the difference between each model’s AIC and the minimum AIC value
within a set of related models. Here, AIC is therefore each models’ AIC minus the AIC from the six correlated factors model, which
by definition receives a AIC of 0. (See text for details of model specification.)
p .01.
CHILDREN’S PERCEPTIONS AND MARITAL CONFLICT 491
six-factor model had the lowest AIC value and the best
model relative to the comparator models. A similar pattern
emerged with the age 6 data.
In terms of absolute model fit, the results were somewhat
less robust. Specifically, at the age 5 assessment, the sixfactor
model yielded an encouraging—although not sufficiently
adequate—fit, 2(260) 475.16, p .001; comparative
fit index (CFI) 0.91, root-mean-square error of
approximation (RMSEA) 0.097, 2/df 1.8. By age 6,
however, the fit indices for the six- factor model were
acceptable and indicative of a good fitting model, 2(260)
364.6, p .001; CFI 0.96, RMSEA 0.043, 2/df 1.4.
Given that each of our scales had between four and five
items apiece, we opted not to utilize modification indices or
to drop, at this early stage of development, what might seem
like weaker items. Rather, our impression was that, at each
age, the reliability of individual items was weaker than what
is often seen with older children and adults. As a test of this
hypothesis, we combined the data from the two ages, averaging
item responses across ages 5 and 6 and submitting
these responses to the same CFA. As expected, the sixfactor
model provided the strongest relative fit and, in
absolute terms, provided a strong accounting of the data,
2(260) 289.0, p .01, CFI 0.98, RMSEA 0.022,
2/df 1.2. In summary, these data are consistent with the
idea that even children as young as 5 years old can provide
a differentiated picture of their parents’ conflict on the BPI.
Criterion Validity of Children’s Reports on the BPI
Marital Relationship Scales
Although children’s perceptions reflect their subjective
realities, we predicted that their descriptions of their parents’
relationship dynamics would correlate with comparable
data obtained from their mothers and fathers. At age 5
and again at age 6, children’s perceptions of their parents’
marital conflict were significantly correlated with mothers’
reports of marital conflict (rs .32 and .36, respectively, ps
.01) and fathers’ reports of marital conflict (rs .45 and
.27, respectively, ps .01). Children’s reports of their parents’
conflict resolution were statistically unrelated to mothers’
and fathers’ reports of spousal conflict resolution at age
5 (rs .19 and .04, respectively, ns) and to their mothers’
reports at age 6 (r .06, ns). However, at age 6, children
who reported higher resolution between their parents tended
to have fathers who reported higher resolution efficacy (r
.33, p .01). At ages 5 and again at age 6, children’s
perceptions of their parents’ affection were significantly
correlated with mothers’ reports of marital affection (rs
.30 and .31, respectively, ps .01) and fathers’ reports of
marital affection (rs .36 and .77, respectively, ps .001).
Support for the specificity of children’s reports was also
found. Specifically, children’s report of marital conflict was
statistically unrelated to either parents’ report of conflict
resolution or affection (rs .16, ns), and children’s reports
of conflict resolution were only modestly correlated with
parents’ reports of affection (rs .23 and .25, ps .05, with
mothers and fathers at age 5, respectively; and rs .20 and
.21, ps .10, with mothers and fathers at age 6, respectively).
As a way to evaluate further the validity of children’s
reports on the conflict property scales, we also compared the
child–mother and child–father correlation coefficients just
described, with the correspondence between spouses’ reports
of marital conflict (rs .39 and .41 for ages 5 and 6,
respectively, ps .001), conflict resolution (for age 5, r
.33, p .01; for age 6, r .15, ns), and affection (rs .53
and .46 at ages 5 and 6, respectively, ps .001). Although
spouses tended to report higher levels of agreement, these
differences were not statistically significant (tested with
Fisher’s r-to-z transformation). As such, young children’s
reports of marital conflict were as strongly correlated with
their mothers and with their fathers, as were spouses’ reports
with each other.
Our next hypothesis examined the potential real-world
consequences of children’s perceptions of their parents’
conflict properties and conflict attributions. Additionally,
we were interested in which source of information— child,
mother, or father—was most predictive of children’s adaptation
to school as rated by teachers. As shown in Table 3,
children’s perceptions of their parents’ conflict and conflict
resolution were significantly associated with teachers’ reports
of children’s internalizing and externalizing at ages 5
and 6. Children reporting higher interparental affection also
were described as having lower levels of internalizing
symptoms at ages 5 and 6, whereas children’s reports of
interparental affection was not significantly associated with
teachers’ reports of externalizing behavior.
At both ages, significant associations were found between
children’s tendency to blame themselves for their parents’
conflict and higher levels of internalizing and externalizing
Table 3
Correlations Between Children’s Reports of Their
Parents’ Marital Relationship With Teachers’ Ratings of
Internalizing and Externalizing Behavior Problems at
Ages 5 and 6
Informant
Teacher report
Internalizing Externalizing
Age 5 Age 6 Age 5 Age 6
Child
Conflict .37 .38 .21 .18
Resolution .35 .34 .39 .27
Affection .29 .26 .15 .17
Self-blame .42 .42 .47 .35
Distress .27 .24 .37 .33
Involvement .09 .14 .45 .38
Mother
Conflict .33 .26 .27 .24
Resolution .13 .24 .08 .16
Affection .09 .11 .03 .12
Father
Conflict .23 .28 .15 .15
Resolution .28 .29 .07 .19
Affection .23 .17 .17 .11
Note. n 93 at age 5; n 89 at age 6.
p .05. p .01. p .001.
492 ABLOW, MEASELLE, COWAN, AND COWAN
symptoms according to teachers. Likewise, children reporting
higher distress because of their parents’ conflict were
rated by their partners as exhibiting greater internalizing and
externalizing symptomatology at ages 5 and 6. It is interesting
that children’s reports of involvement in their parents’
conflict were not related to teachers’ reports of internalizing
problems at either age but were significantly
associated with teacher reports of externalizing problems at
both ages. Although not statistically different (using Fisher’s
r-to-z transformation), the correlations involving mothers
and fathers were neither as strong nor as consistently
related with teachers’ reports as were children’s reports of
their parents’ relationship processes. In summary, compared
with their parents’ reports of their own marital dynamics,
children’s perceptions and appraisals of marital conflict
were better predictors of teachers’ reports of children’s
internalizing and externalizing behavior problems in both
kindergarten and first grade.
Children’s Perceptions as Multivariate Mediators of
Marital Conflict
In our final analyses, we examined the hypothesis that
children’s conflict appraisals would mediate the link between
marital conflict and their classroom adjustment. Mediation
was evaluated by estimating confidence intervals
around the indirect effects, with regression-based path analysis
(Preacher & Hayes, 2004) and nonparametric resampling
(bootstrapping with bias correction; MacKinnon,
1994; Preacher & Hayes, 2004). This procedure yields a
path model that directly estimates the significance of the
indirect effects appropriately for relatively small samples
(MacKinnon, 1994; Preacher & Hayes, 2004). Therefore, it
is similar to, but more powerful than, the older procedure
recommended by Baron and Kenny (1986), in that it allows
for multivariate tests of mediation. Specifically, we examined
the extent to which all three conflict appraisal scales
mediated the link between couples’ reports of their marital
conflict (we averaged mothers’ and fathers’ reports of conflict
to reduce the number of analyses) and teachers’ reports
of children’s internalizing, in a first model, and teachers’
reports of children’s externalizing in a second model. Because
children’s reports on the three mediators were correlated,
the path model specified covariances among the predictors
so that each indirect effect simultaneously controlled
for the indirect effect of the other two possible mediators.
Given that teachers’ reports of internalizing and externalizing
problems were significantly correlated (r .38, p
.001), we entered each behavior problem scale as a covariate
to examine the specific effects of children’s perceptions
on each behavior problem, free from the effects of the other
problem behavior.
We conducted separate models of the type just described
using the age 5 data (parent and child reports at age 5 and
kindergarten teacher reports) and the age 6 data (parent and
child reports at age 6 and first-grade teacher reports). The
specific path and overall model results were statistically
equivalent, reflecting no clear developmentally linked
change between 5 and 6 years of age in the mediational role
of children’s perceptions. To simplify presentation, we restrict
our comments to the age 5 results to underscore the
lower range developmentally of these data.
As shown in the top portion of Figure 1, spouses’ combined
report of marital conflict was directly associated with
kindergarten teachers’ reports of young children’s internalizing
problems. However, as shown in the lower portion of
Figure 1, this effect was partially mediated by children’s
conflict appraisals. Specifically, the overall indirect path
was significant (p .01; see Figure 1’s note) and partially
mediated the effect of marital conflict on kindergarten
teachers’ reports of internalizing while controlling for the
effects of externalizing. Parents’ conflict continued to show
a significant association with children’s internalizing problems,
although portions of this association were better accounted
for by the indirect paths. Specifically, as shown in
Figure 1, children’s self-blame and distress were both significant
mediators, whereas their reports of involvement
were not.
As shown in the top portion of Figure 2, spouses’ combined
report of marital conflict was directly associated with
kindergarten teachers’ ratings of young children’s externalizing
problems. However, as shown in the lower portion of
Figure 2, this effect of parents’ conflict on children’s externalizing
behavior was partially mediated by children’s appraisals
of conflict (p .01; see Figure 2’s note). Specifically,
neither children’s reports of self-blame nor their
reports of distress were significant mediators in this model
when considered simultaneously with child reports of involvement.
However, children’s tendency to report involvement
in their parents’ conflict partially mediated the effect
of marital conflict on children’s externalizing so that high
conflict was related to children’s reports of greater involvement,
which was, in turn, related to higher externalizing
problems as reported by teachers.
Discussion
Increasingly, researchers are including the important role
of children’s perceptions of their parents’ relationship to
advance understanding of the mechanisms that link marital
relationship processes and children’s adjustment. With few
exceptions, however, most of this work has been limited to
including the perceptions of children who are 9 years of age
and older. The overarching goals of the present investigation
were to determine whether the perceptions of younger
children, who were assessed at age 5 and again at age 6,
could be measured reliably and validly and whether their
conflict attributions might mediate the effects of interparental
conflict on their adjustment, as has been shown with
older children (Grych & Fincham, 1990; McDonald &
Grych, 2006).
Psychometric Properties of Young Children’s
Perceptions of Their Parents’ Marital Relationship
The results of this study suggest that young children’s
perceptions of their parents’ relationship can be measured
reliably with the BPI (Ablow & Measelle, 1993). Children’s
CHILDREN’S PERCEPTIONS AND MARITAL CONFLICT 493
perceptions of their parents’ relationship were assessed on
multiple BPI scales, including three scales assessing conflict
properties (marital conflict, resolution of marital conflict,
and spousal affection) and three scales assessing appraisals
of marital conflict (perceived self-blame for marital conflict,
perceived distress due to marital conflict, and perceived
involvement in parents’ conflict). All six scales demonstrated
acceptable levels of internal consistency as well as
distinctiveness, with the average intercorrelations among
scales at age 5 and 6 years not exceeding .20. More important,
the CFA results provided additional evidence of the
dimensionality of children’s perceptions of interparental
conflict. The strength of the six-factor solution at both ages
relative to models with fewer factors suggests that children
as young as 5 years of age can differentiate among multiple
dimensions of interparental conflict. Finally, although reliability
is usually assessed over a short period, children’s
reports at age 5 and again at age 6 were significantly
correlated, suggesting that how children perceived and felt
about their parents’ relationship was reasonably consistent
across a year.
These data suggest that, when interviewed with ageappropriate
methods that use meaningful item content,
young children can provide internally consistent information
about specific dynamics of their parents’ relationship—in particular,
levels of marital conflict—which previous research has
identified as emotionally salient for children as young as 1 year
old (Crockenberg et al., 2007; Porter et al., 2003). These data
may also speak to the effort that has been devoted to
improving the BPI’s age-appropriateness and standardization
as a self-report instrument. Standardized training, administration,
and coding procedures may all contribute to a
method that supports young children’s ability to provide
psychometrically sound information about parents’ relationships.
Criterion Validity of Young Children’s Perceptions
of Their Parents’ Marital Relationship
The validity of the children’s conflict perceptions and appraisals
was supported by testing different cross-informant
patterns of association. As predicted, children’s perceptions of
their parents’ conflict properties were related to conceptually
similar scales provided by each spouse. In particular, children’s
perceptions of their parents’ marital conflict were
related to spouse’s reports of marital conflict at both ages.
Just as meaningful in terms of validity, the child–parent
correlation coefficients were as large as the coefficients of
Figure 1. Path diagram showing unmediated and multivariate mediation of the effect of marital
conflict on kindergarten teachers’ reports of children’s internalizing problems at age 5. Total indirect
effect B 0.21 (SE 0.06), p .01 (95% confidence interval .09 to .33). Numbers in the figure
are path coefficients (standardized regression coefficients). Although not drawn, the model accounted
for covariation among the mediators.
494 ABLOW, MEASELLE, COWAN, AND COWAN
spouse’s reports with one another. Although bias and
gender-linked differences in the salience of conflict dimensions
have helped to explain attenuated levels of agreement
between spouses (Cowan & Cowan, in press), these data
suggested that young children are perceiving aspects of their
parents’ conflict and relationship dynamics that map onto
their parents’ experience.
Recent studies suggest that children’s appraisals of marital
conflict—in particular, whether they blame themselves
or feel threatened—mediate the links between exposure to
marital conflict and children’s behavior problems (Grych et
al., 1992). Because this domain of inquiry may sharpen our
understanding of how psychopathology develops, we examined
patterns of association between the BPI conflict appraisal
scales and teachers’ reports of children’s internalizing
and externalizing behavior problems at both time points.
Children’s attributions of self-blame and distress at ages 5
and 6 were significantly related to teachers’ ratings of
children’s internalizing and externalizing behavior problems.
Research has shown that, for reasons of cognitive
egocentrism, young children (Ablow, 2005; Covell &
Abramovich, 1987) tend to blame themselves for interparental
conflict. These data suggest that both the conflict
itself and young children’s tendency to blame themselves
may predispose some children to elevated mood and conduct
problems in the classroom. Similarly, children’s experience
of distress because of their parents’ conflict appeared
to compromise their sense of security. Indeed, Cummings
and colleagues have demonstrated that this lack of emotional
security, be it objectively or subjectively measured, is
a strong predictor of impaired child outcomes (Cummings
& Davies, 1994).
Children’s reports of involvement in their parents’ conflict
was unrelated to teachers’ reports of internalizing problems
but was significantly associated with teachers’ reports
of externalizing problems at ages 5 and 6. Although the
causal direction in these data cannot be established, we
speculate that young children’s attempts to intervene in their
parents’ conflict may contribute to a sense of frustration that
spills over into the classroom context.
Consistent with Grych and colleagues’ (Fosco & Grych,
2007; McDonald & Grych, 2006) results with older children,
in the present study, young children’s conflict perceptions
and attributions predicted teachers’ ratings of adjustment
better than parents’ reports about their marriage. These
results provide additional evidence that the BPI is a viable
tool for assessing young children’s perceptions of their
parents’ relationship. More germane, these data suggest that
Figure 2. Path diagram showing unmediated and multivariate mediation of the effect of marital
conflict on kindergarten teachers’ reports of children’s externalizing problems at age 5. Total
indirect effect B 0.22 (SE 0.08), p .01 (95% confidence interval .10 to .36). Numbers in
the figure are path coefficients (standardized regression coefficients). Although not drawn, the model
accounted for covariation among the mediators.
CHILDREN’S PERCEPTIONS AND MARITAL CONFLICT 495
researchers might gain a fuller picture of the association
between marital conflict and child adjustment by including
young children’s reports.
Evidence That Young Children’s Appraisals Mediate
the Marital Conflict–Child Adjustment Link
The cognitive– contextual theory proposed by Grych and
Fincham (1990) states that children’s conflict perceptions
and attributions mediate the impact of conflict on their
adjustment. Evidence in support of this mediation has accumulated
in samples of children age 8 and older (Fosco &
Grych, 2007; McDonald & Grych, 2006). A core aim of the
present investigation was to determine whether the social–
cognitive characteristics of even younger children might
function similarly when exposed to interparental conflict
dynamics. The literature has yet to determine the age at
which children’s appraisals of threat and self-blame begin to
act as processes by which exposure to conflict contributes to
maladjustment. It has been speculated (Jouriles et al., 2000)
that children younger than age 7 may not engage in the type
of cognitive processing involved in making self-attributions
(e.g., self-blame) for observed events (Grych & Fincham,
1990; see also a brief review by McDonald & Grych, 2006).
Our results suggest otherwise.
In our multivariate path models, we found that young
children’s appraisals of self-blame and distress partially
mediated the effect of spouses’ combined reports of conflict
with teachers’ reports of internalizing problems. By comparison,
children’s reports of involvement in their parents’
conflict did not mediate this effect. In keeping with
McDonald and Grych’s (2006) investigation with 7- to
9-year-olds, we found that children’s attributions of selfblame
and distress when they were 5 and again 6 years of
age partially mediated the association between marital conflict
and internalizing problems.
In their study, McDonald and Grych (2006) demonstrated
mediation using children’s reports of both their appraisal
processes and their internalizing symptoms. In the present
study, however, the mediated pathway linking parentreported
conflict, children’s appraisal, and teachers’ ratings
of internalizing problems were derived through independent
sources. Turner and Cole (1994) hypothesized that young
children probably lack the capacity to perceive and explain
emotional events with much stability until adolescence;
consequently, their reports of self-blame might have less
effect on their adjustment. In contrast, the data from the
present study suggest that when young children are interviewed
in a developmentally appropriate manner, their tendency
to blame themselves for their parents’ conflict and to
feel distress were both stable across time and were implicated
in their emotional adjustment. The fact that our path
models yielded similar mediation at both ages serves as a
form of replication that increases our confidence in the
implications of these findings.
The specificity of young children’s attribution processes
was also borne out in the path model predicting externalizing
behavior. Here, neither children’s reports of self-blame
nor their reports of distress mediated the effect of conflict on
teachers’ ratings of externalizing problems. However, children’s
reports that they become involved in their parents’
conflict partially mediated the effect of marital conflict
on externalizing problems. Among 7- to 9-year olds,
McDonald and Grych (2006) found no evidence of mediation
for self-blame or perceived threat, and they did
not include an assessment of children’s perceived involvement
in their model. Unlike the internalizing
model—which, we conjecture, may show continuity from
the age range studied in the present investigation to the
older age ranges studied by Grych, Fincham, and
colleagues—it is possible that young children’s sense that
they do or should attempt to resolve their parents’ conflict
may be a time-limited effect. It is interesting that, in their
study of both pre- and postadolescent children, Davies and
colleagues (Davies et al., 1996) found that the older children
reported an inclination to intervene in their parents’ conflict
more than the younger children did. Critically, this study did
not measure whether adolescents actually did intervene
more than the younger age group. As children acquire a
more sophisticated sense of conflict responsibility and resolution,
this knowledge, coupled with what is most likely a
history of ineffectual attempts at becoming involved in and
attempting to resolve their parents’ conflict, could predict
diminishing attempts at involvement. Alternatively, with
age and experience, adolescents may adopt more sophisticated
forms of involvement that might actually ameliorate
their parents’ conflict (Cummings et al., 1991; Davies et al.,
1996). Nevertheless, consistent with earlier work, these data
suggest that children’s involvement in their parents’ problems
do not typically bode well for their psychological
adjustment (Cummings & Davies, 1994).
Limitations
There are several limitations to the present study. Because
families in this sample were two-parent, predominantly
middle-class, and mostly European American, further
study is necessary to determine how reflective these
findings are of other socioeconomic groups and family
structures. At present, researchers in this and other laboratories
around the country are exploring the validity of the
BPI in samples with greater ethnic and familial diversity. A
related issue concerns the restricted range in the level of
conflict reported by all informants in the present sample. As
such, the present study could not address the effects on
children’s adjustment of the most extreme forms of interparental
conflict. Nevertheless, the strength of our primary
findings, coupled with the fact that young children’s reports
mediated the link between marital conflict and their adjustment,
increases our confidence that the BPI might be a
useful tool in more distressed samples. Indeed, although
generally low to moderate in their level of reported conflict,
approximately 20% of the couples in this sample entered the
study reporting clinically significant levels of marital distress.
As well, during the course of several BPI interviews,
children indicated that one or both of their parents engaged
in physically violent conflict. Accordingly, we believe that
496 ABLOW, MEASELLE, COWAN, AND COWAN
even children in higher risk circumstances will be able to
validly describe their parents’ marital conflict.
The present sample was drawn from a prospective, longitudinal
study; nevertheless, with the exception of our
question about the 1-year stabilities of children’s perceptions
and appraisals of marital conflict, our analyses did not
examine predictive associations over time. Consequently,
the mediation results can be only considered tentative, given
that a temporal order among these variables was not established.
Although other investigations designed to test the
cognitive– contextual theory also have been cross-sectional
(e.g., McDonald & Grych, 2006), like these other studies,
the present investigation provides a statistical test of mediation
at best. Future investigations of young children’s
perceptions of interparental conflict will need to establish
temporality to assess mediation formally.
Perhaps the most significant implication of this study is
its demonstration that young children, when assessed with
age-appropriate methodology, can be reliable reporters of
their parents’ relationship and marital conflict. Furthermore,
children’s perceptions of interparental conflict and, more
specifically, the appraisals they form because of this conflict,
appear relevant to their behavioral development. It is
clear that, to understand fully how exposure to marital
conflict is linked to children’s adjustment, researchers must
consider children’s perceptions and appraisals of their parents’
conflict. To date, this line of research has largely been
pursued with older school-age children or adolescents.
However, how children perceive and respond to daily stressors
such as family conflict varies considerably with their
age. To understand these developmental differences, researchers
have pointed to the need for methods that can
facilitate the assessment of children’s perceptions throughout
the childhood years. Although the utility of the BPI
continues to be explored, the results from the present study
suggest that it is a promising measure of young children’s
perceptions of interparental conflict.
References
Ablow, J. C. (2005). When parents conflict or disengage: Children’s
perceptions of parents’ marital distress predict school
adaptation. In P. A. Cowan, C. P. Cowan, J. C. Ablow, V.
Kahen-Johnson, & J. R. Measelle (Eds.), The family context of
parenting in children’s adaptation to elementary school: Monographs
in parenting (pp. 189–208). Mahwah, NJ: Erlbaum.
Ablow, J. C., & Measelle, J. R. (1993). Berkeley Puppet Interview:
Administration and scoring system manuals. Berkeley: University
of California, Berkeley.
Ablow, J. C., Measelle, J. R., Kraemer, H. C., Harrington, R.,
Luby, J., Smider, N., et al. (1999). MacArthur Research Network
on Psychopathology and Development: A three-site study of
adjustment and maladjustment among 4 1/2- to 7 1/2-year olds.
Journal of the American Academy of Child and Adolescent
Psychiatry, 38, 1580–1590.
Achenbach, T. M., & Edelbrock, C. S. (1983). Manual for the
Child Behavior Checklist and revised child behavior profile.
Burlington, VT: Author.
Akaike, H. (1987). Factor analysis and AIC. Psychometrika, 52,
317–332.
Arseneault, L., Moffitt, T. E., Caspi, A., Taylor, A., Rijsdijk, F. V.,
Jaffee, S., et al. (2003). Strong genetic effects on antisocial
behavior among 5-year-old children according to mothers, teachers,
examiner-observers, and twins’ self-reports. Journal of Child
Psychology and Psychiatry, 44, 832–848.
Atzaba-Poria, N., & Pike, A. (2008). Correlates of parental differential
treatment: Parental and contextual factors during middle
childhood. Child Development, 79, 217–232.
Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator
variable distinction in social psychological research: Conceptual,
strategic, and statistical considerations. Journal of Personality
and Social Psychology, 51, 1173–1182.
Buchanan, C. M., Maccoby, E. E., & Dornbusch, S. M. (1991).
Caught between parents: Adolescents’ experience in divorced
homes. Child Development, 62, 1008–1029.
Covell, K., & Abromovitch, R. (1987). Understanding emotion in
the family: Children’s and parents’ attributions of happiness,
sadness, and anger. Child Development, 58, 985–991.
Cowan, C. P., & Cowan, P. A. (1987). A preventive intervention
for couples becoming parents. In C. F. Z. Boukydis (Ed.), Research
on support for parents and infants in the postnatal period
(pp. 225–251). New York: Ablex.
Cowan, C. P., & Cowan, P. A. (in press). Working with couples
during stressful transitions. In S. Dreman (Ed.), The family on the
threshold of the 21st century. Mahwah, NJ: Erlbaum.
Cowan, P. A., & Cowan, C. P. (1990). Becoming a family:
Research and intervention. In I. Sigel & G. H. Brody (Eds.),
Normal families, Vol. 1. Method of family research: Biographies
of research projects (pp.1–51). Hillsdale, NJ: Erlbaum.
Cowan, P. A., Cowan, C. P., Ablow, J. C., Kahen-Johnson, V., &
Measelle, J. R. (2005). Recruitment, design, and measures. In
P. A. Cowan, C. P. Cowan, J. C. Ablow, V. Kahen-Johnson, J. R.
Measelle (Eds.), The family context of parenting in children’s
adaptation to elementary school: Monographs in parenting (pp.
33–78). Mahwah, NJ: Erlbaum.
Cowan, P. A., Cowan, C. P., Schulz, M., & Heming, G. (1994).
Prebirth to preschool family factors predicting children’s adaptation
to kindergarten. In R. Parke & S. Kellam (Eds.), Exploring
family relationships with other social contexts: Advances in
family research (pp. 75–113). Hillsdale, NJ: Erlbaum.
Crockenberg, S. C., Leerkes, E. M., & Lekka, S. K. (2007).
Pathways from marital aggression to infant emotion regulation:
The development of withdrawal in infancy. Infant Behavior and
Development, 30, 97–113.
Cummings, E. M. (1987). Coping with background anger in early
childhood. Child Development, 58, 976–984.
Cummings, E. M., Ballard, M., El-Sheikh, M., & Lake, M. (1991).
Resolution and children’s responses to interadult anger. Developmental
Psychology, 27, 462–470.
Cummings, E. M., & Davies, P. (1994). Children and marital
conflict: The impact of family dispute and resolution. New York:
Guilford Press.
Cummings, E. M., & Davies, P. (2002). Effects of marital conflict
on children: Recent advances and emerging themes in processoriented
research. Journal of Child Psychology and Psychiatry,
43, 31–63.
Cummings, E. M., Schermerhorn, A. C., Davies, P. T., Goek-
Morey, M. C., & Cummings, J. S. (2006). Interparental discord
and child adjustment: Prospective investigations of emotional
security as explanatory mechanism. Child Development, 77,
132–152.
Davies, P. T., & Forman, E. M. (2002). Children’s patterns of
preserving emotional security in the interparental subsystem.
Child Development, 73, 1880–1903.
Davies, P. T., Myers, R. L., & Cummings, E. M. (1996). Re-
CHILDREN’S PERCEPTIONS AND MARITAL CONFLICT 497
sponses of children and adolescents to marital conflict scenarios
as a function of the emotionality of conflict endings. Merrill-
Palmer Quarterly, 42, 1–21.
Eder, R. A. (1990). Uncovering young children’s psychological
selves: Individual and developmental differences. Child Development,
61, 849–863.
Erel, O., & Burman, B. (1995). Interrelatedness of marital relations
and parent-child relations: A meta-analytic review. Psychological
Bulletin, 18, 108–132.
Fincham, F. (1998). Marital conflict and child development. Child
Development, 69, 543–574.
Flavell, J., & Green, F. L. (1999). Development of intuitions about
the controllability of different mental states. Cognitive Development,
14, 133–146.
Fosco, G. M., & Grych, J. H. (2007). Emotional expression in the
family as a context for children’s appraisals of interparental
conflict. Journal of Family Psychology, 21, 248–258.
Gottman, J. M. (1993). The roles of conflict engagement, escalation,
and avoidance in marital interaction: A longitudinal view of
five types of couples. Journal of Consulting and Clinical Psychology,
61, 6–15.
Gottman, J. M. (1994). What predicts divorce? The relationship
between marital processes and marital outcomes. Hillsdale, NJ:
Erlbaum.
Grych, J. H. (1998). Children’s appraisals of interparental conflict:
Situational and contextual influences. Journal of Family Psychology,
12, 437–453.
Grych, J. H., & Fincham, F. D. (1990). Marital conflict and
children’s adjustment: A cognitive– contextual framework. Psychological
Bulletin, 108, 267–290.
Grych, J. H., & Fincham, F. D. (1993). Children’s appraisals of
marital conflict: Initial investigations of the cognitive-contextual
framework. Child Development, 64, 215–230.
Grych, J. H., Fincham, F. D., Jouriles, E. N., & McDonald, R.
(2000). Interparental conflict and child adjustment: Testing the
meditational role of appraisals in the cognitive-contextual framework.
Child Development, 71, 1648–1661.
Grych, J. H., Seid, M., & Fincham, F. D. (1992). Assessing marital
conflict from the child’s perspective: The Children’s Perception
of Interparental Conflict Scale. Child Development, 63, 558–572.
Grych, J. H., Wachsmuth-Schlaefer, T., & Klockow, L. L. (2002).
Interparental aggression and young children’s representation of
family relationships. Journal of Family Psychology, 16, 259–
272.
Harter, S., & Pike, R. (1994). The pictorial scale of perceived
competence and social acceptance for young children. Child
Development, 55, 1969–1982.
Hughes, H. (1984). Measures of self-concept and self-esteem for
children ages 3–12 years: A review and recommendations. Clinical
Psychology Review, 4, 657–692.
Irwin, E. C. (1985). Puppets in therapy: An assessment procedure.
American Journal of psychotherapy, 39, 389–399.
Jouriles, E. N., Spiller, L. C., Stephens, N., McDonald, R., &
Swank, P. (2000). Variability in adjustment of children of battered
women: The role of child appraisals of interparent conflict.
Cognitive Therapy and Research, 24, 233–249.
Katz, J. F., & Gottman, J. M. (1993). Patterns of marital conflict
predict children’s internalizing and externalizing behaviors. Developmental
Psychology, 29, 940–950.
Kerig, P. K. (1995). Triangles in the family circle: Effects of
family structure on marriage, parenting, and child adjustment.
Journal of Family Psychology, 9, 28–43.
Kerig, P. K. (1996). Assessing the links between interparental
conflict and child adjustment: The conflicts and problem-solving
scales. Journal of Family Psychology, 10, 454–473.
Kurdek, L. A. (1986). Children’s reasoning about parental divorce.
In R. D. Ashmore & D. M. Brodzinsky (Eds.), Thinking about
the family: Views of parents and children (pp. 233–276). Hillsdale,
NJ: Erlbaum.
Luby, J. L., Heffelfinger, A., Measelle, J. R., Ablow, J. C., Essex,
M. J., Dierker, L., et al. (2002). Differential performance of the
MacArthur HBQ and DISC-IV in identifying DSM-IV internalizing
psychopathology in young children. Journal of the American
Academy of Child & Adolescent Psychiatry, 41, 458–466.
MacKinnon, D. P. (1994). Analysis of mediating variables in
prevention and intervention research. In A. Ca´zares & L. Beatty
(Eds.), Scientific methods for prevention intervention research
(NIDA Monograph No. 139, pp. 127–153). Rockville, MD:
National Institute on Drug Abuse.
McDonald, R., & Grych, J. H. (2006). Young children’s appraisals
of interparental conflict: Measurement and links with adjustment
problems. Journal of Family Psychology, 20, 88–99.
Measelle, J. R. (2005). Children’s self-perceptions as a link between
family relationship quality and social adaptation to school.
In P. A. Cowan, C. P. Cowan, J. C. Ablow, V. Kahen-Johnson,
& J. R. Measelle (Eds.), The family context of parenting in
children’s adaptation to elementary school: Monographs in parenting
(pp. 189–208). Mahwah, NJ: Erlbaum.
Measelle, J. R., Ablow, J. A., Cowan, P. A., & Cowan, C. P.
(1998). Assessing young children’s views of their academic,
social, and emotional lives: An evaluation of the Berkeley Puppet
Interview. Child Development, 69, 1556–1576.
Measelle, J. R., John, O. P., Ablow, J. C., Cowan, P. A., & Cowan,
C. A. (2005). Can children provide coherent, stable, and valid
self-reports on the Big 5 dimensions? A longitudinal study from
ages 5 to 7. Journal of Personality and Social Psychology, 89,
90–106.
Miller, P. H., & Aloise, P. A. (1989). Young children’s understanding
of the psychological causes of behavior: A review.
Child Development, 60, 257–285.
Minuchin, S., & Fishman, H. C. (1981). Family therapy techniques.
Cambridge, MA: Harvard University Press.
Muthe´n, L. K., & Muthe´n, B. O. (2007). Mplus 5.1 user’s guide.
Los Angeles: Authors.
Neal, J. H. (1983). Children’s understanding of their parents’
divorce. In L. A. Kurdek (Ed.), Children and divorce (pp. 3–14).
San Francisco: Jossey-Bass.
O’Donnell, J. P., & VanTuinen, M. V. (1979). Behavior problems
of preschool children: Dimensions and congenital correlates.
Journal of Abnormal Child Psychology, 7, 61–75.
Pike, A., Coldwell, J., Dunn, J. F. (2005). Sibling relationships in
early/middle childhood: Links with individual adjustment. Journal
of Family Psychology, 19, 523–532.
Porter, C. L., Wouden-Miller, M., Silva, S. S., & Porter, A. E.
(2003). Marital harmony and conflict: Linked to infants’ emotional
regulation and cardiac vagal tone. Infancy, 4, 297–307.
Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for
estimating indirect effects in simple mediation models. Behavior
Research Methods, Instruments & Computers, 36, 717–731.
Radloff, L. S. (1977), The CES-D scale: A self-report depression
scale for research in the general population. Applied Psychological
Measurement, 1, 385–401.
Schaefer, C., & O’Conner, K. (1983). Handbook of play therapy.
New York: Wiley.
Schaefer, E. S., & Hunter, W. M. (1983). Classroom behavior
inventory. Unpublished instrument, University of North Carolina
at Chapel Hill.
498 ABLOW, MEASELLE, COWAN, AND COWAN
Shelton, K. H., & Harold, G. T. (2007). Marital conflict and
children’s adjustment: The mediating and moderating role of
children’s coping strategies. Social Development, 16, 497–512.
Spanier, G. B. (1979). The measurement of marital quality. Journal
of Sex & Marital Therapy, 5, 288–300.
Turner, J. E., & Cole, D. A. (1994). Developmental differences in
cognitive diatheses in child depression. Journal of Abnormal
Child Psychology, 22, 15–32.
Wallerstein, J. S., & Kelly, J. (1980). Surviving the breakup: How
children and parents cope with divorce. New York: Basic Books.
Woltmann, A. G. (1952). Puppet shows as a psychotherapeutic
method. In L. Bender (Ed.), Child psychiatric techniques: Diagnostic
and therapeutic approach to normal and abnormal development
through patterned, expressive, and group behavior (pp.
238–257). Springfield, IL: Charles C Thomas.
Received April 28, 2008
Revision received February 9, 2009
Accepted February 13, 2009
Call for Nominations
The Publications and Communications (P&C) Board of the American Psychological Association
has opened nominations for the editorships of Experimental and Clinical Psychopharmacology,
Journal of Abnormal Psychology, Journal of Comparative Psychology, Journal of Counseling
Psychology, Journal of Experimental Psychology: Human Perception and Performance,
Journal of Personality and Social Psychology: Attitudes and Social Cognition, PsycCRITIQUES,
and Rehabilitation Psychology for the years 2012–2017. Nancy K. Mello, PhD, David
Watson, PhD, Gordon M. Burghardt, PhD, Brent S. Mallinckrodt, PhD, Glyn W. Humphreys, PhD,
Charles M. Judd, PhD, Danny Wedding, PhD, and Timothy R. Elliott, PhD, respectively, are the
incumbent editors.
Candidates should be members of APA and should be available to start receiving manuscripts in
early 2011 to prepare for issues published in 2012. Please note that the P&C Board encourages
participation by members of underrepresented groups in the publication process and would particularly
welcome such nominees. Self-nominations are also encouraged.
Search chairs have been appointed as follows:
? Experimental and Clinical Psychopharmacology, William Howell, PhD
? Journal of Abnormal Psychology, Norman Abeles, PhD
? Journal of Comparative Psychology, John Disterhoft, PhD
? Journal of Counseling Psychology, Neil Schmitt, PhD
? Journal of Experimental Psychology: Human Perception and Performance,
Leah Light, PhD
? Journal of Personality and Social Psychology: Attitudes and Social Cognition,
Jennifer Crocker, PhD
? PsycCRITIQUES, Valerie Reyna, PhD
? Rehabilitation Psychology, Bob Frank, PhD
Candidates should be nominated by accessing APA’s EditorQuest site on the Web. Using
your Web browser, go to http://editorquest.apa.org. On the Home menu on the left, find
“Guests.†Next, click on the link “Submit a Nomination,†enter your nominee’s information,
and click “Submit.â€
Prepared statements of one page or less in support of a nominee can also be submitted by
e-mail to Emnet Tesfaye, P&C Board Search Liaison, at emnet@apa.org.
Deadline for accepting nominations is January 10, 2010, when reviews will begin.
CHILDREN’S PERCEPTIONS AND MARITAL CONFLICT 499