Content uploaded by Patrick T Davies
Author content
All content in this area was uploaded by Patrick T Davies on Nov 19, 2018
Content may be subject to copyright.
The role of marital discord and parenting in
relations between parental problem drinking
and child adjustment
Peggy S. Keller,
1
E. Mark Cummings,
1
and Patrick T. Davies
2
1
University of Notre Dame, USA;
2
University of Rochester, USA
Background: Research suggests that children exposed to parental drinking problems are at risk for
maladjustment. However, the potential impact of drinking problems in a community sample and the
processes involved in the relationship between parental drinking and child outcomes have rarely been
examined. Method: A community sample of 235 mothers and fathers of kindergarten children com-
pleted measures of problem drinking symptoms, family functioning and child adjust-
ment. Results: Model tests indicate that problem drinking was associated with greater marital conflict,
and that marital conflict was related to ineffective parenting which was in turn related to poorer child
adjustment. Conclusions: Even in a community sample, parental problem drinking behaviors are
associated with reduced family functioning that relates to child outcomes. Keywords: Alcohol, par-
enting, marital relations, child adjustment.
Studies of children and adolescents report links be-
tween parental drinking and adjustment problems
(Christensen & Bilenberg, 2000). Toddlers of parents
suffering from alcohol dependence have lower be-
havioral compliance than toddlers of healthy parents
(Das Eiden, Leonard, & Morrisey, 2001) and 6–9-
year-old children of alcoholics (COAs) experience
greater anxiety and depression than non-COAs
(Roosa, Dumka, & Tein, 1996). COAs have more
problems with oppositional and conduct disorders
than non-COAs (Reich, Earls, Frankel, & Shayka,
1993). Adolescent COAs are more likely to suffer
from substance use problems than the general
population (Chassin, Pitts, & Prost, 2002).
Parental alcohol dependence thus has deleterious
effects on children. However, little is known about
sub-clinical levels of drinking problems and child
outcomes. Studies have found that measures of both
alcohol dependence and abuse are related to per-
sonal and family functioning (Harmer, Sanderson, &
Martin, 1999; Leonard & Roberts, 1998a). In-
dividuals with sub-clinical levels of drinking prob-
lems may also be characterized by impaired
functioning. These sub-clinical problems are more
prevalent than clinical drinking problems and may
have a more pervasive impact on child adjustment
than strictly defined alcohol dependence.
Another gap concerns the processes involved in
the effects of parental alcohol problems on children.
Few studies have tested models accounting for fam-
ily processes that contribute to this relationship.
Roosa, Tein, Groppenbacher, Michaels, and Dumka
(1993) identified a pathway from parental drinking
problems and negative life events to unsupportive
parenting and inconsistent discipline, which related
to child depression. El-Sheikh and Flanagan (2001)
found that the frequency and intensity of parent–
child conflict mediated the relationship between
parental problem drinking and internalizing and
externalizing problems in children. In one of the few
studies exploring marital functioning as a possible
mediator, they found that marital aggression medi-
ated the effects of paternal problem drinking on
children’s internalizing, externalizing and social
adjustment problems.
The purpose of this study is to examine the role
of marital conflict, in addition to parenting, in
relations between parental problem drinking and
child internalizing and externalizing problems.
Specifically, this study is designed to consider a
possible pathway from drinking to marital conflict
to child adjustment as well as a more complicated
pathway in which marital conflict is indirectly
related to child adjustment through parenting.
Previous research suggests that marital conflict
may be particularly important for the study of
parental problem drinking in families. Couples
characterized by problem drinking exhibit poor
communication practices such as interruption,
difficulties listening and speaking ineffectively
(Kelly, Halford, & Young, 2002). A study of alco-
holic men and their wives found that both spouses
engaged in more verbal aggression than normed
samples (O’Farrell, Murphy, Neavins, & Van Hut-
ton, 2000). A notable study found a positive re-
lationship between men’s drinking and use of
physical aggression within the marital relationship,
even after controlling for demographics, marital
hostility and marital satisfaction (Leonard & Rob-
erts, 1998b). Exposure to this type of marital con-
flict is associated with children’s adjustment
problems, including internalization and externali-
zation (Cummings & Davies, 1994; Cummings,
Davies, & Campbell, 2000).
Journal of Child Psychology and Psychiatry 46:9 (2005), pp 943–951 doi: 10.1111/j.1469-7610.2004.00399.x
ÓAssociation for Child Psychology and Psychiatry, 2004.
Published by Blackwell Publishing, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA
Marital relations may also affect child outcomes
through parenting. Marital difficulties and poor
parenting have been repeatedly linked (Erel & Bur-
man, 1995). Fathers are less supportive and engaged
with their children following a conflictual discussion
with their wives than following a non-conflictual
discussion (Kitzmann, 2000). Recent work also
shows that marital conflict is a predictor of sub-
sequent increases in parenting difficulties (Davies,
Sturge-Apple, & Cummings, in press). Researchers
have theorized that marital conflict acts as a stressor
that decreases effective parenting, or that negativity
spills over into the parent–child relationship (Grych,
2002). A recent study of over 2500 families with
target children ranging from 2 to 18 years found that
harsh discipline, parental involvement, parental
presence and parent–child conflict acted as medi-
ators of the relationship between marital discord and
child adjustment (Buehler & Gerard, 2002).
Parental problem drinking may also be directly
associated with poor parenting. A recent review
suggests that substance abuse disorders are asso-
ciated with harsh discipline, authoritarian or per-
missive parenting styles and unrealistic expectations
of children’s abilities (Mayes & Truman, 2002).
Problem drinking fathers are less sensitive, positive
and verbal with their infants than control fathers
(Das Eiden, Chavez, & Leonard, 1999). However,
especially in the case of community levels of problem
drinking, the marital relationship may have greater
sensitivity to factors that undermine communication
and confidence in partners than the parent–child
relationship. Children may not be aware of low levels
of symptoms that do not generally damage parenting
abilities until they become a source of conflict in the
interparental relationship. One instance of intoxica-
tion resulting in regrettable behavior may be an
ongoing source of conflict within the marital rela-
tionship, although the ability to parent was only
affected in the immediate aftermath. However, the
conflict resulting from the problem drinking may
then create ongoing parenting difficulties. Thus,
everyday alcohol problems may more often have an
indirect effect on parenting through marital func-
tioning rather than a direct one.
This study examines these hypotheses. Specific-
ally, it is proposed that parental problem drinking is
best examined from a family process perspective and
will be associated with marital conflict; this conflict
will be both directly and indirectly (through parent-
ing) related to child adjustment. Although the data
used in this study are correlational and formal tests
of causality are not possible, the proposed model
represents an important early step in delineating the
role of problem drinking in families. For the pur-
poses of this study, problem drinking is a continu-
ous variable designed to capture the entire spectrum
of alcohol problems, both clinical and sub-clinical,
including dependence, other maladaptive drinking
patterns and social, work-related or legal problems.
By operationalizing problem drinking in this way, the
current study examines the impact of symptoms in
the everyday life of community families, avoiding
distinctions between health and disorder.
This study also advances research by considering
both maternal and paternal problem drinking. Many
previous studies do not examine the role of maternal
alcohol problems, partly because women are less
likely to engage in problem drinking. However,
maternal alcohol problems also pose risks to chil-
dren’s development. Recent studies indicate that
maternal drinking is associated with children’s
social and behavioral problems and that this rela-
tionship is mediated by marital conflict and parent–
child conflict (El-Sheikh & Flanagan, 2001). The
current study considers maternal and paternal
problem drinking together, rather than separately,
because each often occurs in the context of the other,
with maternal and paternal problem drinking having
similar effects on child and family functioning
(El-Sheikh & Flanagan, 2001). Thus, children’s
exposure to problem drinking may best be thought of
as a family-wide risk factor rather than attempting to
separate the effects of maternal and paternal drink-
ing by focusing on one parent’s problems.
Methods
Participants
Data are from a sample of 236 families from South
Bend, IN, Rochester, NY and their surrounding areas
taking part in a larger research study. Families were
recruited using flyers sent home with students in local
schools, postcards sent to several local neighborhoods,
flyers placed at daycare agencies, booths at community
functions and referrals from other participants. The
resulting sample reflects similar demographic charac-
teristics as the counties from which the sample was
drawn. Families were eligible to participate if they met
the following criteria: couples must have been living
together for at least three years, have one child cur-
rently enrolled in kindergarten, and could complete
questionnaires in English. One family was excluded
because they were not in a stable relationship.
Of the participating families, 88.5% were married.
Couples had been living together for an average of
11 years (SD ¼4.9). Mean age of fathers was 36.8 years
(SD ¼6.2); mean age of mothers was 35.0 years (SD ¼
5.6). Children ranged in age from 4.9 to 7.1 years old
(M¼6.0, SD ¼.5). Participants were 77% White, 16.2%
Black and 6.8% other races. The median and mean
family income was between $40,000 and $54,999 a
year. Approximately 95% of female parents and 89% of
male parents were the participating children’s biological
parents. For ease of discussion, all parents will be re-
ferred to as mothers or fathers.
Procedure
These data were drawn from a larger study and were
collected during two laboratory visits. The research was
944 Peggy S. Keller, E. Mark Cummings, and Patrick T. Davies
conducted under the oversight of the Institutional
Review Boards (IRBs) at both data collection sites; all
procedures and consent and assent forms were
approved by the IRBs before the study began.
Measures
Parental problem drinking. Both parents completed
the Parental Alcohol Experiences Scale (PAE; Windle,
1997). The PAE is a 15-item inventory of symptoms
associated with alcohol dependence or abuse and was
designed to provide a continuous measure of clinical
and sub-clinical levels of problem drinking. Items in-
clude assessments of drinking before or during work or
school, missing work or school because of drinking,
drinking that resulted in a fight or disagreement with a
family member, with the spouse, or with a stranger,
regretting afterward things that were done while
drinking, thinking about cutting down on drinking,
getting drunk several days in a row, passing out from
drinking, getting into trouble with the law because of
drinking, drinking alone, drinking to get rid of a hang-
over, drinking to forget one’s troubles, being charged
with drinking and driving, and having a drinking and
driving related accident. These experiences have been
indicated as important in previous studies of alcohol
use (Davies & Windle, 1997) and were chosen because
of the wide range of domains assessed (social problems,
legal problems, work/school problems, etc.). Particip-
ants rated how often they have experienced each
symptom within the past year on a scale from 1 (never)
to 5 (frequently/more than 10 times in the past
12 months). Thus, scores range from 15 to 75, with
higher scores reflecting greater problem drinking. This
measure has high six-month retest reliability (r¼.71),
and internal consistency (a¼.81; Davies & Windle,
1997; a¼.83; Windle, 1996). Concerning convergent
validity, higher scores on the PAE are associated with
less drinking inhibition, early onset drinking and
drinking to cope with problems.
Participants completed the inventory with regard to
their own and their partners’ drinking. Four measures
were obtained: mother self-report, mother report of
father problem drinking, father self-report and father
report of mother problem drinking, a¼.73, .86, .87 and
.81, respectively. Mother self-report was significantly
correlated with father report of mother problem
drinking, r¼.55, p¼000. Similarly, father self-report
was correlated with mother report of father drinking,
r¼.65, p< .001.
Marital conflict. Three dimensions were assessed:
marital hostility based on the O’Leary Porter Scale
(OPS; Porter & O’Leary, 1980), frequency and severity of
conflict based on the Conflict and Problem-Solving
Scales (CPS; Kerig, 1996) Frequency and Severity sub-
scale, and resolution of conflict based on the CPS
Resolution sub-scale. All are widely used measures of
marital conflict. Mothers and fathers reported on each
separately. The OPS consists of 10 items rated on a five-
point Likert scale. Possible scores range from 0 to 40,
with higher scores reflecting greater marital hostility. In
this sample, a¼.81 for mothers and .80 for fathers.
Scores on the Frequency and Severity subscale range
from 3 to 18 with higher scores indicating more fre-
quent and severe conflict. Scores on Resolution range
from )48 to 24, with higher scores reflecting greater
conflict resolution. The CPS has been shown to have
internal consistencies ranging from .75 to .98 (Kerig,
1996).
Parenting. Two dimensions were assessed: discipline
and psychological control. Discipline was measured
using the Inconsistent Discipline Scale of the Alabama
Parenting Questionnaire (APQ; Shelton, Frick, &
Wootton, 1996). The APQ is a widely used measure, and
contains six items describing disciplinary behaviors
rated on a five-point Likert scale (1 ¼‘never’, 5 ¼‘al-
ways’). Scores thus range from 6 to 30, with higher
scores reflecting more inconsistent discipline. The reli-
ability coefficients were .71 for mothers and .66 for
fathers. Psychological control was assessed using the
parent version of the Children’s Report of Parental
Behavior (CRPBI; Margolies & Weintraub, 1977). The
CRPBI consists of 15 items describing parental intru-
siveness, control through guilt and instilling persistent
anxiety that are rated on a five-point Likert scale (1 ¼
‘never’, 5 ¼‘always’). Thus, scores range from 15 to 75,
with higher scores indicating greater psychological
control. The reliability coefficients were .84 for both
mothers and fathers. Mothers and fathers completed
each questionnaire with regard to their practices with
the targeted child.
Child adjustment problems. Child adjustment
problems were measured using the Child Behavior
Checklist (CBCL; Achenbach et al., 1991) Internalizing
and Externalizing scales. The CBCL is a widely used
measure, consisting of 58 items. Parents rate on a
three-point Likert scale how much their children
engage in a number of aggressive, delinquent, with-
drawing, anxious or depressed behaviors. Scores range
from 0 to 66 for the externalizing score and from 0 to
50 for the internalizing score. In the current sample,
alphas were .88 and .84 for mothers and .90 and .88
for fathers on the externalizing and internalizing
scales, respectively.
Results
Data reduction
Scores on all measures were aggregated by avera-
ging across reporters because of the focus on
family-wide rather than individual functioning.
Correlations between reports on the various meas-
ures support averaging across reports: The corre-
lation between reports on the OPS is .59, p< .001;
on frequency and severity of conflict was .46,
p< .001; and on conflict resolution was .53,
p< .001. The correlations between reports were
significant on both dimensions of parenting: r¼
.31, p< .001 for psychological control and r¼.26,
p< .001 for inconsistent discipline. Correlations
between reports of child adjustment were also
significant, r¼.58, p< .001 for internalizing and
r¼.51, p< .001 for externalizing.
Problem drinking and child adjustment 945
Descriptive and preliminary analyses
Table 1 presents means, standard deviations and
intercorrelations between aggregated variables.
Mean scores on the PAE indicate relatively low
levels of problem drinking, consistent with a com-
munity sample. However, fully two-thirds of the
sample reported engaging in at least one problem
drinking behavior. In addition, there was a wide
range of scores, from no problem drinking (PAE ¼
15) to high levels of problem drinking (PAE ¼46).
Some of the more commonly endorsed items
included drinking alone (29.6% of women and
45.9% of men), thinking about cutting down on
drinking (16.1% of women and 30.7% of men),
having an argument with one’s partner because of
drinking (12.4% of women and 23.2% of men),
drinking to forget one’s troubles (15.9% of women
and 23.6% of men) and regretting afterward the
things done (14.6% of women and 23.6% of men).
Twenty-one percent of men and 9.3% of women
endorsed five or more symptoms, indicating
potentially serious levels of problem drinking. It is
important to examine relations between these
community levels of problem drinking for possible
relations to family and child functioning because
they represent the pattern of drinking to which
many children are likely exposed.
Inspection of the correlations between variables
provides an indication of the utility of creating
latent variables. All of these correlations within
constructs were in the moderate to high range, and
were significant at a¼.001. Correlations between
variables indicating different latent constructs
provide preliminary evidence for the proposed
models. These analyses revealed that both mater-
nal and paternal drinking were related to measures
of marital conflict but not to parenting. Marital
conflict was related to measures of parenting: all
measures of marital functioning were associated
with more inconsistent discipline and marital hos-
tility was associated with greater parental psycho-
logical control. Marital hostility, frequency and
severity of conflict, parental psychological control
and inconsistent discipline were associated with
child adjustment problems.
No significant relationships were detected between
parental problem drinking and child outcomes.
Thus, any relationship between parental problem
drinking and child adjustment is likely indirect
through family processes. Previous research has
examined indirect relationships between variables
based on the notion that relations may be contingent
on intervening family processes (Harold & Conger,
1997). In such instances one looks for evidence of
relationships between the predictor variable and the
intervening variable and between the intervening
variable and the outcome (Davies, Harold, Goeke-
Morey, & Cummings, 2002). Within the current data
set, it is clear that parental drinking is associated
with dysfunctional family processes, particularly
marital conflict, and that this dysfunction may play a
role in the development of adverse child outcomes.
Thus, preliminary analyses support the proposed
model.
Structural equation models
Model tests were conducted using the AMOS 4.0
statistical package (Arbuckle & Wothke, 1999). This
program uses maximum likelihood estimation.
Models examining internalizing and externalizing
problems separately yielded extremely similar re-
sults, thus both measures of child adjustment were
included in an overall model. A first test including all
pathways was conducted and is presented in Fig-
ure 1. Fit indices supported the model as a reason-
able fit for the data, v
2
(22) ¼42.86, p< .01. It is not
unusual to have a high chi-square statistic when
testing a complex model because the chi square test
is biased toward detecting differences between ob-
served and null covariance structures with large
sample sizes (Widaman & Thompson, 2003). Thus,
alternative fit indices are recommended as important
evidence of fit (Bentler, 1990) and are commonly
employed in assessing models (Duncan, Duncan,
Stryker, Li, & Alpert, 1999). For example, if the v
2
/df
ratio is between one and three (Arbuckle & Wothke,
1999), the goodness of fit index (GFI) is greater than
.90, the comparative fit index (CFI) is above .95 and
the root mean square error of approximation
(RMSEA) is below .08 (Browne & Cudek, 1993), the
Table 1 Means, standard deviations and intercorrelations among all primary variables
12345678MSD
1. Maternal drinking 16.20 2.49
2. Paternal drinking .549*** 17.87 4.97
3. Marital hostility .083 .271*** 11.97 4.71
4. Frequency and severity of conflict .076 .182** .615*** 8.11 2.78
5. Resolution of conflict ).171** ).266*** ).682*** ).547*** 4.13 10.62
6. Parental psychological control .028 .046 .169** .051 ).114 44.26 7.59
7. Inconsistent discipline .128 .056 .291*** .137* ).145* .250*** 13.44 2.49
8. Maternal report of child problems .076 .108 .187** .137* ).090 .106 .335*** 7.13 5.35
9. Paternal report of child problems .075 .048 .153* .134* ).085 .259*** .372*** .567*** 10.28 6.70
Note: Correlations among indicators within latent constructs are denoted in bold. *p< .05; **p< .01; ***p< .001.
946 Peggy S. Keller, E. Mark Cummings, and Patrick T. Davies
model is considered a good fit for the data. The model
presented in Figure 1 meets all these criteria, v
2
/
df ¼1.948, GFI ¼.962, CFI ¼.961, and RMSEA ¼
.064.
An examination of the path coefficients reveals that
parental problem drinking was associated with
greater marital conflict, b¼.17, p< .05, marital
conflict was related to poor parenting, b¼.40,
p< .01, and poor parenting was associated with
greater child problems, b¼.74, p< .01. The re-
lationships between parental drinking problems and
parenting, b¼.12, marital functioning and child
problems, b¼).09, and parental drinking problems
and child adjustment, b¼).01, were all nonsignific-
ant. Thus, the model supports the hypothesis that
parental problem drinking is associated with child
problems through both marital conflict and poor
parenting.
To further test this hypothesis, the insignificant
pathways in the above model were removed and the
model was run again. Results are presented in
Figure 2. This model was also a reasonable fit for
the data, v
2
(25) ¼45.05, p< .01, v
2
/df ¼1.802,
GFI ¼.961, CFI ¼.963, RMSEA ¼.059. To test
whether there is a significant difference in fit be-
tween the two models, the comparison chi-square
was calculated following Dunn, Everitt, and Pickles
(1993): v
2
¼45.05–42.86 ¼2.19; df ¼25–22 ¼3.
The resulting statistic, v
2
(3) ¼2.19, was nonsig-
nificant, p¼.466, indicating that there is no sig-
nificant difference between the two models in terms
of fit. Examination of alternative fit indices shows
slight improvement in the CFI and RMSEA and a
very slight decrease in the GFI. Thus, the model
excluding pathways from parental problem drink-
ing to parenting and child outcomes and between
marital conflict to child problems fit just as well as
the complete model and, because of parsimony, is
preferred.
An examination of the path coefficients in this
model reveals additional support for the hypothesis
that parental problem drinking places children at
risk because of relations with marital conflict. Par-
ental problem drinking was associated with greater
marital conflict, b¼.18, p< .05; marital conflict was
related to poor parenting, b¼.38, p< .01; poor
parenting was related to greater child problems, b¼
.66, p< .001.
Ineffective
Parenting
Marital
Conflict
Child
Problems
discipli
internal
external
ops freqsev
psycntrl
resolve
Parental
Problem
Drinking
pdrink
mdrink
.40**
–.01
.12
–.09
.58f
.69***
.90f-.77***
.91f
.66*** .38f
.17*
.74***
.68f
.83***
2(22) = 43.175, p= .004
2/df = 1.963
CFI = .960
GFI = .962
RMSEA = .064
Figure 1 Complete model. Note: mdrink ¼maternal drinking; pdrink ¼paternal drinking; OPS ¼O’Leary Porter
Scale; freqsev ¼frequency and severity of conflict; resolve ¼resolution of conflict; discipli ¼inconsistent discipline;
psycntrl ¼psychological control; internal ¼parental report of child internalizing problems; external ¼parental
report of child externalizing problems; f denotes fixed path; *p< .05; **p< .01; ***p< .001.
Problem drinking and child adjustment 947
Discussion
This study examined how community levels of alco-
hol problems might affect child adjustment through
increased marital conflict and, in turn, less effective
parenting. Results indicate that problem drinking
may harm children through its association with
marital and parenting difficulties. Results further
indicate that marital conflict has an indirect, rather
than direct, relationship with child internalization
and externalization in this context: Marital conflict is
related to ineffective parenting, specifically incon-
sistent discipline and psychological control. These
parenting problems are associated with child inter-
nalizing and externalizing problems.
These results highlight the potential importance of
marital functioning in this context. Many research-
ers have not considered the role of the interparental
relationship in their speculations about causal pro-
cesses (Edwards, Leonard, & Das Eiden, 2001; Gal-
lant, Gorey, Gallant, Perry, & Ryan, 1998; Moser &
Jacob, 1997). However, the current study suggests
that marital conflict may be one reason why some
children develop adjustment problems and that
models of the possible effects of problem drinking
that do not include marital functioning are in-
complete. The marital relationship may be particu-
larly sensitive to symptoms of alcohol problems. For
example, instances of problem drinking may be more
likely to occur in the presence of a spouse (e.g., while
enjoying an evening out) and involve behavior that is
offensive to a spouse (e.g., flirting with others, saying
something embarrassing, engaging in criminal be-
havior). Research on the consequences of intoxica-
tion supports this hypothesis (Curtin & Fairchild,
2003; McClelland & Teplin, 2001). This offensive
behavior may become the topic of conflict within the
marital relationship. In the current sample, marital
conflict was not directly related to children’s
adjustment. This is contrary to hypotheses and
somewhat at odds with previous research (Cum-
mings & Davies, 1994). It is possible that in this
community sample, rates of marital conflict were not
severe enough to directly relate to children’s behav-
ior. However, the parent-child relationship may have
been more sensitive to low levels of marital conflict.
Results build upon previous studies. El-Sheikh
and Flanagan (2001) found support for both
Ineffective
Parenting
Marital
Conflict
Child
Problems
discipli
internal
external
ops freqsev
psycntrl
resolve
Parental
Problem
Drinking
pdrink
mdrink
.38**
.58f
.69***
.89f-.77***
.91f
.70*** .38f
.18*
.66***
.69f
.82***
2(25) = 45.049, p= .008
2/df = 1.802
CFI = .963
GFI = .961
RMSEA = .059
Figure 2 Final model. Note: mdrink ¼maternal drinking; pdrink ¼paternal drinking; OPS ¼O’Leary Porter Scale;
freqsev ¼frequency and severity of conflict; resolve ¼resolution of conflict; discipli ¼inconsistent discipline; psy-
cntrl ¼psychological control; internal ¼parental report of child internalizing problems; external ¼parental report of
child externalizing problems; f denotes fixed path; *p< .05; **p< .01; ***p< .001.
948 Peggy S. Keller, E. Mark Cummings, and Patrick T. Davies
parenting and marital functioning as mediators in
child adjustment with a sample of families suffering
from clinical alcohol problems. El-Sheikh and Flan-
agan’s (2001) study suggested that parent–child
conflict may be the more robust mediator of the two,
at least in the case of children’s externalizing
symptoms. The present study suggests that a more
complex pathway from parental alcohol problems to
child functioning may be at work. That is, the par-
ent–child relationship may be one crucial link in a
chain of events that results in increased child prob-
lems. Specifically, parenting difficulties may be the
result not just of the direct impact of parental
drinking, but of the indirect impact of parental
drinking through increased marital conflict. Thus,
parenting and marital functioning may not be two
separate and distinct factors.
This study supports measuring parental adjust-
ment problems such as alcohol problems on a
continuum. Evidence was found for a possible
chain reaction from occasional problem drinking to
marital and then parenting problems and, finally,
child adjustment problems. Although the relation-
ship between drinking and marital conflict was
small, it was in the hypothesized direction, of a
magnitude often found in social science research,
and of interpretable significance. The implication is
that even occasional problem drinking, if it involves
behavior the spouse finds offensive, could provide a
significant source of conflict. Unless the conflict is
handled appropriately, the offended spouse may
continue to feel hurt, angry or unable to trust and
the offending spouse may become defensive or
withdraw from problem-solving attempts. The
tendency for some couples’ conflicts to escalate in
this way has been well documented (Gottman,
1993). It is this escalated, unresolved conflict that
may lead to parenting difficulties and ultimately to
child problems (Erel & Burman, 1995). It is
important to note that the majority of children in
the current study are not suffering from clinical
levels of adjustment problems. However, even when
increased behavior problems do not reach clinical
thresholds, they may represent a potential for
developing more serious problems.
This model of the effects of parental problem
drinking may not generalize to clinical samples. For
example, severe alcohol dependence may be directly
related to parenting (Mayes & Truman, 2002).
Individuals who are frequently intoxicated are likely
unable to develop consistent disciplinary strategies
or be emotionally available for their children. Fur-
thermore, clinical levels of alcohol problems may be
associated with other forms of psychopathology,
such as depression (El-Sheikh & Flanagan, 2001),
economic distress (Mullahy & Sindelar, 1992) and
poor physical health (Adrian & Barry, 2003) that help
account for adverse child outcomes. Findings for age
and developmental status are small and inconsistent
for early to middle childhood (Cummings & Davies,
1994), providing little basis for understanding these
results as a reflection of the sample’s age. These
questions merit study in future research.
It should also be noted that neither the traditional
method of Baron and Kenny (1986) nor the more
recent methods (MacKinnon, Lockwood, Hoffman,
West, & Sheets, 2002) were employed to test for
mediation. The authors instead used a more ex-
ploratory modeling method (Allison, 1995; Kenny,
Kashy, & Bolger, 1998) to assess a range of roles of
problem drinking in families that included multiple
constructs simultaneously and a complex pathway
from drinking to child adjustment through both
conflict and parenting. Future research utilizing
more stringent tests of mediation is needed to further
clarify these relationships. An additional limitation is
the use of cross-sectional data. Thus, the direction of
relationships is speculative. It is also likely that
these relationships are bidirectional, with parental
drinking associated with difficulties in the marital
relationship that then result in greater parental
drinking as a method for coping with the increased
marital disharmony. Ineffective parenting strategies
may themselves be a source of marital conflict, if one
parent disagrees with the other’s practices. Fur-
thermore, difficult child temperament or adjustment
problems may cause parenting problems, marital
conflict, and increased parental drinking if parents
cannot adequately manage their children. For
example, parents consume more alcohol after inter-
acting with a child confederate manifesting behavior
problems (Pelham et al., 1998). Future research
should employ longitudinal designs, and data be-
yond parental reports, to further test causal chains.
Another important avenue for additional research
involves tests of the specific forms of conflict most
linked to drinking and most linked to parenting
problems associated with drinking. Recent studies
indicate that conflict is not a unidimensional phe-
nomenon; some conflict tactics should be considered
destructive and harmful to children while other tac-
tics considered constructive and potentially benefi-
cial for children (Goeke-Morey, Cummings, Harold,
& Shelton, 2003). Future research should therefore
consider the unique characteristics of marital con-
flict in the context of parental drinking. Previous
research has found that these characteristics may
include increased aggression and hostility (Leonard
& Roberts, 1998b; O’Farrell et al., 2000).
Acknowledgements
Preparation of this paper was supported in part by a
grant from the National Institute of Mental Health
(Project R01 MH57318) awarded to Patrick Davies
and E. Mark Cummings. The authors are grateful to
the children and parents who participated in this
project. Their gratitude is also expressed to the staff
and students who assisted on various stages of the
Problem drinking and child adjustment 949
project, including: Courtney Forbes, Marcie Goeke-
Morey, Amy Keller, and Michelle Sutton.
Correspondence to
E. Mark Cummings, Department of Psychology,
Haggar Hall, University of Notre Dame, Notre Dame,
IN 46556, USA; Tel: (574) 631-3404; Fax: (574) 631-
8883; Email: Cummings.10@nd.edu
References
Achenbach, T.M., Howell, C.T., Quay, H.C., & Con-
nors, C.K. (1991). National survey of problems and
competencies among four to sixteen-year-olds.
Monographs of the Society for Research in Child
Development,56.
Adrian, M., & Barry, S.J. (2003). Physical and mental
health problems associated with the use of alcohol
and drugs. Substance Use and Misuse,38, 1575–
1614.
Allison, P.D. (1995). The impact of random predictors
on comparison of coefficients between models: Com-
ment on Clogg, Petkova, and Haritou. American
Journal of Sociology,100, 1294–1305.
Arbuckle, J.L., & Wothke, W. (1999). Amos 4.0 Users’
Guide. Chicago: Smallwaters.
Baron, R.M., & Kenny, D.A. (1986). The moderator–
mediator variable distinction in social psychological
research: Conceptual, strategic, and statistical con-
siderations. Journal of Personality and Social Psy-
chology,51, 1173–1182.
Bentler, P.M. (1990). Comparative fit indices in struc-
tural equation in structural models. Psychological
Bulletin,107, 238–246.
Browne, M.W., & Cudeck, R. (1993). Alternative ways of
assessing model fit. In K.A. Bollen & J.S. Long (Eds.),
Testing structural equation models (pp. 136–162).
Newbury Park, CA: Sage.
Buehler, C., & Gerard, J.M. (2002). Marital conflict,
ineffective parenting, and children’s and adolescents’
maladjustment. Journal of Marriage and Family,64,
78–92.
Chassin, L., Pitts, S.C., & Prost, J. (2002). Binge
drinking trajectories from adolescence to emerging
adulthood in a high-risk sample: Predictors and
substance use outcomes. Journal of Consulting and
Clinical Psychology,70, 67–78.
Christensen, H.B., & Bilenberg, N. (2000). Behavioral
and emotional problems in children of alcoholic
mothers and fathers. European Child and Adolescent
Psychiatry,9, 219–226.
Cummings, E.M., & Davies, P.T. (1994). Children and
marital conflict: The impact of family dispute and
resolution. New York: Guilford.
Cummings, E.M., Davies, P.T., & Campbell, S.B.
(2000). Developmental psychopathology and family
process: Theory, research and clinical implications.
New York: Guilford.
Curtin, J.J., & Fairchild, B.A. (2003). Alcohol and
cognitive control: Implications for regulation of beha-
vior during response to conflict. Journal of Abnormal
Psychology,112, 424–436.
Das Eiden, R., Chavez, F., & Leonard, K.E. (1999).
Parent–infant interactions among families with alco-
holic fathers. Development and Psychopathology,11,
745–762.
Das Eiden, R., Leonard, K.E., & Morrisey, S. (2001).
Paternal alcoholism and toddler noncompliance.
Alcoholism: Clinical and Experimental Research,25,
1621–1633.
Davies, P.T., Harold, G.T., Goeke-Morey, M.C., & Cum-
mings, E.M. (2002). Child emotional insecurity and
interparental conflict. Monographs of the Society for
Research in Child Development,67(3).
Davies, P.T., Sturge-Apple, M.L., & Cummings, E.M. (in
press). Interdependencies among interparental dis-
cord and parenting styles: The role of adult attributes
and relationship characteristics. Development and
Psychopathology.
Davies, P.T., & Windle, M. (1997). Gender-specific
pathways between maternal depressive symptoms,
family discord, and adolescent adjustment. Develop-
mental Psychology,33, 657–668.
Duncan, T.E., Duncan, S.E., Strycker, L.A., Li, F., &
Alpert, A. (1999). An introduction to latent variable
growth curve modeling: Concepts, issues, and ap-
plications. Mahwah, NJ: Lawrence Erlbaum.
Dunn, G., Everitt, B., & Pickles, A. (1993). Modeling
covariances and latent variables using EQS. New
York: Chapman & Hall.
Edwards, E.P., Leonard, K.E., & Das-Eiden, R. (2001).
Temperament and behavioral problems among in-
fants in alcoholic families. Infant Mental Health
Journal,22, 374–392.
El-Sheikh, M., & Flanagan, E. (2001). Parental problem
drinking and children’s adjustment: Family conflict
and parental depression as mediators and modera-
tors of risk. Journal of Abnormal Child Psychology,29,
417–432.
Erel, O., & Burman, B. (1995). Interrelatedness of
marital relations and parent–child relations: A meta-
analytic review. Psychological Bulletin,118, 108–
132.
Gallant, W.A., Gorey, K.M., Gallant, M.D., Perry, J.L., &
Ryan, P.K. (1998). The association of personality
characteristics with parenting problems among alco-
holic couples. American Journal of Drug and Alcohol
Abuse,24, 119–129.
Goeke-Morey, M.C., Cummings, E.M., Harold, G.T., &
Shelton, K.H. (2003). Categories and continua of
destructive and constructive marital conflict tactics
from the perspective of U.S. and Welsh children.
Journal of Family Psychology,17, 327–338.
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.
Grych, J.H. (2002). Marital relationships and parent-
ing. In M.H. Bornstein (Ed.), Handbook of parenting:
Vol. 4: Social conditions and applied parenting (2nd
edn, pp. 203–225). Mahwah, NJ: Lawrence Erlbaum
Associates.
Harmer, A.L.M., Sanderson, J., & Martin, P. (1999).
Influence of negative childhood experiences on psy-
chological functioning, social support, and parenting
for mothers recovering from addiction. Child Abuse
and Neglect,23, 421–433.
950 Peggy S. Keller, E. Mark Cummings, and Patrick T. Davies
Harold, G.T., & Conger, R.D. (1997). Marital conflict
and adolescent distress: The role of adolescent
awareness. Child Development,68, 333–350.
Kelly, A.B., Halford, W.K., & Young, R.M. (2002).
Couple communication and female problem drinking:
A behavioral observation study. Psychology of Addic-
tive Behaviors,16, 269–271.
Kenny, D.A., Kashy, D.A., & Bolger, N. (1998). Data
analysis in social psychology. In D.T. Gilbert, S.T.
Fiske, & G. Lindzey (eds.), The handbook of social
psychology (pp. 233–265). Boston: McGraw-Hill.
Kerig, P.K. (1996). Assessing the links between inter-
parental conflict and child adjustment: The Conflicts
and Problem-Solving Scales. Journal of Family Psy-
chology,10, 454–473.
Kitzmann, K.M. (2000). Effects of marital conflict on
subsequent triadic family interactions and parenting.
Developmental Psychology,36, 3–13.
Leonard, K.E., & Roberts, L.J. (1998a). Marital aggres-
sion, quality, and stability in the first year of
marriage: Findings from the Buffalo Newlywed Study.
In T.N. Bradbury (Ed.), The developmental course of
marital dysfunction (pp. 44–73). New York: Cam-
bridge University Press.
Leonard, K.E., & Roberts, L.J. (1998b). The effects of
alcohol on the marital interactions of aggressive and
nonaggressive husbands and their wives. Journal of
Abnormal Psychology,107, 602–615.
MacKinnon, D.P., Lockwood, C.M., Hoffman, J.M.,
West, S.G., & Sheets, V. (2002). A comparison of
methods to test mediation and other intervening
variable effects. Psychological Methods,7, 83–104.
Margolies, P.J., & Weintraub, S. (1977). The revised 56-
item CRPBI as a research instrument: Reliability and
factor structure? Journal of Clinical Psychology,33,
472–476.
Mayes, L.C., & Truman, S.D. (2002). Substance abuse
and parenting. In M.H. Borstein (Ed.), Handbook of
parenting. Mahwah, NJ: Lawrence Erlbaum Associ-
ates.
McClelland, G.M., & Teplin, L.A. (2001). Alcohol
intoxication and violent crime: Implications for public
health policy. American Journal of Addictions,10, 70–
85.
Moser, R.P., & Jacob, T. (1997). Parent–child interac-
tions and child outcomes as related to gender of
alcoholic parent. Journal of Substance Abuse,9, 189–
208.
Mullahy, J., & Sindelar, J.L. (1992). Effects of alcohol
on labor market success: Income, earnings, labor
supply, and occupation. Alcohol Health and Research
World,16, 134–139.
O’Farrell, T.J., Murphy, C.M., Neavins, T.M., & Van
Hutton, V. (2000). Verbal aggression among male
alcoholic patients and their wives in the year before
and two years after alcoholism treatment. Journal of
Family Violence,15, 295–310.
Pelham, W.E., Lang, A.R., Atkeson, B., Murphy, D.A.,
Gnagy, E.M., Greiner, A.R., Vodde-Hamilton, M., &
Greenslade, K.E. (1998). Effects of deviant child
behavior on parental alcohol consumption: Stress-
induced drinking in parents of ADHD children.
American Journal of Addictions,7, 103–114.
Porter, B., & O’Leary, D. (1980). Marital discord and
childhood behavior problems. Journal of Abnormal
Child Psychology,8, 287–295.
Reich, W., Earls, F., Frankel, O., & Shayka, J.J. (1993).
Psychopathology in children of alcoholics. Journal of
the American Academy of Child and Adolescent
Psychiatry,32, 995–1002.
Roosa, M.W., Dumka, L., & Tein, J.Y. (1996). Family
characteristics as mediators of the influence of
problem drinking and multiple risk status on child
mental health. American Journal of Community Psy-
chology,24, 607–624.
Roosa, M.W., Tein, J.Y., Groppenbacher, N., Michaels,
M., & Dumka, L. (1993). Mothers’ parenting behavior
and child mental health in families with a problem
drinking parent. Journal of Marriage and the Family,
55, 107–118.
Shelton, K.K., Frick, P.J., & Wootton, J. (1996).
Assessment of parenting practices in families of
elementary school-age children. Journal of Clinical
Child Psychology,25, 317–329.
Widaman, K.F., & Thompson, J.S. (2003). On specifying
the null model for incremental fit indices in structural
equation modeling. Psychological Methods,8, 16–37.
Windle, M. (1996). On the discriminative validity of a
family history of problem drinking index with a
national sample of young adults. Journal of Studies
on Alcohol,57, 378–386.
Windle, M. (1997). Parental Alcohol Experiences Scale.
Unpublished inventory, Research Institute on Addic-
tions, Buffalo, NY.
Manuscript accepted 20 July 2004
Problem drinking and child adjustment 951
