Using a Vulnerability-Stress-Adaptation Framework to Predict Physical Aggression Trajectories in Newlywed Marriage

Abstract

The authors used a vulnerability-stress-adaptation framework to examine personality traits and chronic stress as predictors of the developmental course of physical aggression in the early years of marriage. Additionally, personality traits and physical aggression were examined as predictors of the developmental course of chronic stress. Data from 103 couples collected 4 times over the first 3 years of marriage were analyzed with an actor-partner interdependence model and structural equation modeling techniques. Personality traits of husbands predicted their own physical aggression and stress trajectories, as well as their wives' levels of stress and physical aggression. Personality traits of wives predicted their levels of stress and physical aggression and predicted changes in their physical aggression over time. Both husbands' and wives' changes in stress predicted changes in physical aggression over time. Implications for employment of a vulnerability-stress-adaptation model in the study of physical aggression and for improvement of the efficacy of therapies targeting physical aggression in intimate relationships are delineated.

Full text

Using a Vulnerability–Stress–Adaptation Framework to Predict Physical
Aggression Trajectories in Newlywed Marriage
Amie Langer, Erika Lawrence, and Robin A. Barry
University of Iowa
The authors used a vulnerability–stress–adaptation framework to examine personality traits and chronic
stress as predictors of the developmental course of physical aggression in the early years of marriage.
Additionally, personality traits and physical aggression were examined as predictors of the developmen-
tal course of chronic stress. Data from 103 couples collected 4 times over the first 3 years of marriage
were analyzed with an actor–partner interdependence model and structural equation modeling techniques.
Personality traits of husbands predicted their own physical aggression and stress trajectories, as well as
their wives’ levels of stress and physical aggression. Personality traits of wives predicted their levels of
stress and physical aggression and predicted changes in their physical aggression over time. Both
husbands’ and wives’ changes in stress predicted changes in physical aggression over time. Implications
for employment of a vulnerability–stress–adaptation model in the study of physical aggression and for
improvement of the efficacy of therapies targeting physical aggression in intimate relationships are
delineated.
Keywords: APIM, couples, marital, physical aggression, VSA model
Physical aggression in romantic relationships is surprisingly
common; rates range from 25% to 57% in studies of dating,
cohabiting, engaged, and married couples (e.g., O’Leary et al.,
1989; Schumacher & Leonard, 2005) and from 10% to 20% in
nationally representative surveys (e.g., Straus & Gelles, 1990).
Men and women are equally likely to engage in physical aggres-
sion against their partners. The most frequently employed behav-
iors include grabbing, pushing, and slapping, whereas more severe
behaviors, such as punching and kicking, are less common (e.g.,
Leonard & Roberts, 1998). However, research has established that
even mild forms of physical aggression have implications for both
individual well-being (e.g., depression, anxiety, substance use,
global physical health; Coker et al., 2002; Umberson, Anderson,
Glick, & Shapiro, 1998) and family functioning (e.g., child delin-
quent behaviors and psychopathology; Fantuzzo, DePaola, Lam-
bert, & Martino, 1991). Several theories of partner aggression have
been tested to explicate the presence or onset of physical aggres-
sion (e.g., social learning models; O’Leary, 1988) but not neces-
sarily to explicate its longitudinal course. Moreover, only a few
empirically supported interventions have targeted physical aggres-
sion in intimate relationships, and those that have done so are
limited in their efficacy (e.g., Babcock, Green, & Robie, 2004;
Murphy & Eckhardt, 2005). Our purpose in the present study was
to improve understanding of the longitudinal course of physical
aggression in intimate relationships by employing an existing
theoretical model to examine the factors that lead to changes in
aggression over time.
The Vulnerability–Stress–Adaptation Model
The vulnerability–stress–adaptation (VSA) model of marital
dysfunction, which emerged from a diathesis-stress model of in-
dividual psychopathology (e.g., Zubin & Spring, 1977), provides a
framework for clarifying how marriages change over time (Karney
& Bradbury, 1995). According to the VSA model (see Figure 1),
individuals bring preexisting vulnerabilities to their marriages that
may take the form of personality traits (e.g., neuroticism) and/or
experiential factors (e.g., parental divorce). Such vulnerabilities
are expected to be relatively stable. Marriages are also impacted by
stressful events (e.g., loss of a job) and circumstances (e.g., pov-
erty, chronic illness) that can occur due to chance factors, spouses’
vulnerabilities, and/or adaptive processes. Finally, adaptive pro-
cesses represent interactions between spouses that evolve as cou-
ples respond to stress and are conceptualized as behavioral ex-
changes that may be positive or negative in valence (e.g., conflict
management skills, partner support). Links among vulnerabilities,
stressors, and behaviors, both adaptive and maladaptive, are ex-
pected to lead to changes in marital satisfaction and dissatisfaction
and, ultimately, marital stability and instability. In short, the VSA
model has the potential to explain between-couple variability and
within-couple longitudinal change in marriage.
Regardless of whether they have studied relationship dysfunc-
tion within a VSA framework or within a more traditional behav-
ioral or social learning framework, researchers and therapists have
focused on negative affect and behaviors exhibited during
Amie Langer, Erika Lawrence, and Robin A. Barry, Department of
Psychology, University of Iowa.
Portions of this article were presented at a meeting of the Association for
the Behavioral and Cognitive Therapies in November 2007. Collection and
analysis of these data were supported in part by the following grants to
Amie Langer: Centers for Disease Control and Prevention Grants R49/
CCR721682 and R49/CE721682, part of National Institute for Child and
Human Development Grant RO1 HD046789, and a research grant from the
University of Iowa.
Correspondence concerning this article should be addressed to Amie
Langer, Department of Psychology, University of Iowa, 11 Seashore Hall
East, Iowa City, IA 52242. E-mail: amie-langer@uiowa.edu
Journal of Consulting and Clinical Psychology Copyright 2008 by the American Psychological Association
2008, Vol. 76, No. 5, 756–768 0022-006X/08/$12.00 DOI: 10.1037/a0013254
756

problem-solving interactions (e.g., negative escalation, negative
reciprocity; Gottman & Levenson, 2002). However, research has
established that physical aggression is another process that leads to
relationship dissatisfaction and instability (e.g., Arias, Lyons, &
Street, 1997) and that aggression is a significantly stronger pre-
dictor of divorce than are other behaviors exhibited during conflict
interactions (e.g., Rogge & Bradbury, 1999). Given the research
demonstrating that two thirds of couples who seek couple therapy
report the presence of physical aggression (e.g., O’Leary, Vivian,
& Malone, 1992), we believed that interventions designed to treat
marital discord would likely be more effective to the extent that
physical aggression in particular was a target of change. In light of
these facts, we employed a unique approach to the VSA model by
conceptualizing physical aggression as a key dyadic process in the
model.
Before couple therapies can be revised to include treatment
components that target physical aggression, theoretically guided
empirical research is needed for clarification of the key predictors
of its longitudinal course. Although there is a wealth of evidence
regarding the predictors of physical aggression in general, rela-
tively little has been established about their influence on its de-
velopment over time. The existing research on trajectories of
physical aggression in marriage is equivocal regarding its system-
atic change over time, but there is clear evidence of significant
variability in the perpetration of physical aggression both between
couples in general and within couples over time (e.g., O’Leary et
al., 1989; Schumacher & Leonard, 2005). Thus, it is important to
account for the patterns underlying this variability, as different
factors may explain systematic changes in aggression trajectories.
Using a VSA framework in the present study, we conceptualized
physical aggression as a temporally dynamic process that is influ-
enced by individual vulnerabilities and contextual factors. Within
this premise, trajectories of husbands’ and wives’ physical aggres-
sion were predicted from the direct and indirect influences of two
critical risk factors: personality traits and stress.
Personality Characteristics: Impulsivity,
Manipulativeness, and Aggressiveness
Personality traits influence the way individuals experience in-
terpersonal relationships (e.g., Simpson, Winterheld, & Chen,
2006), and there is accumulating evidence that personality may be
a significant factor in the onset of physical aggression in romantic
relationships. Diagnostic and Statistical Manual of Mental Disor-
ders (4th ed.; DSM–IV; American Psychiatric Association, 1994)
Axis II personality pathology in general and, more specifically,
antisocial personality disorder (ASPD) and borderline personality
disorder (BPD) are associated with the perpetration of partner
aggression (e.g., Andrews, Foster, Capaldi, & Hops, 2000; Capaldi
& Owen, 2001; Edwards, Scott, Yarvis, Paizis, & Panizzon, 2003).
However, the examination of personality within the context of
DSM–IV diagnostic categories is likely to lead to a loss of relevant
information, because the diagnostic categories are highly comorbid
and heterogeneous (e.g., Widiger & Samuel, 2005). Alternatively,
the facet-level personality traits underlying these disorders have
superior psychometric properties and may predict behavioral out-
comes better than do broad diagnostic categories (e.g., Paunonen
& Ashton, 2001; Reynolds & Clark, 2001). Further, whereas
personality disorder diagnoses in a community sample may be
relatively rare, there is variability in the continuously measured
underlying traits (Clark, Simms, Wu, & Casillas, in press). Thus,
it may be more fruitful to consider the fundamental dimensions of
DSM personality disorders as opposed to diagnostic categories, as
there likely are commonalities within these broader trait domains
that account for the associations between these disorders and
physical aggression.
ASPD and BPD are subsumed under DSM–IV Cluster B (“dramatic/
erratic”) disorders and are differentiated from other personality clus-
ters not only by their associations with high emotional dysregulation,
low empathy, and stress reactivity (e.g., Crawford et al., 2006;
Kraus & Reynolds, 2001) but by their extreme “action-oriented”
features (e.g., Fossati et al., 2007). Not surprisingly, the personal-
ity facets subsumed under Cluster B—in particular, impulsivity,
manipulativeness, and aggressiveness—are associated with mal-
adaptive social conduct and physically aggressive behavior (Clark
et al., in press). For example, impulsivity is linked to difficulties in
regulating thoughts and behaviors, and manipulativeness is asso-
ciated with a lack of empathy and exploitative interpersonal inter-
actions. All of these personality facets may lead to a greater
likelihood of engaging in physically aggressive behavior (e.g.,
Stafford & Cornell, 2003). Not surprisingly, trait aggressiveness,
Vulnerabilities
Personality
Traits
B
H
F
(Mal)Adaptive
Processes
Physical Aggression
Marital
Quality
Marital
Stability
C
Stress
Chronic Stress
E
G
A
D
Figure 1. The vulnerability–stress–adaptation (VSA) model (Karney & Bradbury, 1995). It was not within the
scope of the current study to test the entire model. Paths examined are denoted by bold lines (A, B, C, E). Paths
not examined are denoted by dashed lines (D, F, G, H). In accord with full actor–partner independence modeling
(APIM) and growth curve modeling guidelines, paths were estimated separately and simultaneously for husbands
and wives; within and across spouses (i.e., actor and partner paths); and for intercepts and slopes for stress and
physical aggression.
757
PREDICTORS OF PHYSICAL AGGRESSION IN MARRIAGE

defined as the tendency to feel argumentative and vengeful and the
desire to engage in destructive behaviors, also facilitates aggres-
sive actions (Buss & Perry, 1992).
The examination of personality in the context of intimate partner
aggression has revealed significant associations between male
physical aggression and these personality traits (e.g., Dutton &
Bodnarchuk, 2005; Stuart & Holtzworth-Munroe, 2005). Unfortu-
nately, most of these studies have been limited to (a) clinical
samples (e.g., batterers), which limit the generalizability of the
findings; (b) male samples, which limit our ability to understand
physically aggressive behavior among women; and (c) cross-
sectional investigations, which prohibit us from explicating the
longitudinal course of physical aggression. There have been im-
portant exceptions to these limitations, such as occasional inves-
tigations of physical aggression in community samples of men and
women; however, the results of these studies have been mixed. For
example, in one longitudinal study of community newlywed cou-
ples, impulsivity was associated with physical aggression at 30
months of marriage for wives but not for husbands (O’Leary,
Malone, & Tyree, 1994). In a separate cross-sectional community
sample, both male and female perpetrators of partner violence
evidenced more trait aggressiveness than did nonviolent individ-
uals (Ehrensaft, Moffitt, & Caspi, 2004). A key limitation of prior
research on partner physical aggression and these personality traits
is that such research has been limited almost exclusively to cross-
sectional data, to males, and to clinical populations. In addition,
although there is evidence for women’s antisocial behavior pre-
dicting men’s physical aggression (Kim & Capaldi, 2004), few
studies have examined the impact of partner personality traits on
an individual’s physical aggression. Moreover, the impact of male
characteristics on female physical aggression has received even
less attention. In the present study, we sought to build on past
research by examining the personality traits of impulsivity, ma-
nipulativeness, and aggressiveness as predictors of one’s own and
one’s partner’s longitudinal courses of physical aggression in a
community sample of men and women.
Stress
In addition, we sought to examine stress as a predictor of
physical aggression trajectories. Stress is associated with disrup-
tions in cognitive and behavioral control (e.g., Rutledge & Linden,
1998). It activates negative affective responses and processes, such
as the fight-or-flight response, that are linked to physically aggres-
sive behavior (Berkowitz, 1990). Given that one’s ability to re-
solve marital conflict effectively requires particular cognitive and
affective resources, it seems plausible that access to such internal
resources would be hindered in times of stress. Past research has
established that stressful circumstances external to one’s marriage
can negatively affect dyadic processes between spouses (e.g.,
withdrawal from marital interaction; Story & Repetti, 2006). Al-
though most of the research on stress and partner physical aggres-
sion has focused on males, there is evidence linking life stressors
and physical aggression for both sexes (Barling & Rosenbaum,
1986; Cano & Vivian, 2003).
In light of research demonstrating that chronic stress can influ-
ence how well couples react to or cope with acute stressors
(Karney, Story, & Bradbury, 2005), we were particularly inter-
ested in the influence of chronic stress on physical aggression. For
example, in a study of newlyweds over the first 3 years of mar-
riage, Frye and Karney (2006) examined the role of chronic stress
in the perpetration of physical aggression and found that, under
higher levels of chronic stress, both husbands and wives were more
likely to be physically aggressive. As the only longitudinal exam-
ination of chronic stress and physical aggression among both
husbands and wives, this study represents an important contribu-
tion to the literature. However, there are many ways in which this
work can be built upon in future research. For example, modeling
physical aggression continuously (rather than dichotomously), ex-
amining associations among changes in stress and aggression over
time (rather than at each time point separately), analyzing hus-
bands’ and wives’ data simultaneously, and including cross-spouse
associations are ways in which we have attempted to build on these
findings.
Overview of the Present Study
Although the consequences of physical aggression have been
well documented, many questions remain regarding the predictors
of the longitudinal course of physical aggression in marriage. With
the advent of statistical techniques that increase our ability to
model change over time and that better account for the interde-
pendence of dyadic data, we attempted to overcome prior meth-
odological limitations by analyzing men’s and women’s physically
aggressive behavior toward their partners four times over the first
3 years of marriage. Physical aggression was scaled as a continu-
ous and dynamic phenomenon and was calculated separately for
husbands and wives. We used both self and partner reports to
arrive at a total physical aggression score for each spouse. Capaldi
and Kim (2007) called for a dynamic approach that conceptualizes
physical aggression as (a) involving the interaction of two indi-
viduals, each with influence on the other, and (b) being embedded
in contextual influences that change over time. Consistent with this
approach, we utilized growth curve modeling techniques to ana-
lyze and predict average levels (intercepts) and rates of change
(slopes) in stress and aggression and employed actor–partner in-
terdependence modeling (APIM; Kenny, Kashy, & Cook, 2006;
Raudenbush, Brennan, & Barnett, 1995) and structural equation
modeling (SEM) techniques to predict within- and cross-spouse
paths of husbands’ and wives’ trajectories. Finally, we employed
the VSA model to frame our investigation of the individual and
contextual factors that predict the longitudinal course of physical
aggression in marriage.
We had two aims in the present study. The first was to replicate
and extend prior research on the natural course of husbands’ and
wives’ physical aggression over the early, high-risk period for
divorce (Cherlin, 2004). In view of past findings (e.g., O’Leary et
al., 1989), we hypothesized that approximately one half of couples
would report engaging in physical aggression, that the aggression
would consist largely of moderately aggressive acts (e.g., grab-
bing, pushing, and slapping), and that the majority of aggressive
couples would report mutual aggression (i.e., both the husband and
the wife would report engaging in aggression). We also examined
rates of change in aggression over the early years of marriage.
There is little prior research on trajectories of aggression, but the
data that do exist provide evidence for increases (e.g., Quigley &
Leonard, 1996), decreases (e.g., O’Leary et al., 1989), and no
linear change (e.g., Lawrence & Bradbury, 2007) in aggression
758 LANGER, LAWRENCE, AND BARRY

over time. We tentatively predicted that rates of change in hus-
bands’ and wives’ physical aggression would vary widely across
spouses but that there would be no average systematic change.
Our second aim was to test the relevant components of the VSA
model as presented by Karney and Bradbury (1995) to explicate
key predictors of husbands’ and wives’ physical aggression tra-
jectories. As demonstrated in Figure 1, we tested four paths of the
VSA model: (A) stress 3physical aggression; (B) personality 3
physical aggression; (C) personality 3stress; and (E) physical
aggression 3stress. All of the paths we hypothesized a priori were
based on within-spouse associations. We also explored potential
gender differences and cross-spouse associations for all paths but,
given that the existing literature reveals equivocal findings, we did
not make specific predictions.
Path A: Stress 3physical aggression. On the basis of re-
search that demonstrated the emotional and behavioral con-
sequences of chronic stress, we expected that individuals who
reported higher levels of stress would be more likely to
engage in physically aggressive behavior. We hypothesized
that greater average levels of stress would be associated with
greater average levels of physical aggression in the first 3
years of marriage and that increases in stress over time would
predict increases in physical aggression over time.
Path B: Personality 3physical aggression. Taking past re-
search on personality dimensions into consideration, we were
particularly interested in the specific personality traits asso-
ciated with physical aggression. We hypothesized that
spouses who were more impulsive, manipulative, and aggres-
sive would report greater average perpetration of physical
aggression and greater increases in physical aggression over
time.
Path C: Personality 3stress. Given that the personality traits
identified are associated with deficits in emotional and be-
havioral control, these traits are likely also associated with the
tendency to experience more stress. Thus, we predicted that
each of these traits would be associated with the experience of
stress, such that spouses who were more impulsive, manipu-
lative, and aggressive would report greater chronic stress on
average and increases in chronic stress over time.
Path E: Physical aggression 3stress. Few studies have
examined stress as a consequence of physical aggression
(e.g., Testa & Leonard, 2001), so the examination of this path
was exploratory in nature. However, it is likely that the
negative consequences of physical aggression would lead to
greater stress in areas beyond the marriage (e.g., relations
with friends, work, and one’s own health). Therefore, we
included this path to allow for the investigation of the influ-
ence that physical aggression trajectories may have on stress
trajectories. However, we hypothesized that stress would
demonstrate predictive dominance over physical aggression,
such that the effects of stress on aggression would be stronger
than the effects of aggression on stress.
We also tested the same paths of the VSA model with person-
ality conceptualized as the diagnostic category of ASPD in order to
assess the utility of (higher order) diagnostic categories versus
(lower order) personality facets in the prediction of physical ag-
gression trajectories. We argue that research on lower order traits
as predictors of physical aggression in intimate relationships
would be more clinically useful and would allow for greater
generalizability of our findings. Therefore, we hypothesized that
the associations revealed in Paths B and C would be stronger when
personality was operationalized in terms of traits and, specifically,
that manipulativeness, impulsivity, and aggressiveness would
more strongly predict stress and physical aggression than would
ASPD.
Method
Participants and Procedures
Participants were recruited through marriage license records in
the Midwest. Newlyweds between 18 and 55 years of age were
mailed letters that explained the study and invited them to partic-
ipate. Of the 1,698 letters that were sent, 358 (21%) were answered
by couples who expressed interest in participation by sending an
e-mail, leaving a telephone message, or returning the stamped
postcard we included with the letter. Telephone screening was
completed on 189 of the interested couples to ensure they met the
following eligibility requirements: relatively fluent in English,
married less than 6 months, and in their first marriages. The
remaining 169 couples did not complete the telephone screen (e.g.,
1 spouse did not want to participate, the couple could not be
reached, the couple was moving out of state). Of the 189 couples
who completed the telephone screen, 42 were ineligible because
they were not in their first marriages, and 6 declined to participate
after completion of the telephone screen. Questionnaire packets
were sent to the remaining 141 couples, and the first 105 couples
who completed their packets and kept their initial laboratory
appointments were included in the present study. One couple’s
data were deleted because it was revealed during the laboratory
session that it was not the wife’s first marriage. The data from the
husband of another couple were removed because his responses
were deemed unreliable. By the end of the fourth wave of data
collection, 3 couples had been lost to attrition and 6 couples had
divorced; however, available data from these couples were in-
cluded in the present study. Thus, data analyses were conducted on
103 couples.
1
Couples had dated an average of 44 months (SD ⫽27) prior to
marriage, and 80% of them had cohabited premaritally. Average
estimated annual joint income was between $40,000 and $50,000,
and modal education was 14 years. Husbands’ average age was
26.23 years (SD ⫽3.60), and wives’ average age was 24.99 years
(SD ⫽3.80). For 15% of our sample, at least 1 spouse self-
identified as a member of a racial minority group. Of spouses, 5%
self-identified as African American, 7% as Asian American, 4% as
Hispanic, 1% as Middle Eastern, and 2% as “Other.” (The pro-
portion of non-Caucasian individuals in Iowa is 7%; U.S. Census
Bureau). By Time 4 of the study, 35% of the couples were parents.
1
Although data from this sample have been published elsewhere (e.g.,
Brock & Lawrence, 2008; Ro & Lawrence, 2007), this is the first article to
examine personality variables and the first to examine predictors of phys-
ical aggression.
759
PREDICTORS OF PHYSICAL AGGRESSION IN MARRIAGE

Demographic variables were not significantly related to the key
variables in this study.
Spouses participated in four waves of data collection: at 3– 6
months (Time 1), 12–15 months (Time 2), 21–24 months (Time 3),
and 30 –33 months (Time 4) of marriage. Time 1 included the
completion of questionnaires at home and a laboratory session that
involved procedures beyond the scope of the present study. Times
2, 3, and 4 comprised the completion of questionnaires at home.
Measures of personality traits were completed once (at Time 2),
and measures of chronic stress and physical aggression were
completed at all four time points. Couples were paid $100 at Time
1 and $50 at each time point for Times 2, 3, and 4.
Measures
Physical aggression. The Revised Conflict Tactics Scales
(Straus, Hamby, Boney-McCoy, & Sugarman, 1996) is a 78-item
self-report scale of aggression that has occurred in the context of
conflicts (e.g., physical, psychological, and sexual tactics) with a
partner during the previous 6 months. It has moderate internal
consistency and yields significant interpartner agreement (Straus et
al., 1996). In the current study, only the Physical Assault Scale was
used and partners reported on husband-to-wife and wife-to-
husband aggression. Examples of items include throwing some-
thing at partner; pushing, grabbing, or shoving partner; and punch-
ing or hitting partner. Items are rated on 7-point scales, ranging
from never to more than 20 times. We calculated composite scores
by adding the midpoints for each response category across tactics
(e.g., the midpoint 4 for 3–5 times), as recommended by Straus et
al. (1996). Interspousal agreement (i.e., husbands’ and wives’
reports on husbands’ aggression and husbands’ and wives’ reports
on wives’ aggression) was uniformly strong and significant (rs
ranged from .61 to .74, all ps⬍.01). We calculated the total score
for each participant by averaging his or her self-reported aggres-
sion and the partner’s report of his or her aggression; this method
is more reliable than use of data from only one informant (Strahan,
1980). Alphas ranged from .72 to .85 across husband and wife
reports and across time.
Personality traits. The Schedule for Nonadaptive and Adap-
tive Personality–2 (SNAP-2; Clark et al., in press) is a 390-item,
factor-analytically-derived self-report inventory designed to assess
personality traits that extend from the normal to the pathological
range. It comprises 3 temperament scales, 12 trait scales, and 13
diagnostic scales and demonstrates good internal consistency, dis-
criminant validity, and test–retest reliability across multiple sam-
ples (Reynolds & Clark, 2001). Three of the trait scales—
Impulsivity, Manipulativeness, and Aggressiveness—and the
Antisocial Personality diagnostic scale were included in our study.
The 3 trait scales each comprise 20 true/false items, and the ASPD
scale contains 34 true/false items. Impulsivity on the SNAP-2 is
defined as the “tendency to act on a momentary basis . . . versus
the tendency to stop and think things over before acting” (Clark et
al., in press, p. 50). Manipulativeness is defined as “an egocentric
willingness to use people for personal gain without regard for the
rights or feelings of others or for abstract ideals such as fairness”
(Clark et al., in press, p. 47). It is measured as the extent to which
a person enjoys exploiting others and views this behavior as a skill.
Aggressiveness on the SNAP-2 measures individual differences in
the frequency and intensity of the experience of anger. According
to Clark et al., “High scorers anger easily with slight provocation,
have difficulty controlling their anger both internally and exter-
nally, stay angry longer, to the point of holding grudges and
seeking revenge, and derive pleasure from violence” (p. 48). The
ASPD scale was included to assess clusters of traits that represent
the DSM–IV diagnosis. This scale contains items assessing conduct
disorder before the age of 15, irresponsibility, lying, and illegal
acts. For husbands and wives, alphas ranged from .70 to .76 across
the scales.
Chronic stress. We used a modification of an interview pro-
tocol developed by Hammen et al. (1987) to assess chronic stress
via a paper-and-pencil, self-report method; similar modifications
have been used previously (e.g., Frye & Karney, 2006). The
measure covers nine domains that may generate chronic stress,
including relationship domains (relations with spouse, family, in-
laws, and friends) and other life domains (school, work, home,
financial status, and health). The present study excluded “relation-
ship with spouse”; the remaining eight domains were deemed
nonmarital domains to account for chronic stress external to the
marriage. The measure assesses a type of chronic stress resulting
from role occupancy (Wheaton, 1997) and has implications for
individual and dyadic functioning (e.g., Karney et al., 2005). At
each assessment, spouses chose the number that best represented
their experiences in these areas on 9-point scales; higher numbers
indicated greater stress. Scores at each time point were obtained by
averaging the scores across the eight domains, because not all
domains applied to every participant (e.g., school). Alphas ranged
from .72 to .81 for husbands and wives across time.
Data Analyses
We employed APIM for mixed independent variables (Kenny et
al., 2006). In an APIM, there are two dyad members and at least
two variables, Xand Y, for each. When dyad members are distin-
guishable, as is the case in our sample of heterosexual married
couples, there are potentially two actor effects: one for the effects
of the husband’s predictor on the husband’s outcome and one for
the effect of the wife’s predictor on the wife’s outcome. There are
also potentially two partner effects, one for the effect of the
husband’s predictor on the wife’s outcome and one for the effect
of the wife’s predictor on the husband’s outcome. Finally, there are
at least two correlations in an APIM. First, husbands’ and wives’
predictors may be correlated. Second, the residual nonindepen-
dence in outcome scores is represented by the correlation between
the error terms in husbands’ and wives’ outcomes. This second
correlation represents the nonindependence not explained by the
APIM.
Hypotheses were tested with SEM techniques and Mplus soft-
ware (Muthe´n & Muthe´n, 2006) with maximum likelihood esti-
mation. SEM facilitated the simultaneous modeling of all variables
(albeit with some statistical constraints, i.e., model identification;
for a discussion, see Bollen, 1989), which allowed us to use APIM
and to model a variable simultaneously as an outcome and as a
predictor (Olsen & Kenny, 2006). We estimated growth trajecto-
ries of the longitudinal variables using latent growth modeling.
This statistical procedure uses variables measured over time to
develop a latent intercept and a latent slope variable for each
longitudinal variable (i.e., physical aggression and chronic stress).
The intercepts were modeled at the midpoint of the assessment
760 LANGER, LAWRENCE, AND BARRY

period and provided a reference of average level for an individu-
al’s trajectory of change over time, whereas the slope represented
an individual’s pattern of linear change over time. The latent
intercepts and slopes are linked to the occasion-based measures for
dyad members through model-specific factor loading. SEM allows
estimation of the average value of the intercepts and slopes as well
as the degree of variability in the intercepts and slopes among
individuals. Once a model is deemed to fit the data adequately,
parameter estimates can be interpreted. Goodness-of-fit measures
used in the present study were the model chi-square statistic
(nonsignificant at p⬍.05), the incremental fit index (IFI ⬎.90;
Bollen, 1989), and the root-mean-square error of approximation
(RMSEA ⬍.10; Kline, 2005).
Results
Descriptive and Preliminary Analyses
During the first 3 years of marriage, 44% of husbands, 54%
of wives, and 62% of couples were physically aggressive during
marital conflict; spouses’ prevalence rates ranged from 20% to
38% across the four time points (see Table 1). Using Straus et
al.’s (1996) categories, if either respondent endorsed any mod-
erate item, we classified the spouse as moderately aggressive. If
either respondent endorsed any severe item, the spouse was
classified as severely aggressive. If both types of items were
endorsed, the spouse was classified as severely aggressive. The
most frequent behaviors endorsed were moderately aggressive
acts, such as pushing, grabbing, and slapping; however, 41% of
aggressive husbands and 60% of aggressive wives engaged in
severely aggressive acts. On average, aggressive husbands and
wives had engaged in five to six acts of physical aggression in
the 6 months prior to each assessment (see Table 2).
2
The
majority of the aggression was mutual (both spouses were
aggressive), and both husbands and wives reported greater
frequency of aggressive behavior for wives; however, this gen-
der difference was significant only at Time 1, t(201) ⫽2.79,
p⬍.01. Levels of personality traits and chronic stress were in
the expected ranges (low-to-moderate levels), and husbands
scored significantly higher on antisocial personality traits than
did wives, t(201) ⫽3.70, p⬍.01.
We conducted an inverse transformation on physical aggression
scores (and added ⫹1 because some participants’ original scores
equaled 0), and this greatly improved the distributions. Once
transformed, greater physical aggression was indicated by rela-
tively lower scores. Interspousal correlations were small for per-
sonality variables, moderate for stress, and moderate to high for
physical aggression (see Table 3). Within-spouse correlations were
uniformly small to moderate in size. This indicated that these
variables were related but were sufficiently distinct to support their
investigation as separate constructs.
Measurement Models
Next, we estimated three SEM measurement models. For the
first two models, we estimated growth curves for each of the
time-varying variables (i.e., physical aggression and stress) in
order to examine (a) whether there was systematic, average linear
change in either of these variables and (b) whether there was
significant variability for each of the parameters (i.e., intercepts
and slopes) to warrant prediction of these parameters. In Model 1
we included husbands’ and wives’ time-varying chronic stress, and
in Model 2 we included husbands’ and wives’ time-varying phys-
ical aggression. All intercepts and slopes were allowed to covary.
Both models fit the data very well: chronic stress, ␹
2
(22, N⫽
103) ⫽21.04, p⫽.52, IFI ⫽1.0, RMSEA ⫽.00; physical
aggression, ␹
2
(22, N⫽103) ⫽18.07, p⫽.45, IFI ⫽1.0,
RMSEA ⫽.01. The variance components for all parameters were
significant for both spouses; this suggested that it was appropriate
for us to interpret and attempt to predict the parameter estimates.
For chronic stress, the variance estimates were .64 (SE ⫽.10) for
husbands’ intercepts and .53 (SE ⫽.08) for wives’ intercepts and
.01 (SE ⫽.006) for husbands’ slopes and .01 (SE ⫽.05) for wives’
slopes ( ps⬍.05). For physical aggression, the variance estimates
were .05 (SE ⫽.01) for husbands’ intercepts and .005 (SE ⫽.002)
for wives’ intercepts and .06 (SE ⫽.01) for husbands’ slopes and
.01 (SE ⫽.004) for wives’ slopes ( ps⬍.01).
In Model 3, we tested the adequacy of the measurement model
that included all variables of interest. Husbands’ and wives’ time-
varying stress and aggression were specified, so that there were
four latent variables for husbands and four latent variables for
wives (i.e., stress intercepts, stress linear slopes, aggression inter-
cepts, and aggression linear slopes). Personality variables were
modeled as six observed variables: husbands’ and wives’ impul-
sivity, manipulativeness, and aggressiveness. The latent variables
and observed variables were allowed to covary within and between
spouses. The model was an adequate fit for the data, ␹
2
(142, N⫽
103) ⫽225.95, p⬍.01, IFI ⫽.92, RMSEA ⫽.07.
Stress 3aggression and aggression 3stress. The VSA
model as originally published (Karney & Bradbury, 1995) is
nonrecursive and therefore underidentified when specified in its
entirety. To test all of the VSA paths of interest in our study, we
could include only one of these two paths (aggression 3stress
or stress 3aggression). To guide this decision, we ran two
more models— one testing Path A (stress 3aggression, Model
4) and one testing Path E (aggression 3stress, Model 5)—and
compared each of these models to our final measurement model
(Model 3, in which stress and aggression linear slopes were
modeled as covariances rather than as directional paths). Model
4 produced a marginally significant reduction in model fit,
␹
2
(10, N⫽103) ⫽17.05, p⫽.07, whereas Model 5 did not,
␹
2
(10, N⫽103) ⫽9.41, p⫽.49. We also examined the four
individual paths in Models 4 and 5 (e.g., in Model 4, changes in
husbands’ and wives’ stress predicting changes’ in husbands’
and wives’ aggression). Two of these four paths were signifi-
cant in Model 4, whereas none of these paths were significant
in Model 5. In sum, associations between changes in stress and
changes in aggression were best represented by a model in
which changes in stress predicted changes in physical aggres-
sion (Path A) rather than one in which changes in physical
aggression predicted changes in stress (Path E).
2
Mean number of aggressive acts over a 6-month period was calculated
based on the midpoints of ranges (e.g., 3–5 acts are coded as a 4; Straus et
al., 1996) and, therefore, should be considered estimates.
761
PREDICTORS OF PHYSICAL AGGRESSION IN MARRIAGE

Testing the Relevant Components of the VSA Model
Physical aggression slopes were regressed onto stress slopes,
and all of the parameters for stress and physical aggression were
regressed onto the personality predictors (impulsivity, manipula-
tiveness, and aggressiveness; Model 6). Personality variables were
allowed to covary both within and between spouses. The model
was found to fit the data adequately, ␹
2
(187, N⫽103) ⫽290.65,
p⬎.001, IFI ⫽.91, RMSEA ⫽.08. We were able to explain 17%
and 31% of the variance in husbands’ and wives’ average levels of
stress (intercepts), respectively; we did not explain a significant
proportion of the variance in spouses’ stress slopes. We explained
20% and 15% of the variance in husbands’ and wives’ average
physical aggression, respectively, and explained 50% and 24% of
the variance in husbands’ and wives’ physical aggression slopes,
respectively. (See Figure 2 for a visual depiction of all significant
paths in this model).
Path A: Stress 3aggression. Changes in stress predicted
changes in physical aggression for husbands (␥⫽⫺.54, SE ⫽.27,
p⬍.05) and for wives (␥⫽⫺1.25, SE ⫽.55, p⬍.05), even after
controlling for the effects of personality on stress and on physical
aggression. Husbands’ and wives’ paths did not differ signifi-
cantly: nested chi-square, ␹
2
(1, N⫽103) ⫽1.79, p⫽.18.
Cross-spouse paths were not significant. Further, husbands’ phys-
ical aggression intercepts correlated significantly with wives’
physical aggression slopes (r⫽⫺.27, p⬍.05); to the extent that
husbands were more physically aggressive on average, their wives
became increasingly physically aggressive over time.
Path B: Personality traits 3physical aggression. We present
the significant paths for predictors of husbands’ and wives’ phys-
ical aggression intercepts and wives’ physical aggression slopes.
First, husbands’ trait aggressiveness and impulsivity predicted
husbands’ physical aggression intercepts; to the extent that husbands
had more aggressive and impulsive personalities, they were more
physically aggressive with their wives (␥⫽⫺.03, SE ⫽.01, and ␥⫽
⫺.01, SE ⫽.01, respectively; ps ⬍.01). Second, wives’ physical
aggression intercepts were predicted by husbands’ trait aggressiveness
(␥⫽⫺.04, SE ⫽.01, p⬍.01) and by wives’ impulsivity (␥⫽⫺.03,
SE ⫽.01, p⬍.01); wives who were more physically aggressive
toward their husbands had more impulsive personalities and were
married to husbands with more aggressive personalities. Third,
wives’ trait aggressiveness predicted wives’ physical aggression
slopes; wives with more aggressive personalities became increas-
ingly physically aggressive toward their husbands over time (␥⫽
⫺.02, SE ⫽.01, p⬍.05).
We conducted nested chi-square tests to explore potential gen-
der differences. First, we compared significant predictors of hus-
bands’ and wives’ physical aggression intercepts. We set the paths
significantly predicting husbands’ physical aggression intercepts
(husbands’ trait aggressiveness and impulsivity) equal to the paths
significantly predicting wives’ physical aggression intercepts (hus-
bands’ trait aggression and wives’ impulsivity). These constraints
did not produce a significant drop in model fit, ␹
2
(3, N⫽103) ⫽
4.1, p⫽.25, so there was no evidence of gender differences. Next,
wives’ trait aggressiveness predicted wives’ physical aggression
slopes, but husbands’ trait aggressiveness did not predict hus-
bands’ physical aggression slopes; however, that difference does
not necessarily imply a significant gender difference in these
associations. We examined whether that path was significantly
stronger than the same path for husbands. To examine whether this
gender difference was significant, the within-spouse effects of trait
aggressiveness on physical aggression slopes were constrained to
equality. The constraints did not yield a significant reduction in
model fit, ␹
2
(1, N⫽103) ⫽1.58, p⫽.21, so there was no
evidence of a gender difference.
Table 1
Descriptives of Husbands’ and Wives’ Physical Aggression Over the First 3 Years of Marriage
Variable
At each time point Aggregated across time
1234AnyModerate Severe Directionality
Husbands 24 20 24 25 44 26 18 H3W: 13
Wives 38 27 36 33 54 22 32 W3H: 30
Couples 42 31 40 37 62 26 36 H7W: 57
Note. All data are percentages. Any ⫽engaged in any aggressive act; moderate ⫽engaged in only moderately
aggressive acts; severe ⫽engaged in any severely aggressive act; H3W⫽unidirectional husband-to-wife
aggression; W3H⫽unidirectional wife-to-husband aggression; H7W⫽both partners were aggressive.
Table 2
Mean Aggression Scores Among Physically Aggressive Spouses
Time
Husbands Wives
t(103)MSDRange MSDRange
Time 1 (3–6 months) 2.58 2.18 1–8 4.72 6.25 1–33 ⫺2.788
ⴱ
Time 2 (12–15 months) 4.50 6.00 1–20 6.12 8.04 1–30 ⫺1.855
Time 3 (21–24 months) 4.93 8.06 1–33 5.02 4.26 1–18 ⫺1.299
Time 4 (30–33 months) 8.05 9.78 1–34 6.72 7.15 1–32 ⫺.163
ⴱ
p⬍.05.
762 LANGER, LAWRENCE, AND BARRY

Path C: Personality traits 3stress. Husbands’ stress inter-
cepts were positively predicted by husbands’ trait aggressiveness
(␥⫽.11, SE ⫽.04, p⬍.05), such that husbands with more
aggressive personalities experienced more stress on average.
Wives’ stress intercepts were positively predicted by husbands’
and wives’ trait aggressiveness (␥⫽.09, SE ⫽.03, and ␥⫽.17,
SE ⫽.03, respectively; ps⬍.01); to the extent that either hus-
bands or wives had more aggressive personalities, wives experi-
enced more stress on average. We conducted two nested chi-square
tests to explore potential gender differences. First, all four paths—
the within- and cross-spouse effects of trait aggressiveness on
stress intercepts—were constrained to be equal. The constraints
Table 3
Bivariate Correlations Among Husbands’ and Wives’ Predictors and Outcomes
Wives’ variables
Husbands’ variables
ASPD IMP MNP AGG Stress 1 Stress 2 Stress 3 Stress 4 PA1 PA2 PA3 PA4
Personality
ASPD .15 .55
ⴱⴱ
.49
ⴱⴱ
.56
ⴱⴱ
.18 .35
ⴱⴱ
.23 .31
ⴱⴱ
.20 .18 .44
ⴱⴱ
.65
ⴱⴱ
IMP .60
ⴱⴱ
ⴚ.20 .38
ⴱⴱ
.46
ⴱⴱ
.14 .18 .19 .26
ⴱ
.14 .16 .20 .17
MNP .40
ⴱⴱ
.34
ⴱⴱ
.19 .57
ⴱⴱ
.29
ⴱⴱ
.36
ⴱⴱ
.30
ⴱⴱ
.39
ⴱⴱ
.10 .28
ⴱ
.22
ⴱ
.19
AGG .36 .24
ⴱ
.22
ⴱ
ⴚ.03 .27
ⴱⴱ
.36
ⴱⴱ
.41
ⴱⴱ
.40
ⴱⴱ
.14 .26
ⴱ
.37
ⴱⴱ
.36
ⴱⴱ
Chronic stress
Time 1 .16 .10 .11 .26
ⴱ
.22
ⴱ
.57
ⴱⴱ
.64
ⴱⴱ
.63
ⴱⴱ
.27 .13 .11 .01
Time 2 .22 .13 .32
ⴱⴱ
.28
ⴱ
.71
ⴱⴱ
.29
ⴱⴱ
.76
ⴱⴱ
.72
ⴱⴱ
.19 .12 ⫺.02 .02
Time 3 .27
ⴱ
.18 .08 .51
ⴱⴱ
.65
ⴱⴱ
.58
ⴱⴱ
.44
ⴱⴱ
.71
ⴱⴱ
.24 .08 .01 ⫺.01
Time 4 .31
ⴱⴱ
.18 .21 .36
ⴱⴱ
.60
ⴱⴱ
.67
ⴱⴱ
.68
ⴱⴱ
.39
ⴱⴱ
.31
ⴱⴱ
.23
ⴱ
⫺.02 .06
Physical aggression
Time 1 ⫺.02 .12 .32
ⴱⴱ
.03 .28
ⴱⴱ
.30
ⴱⴱ
.19 .27
ⴱ
.32
ⴱⴱ
.07 .24
ⴱ
.31
ⴱⴱ
Time 2 .05 .02 .22
ⴱ
.24
ⴱ
.20 .11 .29
ⴱⴱ
.15 .09 .89
ⴱⴱ
.01 .22
Time 3 .22 .13 .09 .22
ⴱ
.07 .15 .06 .29
ⴱ
.28
ⴱⴱ
.05 .62
ⴱⴱ
.71
ⴱⴱ
Time 4 .37
ⴱⴱ
.19 .11 .20 .19 .19 .17 .21 .11 .10 .21 .63
ⴱⴱ
Note. For bivariate correlations, within-husband correlations are above the diagonal; within-wife correlations are below the diagonal; interspousal
correlations are along the diagonal and are in bold. ASPD ⫽antisocial personality disorder; IMP ⫽impulsivity; MNP ⫽manipulativeness; AGG ⫽
aggressiveness; PA ⫽physical aggression.
ⴱ
p⬍.05.
ⴱⴱ
p⬍.01, two-tailed.
Husbands’
Aver age PA
Husbands’
Impulsivity
Husbands’
Change PA
Wives’
Average PA
Wives’
Change PA
Wives’
Change Stress
Wives’
Average Stress
Husbands’
Change Stress
Husbands’
Aver age Stress
Husbands’
Aggressiveness
Wives’
Impulsivity
Wives’
Aggressiveness
A: -1.25 (.55)
A: -.03 (.01)
B: -.01 (.01)
A: -.04 (.01)
B: -.03 (.01)
B: -.02 (.01)
C: .11 (.04)
C: .09 (.03)
C: .17 (.03)
A: -.54 (.27)
Figure 2. Estimated significant paths between husbands’ and wives’ personality traits, chronic stress, and
physical aggression. Nonsignificant paths from the full vulnerability–stress–adaptation (VSA) model are not
shown. VSA paths (A, B, or C), standardized regression coefficients, and standard errors are denoted above the
corresponding path. All paths were significant at p⬍.05, one-tailed. Husbands and wives’ manipulativeness are
not shown because they did not significantly predict any parameters of stress or physical aggression. PA ⫽
physical aggression.
763
PREDICTORS OF PHYSICAL AGGRESSION IN MARRIAGE

yielded a significant drop in model fit, ␹
2
(3, N⫽103) ⫽16.90,
p⫽.001. Second, we removed the constraint on the one nonsig-
nificant path: wives’ trait aggressiveness on husbands’ stress in-
tercept. The remaining constraints did not result in a reduction in
the fit compared with the original model, ␹
2
(2, N⫽103) ⫽1.15,
p⫽.56. In sum, husbands’ trait aggressiveness had a significantly
greater impact on wives’ stress than wives’ trait aggressiveness
had on husbands’ stress. (Within-spouse effects were also signif-
icantly greater than was the cross-spouse effect of wives’ aggres-
siveness on husbands’ stress.)
Additional analyses for Paths B and C: ASPD 3stress and
aggression. Finally, we examined whether the effects of person-
ality on stress and physical aggression differed when personality
was operationalized as the diagnostic cluster of ASPD rather than
in terms of the lower order personality facets we had tested
previously. Specifically, we revised Model 6 by removing hus-
bands’ and wives’ impulsivity, aggressiveness, and manipulative-
ness and adding husbands’ and wives’ ASPD (Model 7). First, we
tested the overall fit of the new model and found it to adequately
fit the data, ␹
2
(140, N⫽103) ⫽225.16, p⬎.001, IFI ⫽0.92,
RMSEA ⫽.08. Second, we examined whether the new paths were
significant. Husbands’ ASPD significantly predicted husbands’
stress intercepts (␥⫽.07, SE ⫽.02, p⬍.01), husbands’ physical
aggression intercepts (␥⫽⫺.01, SE ⫽.003, p⬍.01), and wives’
physical aggression intercepts (␥⫽⫺.01, SE ⫽.01, p⬍.05).
Wives’ ASPD significantly predicted wives’ stress intercepts (␥⫽
.07, SE ⫽.03, p⬍.01) and wives’ physical aggression slopes (␥⫽
⫺.03, SE ⫽.01, p⬍.05). Third, we examined the percent of
variance accounted for in this new model. The R
2
values indicated
that the post hoc model explained 17% of the variance in hus-
bands’ initial stress and 20% and 24% of the variance in husbands’
and wives’ physical aggression slopes, respectively. However, it
did not explain a significant amount of the variance in husbands’
or wives’ physical aggression intercepts, husbands’ or wives stress
slopes, or wives’ stress intercepts. In sum, the post hoc model
demonstrated some predictive utility.
Fourth, to examine whether Model 7 (in which we operational-
ized personality in terms of the ASPD cluster) was superior to
Model 6 (in which we operationalized personality in terms of
lower order traits), we compared the confidence intervals of the
proportions of variance explained in each of the parameters of
interest across the two models. Model 6 explained significantly
more variance in husbands’ physical aggression slopes than did
Model 7; all other confidence intervals overlapped. These results
suggest that the model that used lower order traits was superior for
prediction of change in husbands’ physical aggression.In sum,
operationalizing personality in terms of the lower order traits
yielded more explanatory power than did operationalizing it as
ASPD.
Discussion
Summary and Interpretation of Results
Fully 62% of couples (44% of husbands and 54% of wives) in
the present sample were physically aggressive over the early years
of marriage; this finding is consistent with the prevalence rates of
52% and 57% reported by Lawrence and Bradbury (2001) and
O’Leary et al. (1989), respectively. As expected, the majority of
the couples engaged in mutual aggression and more wives than
husbands were aggressive. Notably, 18% of husbands, 32% of
wives, and 36% of couples engaged in severe aggression at some
point in the first 3 years of marriage. These rates were higher than
those reported in previous studies of newlywed couples (e.g.,
Lawrence & Bradbury, 2001). However, moderate acts, such as
grabbing and pushing, were the most frequent among both hus-
bands and wives, and severe acts were relatively rare by compar-
ison. Consistent with Lawrence and Bradbury (2007), there was
significant variability among spouses’ longitudinal courses of ag-
gression over time but no average systematic change over time.
As hypothesized, personality traits were associated with physi-
cal aggression for both husbands and wives. Husbands higher in
trait impulsivity and trait aggressiveness were more physically
aggressive toward their partners on average; these findings were
consistent with prior research on male batterers (e.g., Stuart &
Holtzworth-Munroe, 2005). With few exceptions, relatively little
research has examined these traits in the context of female partner
aggression in the early years of marriage. Consistent with the
findings of O’Leary et al. (1994), our model revealed that impul-
sive wives were more physically aggressive on average, even after
it had accounted for their trait aggressiveness and manipulative-
ness and for their husbands’ personality traits. In addition, wives
who were higher on trait aggressiveness became increasingly
physically aggressive over time; this finding was similar to those
of Ehrensaft et al. (2004). These results add empirical support for
the role of personality in prediction of female partner aggression.
When we examined the relations between personality and
chronic stress, we found that trait aggressive husbands and trait
aggressive wives experienced more stress on average but that these
traits did not predict changes in stress over time. These findings are
consistent with those of cross-sectional studies that have indicated
the contribution of personality to stressful circumstances (e.g.,
Poulton & Andrews, 1992). Our analyses also revealed partner
effects of husbands’ personality traits, such that aggressive per-
sonality styles of husbands were associated with their wives being
more physically aggressive and reporting more chronic stress, on
average. Although partner effects have been found for female
antisocial behavior predicting male physical aggression (Kim &
Capaldi, 2004), few studies have examined the impact of partner
personality traits on one’s own physical aggression or stress. As
measured in the present study, trait aggressiveness reflects a ten-
dency to be argumentative and hostile, which may increase the
likelihood of engaging in verbal aggression or other nonconstruc-
tive conflict behaviors. Given this possibility, wives may retaliate
with physical aggression during conflicts in which husbands em-
ploy such tactics (e.g., hostile communication), and wives may
experience greater stress as a result of these interactions.
In sum, the present results suggest that certain personality traits
may act as vulnerabilities that render individuals at greater risk for
experiencing stress and for engaging in physically aggressive
behavior toward their partners. Moreover, there appears to be
evidence that husbands’ trait aggressiveness may be particularly
detrimental to wives; however, future research should replicate and
further expand on the current findings. Finally, trait manipulative-
ness was not associated with physical aggression for husbands or
wives. Given that manipulativeness is a trait underlying psychop-
athy, it is possible that this trait is associated with severe battering
violence and would not be a strong predictor of the situational
764 LANGER, LAWRENCE, AND BARRY

couple violence that characterized the current sample (e.g., John-
son & Leone, 2005).
We found support for the hypothesis that increases in chronic
stress would predict increases in physical aggression perpetration
in the present study, which was the first examination of the
influence of changes in stress on changes in physical aggression
over time. Notably, there was strong evidence for this within-
spouse association among both husbands and wives. The experi-
ence of chronic stress demands more emotional resources, and
individuals under increased strain may have greater difficulty
employing adaptive behaviors (e.g., active listening techniques)
during couple interactions. Our statistical approach allowed us to
rule out the possibility that the links between stress and physical
aggression were artifacts of the underlying maladaptive personal-
ity traits included in the present study. We were also able to
demonstrate that the links between stress and physical aggression
were generally unidirectional; none of the paths in which physical
aggression predicted stress were significant. However, no longitu-
dinal studies to date have documented this link between stress and
physical aggression, and specific mechanisms underlying the as-
sociations must be clarified. For example, affect regulation may
mediate this association, such that repeated exposure to stress
contributes to progressive declines in emotional stability, which in
turn reduces one’s threshold for engaging in physically aggressive
behavior (Umberson, Williams, & Anderson, 2002). Regardless of
whether this association is better accounted for by direct or indirect
links, we strongly recommend that future research include a con-
sideration of the role of stress in predicting the developmental
course of physical aggression in intimate relationships.
Strengths and Limitations of the Present Study
There are several strengths of the present study. First, we used
an established theoretical framework that integrates individual
characteristics and stress into the study of dyadic behavior to
investigate physical aggression over time. Second, our focus on
newly married couples allowed us to study a period of high risk in
marriage (Cherlin, 2004); research on established marriages is
limited because divorced couples are necessarily left out of these
samples (Glenn, 1998). Third, a longitudinal (3-year) design was
employed and multiple (four) waves of data were collected.
Fourth, data were collected from both husbands and wives, and an
APIM was utilized to account for the shared method variance of
husband and wife data. Fifth, both self- and partner reports of physical
aggression were used to yield a more reliable indicator of aggres-
sion. Finally, we tested our hypotheses via SEM techniques; SEM
has several advantages as a modeling approach with dyadic data
(see Olsen & Kenny, 2006).
There are also several important limitations of the present find-
ings. First, although the longitudinal design allows for statements
about temporal order, any causal conclusions should be made with
caution. For example, if increases in stress as predictors of in-
creases in physical aggression are established, this does not nec-
essarily mean that an individual will experience an escalation in
stress right before he or she engages in physically aggressive
behavior. Second, the emphasis placed on the internal rigor in this
study (e.g., all heterosexual married couples) is offset by con-
straints on the generalizability of the findings. For example, as the
majority of the sample consisted of Caucasian and well-educated
couples, our findings may not be generalizable to other samples of
newlywed couples, in particular to severely distressed couples or
to spouses engaging in battering violence. Also, we cannot con-
clude that our results would generalize to same-sex couples or to
dating or cohabiting couples. Finally, the self-selected nature of
our sample raises the possibility that the decision to take part in the
study might reflect characteristics that differ from those of couples
who decided not to participate. Third, data were obtained via
self-report questionnaires, which are vulnerable to memory distor-
tion and social desirability biases. Although these biases would
presumably result in the underreporting of physical aggression,
which would render the reported prevalence rates conservative,
observational studies may provide a more reliable indicator of
aggression by overcoming the biases of self-report data. Fourth,
although we removed the item relevant to marital stress on the
measure of chronic stress and controlled for several personality
traits, it is possible that our measure of stress did not allow us to
truly isolate chronic stress that is external to the relationship and
external to the individual.
A final limitation of the study relates to the exclusive focus on
individual characteristics (personality and stress) of each partner
and on one dyadic process (physical aggression). We chose to take
a parsimonious approach to investigation of physical aggression
within the context of the VSA model; however, this approach
means that we have not fully explicated the longitudinal course of
physical aggression in marriage. Other researchers have suggested
that examining episodic interactions (e.g., Linder & Collins, 2005)
and day-to-day fluctuations in relationship dynamics (e.g., Schu-
macher & Leonard, 2005) will allow us to understand the nature of
physical aggression more fully over time. Predictive accuracy will
be enhanced to the extent that future studies include measures of
other contextual factors and of adaptive and maladaptive processes
(such as psychological aggression or poor problem-solving skills)
not tested here that may significantly affect the course of physical
aggression in marriage.
Implications for Research, Theory, and Couple Therapy
The results of the present study have several implications for
explication of the role of personality in the context of intimate
partner aggression. Although ASPD has emerged as instrumental
in the development of partner aggression for both sexes (e.g.,
Capaldi & Owen, 2001), there are several reasons to consider the
role of lower order personality traits. Often, Axis II diagnostic
categories lack reliability, share diatheses, and are manifested
heterogeneously (e.g., Widiger & Samuel, 2005). There is also
evidence that similar personality traits manifest themselves differ-
ently—and lead to different behaviors—in men and women; thus,
they yield different diagnoses (Paris, 1997). Furthermore, the
results of the current study, as well as evidence that specific
underlying trait dimensions may distinguish different types of
domestic violence (e.g., Swogger, Walsh, & Kosson, 2007), pro-
vide support for the potential use of lower order traits in prediction
of intimate partner aggression. In sum, the above research con-
verges to suggest that the examination of lower order personality
traits may be more reliable and may yield greater explanatory
power than does the examination of diagnostic categories and thus
should be included in future research on personality and partner
aggression.
765
PREDICTORS OF PHYSICAL AGGRESSION IN MARRIAGE

From a theoretical standpoint, the current study provides evidence
of physical aggression as a process resulting from personality traits
and chronic stress. This evidence lends support to our argument for
the embedding of aggression research within a VSA framework.
We recommend further specification of this model and suggest
several issues for future research. First, the novel findings regard-
ing changes in chronic stress and physical aggression suggest that
the examination of facets of stress beyond those assessed in the
current study (e.g., acute stress) may elucidate this putative lon-
gitudinal association. Second, given the emerging evidence that
psychological aggression is a strong predictor of psychopathology
(e.g., Taft et al., 2006) and of physical aggression (e.g., O’Leary et
al., 1994), the examination of this maladaptive process in the VSA
model would improve our understanding of the marital context in
which physical aggression emerges and continues. Finally, an
integration of the current findings on distal predictors of physical
aggression with the inclusion of proximal predictors of aggression,
such as alcohol use, would likely provide information about par-
ticular situations in which at-risk couples might be particularly
likely to experience physical aggression.
In terms of clinical implications, current violence interven-
tion programs present two main problems for the treatment of
intimate partner violence. First, the limited efficacy of batter-
ers’ programs in general suggests that there is a need for
refinement of existing treatments and/or development of new
treatments (Babcock et al., 2004). Second, most intervention
programs were developed on the basis of research that exam-
ined models of male violence (e.g., the Duluth model; Pence &
Paymar, 1993); this limits our confidence in the application of
these same programs to the treatment of female partner aggres-
sion. Moreover, the evidence that the behaviors of both partners
contribute to the risk of clinically significant partner aggres-
sion, regardless of whether one or both are perpetrators (e.g.,
Capaldi & Owen, 2001), suggests that treatment of both part-
ners may be a viable alternative or complement to interventions
that target one partner. Thus, one option is to focus our attention
and resources on enhancing existing couple therapies to more
effectively treat intimate partner aggression.
3
As personality traits are a risk factor associated with physical
aggression and individual psychopathology, they must be ad-
dressed in the treatment of men’s and women’s partner aggression,
regardless of whether treatment is conjoint or sex specific. We
recommend an evaluation of spouses’ personality traits to aid in
treatment planning and to identify obstacles that may emerge
during the therapeutic process. Moreover, men’s and women’s
personality may be differentially associated with severity of part-
ner violence. For example, Ehrensaft et al. (2004) suggested that
women’s psychopathology may be predictive of moderate forms of
partner aggression, whereas men’s psychopathology may be pre-
dictive of severe partner violence. Furthermore, the evidence from
clinical studies that the personality characteristics of physically
violent men tend to be more deviant than are those of physically
violent women provides further indication that men’s psychopa-
thology should be considered carefully.
Many forms of specialized individual therapy have been devel-
oped for the treatment of personality disorders, and there is grow-
ing interest in the implications for couple therapy (e.g., Magnavita,
2000). Different personality traits are connected with different
characteristics that influence perceptions of interpersonal relation-
ships and present unique therapeutic obstacles (e.g., Bender,
2005). The identification of traits associated with emotional and
behavioral dysregulation may provide a basis for the use of spe-
cific techniques to target physical aggression. For example, im-
pulsive and (trait) aggressive individuals should be encouraged to
identify problematic behavioral urges and to identify when these
urges are most likely to occur in their own lives (Black, Blum,
Pfohl, & St. John, 2004). Techniques to reduce reactivity and
promote empathy would be useful (e.g., teaching partners to re-
spect and validate one another by listening, reflecting, and/or
responding with compassion; Goldman & Greenberg, 2006). It is
common for individuals with more aggressive or impulsive per-
sonalities to experience difficulties with trust and closeness; thus,
the development of emotional awareness, accurate expression of
emotion, and nonjudgmental self-disclosure within a supportive
therapeutic environment may be helpful for these individuals
(Magnavita, 2000).
Regardless of the specific techniques employed to treat person-
ality traits as precursors to physical aggression in intimate rela-
tionships, it is widely known that personality disorders are among
the most difficult psychological problems to treat with psychother-
apy. In contrast, there are highly effective therapies for stress
reduction and stress management, which are the other key precur-
sor to intimate partner aggression that we have identified in the
present study. Consequently, couple therapies that target physical
aggression in intimate relationships might be more effective to the
extent that the aggression is linked to stress rather than to person-
ality traits.
Therefore, the assessment of stress and perceived impact of
stressors may help determine whether stress management should
be a component of couple therapy when physical aggression is
present. Targeting chronic stress in particular may be a useful
therapeutic intervention because of its continuous presence (Kar-
ney et al., 2005) and its connection to changes in physical aggres-
sion perpetration over time. Spouses can be taught simple tech-
niques that can be practiced outside of the therapeutic context,
such as progressive muscle relaxation to decrease physiological
arousal and meditation to reduce stress reactivity (e.g, Rausch,
Gramling, & Auerbach, 2006). Emphasis on individual self-care
and the use of available sources of social support to cope with
demands external to the marriage should also be part of stress
management treatment. Finally, working on stress reduction as a
couple may help reduce the reciprocity of dysfunctional behaviors
that serve to keep couples in this cycle of negativity and physical
aggression. Moreover, teaching these techniques within the con-
text of couple therapy may promote “unified detachment” or a
“shared platform” for the couple, a treatment goal that has already
been shown to be effective in couple therapies (i.e., integrative
3
Before beginning conjoint therapy with a couple experiencing physical
aggression, an evaluation should be conducted to determine whether con-
joint therapy is appropriate. If the aggression is severe enough to warrant
medical treatment or if one spouse is fearful of participating in treatment
with his or her partner, couple therapy may be contraindicated. If the
partners plan to stay together and both partners commit to treatment, couple
therapy may have advantages over traditional interventions and over none
at all (O’Leary, 1996).
766 LANGER, LAWRENCE, AND BARRY

behavioral couple therapy; Christensen et al., 2004; collaborative
couple therapy; Wile, 1995).
References
American Psychiatric Association. (1994). Diagnostic and statistical man-
ual for mental disorders (4th ed.). Washington, DC: Author.
Andrews, J. A., Foster, S. L., Capaldi, D., & Hops, H. (2000). Adolescent
and family predictors of physical aggression, communication, and sat-
isfaction in young adult couples: A prospective analysis. Journal of
Consulting and Clinical Psychology, 68, 195–208.
Arias, I., Lyons, C. M., & Street, A. E. (1997). Individual and marital
consequences of victimization: Moderating effects of relationship effi-
cacy and spouse support. Journal of Family Violence, 12, 193–210.
Babcock, J., Green, C., & Robie, C. (2004). Does batterers’ treatment
work? A meta-analytic review of domestic violence treatment outcome
research. Clinical Psychology Review, 23, 1023–1053.
Barling, J., & Rosenbaum, A. (1986). Work stressors and wife abuse.
Journal of Applied Psychology, 71, 346 –348.
Bender, D. S. (2005). The therapeutic alliance in the treatment of person-
ality disorders. Journal of Psychiatric Practice, 11, 73– 87.
Berkowitz, L. (1990). On the formation and regulation of anger and
aggression: A cognitive–neoassociationistic analysis. American Psy-
chologist, 45, 494 –503.
Black, D., Blum, N., Pfohl, B., & St. John, D. (2004). The STEPPS group
treatment program for outpatients with borderline personality disorder.
Journal of Contemporary Psychotherapy, 34, 193–210.
Bollen, K. A. (1989). Structural equations with latent variables. New
York: Wiley.
Brock, R., & Lawrence, E. (2008). A longitudinal investigation of stress
spillover in marriage: Does spousal support adequacy buffer the effects?
Journal of Family Psychology, 22, 11–20.
Buss, A. H., & Perry, M. (1992). The Aggression Questionnaire. Journal
of Personality and Social Psychology, 63, 452– 459.
Cano, A., & Vivian, D. (2003). Are life stressors associated with marital
violence? Journal of Family Psychology, 17, 302–314.
Capaldi, D. M., & Kim, H. K. (2007). Typological approaches to violence
in couples: A critique and alternative conceptual approach. Clinical
Psychology Review, 27, 253–265.
Capaldi, D. M., & Owen, L. D. (2001). Physical aggression in a community
sample of at-risk young couples: Gender comparisons for high fre-
quency, injury, and fear. Journal of Family Psychology, 15, 425– 440.
Cherlin, A. (2004). The deinstitutionalization of American marriage. Jour-
nal of Marriage and Family, 66, 848 – 861.
Christensen, A., Atkins, D. C., Berns, S., Wheeler, J., Baucom, D. H., &
Simpson, L. E. (2004). Traditional versus integrative behavioral couple
therapy for significant and chronically distressed married couples. Jour-
nal of Consulting and Clinical Psychology, 72, 176 –191.
Clark, L. A., Simms, L. J., Wu, K. D., & Casillas, A. (in press). Schedule
for Nonadaptive and Adaptive Personality (2nd ed.). Minneapolis: Uni-
versity of Minnesota.
Coker, A. L., Davis, K. E., Arias, I., Desai, S., Sanderson, M., Brandt,
H. M., & Smith, P. H. (2002). Physical and mental health effects of
intimate partner violence for men and women. American Journal of
Preventive Medicine, 23, 260 –268.
Crawford, T., Shaver, P., Cohen, P., Pilkonis, P., Gillath, O., & Kasen, S.
(2006). Self-reported attachment, interpersonal aggression and person-
ality disorder in a prospective community sample of adolescents and
adults. Journal of Personality Disorders, 20, 331–351.
Dutton, D., & Bodnarchuk, M. (2005). Through a psychological lens:
Personality disorder and spouse assault. In D. Loseke, R. Gelles, & M.
Cavanaugh (Eds.), Current controversies on family violence (pp. 5–18).
Newbury Park, CA: Sage.
Edwards, D., Scott, C., Yarvis, R., Paizis, C., & Panizzon, M. (2003).
Impulsiveness, impulsive aggression, personality disorder, and spousal
violence. Violence and Victims, 18, 3–14.
Ehrensaft, M. K., Moffitt, T. E., & Caspi, A. (2004). Clinically abusive
relationships in an unselected birth cohort: Men’s and women’s partic-
ipation and developmental antecedents. Journal of Abnormal Psychol-
ogy, 113, 258 –271.
Fantuzzo, J. W., DePaola, L. M., Lambert, L., & Martino, T. (1991).
Effects of interpersonal violence on the psychological adjustment and
competencies of young children. Journal of Consulting and Clinical
Psychology, 59, 258 –265.
Fossati, A., Barratt, E. S., Borroni, S., Villa, D., Grazioli, F., & Maffei, C.
(2007). Impulsivity, aggressiveness, and DSM–IV personality disorders.
Psychiatry Research, 149, 157–167.
Frye, N. E., & Karney, B. R. (2006). The context of aggressive behavior in
marriage: A longitudinal study of newlyweds. Journal of Family Psy-
chology, 20, 12–20.
Glenn, N. D. (1998). The course of marital success and failure in five
American 10-year marriage cohorts. Journal of Marriage and the Fam-
ily, 60, 569 –576.
Goldman, R., & Greenberg, L. (2006). Promoting emotional expression
and emotion regulation in couples. In D. K. Snyder, J. A. Simpson, &
J. N. Hughes (Eds.), Emotion regulation in couples and families (pp.
231–248). Washington, DC: American Psychological Association.
Gottman, J. M., & Levenson, R. W. (2002). A two-factor model for
predicting when a couple will divorce: Exploratory analyses using
14-year longitudinal data. Family Process, 41, 83–96.
Hammen, C., Adrian, C., Gordon, D., Burge, D., Jaenicke, C., & Hiroto, D.
(1987). Children of depressed mothers: Maternal strain and symptom
predictors of dysfunction. Journal of Abnormal Psychology, 96, 190 –
198.
Johnson, M. P., & Leone, J. M. (2005). The differential effects of intimate
terrorism and situational couple violence. Journal of Family Issues, 26,
322–349.
Karney, B., & Bradbury, T. (1995). The longitudinal course of marital
quality and stability: A review of theory, method, and research. Psycho-
logical Bulletin, 118, 3–34.
Karney, B., Story, L., & Bradbury, T. (2005). Chronic and acute stress
among newlyweds. In T. Revenson, K. Kayser, & G. Bodenmann (Eds.),
Couples coping with stress (pp. 13–32). Washington, DC: American
Psychological Association.
Kenny, D. A., Kashy, D. A., & Cook, W. L. (2006). Dyadic data analysis.
New York: Guilford.
Kim, H. K., & Capaldi, D. M. (2004). The association of antisocial
behavior and depressive symptoms between partners and risk for ag-
gression in romantic relationships. Journal of Family Psychology, 18,
82–96.
Kline, R. (2005). Principles and practice of structural equation modeling
(2nd ed.) New York: Guilford.
Kraus, G., & Reynolds, D. J. (2001). The “ABC’s” of the cluster B’s:
Identifying, understanding, and treating Cluster B personality disorders.
Clinical Psychology Review, 21, 345–373.
Lawrence, E., & Bradbury, T. N. (2001). Physical aggression as a predictor
of marital dysfunction: A longitudinal analysis. Journal of Family Psy-
chology, 15, 135–154.
Lawrence, E., & Bradbury, T. N. (2007). Trajectories of change in physical
aggression and marital satisfaction, Journal of Family Psychology, 21,
236 –247.
Leonard, K. E., & Roberts, L. J. (1998). Marital aggression, adjustment,
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: Cambridge.
Linder, J. R., & Collins, W. A. (2005). Parent and peer predictors of
physical aggression and conflict management in romantic relationships
in early adulthood. Journal of Family Psychology, 19, 252–262.
767
PREDICTORS OF PHYSICAL AGGRESSION IN MARRIAGE

Magnavita, J. J. (2000). Relational therapy for personality disorders. New
York: Wiley.
Murphy, C. M., & Eckhardt, C. I. (2005). Treating the abusive partner: An
individualized cognitive– behavioral approach. New York: Guilford.
Muthe´n, L. K., & Muthe´n, B. O. (2006). Mplus user’s guide (4th ed.). Los
Angeles: Author.
O’Leary, K. D. (1988). Physical aggression between spouses: A social
learning perspective. In V. B. Van Hasselt, R. L. Morrison, A. S.
Bellack, & M. Hersen (Eds.), Handbook of family violence (pp. 31–55).
New York: Plenum Press.
O’Leary, K. (1996). Physical aggression in intimate relationships can be
treated within a marital context under certain circumstances. Journal of
Interpersonal Violence, 11, 450 – 452.
O’Leary, K. D., Barling, J., Arias, I., Rosenbaum, A., Malone, J., & Tyree,
A. (1989). Prevalence and stability of physical aggression between
spouses: A longitudinal analysis. Journal of Consulting and Clinical
Psychology, 57, 263–268.
O’Leary, K., Malone, J., & Tyree, A. (1994). Physical aggression in early
marriage: Pre-relationship and relationship effects. Journal of Consult-
ing and Clinical Psychology, 62, 594 – 602.
O’Leary, K. D., Vivian, D., & Malone, J. (1992). Assessment of physical
aggression against women in marriage: The need for multimodal assess-
ment. Behavior Assessment, 14, 5–14.
Olsen, J., & Kenny, D. (2006). Structural equation modeling with inter-
changeable dyads. Psychological Methods, 11, 127–141.
Paris, J. (1997). Antisocial and borderline personality disorders: Two
separate diagnoses or two aspects of the same psychopathology? Com-
prehensive Psychiatry, 38, 237–242.
Paunonen, S. V., & Ashton, M. C. (2001). Big Five factors and facets and
the prediction of behavior. Journal of Personality and Social Psychol-
ogy, 81, 524 –539.
Pence, E., & Paymar, M. (1993). Education groups for men who batter:
The Duluth Model. New York: Springer.
Poulton, R. G., & Andrews, G. (1992). Personality as a cause of adverse
life events. Acta Psychiatrica Scandanavica, 85, 35–38.
Quigley, B. M., & Leonard, K. E. (1996). Desistance of husband aggres-
sion in the early years of marriage. Violence and Victims, 11, 335–370.
Raudenbush, S. W., Brennan, R. T., & Barnett, R. C. (1995). A multivar-
iate hierarchical model for studying psychological change within mar-
ried couples. Journal of Family Psychology, 98, 167–174.
Rausch, S., Gramling, S., & Auerbach, S. (2006). Effects of a single
session of a large-group meditation and progressive muscle relaxation
training on stress reduction, reactivity, and recovery. International Jour-
nal of Stress Management, 13, 273–290.
Reynolds, S. K., & Clark, L. A. (2001). Predicting personality disorder
dimensions from domains and facets of the five factor model. Journal of
Personality, 69, 199 –222.
Ro, E., & Lawrence, E. (2007). Comparing three measures of psycholog-
ical aggression: Psychometric properties and differentiation from nega-
tive communication. Journal of Family Violence, 7, 575–586.
Rogge, R. D., & Bradbury, T. N. (1999). Til violence do us part: The
differing roles of communication and aggression in predicting adverse
marital outcomes. Journal of Consulting and Clinical Psychology, 67,
340 –351.
Rutledge, T., & Linden, W. (1998). To eat or not to eat: Affective and
physiological mechanisms in the stress– eating relationship. Journal of
Behavioral Medicine, 21, 221–240.
Schumacher, J. A., & Leonard, K. E. (2005). Husbands’ and wives’ marital
adjustment, verbal aggression, and physical aggression as longitudinal
predictors of physical aggression in early marriage. Journal of Consult-
ing and Clinical Psychology, 73, 28 –37.
Simpson, J., Winterheld, H., & Chen, J. (2006). Personality and relation-
ships: A temperament perspective. In A. Vangelisti & D. Perlman (Eds.),
The Cambridge handbook of personal relationships (pp. 231–250). New
York: Cambridge.
Stafford, E., & Cornell, D. G. (2003). Psychopathy scores predict adoles-
cent inpatient aggression. Assessment, 10, 102–112.
Story, L., & Repetti, R. (2006). Daily occupational stressors and marital
behavior. Journal of Family Psychology, 20, 690 –700.
Strahan, R. F. (1980). More on averaging judges’ ratings: Determining the
most reliable composite. Journal of Consulting and Clinical Psychology,
48, 587–589.
Straus, M., & Gelles, R. (1990). Physical violence in American families:
Risk factors and adaptations to violence in 8,145 families. New Bruns-
wick, NJ: Transaction Books.
Straus, M., Hamby, S., Boney-McCoy, S., & Sugarman, D. (1996). Re-
vised Conflict Tactics Scales: Development and preliminary psychomet-
ric data. Journal of Family Issues, 17, 283–316.
Stuart, G., & Holtzworth-Munroe, A. (2005). Testing a theoretical model
of the relationship between impulsivity, mediating variables, and hus-
band violence. Journal of Family Violence, 20, 291–303.
Swogger, M., Walsh, Z., & Kosson, D. (2007). Domestic violence and
psychopathic traits: Distinguishing the antisocial batterer from other
antisocial offenders. Aggressive Behavior, 33, 253–260.
Taft, C., O’Farrell, T., Torres, S., Panuzio, J., Monson, C., Murphy, M., &
Murphy, C. (2006). Examining the correlates of psychological aggres-
sion among a community sample of couples. Journal of Family Psy-
chology, 20, 581–588.
Testa, M., & Leonard, K. E. (2001). The impact of marital aggression on
women’s psychological and marital functioning in a newlywed sample.
Journal of Family Violence, 16, 115–129.
Umberson, D., Anderson, K. L., Glick, J., & Shapiro, A. (1998). Domestic
violence, personal control, and gender. Journal of Marriage and Family,
60, 442– 452.
Umberson, D., Williams, K., & Anderson, K. (2002). Violent behavior: A
measure of emotional upset? Journal of Health and Social Behavior, 43,
189 –206.
U.S. Census Bureau. (2006). American Community Survey. Retrieved July 16,
2008, from http://factfinder.census.gov/servlet/ACSSAFFFacts?_event⫽
Search&_lang⫽en&_sse⫽on&geo_id⫽04000US19&_state⫽04000US19
Wheaton, B. (1997). The nature of chronic stress. In B. Gottlieb (Ed.),
Coping with chronic stress (pp. 43–73). New York: Plenum.
Widiger, T. A., & Samuel, D. B. (2005). Diagnostic categories or dimen-
sions? A question for the Diagnostic and Statistical Manual of Mental
Disorders—Fifth Edition. Journal of Abnormal Psychology, 114, 494 –
504.
Wile, D. (1995). The ego-analytic approach to couple therapy. In N.
Jacobson & A. Gurman (Eds.), Clinical handbook of couple therapy (pp.
91–120). New York: Guilford.
Zubin, J., & Spring, B. (1977). Vulnerability: A new view of schizophre-
nia. Journal of Abnormal Psychology, 86, 103–112.
Received August 6, 2007
Revision received June 24, 2008
Accepted June 26, 2008 䡲
768 LANGER, LAWRENCE, AND BARRY