Professional Documents
Culture Documents
Thomas N. Bradbury
University of California, Los Angeles
Data from 172 newlywed couples were collected over the first 4 years of marriage to test how
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
Stress generation theory (Hammen, 2006) asserts that unsupportive partner may be prone to generating more
stable vulnerability factors and interpersonal deficits con- stress, thereby perpetuating depressive symptoms. This
tribute to a cyclical association between depressive symp- proposition builds upon a core assumption of Coynes in-
toms and life stress. This model suggests that interpersonal terpersonal theory of depression (Coyne, 1976b), that de-
deficits can heighten the effects of depressive symptoms on pression is in part a consequence of how close others
later stress, while interpersonal strengths can offset these respond to a depressed persons symptoms and behaviors.
effects. Understood in the context of intimate relationships, Using observational data on couple communication and
an individual with depressive symptoms in an emotionally eight assessments of depressive symptoms and life stress,
supportive relationship may be protected against future this research examines interpersonal processes involved in
stressors whereas a similar individual involved with an stress generation and considers whether specific behavioral
moderators affect the relationship between depressive
symptoms and future stressors beyond global marital satis-
Joseph M. Trombello and Thomas N. Bradbury, Department of
faction.
Psychology, University of California, Los Angeles; Dominik In contrast to models implicating stress as a cause of
Schoebi, Department of Psychology, University of Fribourg, Fri- depression, stress generation theory argues that depressive
bourg, Switzerland. symptoms predict future interpersonal stressors that are at
This research was supported by the UCLA Graduate Divisions least partially dependent upon the depressed persons ac-
Graduate Research Mentorship awarded to Joseph M. Trombello, tions (Hammen, 2006). Evidence supports this view (Ham-
a Swiss National Science Foundation Grant PZ00P1_121616 to men, 2005) and the view that future life events are in part a
Dominik Schoebi and a National Institute of Mental Health Grant function of individuals deficits in interpersonal problem
MH48674 to Thomas N. Bradbury. We thank Constance Hammen solving (Davila, Hammen, Burge, Paley, & Daley, 1995)
and Nicole Eberhardt for their consultation on coding stressful
events, and Jennifer Krull for statistical consultation. and competence (Herzberg et al., 1998). Herzberg and col-
Correspondence should be addressed to Joseph M. Trombello, leagues found that the inability to listen empathically and
Department of Psychology, University of California, Los Angeles, provide support was associated with future increases in
1285 Franz Hall, Box 951563, Los Angeles, CA 90095-1563. chronic interpersonal stress, as such deficits may heighten
E-mail: joetrombello@ucla.edu interpersonal conflict when close others solicit but do not
58
RELATIONAL MODERATORS OF STRESS GENERATION 59
receive expected support from the target or provoke the how marital communication moderates the generation of
partner to withdraw, thereby reducing support when the nonmarital conflict stressors. First, Hammens model sug-
target requires it. We aim to build on this idea by using gests that how a depressed or dysphoric individual interacts
observational data on marital interaction to identify possible with his/her partner may be an indicator of how he/she
behavioral moderators of depression-to-stress linkages that interacts more globally. Given the pervasive nature of in-
are most likely to amplify or offset the stress-generating terpersonal dysfunction, deficits in seeking or providing
aspects of depressive symptoms. social support and in problem-solving with ones partner
Coynes interpersonal theory of depression (Coyne, may be an index of general communication deficits that
1976b) provides direction for this aim by highlighting how occur across a variety of social contexts. These global
critical feedback from close others worsens depressive interpersonal deficits may affect stress generation outside of
symptoms. This model suggests that depressed individuals the marriage because the depressed individual is likely to
seek but then reject reassurance from others about their own employ the same dysfunctional communication (e.g. exces-
self-worth, eventually frustrating the partner and precipitat- sively seeking reassurance) with other people as with ones
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
who are high in reassurance-seeking are subsequently re- A second mechanism suggests that depressive behaviors
jected (e.g., Joiner, Metalsky, Katz, & Beach, 1999) sug- elicit negative responses from others, including depression
gests that how depressed individuals behave during social contagion and feelings of burden, that ultimately lead them
interactions merits consideration as an important component to reject the depressed individual (Coyne, 1976a). This
of stress generation, because individuals with depressive direct pathway by which depressive behaviors predict inter-
symptoms may engage in behaviors that are met with anger, personal rejection constitutes a stressful event that heightens
hostility and rejection from others, and this rejection can depressive symptoms.
perpetuate the course or severity of depression. A third mechanism may be that the manner in which
Coyne and Joiners research implies that repeated pat- partners respond to a targets depression affects the targets
terns of interpersonal communication between a depressed self-esteem and social competence in ways that either pro-
person and his/her spouse may affect the extent to which mote or hinder the generation of future stressors. For ex-
depressive symptoms impact the generation of future inter- ample, partners who support rather than reject a burdensome
personal conflict stressors. Particular patterns of responding partner may instill the depressed person with the needed
to a depressed person especially criticism and rejection confidence to believe the feedback they receive (thereby
appear to be important in the stress generation process and reducing excessive reassurance-seeking) and engage in
characterize communication between depressed or dys- problem-solving and support provision/seeking without re-
phoric spouses (Rehman, Gollan, & Mortimer, 2008). These sorting to withdrawal or interpersonal conflict. Furthermore,
well-established findings suggest that displays of hard if depressed individuals turn to their partner for advice with
negative affect like anger and contempt are especially char- interpersonal problems (like a conflict with ones boss),
acteristic of interactions among depressed spouses. adaptive communication within the marriage may also help
Problem-solving discussions involving depressed/dysphoric individuals solve interpersonal conflicts outside of marriage
partners are characterized by fewer positive and more neg- before they become more severe.
ative behaviors (e.g., Du Rocher Schudlich, Papp, & Cum- As studies have not yet examined interpersonal behaviors
mings, 2004), and the longitudinal association between in- alongside general reports of relationship quality as moder-
terpersonal stressors and depressive symptoms is stronger ators of the symptoms-to-stressors link, a secondary aim of
when wives display negative behaviors and emotion while the study is to test the alternative possibility that global
discussing a source of marital conflict (Cohan & Bradbury, reports of distress are primarily responsible for the behav-
1997). Coynes interpersonal model is not limited to cou- iors that emerge as stress generation moderators (cf. Whis-
ples problem-solving, however, and suggests that commu- man, 2007). By controlling for marital satisfaction, our
nication involving the provision and receipt of social sup- analyses will examine whether the relationship between
port may also moderate stress generation. For example, depressive symptoms and later stressors reflect effects of
individuals experiencing stress or depressive symptoms are marital communication rather than marital distress.
impaired in these domains and are especially likely to Using observational data collected from couples discuss-
provide and solicit social support using negative behaviors ing relationship problems and personal concerns, and mul-
and affect (Coyne & DeLongis, 1986; Pasch, Bradbury, & tiwave assessments of depressive symptoms and stressful
Davila, 1997). life events, this study will explore the extent to which
Taken together, this research suggests that repeated pat- specific interpersonal behaviors or general marital quality
terns of communication between depressed/dysphoric indi- moderate the association between depressive symptoms and
viduals and their partners may affect the relationship be- future conflict stressors outside of marriage. Three main
tween depressive symptoms and later stressful events. One hypotheses are tested. First, we expect that symptoms of
key aim of the current study is to expand upon stress depression will predict 6-month increases in life events and
generation research by examining whether any effects as- that this association will be more likely to be significant for
sociated with marital communication predict dependent and women than for men, given past findings (Hammen, 1991).
interpersonal stressors that are explicitly outside of mar- The second and more important hypothesis is that the
riage. A number of possible mechanisms may help explain symptoms-to-stress link will be stronger when spouses and
60 TROMBELLO, SCHOEBI, AND BRADBURY
their partners display lower levels of positive behaviors and four 10-minute discussions; two that centered around a
affect and higher levels of negative behaviors and affect non-marital area that one partner wished to change about
during social-support and problem-solving discussions. Al- him/herself (social support conversation), and two that fo-
though this hypothesis builds on studies linking observed cused on a source of conflict in the marriage (problem-
behavior with depression (Rehman et al., 2008) and our solving discussion). For the social support discussions, one
earlier argument, it remains tentative as relatively few stud- partner (the helpee) was selected to talk about something
ies have examined specific behavioral moderators of stress you would like to change about yourself, while the other
generation. Finally, we predict that any behavioral moder- spouse (the helper) was asked to be involved in the dis-
ation that we identify will remain independent of concurrent cussion and respond in whatever way you wish. Example
reports of relationship satisfaction, in line with several stud- topics included weight loss and finding a more suitable job.
ies that still found communication deficits among depressed In the problem-solving discussions, couples were told to
dyads after controlling for levels of marital satisfaction discuss the topic for 10 minutes and try to work toward a
(Rehman et al., 2008). The last two hypotheses represent mutually satisfying solution. After a break, partners
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
unique research questions with significant implications for switched roles so that each partner was a provider and
This document is copyrighted by the American Psychological Association or one of its allied publishers.
targeting specific patterns of marital communication that recipient of support and the focus and listener of the
may help a depressed or dysphoric individual cope with problem-solving discussion. All conversations were video-
symptoms so as to reduce future stressful event exposure. taped and coded by external raters.
Six months after the initial laboratory session, couples
Method completed mailed questionnaires. After an additional
6-month interval, couples returned for a second laboratory
Participants session that included questionnaires and the discussions.
Across five additional 6-month periods, couples responded
One hundred seventy-two newlyweds were recruited to mailed questionnaires. For all six nonlaboratory sessions,
from Los Angeles County marriage licenses filed between couples were mailed a questionnaire packet and instructed
May 1993 and January 1994. Couples received letters ask- to answer all questions independently of their partner. Cou-
ing them to participate in a longitudinal study of newlywed ples were paid $75 after completing each of the two labo-
couples. The first 172 couples who were eligible and at- ratory sessions and $25 after each of the six nonlaboratory
tended their laboratory appointment comprised the sample.1 assessments. Therefore, the first 4 years of marriages were
Almost all of the initial laboratory sessions occurred during studied across eight time points with a 6-month interval
the first 6 months of marriage. Compared with the couples between each assessment period.
who responded to the letter, nonresponders were signifi-
cantly lower in years of education (effect size r .18 wives
Materials
and r .29 husbands), age (wives only, r .07), job status
(husbands r .20, wives r .18), and rates of cohabi- Stressful life events. A 168-item Survey of Life Events
tation before marriage (r .11). (Bradbury, 1990) was administered at all eight time points
Husbands mean age was 27.9 years (SD 3.9). They and assessed life stressors that occurred over the past six
also averaged 15.6 (SD 2.2) years of education and months within nine domains, including work, marriage, and
earned a median annual income between $21,000 and personal health. Participants endorsed whether each event
$30,000. Sixty-seven percent of husbands were Caucasian, occurred to themselves or their partner and then rated its
15% were Latino/Chicano, 13% were Asian-American/ impact on a seven-point scale ranging from extremely
Pacific Islander, 4% were African-American, and 1% were negative (3) to extremely positive (3).
Middle Eastern. Wives mean age was 26.0 years (SD All checklist items were coded by nine raters, who, in
3.4). They averaged 16.2 (SD 3.4) years of education, accordance with stress generation theory, assessed if each
with a median annual income between $11,000 and 20,000. item involved another person (interpersonal) and was at
Sixty-one percent of wives were Caucasian, 16% were least partially dependent upon the participants behaviors
Latina/Chicana, 15% were Asian-American/Pacific Is- (dependent) after reading a coding manual created by the
lander, 5% were African-American, 2% were Middle East- first author. Items were selected as being dependent and
ern, and 1% identified as other. interpersonal if seven out of nine raters agreed. Items were
excluded if they were symptoms of depression or involved
Procedure
1
Spouses completed questionnaires and marital discus- Our paper uses the same sample as several other publications
sions during an initial laboratory session, with a follow-up on the developmental course of marriage (e.g., Davila et al., 1997;
Johnson et al., 2005; Lavner & Bradbury, 2010; Sullivan et al.,
laboratory session a year later at which 152 of the 172 2010). However, this study represents the first from our laboratory
couples returned, and where 150 couples provided problem- to use social support and problem-solving behaviors at multiple
solving affect data and 148 provided social support data. time points as moderators of stress generation, to consider how
During the laboratory sessions, questionnaires including marital satisfaction is involved in this process, and to use as an
marital satisfaction and depressive symptomatology were outcome variable the combination of interpersonal and dependent
completed separately, after which couples participated in life stressors that are explicitly outside of the marital domain.
RELATIONAL MODERATORS OF STRESS GENERATION 61
ones partner, to remove marital stressors and depressive behaviors were disregarded, and raw counts of the total
symptoms from the acute measure. Thirteen events were number of positive and negative behaviors as both helper
identified as interpersonal and dependent. We then com- and helpee served as the predictor variables, as others have
puted each persons average impact rating of each item, and done (Sullivan et al., 2010).
we constructed 95% confidence intervals around these Problem-solving discussion. Each spouse identified a
means to determine whether the sample average impact source of tension within the marriage which became the
rating of each item was significantly more negative than problem-solving discussion topic. Trained raters focused on
zero (p .05). These procedures resulted in 7 items2 that voice tone and pitch, posture, gestures and facial expres-
were both interpersonal and dependent life events, and the sions to rate affect displayed in five-second intervals using
outcome variable was the number of these items endorsed. the Specific Affect Coding System (SPAFF; Gottman &
The average impact rating for each item was more negative Krokoff, 1989). Each interval was rated with one of five
than .75 (argument with coworkers) and the average im- possible negative affect codes (anger, contempt, whining,
pact rating across persons and items was 1.01. All items
anxiety, or sadness) one of three possible positive affect
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
the correlation between husbands QMI at Time 1 and 3 was ance component was estimated for each parameter. We
r .70, while the corresponding correlation was r .57 for interpreted robust standard errors in all analyses and, due to
wives; both values were significant at the p .001 level. the specific directional hypotheses that we made, we re-
Behavioral codes at the two laboratory sessions were ported one-tailed significance values. All significant results
substantially correlated, with a median value of r .41 for were re-run using the extension for Poisson distributions. As
behaviors/affect during ones own topic (range: r .08 to the results did not differ in any appreciable way, we re-
r .56) and a median correlation of r .44 for behaviors/ ported results from models assuming a normal distribution
affect during the partners topic (range: r .24 to r .63). for clarity of interpretation.
To reduce measurement error and improve reliability, we We found a significant positive association between cur-
averaged behaviors from both assessments. Values of be- rent depressive symptoms and future stressors in husbands
haviors displayed at the first session were used instead of (3 .035, t 2.40, p .009), lending support to stress
this average for the 22 (social support) and 24 (problem- generation theory. For every 1 SD increase in depressive
solving) couples who did not provide behavioral data at the symptoms at any given time point, future experienced life
second session. However, systematic changes in life stress stressors at the next time point are expected to increase by
or depressive symptoms may affect behavioral codes at the .128 events. No significant effect resulted for wives (4
second laboratory session, such that couples who experi- .002, t .30, p .381). Importantly, variance in stress
ence more depressive symptoms or life stressors over the generation effects varied across couples [husbands: SD
one year period engage in marital interactions that are .10; 2(1) 6.99, p .008; wives: SD .03; 2(1) 3.91,
characterized by higher levels of negative behaviors/affect p .048], making the testing of potential moderators mean-
and lower levels of positive behaviors/affect. To account for ingful.
this possibility, in all analyses that included behavioral
moderators, we also controlled for each persons average Behavioral Moderators of the Depressive Symptoms-
level of depressive symptoms across all eight time points. Life Stress Relationship
Test of the Basic Stress Generation Model Social support behaviors. Building upon this model, we
tested for moderator effects of ones own behavior during
In an autoregressive two-level model, we predicted ones social support topic (helpee), and partners behav-
stressful life events at a subsequent time point from depres- iors as a support provider (helper), controlling for the
sive symptoms at the current time point, the number of average level of depressive symptoms across all time points
combined interpersonal and dependent stressful life events and the average marital satisfaction at lab sessions 1 and 2.
at the current time point, and separate predictors for hus- These controls helped determine whether specific behav-
bands and wives. Depressive symptoms thus predicted pro- ioral processes moderated stress generation above and be-
spective change in life events from current depression yond the unique effects of global marital satisfaction. Sep-
scores. In order to capture within-person associations, all arate models examined moderation of the average number
Table 1
Correlations Between Depressive Symptoms and Marital Satisfaction Across All Eight Time Points, for Husbands and
for Wives
Time point
Gender 1 2 3 4 5 6 7 8
Husbands .29 .36 .38 .14 .46 .31 .51 .20
Wives .32 .40 .31 .23 .28 .28 .38 .24
p .05. p .01. p .001.
RELATIONAL MODERATORS OF STRESS GENERATION 63
of positive and negative behaviors displayed as the helpee .003, t 1.20, p .116) negative behaviors during
and helper. Therefore, eight separate combinations were their partners topic significantly moderated the relationship
tested (husband positive support as helper, husband positive between their partners depressive symptoms and subse-
support as helpee, husband negative support as helper, hus- quent life stressors. Neither husbands (31 .000, t
band negative support as helpee, wife positive support as .16, p .437) nor wives (41 .001, t .97, p
helper, wife positive support as helpee, wife negative sup- .168) positive behaviors during their own support topic
port as helper, wife negative support as helpee). The corre- significantly moderated the relationship between their own
sponding level-2 equations were: depressive symptoms and subsequent life stressors. Simi-
larly, neither husbands (31 .000, t .16, p .439)
1j 10 11H/W_BEH nor wives (41 .001, t .49, p .311) negative
12H/W_QMI) 13H/W_ABDI u1j behaviors during their own topic moderated their own stress
generation process.
2j 20 21W/H_BEH 22H/W_QMI Problem-solving affect. As with the support behaviors,
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
reported marital satisfaction and depressive symptoms.4 stress and high levels of marital quality. Therefore, this
Figure 1 illustrates all significant interaction effects, dis- sample may lack some variability in their exposure to stress-
playing stress generation at times when spouses were one ful events, depressive symptoms, and negative marital con-
SD above or below their average levels of depressive symp- versations compared to groups recruited from a depressed/
toms, and when spouses were one SD above and below their distressed population. Even subclinical levels of depression,
mean on the behavioral moderators. We also found no however, can impair an individuals social functioning and
strong correlations between the three significant behavioral physical health (Judd, Paulus, Wells, & Rapaport, 1996) and
moderators (all correlation coefficients were smaller than affect work performance. Subclinical levels of depressive
r .18). symptoms also predict clinical depression and place indi-
It should be noted that for husbands, marital satisfaction viduals at substantial risk for future diagnoses of major
was never a statistically significant moderator of husbands depressive disorder (Wells, Burnam, Rogers, Hays, &
stress generation beyond the behavioral moderators; for all Camp, 1992). Furthermore, an analysis of depressive symp-
eight analyses, the marital satisfaction variable failed to toms across all time points revealed that between 5 and 15%
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
reach statistical significance. These findings lend support to of participants experienced mildmoderate symptoms (a
This document is copyrighted by the American Psychological Association or one of its allied publishers.
that idea that, among this sample, three specific kinds of BDI score of 10 or above) at any time point, demonstrating
interpersonal behaviors moderate the linkage between hus- that a portion of our participants faced clinically relevant
bands depressive symptoms and future life stressors symptoms that may be obscured by only examining mean
whereas general self-reports of marital quality do not. On BDI scores.
the other hand, while in none of the eight analyses did Second, behavioral data were collected in the first 6-18
behaviors moderate wives stress generation, wives self months of marriage, and we have no evidence that the
reports of marital satisfaction moderated her stress genera- interactional processes were stable over longer durations.
tion process in all eight of the analyses, with coefficients Third, while we examined marital quality, we did not assess
ranging from .001 to .005 and p-values ranging from for relationship quality with the friends, relatives or co-
p .006 to p .021. Irrespective of interpersonal behav- workers who were the sources of examined stress. We
iors, wives who reported lower levels of marital satisfac- therefore cannot fully address how relationship quality af-
tion and higher levels of depressive symptoms were espe- fects stress generation and interpersonal communication, as
cially likely to generate subsequent interpersonal conflict there is a mismatch between the dyadic partner who com-
stressors. Taken together, interpersonal communication municates with the target and the partner who contributes to
moderated husbands stress generation while martial satis- the observed stress. Fourth, coding the problem-solving
faction moderated wives stress generation. discussions for non-verbal affect and the support discus-
sions for verbal behaviors confounds the coding system
Discussion with the conversation topic, making it difficult to determine
whether kind of conversation or coded behaviors moderates
The current study used Hammens stress generation the- stress generation. It should be noted, however, that these
ory and Coynes interpersonal model of depression to clar- coding systems are common in previous literature (Sullivan
ify how specific interpersonal processes in marriage govern et al., 2010), that within-person support behaviors and
links between depressive symptoms and 6-month changes in
problem-solving affect were often moderately to highly
stressors outside of marriage across the first 4 years of
correlated, and that conversation type was constant, as all
marriage. Depressive symptoms predicted future stressors
participants took part in four discussions.
for husbands, and three specific forms of marital commu-
Fifth, reported effects were relatively small in clinical
nication moderated husbands stress generation beyond the
terms, with a one SD increase in depressive symptoms
effects of husbands marital satisfaction. Specifically, hus-
predicting a .128 event increase in future experienced stres-
bands one SD above the average level of depression were
sors, and a one SD change in communication behaviors
especially likely to generate future stressors to the extent
predicting between a .077 and .193 unit change in the stress
that husbands made infrequent displays of positive affect
and anger/contempt when problem-solving and wives dis-
played a greater number of positive behaviors during a 4
We also ran a stress generation model that included the three
social support conversation. Husbands marital satisfaction significant behavioral moderators simultaneously alongside the
never moderated husbands stress generation. In contrast, other controls, and significant findings for the behaviors were
behavioral variables never moderated wives stress genera- maintained: husbands hard negative affect during husbands
tion beyond the effects of wives marital satisfaction, problem-solving topics still moderated the association between
whereas wives marital satisfaction moderated their depres- husbands depressive symptoms and husbands future experienced
sive symptoms-to-stress associations. Overall, this work stressors (32 .002, t 2.25, p .013) and husbands
positive affect during husbands problem-solving topics still mod-
suggests that relationship functioning captured by behav- erated the association between his depressive symptoms and sub-
ioral variables in the case of husbands, and reports of sequent stressors (33 .009, t 2.80, p .003), while
relationship satisfaction in the case of wivesmoderates wives positive behaviors during husbands support topics still
stress-generation effects in the first 4 years of marriage. moderated the relationship between husbands depressive symp-
Interpretation of these results is limited in several ways. toms and husbands experienced stressful events (13 .007, t
First, participants initially had low levels of depression and 3.37, p .001).
RELATIONAL MODERATORS OF STRESS GENERATION 65
such that men were more likely to generate stressors to the teractions could examine whether patterns of interaction in
extent that they less frequently displayed anger and con- one domain pervade others.
tempt when problem-solving and had wives who displayed These findings also suggest implications for improving
more positive behaviors when providing support. Cohan the quality of interpersonal conversations so as inhibit the
and Bradburys research (1997) suggests some possible propensity for depressive symptoms to be associated with
explanations for these results, as they found that wives future stressors. Given that husbands positive affect during
frequent displays of anger during problem-solving were support conversations protected against experiencing future
associated with increased marital satisfaction and reduced stressors, interventions that help husbands solve interper-
depressive symptoms among wives with a high occurrence sonal problems using positive affect could create an inter-
of major stressful events. Cohan and Bradbury suggested personal environment that buffers against the stress-
that displays of anger may be beneficial if anger represents inducing effects of depressive symptoms. In the same way
active engagement in problem-solving and facilitates effec- that marital discord and conflict predict dysphoria/
tively solving stressors outside of marriage, and it is possi- depression (e.g., Whisman, 2007), repeated exchanges that
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
ble that anger is similarly constructive in promoting active promote positive affect when problem-solving may protect
This document is copyrighted by the American Psychological Association or one of its allied publishers.
problem-solving among men. Likewise, wives repeated against stress generation that perpetuates the cycle of de-
provisions of positive behaviors like reassurance or encour- pression and relationship discord.
agement when supporting dysphoric men may only encour-
age mens persistence with excessive reassurance-seeking,
References
in that depressed men may continually fail to believe the
positive feedback they receive, in accordance with Coynes Beck, A. T., Steer, R. A., & Garbin, M. G. (1988). Psychometric
model. properties of the Beck Depression Inventory: Twenty-five years
Finally, while marital satisfaction did not moderate mens of evaluation. Clinical Psychology Review, 8, 77100.
stress generation beyond observed behavioral variables, Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J.
womens stress generation was moderated by their self- (1961). An inventory for measuring depression. Archives of
reported marital satisfaction. These findings support past General Psychiatry, 4, 53 63.
research suggesting different pathways between marital sat- Bradbury, T. N. (1990). Survey of life events. Unpublished man-
isfaction and depressive symptoms for men and women, and uscript. University of California, Los Angeles.
Cohan, C. L., & Bradbury, T. N. (1997). Negative life events,
they add to the view that womens mood is linked more
marital interaction, and the longitudinal course of newlywed
strongly with perceived relationship satisfaction or social marriage. Journal of Personality and Social Psychology, 73,
support (e.g. Davila, Bradbury, Cohan, & Tochluk, 1997; 114 128.
Fincham, Beach, Harold, & Osborne, 1997). Pending rep- Coyne, J. C. (1976a). Depression and the response of others.
lication, the present results could implicate different strate- Journal of Abnormal Psychology, 85, 186 193.
gies for inhibiting stress generation in men and women, with Coyne, J. C. (1976b). Toward an interactional description of
a focus on enhancing specific interpersonal behaviors (pro- depression. Psychiatry, 39, 28 40.
moting more frequent displays of positive affect and also Coyne, J. C., & DeLongis, A. (1986). Going beyond social sup-
problem-solving engagement more broadly, even if that has port: The role of social relationships in adaptation. Journal of
Consulting and Clinical Psychology, 54, 454 460.
a negative tone) among men during problem-solving dis- Davila, J., Bradbury, T. N., Cohan, C. L., & Tochluk, S. (1997).
cussions of topics they identify, while suggesting the more Marital functioning and depressive symptoms: Evidence for a
difficult task of improving womens global perceptions of stress generation model. Journal of Personality and Social Psy-
relationship satisfaction. chology, 73, 849 861.
Future research on stress generation processes might ben- Davila, J., Hammen, C., Burge, D., Paley, B., & Daley, S. E.
efit from a sequential analysis of how one partners behav- (1995). Poor interpersonal problem solving as a mechanism of
iors or affect impact the others, in order to clarify the stress generation in depression among adolescent women. Jour-
mechanism by which depressive symptoms and interper- nal of Abnormal Psychology, 104, 592 600.
Du Rocher Schudlich, T. D., Papp, L. M., & Cummings, E. M.
sonal communication predict future stressors. Stress gener-
(2004). Relations of husbands and wives dysphoria to marital
ation research might also examine this process across conflict resolution strategies. Journal of Family Psychology, 18,
shorter time frames; daily diary studies should be used to 171183.
more closely link together symptoms, interpersonal behav- Fincham, F. D., Beach, S. R. H., Harold, G. T., & Osborne, L. N.
iors, and life stressors. This line of research could examine (1997). Marital satisfaction and depression: Different causal re-
how diurnal changes in mood predict specific communica- lationships for men and women? Psychological Science, 8, 351
tion patterns between partners that lead to subsequent stres- 357.
sors or hassles. Daily diary studies could also minimize Gottman, J. M., & Krokoff, L. J. (1989). Marital interaction and
recall errors that may occur when reporting retrospectively satisfaction: A longitudinal view. Journal of Consulting and
Clinical Psychology, 57, 4752.
about stressors. Finally, future work could use naturalistic Hammen, C. (1991). Generation of stress in the course of unipolar
approaches to examine how day-to-day interactions with depression. Journal of Abnormal Psychology, 100, 555561.
friends or employers amplify or modulate the potential for Hammen, C. (2005). Stress and depression. Annual Review of
interpersonal conflict. Research that directly tests these Clinical Psychology, 1, 293319.
mechanisms and collects data from a variety of social in- Hammen, C. (2006). Stress generation in depression: Reflections
RELATIONAL MODERATORS OF STRESS GENERATION 67
on origins, research, and future directions. Journal of Clinical dependent variable. Journal of Marriage and the Family, 45,
Psychology, 62, 10651082. 141151.
Herzberg, D. S., Hammen, C., Burge, D., Daley, S. E., Davila, J., Pasch, L. A., & Bradbury, T. N. (1998). Social support, conflict,
& Lindberg, N. (1998). Social competence as a predictor of and the development of marital dysfunction. Journal of Consult-
chronic stress. Personal Relationships, 5, 207218. ing and Clinical Psychology, 66, 219 230.
Johnson, M. D., Cohan, C. L., Davila, J., Lawrence, E., Rogge, Pasch, L. A., Bradbury, T. N., & Davila, J. (1997). Gender,
R. D., Karney, B. R., . . . Bradbury, T. N. (2005). Problem- negative affectivity, and observed social support behavior in
solving skills and affective expressions as predictors of change in marital interaction. Personal Relationships, 4, 361378.
marital satisfaction. Journal of Consulting and Clinical Psychol- Rehman, U. S., Gollan, J., & Mortimer, A. R. (2008). The marital
ogy, 73, 1527. context of depression: Research, limitations, and new directions.
Joiner, T. E., Metalsky, G. I., Katz, J., & Beach, S. R. H. (1999). Clinical Psychology Review, 28, 179 198.
Depression and excessive reassurance seeking. Psychological Sullivan, K. T., Pasch, L. A., Johnson, M. D., & Bradbury, T. N.
Inquiry, 10, 269 278. (2010). Social support, problem solving, and the longitudinal
Judd, L. L., Paulus, M. P., Wells, K. B., & Rapaport, M. H. (1996). course of newlywed marriage. Journal of Personality and Social
Socioeconomic burden of subsyndromal depressive symptoms Psychology, 98, 631 644.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
and major depression in a sample of the general population. Wells, K. B., Burnam, M. A., Rogers, W., Hays, R., & Camp, P.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
American Journal of Psychiatry, 153, 14111417. (1992). The course of depression in adult outpatients: Results
Karney, B. R., & Bradbury, T. N. (1997). Neuroticism, marital from the Medical Outcomes Study. Archives of General Psychi-
interaction, and the trajectory of marital satisfaction. Journal of atry, 49, 788 794.
Personality and Social Psychology, 72, 10751092. Whisman, M. A. (2007). Marital distress and DSM-IV psychiatric
Kenny, D. A., Kashy, D. A., & Cook, W. L. (2006). Dyadic data disorders in a population-based national survey. Journal of Ab-
analysis. New York: Guilford. normal Psychology, 116, 638 643.
Lavner, J. A., & Bradbury, T. N. (2010). Patterns of change in
marital satisfaction over the newlywed years. Journal of Mar- Received April 30, 2010
riage and Family, 72, 11711187. Revision received October 16, 2010
Norton, R. (1983). Measuring marital quality: A critical look at the Accepted October 21, 2010