Benevolent cognitions as a strategy of relationship maintenance: Don't sweat the small stuff ....But it is not all small stuff

Benevolent cognitions as a strategy of relationship maintenance: Don't sweat the small stuff ....But it is not all small stuff
of 16
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Benevolent Cognitions as a Strategy of Relationship Maintenance:“Don’t Sweat the Small Stuff”. . . . But It Is Not All Small Stuff  James K. McNulty and Erin M. O’Mara University of Tennessee Benjamin R. Karney University of California, Los Angeles To maintain intimate relationships in the face of negative experiences, many recommend cognitivestrategies that minimize the implications of those experiences for global evaluations of the relationship.But are such strategies always adaptive? Suggesting otherwise, 2 longitudinal studies spanning the 1st 4years of 251 new marriages revealed that the effects of benevolent cognitions on relationship develop-ment depended on the initial levels of negativity in the relationship. Cross-sectionally, the tendency tomake positive attributions or otherwise disengage global evaluations of the relationship from negativeexperiences was associated with higher levels of satisfaction in marriages characterized by more frequentnegative behavior and more severe problems. Longitudinally, in contrast, such strategies only demon-strated benefits to healthier marriages, whereas they predicted steeper declines in satisfaction amongspouses in more troubled marriages by allowing marital problems to worsen over time. These findingshighlight the limits of purely cognitive theories of relationship maintenance and suggest that widelyrecommended strategies for improving relationships may harm vulnerable couples by weakening theirmotivations to address their problems directly. Keywords:  attributions, marriage, longitudinal, positive illusions, enhancement Even partners in satisfying relationships occasionally have neg-ative experiences of each other. Over the course of long-termrelationships in particular, partners will at times misunderstandeach other, get irritated, or disagree. Coping with these moments isone of the many challenges that couples must overcome if they areto remain satisfied with their relationships over time.What is the optimal strategy for managing negative experiencesin an intimate relationship? Researchers and practitioners agreethat one effective way for partners to cope with such events isthrough benevolent cognitions, that is, interpreting negative eventsin ways that allow each partner to maintain positive views of therelationship and of each other (Neff & Karney, 2005b). For ex-ample, several decades of research on marital attributions haveshown that spouses who tend to excuse each other’s negativebehaviors enjoy higher and more stable marital satisfaction thanspouses who tend to blame each other for those behaviors (for areview, see Bradbury & Fincham, 1990). Drawing on such find-ings, marital therapists expanded the scope of traditional behav-ioral marital therapies to target not only spouses’ overt actions butalso their cognitive and emotional reactions (e.g., Baucom &Lester, 1986). The idea that there are adaptive ways of thinkingabout problems has even permeated popular culture, as illustratedby the maxim: “Don’t sweat the small stuff.”But what happens when it is not all “small stuff?” Are benev-olent cognitions adaptive even for couples facing severe or intrac-table problems? The overarching goal of the studies described herewas to address these questions by examining whether associationsbetween benevolent cognitions and marital development are mod-erated by the frequency of couples’ negative behaviors and theseverity of their marital problems. To this end, the remainder of this introduction is organized into four sections. The first proposesthat various benevolent cognitions provide short-term relief fromnegative experiences within relationships by minimizing the im-plications of those experiences for immediate evaluations of therelationship. The second reviews research on the long-term impli-cations of such cognitions, noting that previous research has beeninconsistent in suggesting whether benevolent cognitions harm orprotect relationships over time. The third attempts to reconcilethose inconsistencies by suggesting that the long-term implicationsof benevolent cognitions for relationships may depend on the typesof problems couples face. Benevolent cognitions may benefitrelationships in which partners experience mostly mild or infre-quent problems but may ultimately harm relationships in whichpartners face severe or frequent problems that require more activeresponses because such thinking may allow those problems toworsen over time. The final section provides an overview of twolongitudinal studies of married couples that directly examinewhether the severity of self-reported relationship problems and thefrequency of observed negative behavior moderate the effects of  James K. McNulty and Erin M. O’Mara, Department of Psychology,University of Tennessee; Benjamin R. Karney, Department of Psychology,University of California, Los Angeles.Preparation of this article was supported by a Research DevelopmentAward from the College of Liberal Arts and Sciences at the University of Florida awarded to Benjamin Karney, Institute of Mental Health GrantMH59712 awarded to Benjamin Karney, and an award by the FetzerInstitute awarded to Benjamin Karney.We thank Chris Adams, Jessica Baker, Krista Bernard, Mark DaSilva,Nancy Frye, Katherine Leong, Sacha Lindekens, Giovanni Montrone,Kimberly Mosler, Lisa Neff, Rachel Nitzburg, Joanna Sadowski, JenniferSchurman, Jennifer Smith, Kara Sweeney, and Mark Trujillo for theirassistance in data collection, observational coding, and data entry.Correspondence concerning this article should be addressed to James K.McNulty, Department of Psychology, University of Tennessee, 1404 CircleDrive, Knoxville, TN 37996. E-mail: Journal of Personality and Social Psychology Copyright 2008 by the American Psychological Association2008, Vol. 94, No. 4, 631–646 0022-3514/08/$12.00 DOI: 10.1037/0022-3514.94.4.631 631  benevolent cognitions on changes in relationship satisfaction, andwhether changes in the severity of problems themselves mediatethese effects. Benevolent Cognitions and Immediate RelationshipEvaluations A robust literature suggests that satisfied couples use a varietyof cognitive strategies to manage negative experiences within theirrelationships (for a review, see Karney, McNulty, & Bradbury,2001). For example, when asked to describe each other, satisfiedpartners highlight each other’s positive qualities (e.g., Murray,Holmes, & Griffin, 1996a, 1996b) and minimize the importance of each other’s negative qualities (Neff & Karney, 2003). Whenasked to describe their relationships, satisfied partners emphasizetheir most positive experiences (e.g., Murray & Holmes, 1993,1994) and dismiss their more negative experiences (e.g., Eidelson& Epstein, 1982; Epstein & Eidelson, 1981; Neff & Karney,2003). When asked to interpret a specific negative experience inthe relationship, satisfied partners attribute that experience to tran-sient and external sources (for a review, see Bradbury & Fincham,1990). Even in the face of declining relationship quality, satisfiedpartners maintain optimism by recalling the present as an improve-ment over the past (e.g., Karney & Coombs, 2000; Karney & Frye,2002).Although these and other strategies have been described asdistinct processes of motivated cognition, they share a commontheme. In general, benevolent cognitions allow partners to ac-knowledge specific negative experiences of each other while min-imizing the implications of such experiences for their global eval-uations of the relationship (Karney, McNulty, & Frye, 2001;McNulty & Karney, 2001; Neff & Karney, 2005b). A wife with ahusband who fails to be supportive, for example, can attribute thenegative behavior to her husband’s stressful job rather than to hislack of warmth, thereby reconciling the specific negative experi-ence with her globally positive view of the relationship. McNultyand Karney (2001) demonstrated this process directly when theyasked recently married spouses to rate their experiences withvarious specific domains of their relationships (e.g., communica-tion, support, household management) and to report their moreglobal evaluations of those relationships, every day for 7 days.Multilevel modeling revealed that, for spouses who tended to makebenevolent attributions for their partners’ negative behaviors, thecovariance between negative experiences and global evaluationsacross days was significantly weaker than it was for spouses whotended to hold their spouses responsible for negative behaviors.Benevolent attributions, and possibly other benevolent cognitions,thus appear to operate by protecting partners’ global evaluations of their relationships from the broader implications of their specificnegative experiences (cf. Neff & Karney, 2005b). Benevolent Cognitions and Long-Term RelationshipOutcomes To be considered effective strategies of relationship mainte-nance, benevolent cognitions must be more than correlates of globally positive evaluations of relationships, however: They mustprotect global relationship evaluations over time. There are at leasttwo reasons to expect that they do. First, to the extent that spousesview their partners and relationships more positively in the present,they should interpret their partners’ behaviors more positively andbehave more positively themselves in the future (cf. Murray et al.,1996b). Second, minimizing the implications of negative experi-ences should allow spouses to avoid conflicts that could provecostly or painful over time. In line with such reasoning, treatment-outcome studies suggest that supplementing traditional behavioralmarital therapies with cognitive interventions can be more effec-tive at improving marital quality than behavioral marital therapiesalone (e.g., Christensen, Atkins, Yi, Baucom, & George, 2006;Jacobson, Christensen, Prince, Cordova, & Eldridge, 2000). Fur-thermore, several longitudinal studies have demonstrated that, onaverage, couples who think more benevolently about their rela-tionships are more satisfied with those relationships in the future(Fincham & Bradbury, 1987b, 1993; Karney & Bradbury, 2000;Miller, Niehuis, & Huston, 2006; Murray et al., 1996b; Murray &Holmes, 1997; Neff & Karney, 2005b).Nevertheless, other work offers an alternative perspective bysuggesting that benevolent cognitions can be costly over time(Colvin, Block, & Funder, 1995; Crocker, Major, & Steele, 1998;Dillard, McCaul, & Klein, 2006; Major & Schmader, 1998; Robins& Beer, 2001; Steele, Spencer, & Aronson, 2002; Swann, Silvera,& Proske, 1997). For instance, work outside close relationshipssuggests that cognitions that disengage evaluations of the self fromthe implications of failures in specific domains can undermine themotivation to seek improvements in those domains (e.g., Crockeret al., 1998; Major & Schmader, 1998; Steele et al., 2002). Be-nevolent cognitions may have similar implications within relation-ships. For example, women in abusive relationships who makebenevolent attributions for their partners’ negative behaviors areless likely to end their relationships, thereby exposing themselvesto further abuse (Gordon, Burton, & Porter, 2004; Katz, Arias,Beach, Brody, & Roman, 1995; Pape & Arias, 2000; Truman-Schramm, Cann, Calhoun, & Vanwallendael, 2000). Accordingly,though benevolent cognitions may provide immediate relief fromproblems, they may also allow those problems to fester and thusresult in lower levels of satisfaction over time. Indeed, severalstudies demonstrate that, though avoiding conflict is associatedwith more positive feelings toward the relationship in the shortterm, failing to address problems directly can lead to declines insatisfaction over time (Gottman & Krokoff, 1989; Heavey, Layne,& Christensen, 1993; but see Cramer, 2003, and Woody &Costanzo, 1990). Acknowledging the Context of Benevolent Cognitions How can the seemingly inconsistent findings on the longitudinalimplications of benevolent cognitions for relationships be recon-ciled? The answer may lie in recognizing a limitation that iscommon to research that has obtained both sorts of results. Al-though all studies of the effects of benevolent cognitions explicitlyor implicitly address how partners respond to negative experi-ences, research has not yet paid much attention to the nature of thenegative experiences themselves. On the contrary, studies in thisvein often examine partner’s reactions to hypothetical problems(e.g., “Imagine your partner were cool and distant to you”) orexperimentally induced problems (e.g., presenting partners withbogus evidence that a valued trait may in fact be negative). Evenresearch that has examined partners’ reactions to actual problems 632  M C NULTY, O’MARA, AND KARNEY  (e.g., Fletcher & Thomas, 2000) has not examined the extent towhich variability across couples in the sorts of problems they facemay be associated with the effects of cognitive processes withinthose relationships. Yet, research outside relationships suggestssuch variability is likely to play an important role in determiningthe implications benevolent cognitions have for long-term out-comes. Specifically, whereas the previously described process of disengaging self-esteem from weaknesses in specific domains maybe comforting when such weaknesses occur in  unimportant   do-mains, several authors have argued that dismissing negative feed-back in  important   domains can lead to missed opportunities toimprove those domains or respond directly to potentially construc-tive feedback (e.g., Crocker et al., 1998; Major & Schmader, 1998;Steele et al., 2002).In line with contextual models of relationships (e.g., Bradbury &Fincham, 1991; Karney & Bradbury, 1995), the long-term effectsof benevolent cognitions in relationships may similarly depend onthe types of problems to which they are applied. Because minorand infrequent problems may not affect the relationship, and in anycase may resolve on their own with time, benevolent cognitionsthat minimize the implications of such problems may do little harmto relationships. In fact, such cognitions may benefit relationshipscharacterized by minor problems by limiting opportunities fornegative interactions and promoting positive ones. However, whenapplied to severe or frequent problems that are likely to improveonly when addressed directly, the same cognitive strategies mayallow problems to worsen over time. Accordingly, relationshipscharacterized by frequent and severe problems may benefit morefrom cognitions that acknowledge the global implications of spe-cific negative experiences and thus provide the motivation neces-sary to engage in efforts at repair.We are not aware of research that has directly tested whether theeffects of benevolent cognitions on relationship development aremoderated by the nature of the problems experienced in the rela-tionship. However, a closer analysis of the samples addressed byprior research on these issues offers some support for this possi-bility. Studies suggesting that benevolent cognitions may havelong-term costs for intimate relationships have used samplesdrawn primarily from couples in established (e.g., Gottman &Krokoff, 1989; Heavey et al., 1993) or even distressed (Gordon etal., 2004; Katz et al., 1995; Pape & Arias, 2000; Truman-Schramet al., 2000) relationships. In contrast, studies showing that benev-olent cognitions protect relationships have used samples drawnprimarily from undergraduate dating couples and newlyweds (e.g.,Karney & Bradbury, 2000; Miller et al., 2006; Murray & Holmes,1997; Murray et al., 1996b; Neff & Karney, 2005b), populationswhose problems, on average, may be less severe than the problemsexperienced by couples in distressed or established relationships(e.g., Amato & Previti, 2003; Karney, Garvan, & Thomas, 2003).Even among these younger couples, variability in the severity orfrequency of problems may moderate the effects of benevolentcognitions, but the implications of this variability have yet to beexamined directly. Overview of the Present Studies Though interpreting negative experiences benevolently may al-low partners to remain satisfied with their relationships in the shortterm, the long-term benefits of such strategies may depend on thenature of the experiences to which they are applied. Althoughinterpreting minor and infrequent problems positively may avoidunnecessary conflicts, disengaging from more frequent or seriousproblems may reduce couples’ motivations to address problemsdirectly, allowing those problems to worsen over time. Yet, despitethe extensive literature on benevolent cognitions in relationships,we are aware of no studies that have directly examined whether theeffects of these cognitions are moderated by the types of problemsthat couples experience.Research to address this question must (a) sample relationshipsexperiencing problems that vary in severity; (b) directly measurethat variability; and (c) assess relationship outcomes over substan-tial lengths of time to allow analyses of both short-term andlong-term correlates of these cognitive processes. Accordingly, toexamine the role of negativity in moderating the effects of benev-olent cognitions on the course of relationship development, weconducted two studies that met these criteria. To ensure that anyobserved effects were not idiosyncratic to a particular measure-ment technique, both studies defined benevolent cognitions andnegativity in two conceptually similar yet operationally distinctways. To measure benevolent cognitions, both studies assessed (a)partners’ tendencies to make charitable attributions for each oth-er’s hypothetical negative behaviors and (b) the strength of theassociation between partners’ perceptions of specific relationshipexperiences and their global evaluations of the relationship acrossa 7-day diary assessment. To measure negativity within the rela-tionship, both studies assessed (a) the self-reported severity of problems encountered in the relationship and (b) the quality of couples’ behaviors during videotaped problem-solving discus-sions, as rated by objective observers. Both studies also askedcouples to report on their marital satisfaction and problem severityevery 6 months for the first 4 years of the marriage, a total of eightassessments in each study. Analyses drew on these repeated as-sessments to examine how aspects of the relationship and specificcognitive strategies interact to account for trajectories of relation-ship satisfaction and trajectories of problem severity over the first4 years of these marriages. Given the parallel designs of the twostudies, both are described simultaneously below. 1 Both studies sampled first-married newlywed couples, a partic-ularly appropriate sample in which to investigate these issues, forseveral reasons. First, the early years of marriage are a period of dramatic change for couples, during which they are at elevated risk  1 Data describing participants from both samples have been described inpreviously published reports. Data from the couples in Study 1 have beendescribed in several articles (Frye & Karney, 2002, 2004, 2006; McNulty& Karney, 2001, 2002, 2004; Neff & Karney, 2002, 2003, 2004, 2005b),but there has been little overlap between the variables examined in theseprior studies and the variables examined here. The two exceptions are (a)the diary data from this sample, treated here as an independent variable,were treated as a dependent variable in a previous report (McNulty &Karney, 2001) and (b) trajectories of satisfaction from this sample havebeen treated as dependent variables in prior reports (McNulty & Karney,2004; Neff & Karney, 2003). Trajectories of satisfaction have also beenexamined as dependent variables in studies drawing from the couples inStudy 2 (Neff & Karney, 2005a, 2005b). Despite the overlap in thedependent variables, however, this is the first report using data from any of these studies to examine interactions between benevolent cognitions andobservations of behavior or reported severity of problems. 633 BENEVOLENT COGNITIONS IN MARRIAGE  of marital disruption (Bramlett & Mosher, 2002; Cherlin, 1992).Second, couples in the early years of marriage are likely toexperience an especially wide range of relationship problems, as anumber of challenges tend to accompany youth and the transitionto marriage (e.g., relocation, starting a new job). Thus, the couplesexamined in these studies were likely to demonstrate significantvariability in both the predictors and the outcomes of centralinterest.We predicted that, in the short term, a greater tendency to makepositive attributions and a greater tendency to separate globalevaluations of the relationship from negative specific experienceswould be associated with higher levels of satisfaction with therelationship. We expected this association to be particularly pro-nounced among couples who experienced the most frequent andsevere negativity, as such couples should have the most to gainfrom minimizing the implications of their negative experiences.However, we predicted that the opposite would be true in the longterm: Although benevolent cognitions should predict more stablesatisfaction in relationships characterized at the outset by infre-quent and minor problems, these same strategies should predictsteeper declines in satisfaction in relationships characterized at theoutset by more frequent negative behavior or more severe prob-lems. Finally, we predicted that any interactions observed betweenbenevolent cognitions and negativity on changes in satisfactionwould be mediated by changes in problem severity over time suchthat thinking benevolently about frequent or severe problemswould predict declines in satisfaction through direct effects onincreasing problem severity over time. Method Participants Two independent studies were conducted in a Northern Floridacommunity surrounding a major state university. In both studies,couples were recruited using two methods. The first was to placeadvertisements in community newspapers and bridal shops, offer-ing payment to couples willing to participate in a study of newly-weds. The second was to send invitations to eligible couples whohad completed marriage license applications in counties near studylocations. All couples responding to either solicitation werescreened for eligibility in an initial telephone interview. Inclusionrequired that (a) this was the first marriage for each partner, (b) thecouple had been married less than 6 months, (c) each partner wasat least 18 years of age, (d) each partner spoke English and hadcompleted at least 10 years of education (to ensure comprehensionof the questionnaires), and (e) couples did not have children andwives were not older than 35 (to allow a similar probability of transitioning to first parenthood for all couples, as part of the largeraims of the studies). Eligible couples were scheduled for an initiallaboratory session.Descriptive statistics for both samples are presented in Table 1.As the table reveals, participants were of comparable age acrossboth samples, with both spouses in their mid-20s and husbandsbeing slightly older than wives on average. Reflecting the educa-tion level of the community, participants reported relatively highlevels of education, with a large proportion of participants beingfull-time students at the baseline assessment. The income level of participants reflects this fact, with the median total combinedhousehold income being less than $25,000. In both samples, themajority of participants were Caucasian (  80%) and Christian (  60%). Procedure Procedures were nearly identical in both studies. Before theirlaboratory session, participants were mailed a packet of question-naires to complete at home and bring with them to their appoint-ment. This packet included a consent form approved by the localhuman subjects review board; self-report measures of relationshipattributions, marital satisfaction, and the severity of marital prob-lems; and a letter instructing couples to complete all questionnairesindependently of one another and to bring their completed ques-tionnaires to their upcoming laboratory session. Upon arriving tothat session, spouses participated in two problem-solving discus-sions designed to assess the frequency of the negative behaviorsthey tended to exchange during their interactions with each other.Each spouse identified an area of difficulty in the marriage, andthen both spouses participated in two 10-min videotaped discus-sions in which they were left alone to “work towards some reso-lution or agreement” for each area of difficulty. The order of thetwo interactions was determined through a coin flip. If bothspouses chose the same topic, then they first discussed that topicand then discussed a second topic chosen by the spouse whosetopic was designated to be discussed second. After completingtheir interactions, couples were paid for participating in this phaseof the study (Study 1  $50, Study 2  $70).After payment, each spouse was given a stack of seven stamped,addressed envelopes, each containing a one-page questionnaireTable 1 Sample Demographics SpouseAgeYears of education Full-timeemployed(%)Full-timestudent(%)Income groupCaucasian (%) Christian (%)  M SD M SD Mdn. SD Study 1 (  N   82)Husband 25.12 3.32 16.43 2.22 40 54 $5K–$10K $4.83K 83 59Wife 23.67 2.77 16.35 1.77 39 50 $5K–$10K $4.41K 89 59Study 2 (  N   169)Husband 25.53 4.13 16.48 2.33 59 35 $5K–$10K $7.21K 94 66Wife 23.84 3.60 16.32 2.01 45 43 $0K–$5K $5.41K 86 63 634  M C NULTY, O’MARA, AND KARNEY  designed to assess their evaluations of their specific experiences intheir relationship that day as well as their global satisfaction withtheir marriage that day. Spouses were instructed to (a) completeeach page independently of one another every night for the nextseven nights, (b) seal each completed questionnaire in the providedenvelopes, and (c) place the sealed envelope in the mail thefollowing day. Couples received $1.50 for each page they com-pleted and $25 for completing all 14 pages (i.e., a bonus forcouples who completed all pages).At approximately 6-month intervals subsequent to the initialassessment, couples were recontacted by phone and again mailedmarital attributions, marital satisfaction, and marital problemsquestionnaires, along with postage-paid return envelopes and aletter of instruction reminding couples to complete forms indepen-dently of one another. This procedure was used at all follow-upprocedures except at Time 5. The Time 5 assessment resembledTime 1 in that couples completed questionnaires at home andbrought them to the laboratory where they engaged in a variety of tasks beyond the scope of the present study. After completing eachphase, couples were mailed a check for participating (Study 1   $40, Study 2  $40–$50).  Measures  Marital attributions.  Marital attributions were assessed eighttimes over the 4 years of each study, once every 6 months, usingthe Relationship Attributions Measure (Fincham & Bradbury,1992). This 24-item measure presents spouses with four negativestimulus events that are likely to occur in all marriages (e.g., “Yourspouse criticizes something you say”; “Your spouse does not payattention to what you are saying”). For each event, spouses areasked to rate their agreement, on a 7-point scale ranging from 1( agree strongly ) to 6 ( disagree strongly ), with statements thatreflect six attribution dimensions. The measure includes two sub-scales: a Causal Attribution and Responsibility Attribution, eachconsisting of 12 judgments. Given high correlations between thetwo subscales (Study 1, husbands’  r   .60, wives’  r   .60; Study2, husbands’  r     .62, wives’  r     .63) and given our lack of differential predictions regarding the effects of each subscale, all24 items were combined in both studies, resulting in a single scorefor each spouse with possible ranges of 24–168, with higher scoresindicating more negative attributions. The internal consistency of the total scale was adequate in both studies for Study 1, coefficientalpha was .88 for husbands and .90 for wives; for Study 2,coefficient alpha was .92 for husbands and .90 for wives). Seven-day diary.  Spouses’ tendencies to disengage theirglobal evaluations of their relationship from their specific experi-ences in the relationship was assessed at baseline through a pro-cedure described by McNulty and Karney (2001) that draws on thediary assessments. To measure spouses’ daily perceptions of theirexperiences, spouses were asked to evaluate, every day for 7 days,nine specific aspects of the marriage chosen to represent variouspotential sources of negative experiences that could vary daily.Each night, spouses evaluated that day’s experiences of (a) theirsex life, (b) their partner’s physical appearance, (c) their partner’ssocial skills, (d) the way their partner contributed to householdchores, (e) how their partner supported them, (f) their partner’sintellect, (g) their interactions with their partner, (h) the time spenttogether with their partner, and (i) the way disagreements wereresolved. To measure spouses’ daily global evaluation of therelationship, each night spouses also responded to the followingthree items modified from the Kansas Marital Satisfaction Scale(Schumm et al., 1986): (a) “How satisfied were you with yourpartner today?” (b) “How satisfied were you with your relationshipwith your partner today?” and (c) “How satisfied were you withyour marriage today?” Participants responded to all items on a7-point Likert scale, where 1    very unsatisfied   and 7    verysatisfied  . The internal consistency of each subscale was relativelyhigh across both studies (for items measuring evaluations of spe-cific experiences, coefficient alpha ranged between .82 and .94 forhusbands and between .82 and .94 for wives; for items measuringglobal satisfaction, coefficient alpha ranged between .90 and .97for husbands and between .89 and .96 for wives, across bothstudies). To estimate the extent to which spouses’ global andspecific ratings of the marriage covaried across days, the within-person association between the daily sum of their perceptions of their specific experiences and the daily sum of their global eval-uations of the relationship was estimated using the first level of amultilevel model (a more detailed description of the analysisappears in the  Overview of Data Analyses  section), in whichweaker associations indicate greater disengagement of global eval-uations from specific perceptions and stronger associations indi-cate less disengagement of global evaluations from specific per-ceptions and presumably represent less effective relationshipenhancement. Observed negative behavior.  The prevalence of negative be-havior in each marriage was estimated at baseline by codingvideotapes of couples’ problem-solving discussions for negativeversus positive behavior using a modified version of the VerbalTactics Coding Scheme (Sillars, Coletti, Parry, & Rogers, 1982).Each speaking turn from each spouse was coded. A speakerreceived an Avoidant code for speaking turns that were off-topic ormoved the discussion away from the problem at hand. A speakerreceived a Negative code for speaking turns that either directlyfaulted, rejected, or criticized the partner, or indirectly criticizedthe partner through presumptive attributions, avoiding responsibil-ity, or hostile questions. A speaker received a Constructive codefor speaking turns that were on-topic and not negative (Study 1) orfurthered the resolution of the conflict (Study 2).To determine the reliability of our coding, a subset of thediscussions (30% in Study 1 and 25% in Study 2) were randomlychosen to be coded by a second rater, and agreement betweencoders was assessed by calculating intraclass correlation coeffi-cients (ICC) between the proportions of speaking turns coded asconstructive and negative by each coder. Reliability was adequatein both studies (in Study 1, ICC for Constructive    .96, ICC forNegative  .75; in Study 2, ICC for Constructive  .82, ICC forNegative  .89).A total proportion of negative behavior and a total proportion of constructive behavior exhibited by each husband and each wifewas computed for each conversation by dividing the number of codes for each spouse by the total number of speaking turns forthat spouse in that conversation. We then obtained an estimate of the relative negativity exhibited by each spouse in each conversa-tion by subtracting the proportion of Constructive codes from theproportion of Negative codes exhibited by that spouse. Accord-ingly, scores on each conversation could range from   1.0, indi-cating that every speaking turn was positive, to 1.0, indicating that 635 BENEVOLENT COGNITIONS IN MARRIAGE
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks