2004-disclosure-and-relationship-satisfaction.pdf

CATRIN FINKENAUER Free University of AmsterdamNetherlands

RUTGER C. M. E. ENGELS University of NijmegenNetherlands*

SUSAN J. T. BRANJE AND WIM MEEUSUtrecht UniversityNetherlands**

Disclosure and Relationship Satisfaction in Families

The present study developed a theoretical frame-work for understanding the social mechanismsunderlying disclosure and its link with relationshipsatisfaction in a full family design. A study among262 intact families, consisting of 2 children and 2adults and involving 1,048 individuals, appliedthe social relations model. Results showed thatdisclosure was more important to satisfactionin horizontal relationships than vertical ones.Further, relationship-specific disclosure wasmore important to satisfaction than dispositionaldisclosure. These results have implications forthe examination of relationship regulation andmaintenance in (non)voluntary relationships andthe development of psychosocial problems in parent-child relationships.

Disclosure is at the heart of most relationships(e.g., Rubin, 1973). People strategically discloseinformation about themselves to develop and

maintain relationships (e.g., Canary, Stafford,Hause, & Wallace, 1993). Greater disclosurein relationships is related to greater emotionalinvolvement (e.g., Rubin, Hill, Peplau, & Dunkel-Schetter, 1980), liking (Collins & Miller, 1994),feelings of intimacy (Laurenceau, FeldmanBarrett, & Pietromonaco, 1998), and relationshipsatisfaction (Derlega, Metts, Petronio, & Margulis,1993). In fact, disclosure and indicators ofrelationship quality seem so intertwined thatDerlega and his colleagues called disclosure andrelationships mutually transformative (p. 9). Dis-closure affects the definition and quality ofthe relationship, which in turn affect the content,meaning, and impact of disclosure.Most theorists agree that disclosure should be

conceptualized as a multifaceted social processthat combines both dispositional and relationalaspects (Dindia, Fitzpatrick, & Kenny, 1997).Further, as a general rule, more disclosure shouldbe associated with more liking for the discloser(Collins & Miller, 1994). It remains unclear,however, whether dispositional aspects (we likepeople who disclose a lot) or relational aspects(we like people who disclose to us) of disclosureare responsible for the link between disclosureand relationship quality (Derlega et al., 1993).Further, little is known about whether findingson disclosure and the link between disclosure andquality of relationships in one type of relationshipgeneralize to other relationships. The presentstudy aims to enhance our understanding of dis-closure, satisfaction, and their relation infamilies.

Department of Social Psychology, Free University, van derBoechorststraat 1, 1081 BT Amsterdam, Netherlands([email protected]).

*University of Nijmegen, Institute of Family and ChildCare Studies, P.O. Box 9104, 6500 HE Nijmegen,Netherlands.

**Utrecht University, Child and Adolescent Studies, P.O.Box 80140, 3508 TC Utrecht, Netherlands.

Key Words: communication, disclosure, family, relationshipsatisfaction.

Journal of Marriage and Family 66 (February 2004): 195209 195

DISCLOSURE IN RELATIONSHIPS

Disclosure refers to the verbal communication ofinformation about the self, including personalstates, dispositions, events in the past, and plansfor the future (Jourard, 1971). Research confirmsthat disclosure can be conceptualized as both apersons disposition to disclose (Jourard; Miller,Berg, & Archer, 1983) and as a dynamic processbetween two partners in a unique relational context(e.g., Dindia, 1994; Laurenceau et al., 1998). Toillustrate, Dindia and her colleagues (1997) appliedsocial relations modeling, a statistical tool devel-oped to analyze dyadic data (Kenny & La Voie,1984), to examine disclosure in adult relationships.Their results revealed that the level of disclosure ina relationship is a function of both partners dis-position to disclose (i.e., their typical baseline ofdisclosure), both partners disposition to elicit dis-closure from others, and their unique relationship(i.e., wives disclosedmore to their husbands than toa stranger). Further, a dyadic reciprocity effect ofdisclosure emerged: Partners who disclosed moreto others also received more disclosure from others(Jourard). Thus, by applying social relations mod-eling to dyadic relationships between adults, thisstudy elegantly showed that dispositional and rela-tional processes are simultaneously at work in dis-closure in relationships.Despite the abundance of research on disclosure,

to our knowledge, there is no research examiningdisclosure in whole families. This is not to say thatthere is no research on disclosure in families. Toillustrate, there exists a considerable body ofresearch on adolescents disclosure to their parents(e.g., Buhrmester & Prager, 1995; Noller & Callan,1990), parents disclosure to their adolescent chil-dren (e.g., Dolgin, 1996), adolescents perceptionof their parents disclosure (Dolgin & Berndt,1997), and disclosure in sibling relationships (e.g.,Howe, Aquan-Assee, & Bukowski, 1995). Thisresearch consistently shows that children disclosemore to their mothers than to their fathers (for anoverview see Buhrmester & Prager) and that chil-dren report feeling closer to parents to whom theydisclose more (Miller & Lane, 1991). Because thisresearch often remains descriptive, however, andbecause it focuses on selected subsystems withinfamilies, it draws an incomplete picture of howdisclosure varies across different types of familyrelationships. Questions regarding disclosure asa social process occurring between family mem-bers therefore remain unanswered. Hence, the firstgoal of the present study was to provide more

complete evidence on disclosure processes infamilies.How should disclosure in families be concep-

tualized? Based on existing evidence (Dindiaet al., 1997), we predict that disclosure variesacross each family members disposition to dis-close, across each family members disposition toelicit disclosure from others (e.g., in somefamilies mothers may elicit more disclosurethan in other families), and across relationships(in some families, two family members disclosemore to each other than in other families, therebyfacilitating the development of coalitions [Jacob-vitz & Bush, 1996]). Additionally, we expectdisclosure to vary across families as a group.Does disclosure reciprocity occur across all

types of family relationships? We argue thatit does not. Rather, we propose that disclosurereciprocity is dependent on the type of familyrelationship. Research shows that child parentifi-cation, the reversal of parent and child roles, maylead to adjustment problems among children (foran overview see Chase, 1999). High disclosurereciprocity between parents and children maybe considered an indicator for parentification.Parents consider their children as equal partnersand use them as emotional confidants. This dis-closure reciprocity may increase trust and close-ness in the parent-child relationship (Miller &Lane, 1991), but at the same time, it may putchildren at risk for psychosocial problems,because parents burden their children withtheir own worries and concerns (Lehman &Silverberg-Koerner, 2002; Minuchin, Rosman,& Baker, 1978). This reasoning led us to predictthat disclosure reciprocity in intact families ismore likely to occur in horizontal relationshipsbetween family members of equal status (parent-parent and sibling-sibling relationships) than invertical relationships where partners haveunequal status (parent-child relationships).

DISCLOSURE AND RELATIONSHIPSATISFACTION IN FAMILIES

Disclosure and relationship quality are stronglylinked (Derlega et al., 1993). A meta-analysis byCollins and Miller (1994) supports a linear rela-tion between self-disclosure and liking in adultrelationships. Their analysis revealed (a) that wedisclose more to people we like, (b) that we likepeople more after having disclosed to them, and(c) that we like people who disclose more. In

196 Journal of Marriage and Family

adult relationships where partners have compar-able status, disclosure seems to help partners tocreate the mutual liking necessary to developintimate relationships (e.g., Lippert & Prager,2001). Liking and intimacy, in turn, appear to benecessary conditions for the disclosure of moreintimate personal information in relationships(e.g., Hendrick, 1981). Generalizing these findingsto family relationships, family members who dis-close more should be more satisfied with theirrelationships, family members should disclosemore to specific family members with whom theyhave good relationships, and family membersshould be more satisfied with relationships withother family members who disclose more. Despitetheir appeal, these assumptions have not been sub-stantiated because past research has included onlyone relationship partner or subsystem in families.Thus, the second aim of our study was to investi-gate the link between disclosure and relationshipsatisfaction in a full family design.The strength of the proposed link between dis-

closure and satisfaction should be moderated bytwo factors. First, it should vary across types offamily relationships. In horizontal relationships(i.e., parent-parent and sibling relationships),partners interact on an egalitarian and reciprocalbasis (e.g., Hinde, 1979). In these types of rela-tionships, disclosure patterns should parallelthose found in earlier research (e.g., Dindiaet al., 1997). On the contrary, in vertical relation-ships (i.e., parent-child and child-parent relation-ships), partners interact on a complementarybasis. Children turn to their parents for help,support, and guidance. Parents usually providehelp, support, and guidance to their children,but do not (and perhaps should not), in return,require help, support, and guidance from theirchildren. Parents thereby respect implicit ruleson generational boundaries and parental behavior(e.g., Minuchin et al., 1978). Hence, in well-adjusted families, disclosure should be moreimportant to relationship satisfaction in horizon-tal relationships than in vertical ones.Second, the link between disclosure and satis-

faction should be moderated by type of disclo-sure. Disclosure in relationships may bedispositional (i.e., the discloser is a generallyopen person who commonly reveals personalinformation to everybody) or relational (i.e.,intended only for this specific disclosure recip-ient; Dindia, 1994). Surprisingly few studies haveexamined which type of disclosure determinesrelationship quality. In the literature, two differ-

ent perspectives exist but have rarely been com-pared in empirical studies. The first perspectiveimplicitly assumes that the more people discloseabout themselves, the more others are satisfiedwith the relationship (Altman & Taylor, 1973). Incontrast, the second perspective argues thatrelationship satisfaction will be high only whendisclosure is exclusively directed at one partner(e.g., Archer & Burleson, 1980; Taylor, Gould, &Brounstein, 1981; Wortman, Adesman, Herman, &Greenberg, 1976). If disclosure is directed ateveryone, because the person is a high discloser,it should not be linked to relationship satisfaction.Hence, an additional goal of the present study wasto provide evidence about the relation betweendisclosure and satisfaction so as to determine therelative importance of dispositional and relationaldisclosure for relationship satisfaction in families.

ANALYTIC STRATEGIES TO STUDYDISCLOSURE AND RELATIONSHIP

SATISFACTION

Because family members continuously influenceeach other, the study of disclosure and relation-ship satisfaction in families poses a challenge forresearchers. How can we investigate disclosureand relationship satisfaction when family mem-bers coordinate their thoughts, feelings, andbehaviors, and when these patterns of interdepen-dence are of fundamental importance to ourunderstanding of these phenomena (e.g., Kenny& Kashy, 1991; Kenny, Kashy, & Bolger, 1998)?The most obvious answer is to investigate entirefamilies, yet in this case, the collected data willbe statistically interdependent.Most commonly used statistical analyses require

independence of observations, which is clearly anissue in studying families. The social relationsmodel offers one solution to this problem (e.g.,Cook, 1994; Kenny & La Voie, 1984). It allowsresearchers to treat interdependence as an interest-ing phenomenon in its own right rather than astatistical nuisance (Kenny & Kashy, 1991,p. 277). Specifically, it allows one to identify thatpart of the variance in complex social behaviorsuch as disclosure (or relationship satisfaction)that is due (a) to peoples disposition to discloseto others, the so-called actor effect (one familymember dispositionally discloses to all otherfamily members), (b) to peoples disposition toelicit disclosure from others, the so-called targeteffect (all family members disclose to one particularfamily member), (c) to the relationship between two

Disclosure and Satisfaction in Families 197

specific family members, the so-called relationshipeffect (a family member discloses to one specificpartner, but not to others), and (d) to families as awhole, the so-called family effect (the extent towhich a family as a group shows a tendency todisclose).Further, social relations modeling allows us to

assess processes of reciprocity. As Kenny andcolleagues (1998) point out, reciprocity is oneof the most fundamental elements in the investi-gation of social processes such as disclosure. Notonly do we disclose to others but others alsodisclose back to us (Jourard, 1971). Social rela-tions modeling allows us to consider this socialexchange by recognizing that disclosure recipro-city can operate at the dyadic level (one familymember discloses to another who discloses inreturn). This reciprocity effect reflects the uniquedisclosure that occurs between two family mem-bers and emerges when the relationship effects oftwo family members are correlated.In short, the social relations model allows one

to analyze interdependence in social interactiondata. It recognizes that behavior occurs at differ-ent levels of a social unit by distinguishingeffects due to individuals, nested in dyads, nestedin families. Finally, it allows one to identifyreciprocity at the dyadic level, so it is a uniquetool for the examination of social processes andbehavior in social systems such as families.

OVERVIEW

To test the proposed hypotheses, data were col-lected among 285 two-parent, two-(adolescent)children families. All four family members com-pleted questionnaires assessing the extent to whichthey disclose self-relevant information to eachother and the extent to which they are satisfiedwith their relationships with each other. Applyingthe social relations model, this design allows us toexamine three questions that are at the core of thepresent study. First, how should disclosure infamilies be conceptualized? Second, how doesrelationship satisfaction in families vary as a func-tion of disclosure? Third, is the importance of dis-closure for relationship satisfaction in families afunction of the type of family relationships andthe type of disclosure?By using social relations modeling in a full

family design that includes both disclosure andrelationship satisfaction, the present study isnovel in several respects. First, it examines dis-closure as a truly social process that occurs

between family members who influence eachother. Second, it is the first study to investigatedisclosure and its link with relationship satisfac-tion in different types of family relationships.Third, it is the first study to examine the relativeimportance of different types of disclosure forrelationship satisfaction.

METHOD

Participants

Data were obtained from a study on communica-tion, personality, and social relationships among285 Dutch two-parent families with two adoles-cent children (Haselager & Van Aken, 1999,Family and Personality Research Project). Theinitial sample of eligible families with two ado-lescent children was stratified, with equal num-bers of children in all age groups between 13years and 17 years and equal numbers of boysand girls, and was randomly taken from the reg-isters of 23 municipalities in The Netherlands.After sending a letter announcing the study, inter-viewers contacted the families by phone. Of thefamilies approached in this manner, 50% agreedto participate. Frequently given reasons for notwanting to participate were that families had nointerest in the topic of the project or that onefamily member did not want to collaborate.The total sample included 285 families.

Because social relations modeling deletes miss-ing cases listwise, this sample was reduced to 262families, including 1,048 individuals. Attritionanalyses revealed that the exclusion of caseswas merely due to single missing data entries(e.g., one younger adolescent did not rate rela-tionship satisfaction with sibling but did com-plete all other variables) and not to systematicvariations across families or individuals on any ofthe assessed variables. Given that social relationsanalyses were restricted to those 262 families thathad complete data, however, all reported analyseswere conducted on this sample.The sample comprised 140 families (54%)

with two children, 69 families (27%) with threechildren, 31 families (12%) with four children,and 22 families with five or more children. In 224families (79%), the older child who participatedin the study was the oldest child in the family.In 219 families (77%), the younger child hadonly one older sibling. In 22% of the families,more than two children fell in the age groupseligible for participation. In this case, the families


themselves decided which two children partici-pated in the study. For reasons of clarity andstatistical necessity (see below), the two childrenin each family are labeled older and youngerchild throughout the manuscript.Fathers mean age was 46.0 years (SD 3.71)

and mothers was 43.7 years (SD 3.73). Theolder childrens (137 boys, 125 girls) mean agewas 16.6 years (SD 0.83), and the youngerchildrens (124 boys, 138 girls) mean age was14.4 years (SD 0.75). Ninety-six percent of thefamilies were of Dutch origin. A small proportionof the parents, 17% of the mothers and 19% ofthe fathers, had finished primary or low second-ary education. Forty-six percent of the fathers and28% of the mothers had attended college or haduniversity education. All children lived with bothbiological parents.

Procedure

Trained interviewers visited the families at home,where mother, father, and each of the two chil-dren filled out a battery of questionnaires. Thepresence of the interviewer served to encouragethorough responding and to prevent discussionsamong family members during completion of thequestionnaires. Both children were given a com-pact disc gift certificate after completion of thequestionnaires. Further, families who filled outall questionnaires took part in a lottery to winone of 10 travel cheques (value of about $900) asrewards.

Measures

Only those questionnaires relevant to the presentstudy are described here. Given the full familydesign, measures were adapted to be appropriateand adequate in all types of family relationships.Relationship specificity was achieved by formu-lating items as statements and instructing partici-pants to imagine each specific family member(i.e., father, mother, sibling) before rating eachstatement for the specific partner.

Disclosure. To assess disclosure to different part-ners, we adapted the Self-Disclosure Index(Miller, Berg, & Archer, 1983). The originalscale consists of 10 items assessing general self-disclosure in same-sex relationships, and thenecessary adjustments were twofold. First, familymembers separately rated the frequency withwhich they disclose to each specific partner.

Second, the topics of disclosure were adapted tofit all types of family relationships. To illustrate,gender may be commonly discussed among mar-ried partners, but it is not high on the priority listin conversations between parents and children(see Dolgin & Berndt, 1997). Based on the topicsin the Self-Disclosure Index and topics identifiedin Dolgin and Berndts (1997) study on informa-tion that parents disclose to their children, wederived nine topics relevant in all family relation-ships: health, finances, positive things that hap-pened during the day, disappointments andsetbacks, other family members, future plans,friends, secrets, and fears and insecurities. Familymembers rated the extent to which they disclosedthese topics to each other on five-point scales(1 never, 2 rarely, 3 sometimes, 4 often,5 almost always). An example item is I talk tomy father about my friends. A pilot study con-firmed that all identified topics were relevanttopics of disclosure in parent-child relationships(Finkenauer, Engels, & Meeus, 2002). In thepresent study, items showed satisfactory internalconsistency, ranging from .84 to .91 (mean Cron-bachs alpha over 12 scales was .88). Each familymembers ratings were averaged to establishthree disclosure scores per person (e.g., fathersdisclosure to mother, to older child, and toyounger child); higher values on these scoresindicated greater disclosure.

Relationship satisfaction. To assess family mem-bers satisfaction with their different relation-ships, participants rated each relationship onfour adjectives (i.e., very good, pleasant, valu-able, difficult [reverse scored]; cf. Campbell,Converse, & Rodgers, 1976). We chose this meas-ure because it does not refer to interpersonalbehaviors related to communication (e.g., I cantalk openly with this person) as do most of thecommonly used relationship satisfaction ques-tionnaires (e.g., Locke & Wallace, 1959). Itthereby reduces the risk of the scale yieldingartificially high correlations with disclosure (seeFincham & Bradbury, 1987 for a detailed discus-sion). Family members rated each of their rela-tionships on five-point scales (e.g., 1 not at all;5 very much). In our study, all satisfactionscales showed a satisfactory internal consistency,with alphas ranging from .74 to .87 (mean alphawas .81). Participants ratings were averaged toestablish three relationship satisfaction scores foreach person (e.g., older childs satisfaction withrelationship with father, mother, and younger


sibling); higher values on these scores indicatedgreater satisfaction with the relationship.

Strategy for Analyses

The most common social relations design is theround-robin research design used in the presentstudy. To reliably estimate relationship effects,the analyses require data ideally from four familymembers (Cook & Dreyer, 1984; Cook, Kenny,& Goldstein, 1991). Data are analyzed from bothmembers of each possible dyad. In our case,social relations modeling analyzes families con-taining two parents and two children. The labelsolder and younger child should therefore not beconsidered theoretically important; rather theyconstitute labels to distinguish between the twomembers of the sibling dyad. The effects of thesocial relations model are, comparable to struc-tural equation modeling, estimated from theseobserved data. (For a detailed description of thesocial relations model with family data and esti-mation procedure, see Kashy & Kenny, 1990;Cook, 1994, 2000.)Social relations modeling was performed on

the 3 4 covariance matrix of each family mem-bers disclosure to and relationship satisfactionwith the three other family members. This analy-sis explores the extent to which variance in dis-closure and relationship satisfaction in each ofthe 12 family relationships is due to actor, target,relationship, and family effects. All effects areestimated separately, controlling for all othereffects. For example, a relationship effect is esti-mated after controlling for actor, target, andfamily effects. The different variances for disclo-sure and relationship satisfaction were simulta-neously estimated using structural equationmodeling with maximum likelihood estimationprocedures (Joreskog & Sorbom, 1993, LISREL8.30). Missing cases were deleted listwise, whichreduced the sample to 262 families. Given thesmall sample size of the subgroups and therequired power, distinguishing sons from daugh-ters or same-sex from mixed-sex sibling pairs insocial relations modeling was not feasible. Toexplore gender differences in disclosure, ancil-lary analyses of variance were conducted, whichwill be described below.For a four-person family, there are 12 unidir-

ectional indicators of disclosure and 12 unidirec-tional indicators of satisfaction. To separatelyestimate relationship effects and error variances,we treated split-half scales of disclosure and

satisfaction as separate indicators of disclosureand satisfaction, which produced 24 observedscores of disclosure and 24 of satisfaction (12relationships 2 scales; see Cook 1993, 1994,2000). Next, a single social relations model ana-lysis was conducted with each of the two disclo-sure and satisfaction scales (the indicators)included to partition the variance in disclosureand satisfaction into actor, target, relationship,and family effects for disclosure and satisfaction,respectively. We allowed for correlations amongmeasurement errors for each indicator per ratingfamily member (e.g., for each indicator offathers disclosure, we allowed fathers measure-ment errors for their disclosure to mother, olderchild, and younger child to correlate).The actor, target, relationship, and family

effects technically constitute separate factors orlatent variables within a confirmatory factor ana-lysis (cf. Cook, 1994). The factor loadings (i.e.,paths from the latent variables to the indicators)were all fixed at 1.0 and the variances were thenestimated. In the same model, we also estimatedthe hypothesized relations between the socialrelations components of disclosure and satisfac-tion.We did not estimate separate models for dis-

closure and satisfaction but instead estimatedboth conceptsas well as the hypothesized rela-tions between themin a single model. The sig-nificance level for the estimated components inthis model was set at p< .01. Thus, we conducteda conservative test of the model. Despite its com-plexity, it showed a good fit in our data set. Thew2 of this model was 1516.60, p< .01, df 963,the Non-Normed Fit Index (NNFI) was .95, andthe root mean square error of approximation(RMSEA) was .04, indicating an acceptable fitof the overall model (Browne & Cudeck, 1989).

RESULTS

The literature is replete with evidence on genderdifferences in disclosure (e.g., Buhrmester &Prager, 1995; Dindia & Allen, 1992). Beforeaddressing the three questions that were at thecore of our study, we describe analyses of variancethat were conducted to explore gender differencesin mean levels of disclosure in our sample.To compare the level of disclosure in same-sex

sibling dyads and mixed-sex sibling dyads, eachchilds disclosure scores were submitted to amultiple analysis of variance (MANOVA) withbirth order as a within-subjects factor, and gender


of discloser and gender of target as between-subjects factors. A marginal effect for birthorder indicated that older children reported dis-closing less information to their younger sibling(M 2.96) than vice versa (M 3.06), F(1,258) 3.13, p .078, E2 .012. Further, a maineffect for gender of discloser revealed that dyadsin which the older child was a girl showed greaterlevels of disclosure (M 3.13) than dyads inwhich the older child was a boy (M 2.90),F(1, 258) 9.40, p .002, E2 .035. This maineffect was moderated by a significant interactionbetween gender of older and gender of youngerchild, F(1, 258) 12.25, p .000, E2 060, indi-cating that female-female dyads reported disclos-ing more to each other than all other dyadconstellations (results not shown). These resultsreplicate findings from earlier research, showingthat girls disclose more to girls. Girl-to-boy dis-closure and boy-to-girl disclosure, however, wasnot greater than boy-to-boy disclosure (Dindia &Allen, 1992).To compare boys and girls levels of disclo-

sure toward mother and father, younger and olderchildrens disclosure scores were submitted to aMANOVA with parent (father versus mother) asa within-subjects factor, and gender of child as abetween-subjects factor. For older children, amain effect for parent yielded that both girlsand boys reported disclosing more informationto their mothers (M 3.55) than to their fathers(M 3.33),F(1, 260) 62.28, p .000, E2 .193.This main effect was moderated by an inter-action with childs gender, F(1, 260) 8.11,p .005, E2 .030, indicating that boys andgirls levels of disclosure were comparable forfathers (M 3.30 and M 3.35, respectively),but girls (M 3.66) disclosed more than boys(M 3.45) to mothers. The same pattern ofresults emerged for younger children. A maineffect for parent, F(1, 260) 94.03, p .000,E2 .266, was moderated by a significant inter-action, F(1, 260) 18.45, p .000, E2 .066.Boys and girls did not differ in their levels ofdisclosure to fathers (M 3.26 and M 3.20,respectively), but girls (M 3.56) disclosedmore to mothers than did boys (M 3.40).Taken together, these results confirm that chil-

dren disclose more information to mothers thanto fathers (for an overview, see Buhrmester &Prager, 1995). Further, they parallel the resultsfound for the gender constellation of sibling dyads.The disclosure level is highest in female pairs, inthis case when daughters disclose to mothers.

How Should Disclosure in FamiliesBe Conceptualized?

Confirming that disclosure in families varies as afunction of each members disposition to dis-close, Table 1 shows that all actor effects fordisclosure (father, mother, older child, youngerchild) were significant. These findings indicatethat individual family members differed in theirdispositions to disclose to other family members,some family members tend to disclose to others,and others are less open, regardless of their par-ticular family roles. Disclosure was also relation-ship specific. All relationship effects for dis-closure were significant (Table 1), indicatingthat the extent to which one family member dis-closes to another depends on the unique relation-ship between those two family members in aspecific family. To illustrate, mothers disclosureto father differs across families (some mothersdisclose much to their partners, others less).Additionally, disclosure varied across families asreflected by a significant family effect (Table 1).Hence, in some families, disclosure is greaterthan in others, independent of individual familymembers dispositions to disclose and their dis-closures in unique relationships. Disclosure doesnot seem to depend on a family members dis-position to elicit disclosure from others. The onlysignificant target effect emerged for fathers, sug-gesting that family members differ in the extentto which they disclose to the father (some fatherselicit much more disclosure from all family mem-bers than others).Understanding the importance of each social

relations component is facilitated by calculatingthe extent to which each component contributesto explaining variance in disclosure. Table 2 pre-sents the contributions of the components (i.e.,percentages of the variance explained by eachsocial relations effect) to the total construct var-iance (i.e., excluding error variance) in disclosurein specific dyadic relationships. To illustrate, thetotal construct variance in mothers disclosure tofather consists of the sum of the variances formothers actor effect, fathers target effect, themother-father relationship effect, and the familyeffect. The contribution of mothers actor effectsto the total variance in mothers disclosure tofathers is computed by dividing mothers actoreffect by the total construct variance.The largest part of the variance in disclosure

was explained by actor (between 27% and 53%)and relationship (between 31% and 54%) effects.


Family effects explained between 11% and 20%of the variance in disclosure. Target effects con-tributed very little to the variance in disclosure indyadic relationships (between 0% and 5%). Inline with the above described results, differences

in disclosure do not seem to depend much onindividual family members dispositions to elicitdisclosure from others.Interestingly, although this observation cannot

be tested, relationship effects contributed more tothe variance in certain relationships than actoreffects, whereas in other relationships, thereverse pattern emerged. Relationship effectswere more important than actor effects forparent-parent disclosure (M 41% versusM 37%), parents disclosure to their children(M 50% versus M 33%), and sibling disclo-sure (M 52% versus M 37%). The patternwas reversed for childrens disclosure to parents,where actor effects explained more of the var-iance in childrens disclosure than relationshipeffects did (M 48% versus M 35%). Thus,parents disclosure in families appears to dependmore strongly on the type of relationships inwhich they disclose (e.g., the extent to whichmothers disclose differs according to whetherthey disclose to their husbands or to their chil-dren) than on parents dispositions to disclose.The same applies to sibling relationships. Withrespect to disclosure from children to parents,however, it seems that childrens disposition todisclose makes more of a difference than thespecific relationships they have with their par-ents. Some children seem to disclose to theirparents (both of them) and others do not.

Is disclosure reciprocal? Dyadic reciprocity cor-relations were estimated by correlating the rela-tionship effects for disclosure for two familymembers. As predicted, dyadic disclosure

TABLE 1. SOCIAL RELATIONS MODEL VARIANCEESTIMATES FOR DISCLOSURE AND SATISFACTION

(N 262 FAMILIES)

Disclosure Satisfaction

Actor effect F .150* .090*M .090* .069*O .174* .208*Y .212* .204*

Target effect F .017* .008M .000 .007O .002 .002Y .003 .014*

Relationship effect FM .098* .135*FO .214* .086*FY .209* .038*MF .171* .168*MO .145* .075*MY .158* .040*OF .151* .101*OM .132* .062*OY .267* .279*YF .127* .064*YM .129* .029*YO .257* .216*

Family effect .063* .069*

Note: F father; Mmother; O older adolescent;Y younger adolescent; FM disclosure of father tomother (or fathers satisfaction with relationship withmother); FO disclosure of father to the older adolescent;FY disclosure of father to the younger adolescent;MF disclosure of mother to father, and so on.*p< .01.

TABLE 2. PERCENTAGE OF VARIANCE EXPLAINED BY THE SOCIAL RELATIONS MODEL VARIANCE ESTIMATES(N 262 FAMILIES)

Disclosure SatisfactionActor Target Relationship Family Actor Target Relationship Family

FM 48 0 32 19 29 2 46 23FO 36 0 50 14 36 1 36 28FY 36 0 50 14 42 7 19 33MF 27 5 50 18 22 3 53 22MO 30 1 50 20 32 1 36 32MY 29 0 52 20 36 7 21 36OF 43 4 38 15 54 2 26 18OM 47 0 36 17 61 2 17 20OY 34 0 54 12 37 2 49 12YF 50 4 31 14 59 2 18 21YM 53 0 33 15 65 2 10 23YO 39 0 49 11 41 0 45 14

Note: FM disclosure of father to mother (or fathers satisfaction with relationship with mother); FO disclosure offather to the older child; FY disclosure of father to the younger child; MF disclosure of mother to father, and so on.


reciprocity was stronger in horizontal family rela-tionships (i.e., parent and sibling relationships;see Table 3, upper panel) than in vertical rela-tionships (i.e., parent-child and child-parentrelationships), where it was not significant orvery low. For example, the more mother reporteddisclosing to father, the more father reported dis-closing to mother. Disclosure reciprocity washigher in sibling relationships than in parent rela-tionships. Thus, married partners and siblingsappear to match their level of disclosure. Thisdoes not appear to be the case in parent-childrelationships, however.

Social Relations Model Findings for Satisfaction

For reasons of consistency and to test thehypothesized disclosure-satisfaction relations,the model for relationship satisfaction was testedin the same fashion as the one for disclosure. Ascan be seen in Table 1, the results for relationshipsatisfaction largely parallel those for disclosure.All actor effects were significant, indicating thatthere are individual differences in how satisfiedindividual members are with their relationshipswith other family members. In addition, all rela-tionship effects were significant, indicating thatthe extent to which family members are satisfiedwith their family relationships depends on theunique relationship between two family mem-bers. The significant family effect indicates thatthere are between-family differences in relation-ship satisfaction. Similar to the results for disclo-sure, satisfaction does not seem to depend onrelationship partners. The only significant targeteffect emerged for younger children, indicating

that families differ in the extent to which allmembers are satisfied with their relationshipswith the younger child.The relative importance of social relations

components in accounting for variance in satis-faction again paralleled those for disclosure. Thelargest part of the variance was explained byactor and relationship effects (see Table 2; 22%to 65% and 10% to 53%, respectively). Relation-ship effects were more important in horizontalrelationships, and actor effects were more impor-tant in vertical relationships. Relationship effectscontributed more than actor effects to the var-iance in parents satisfaction with their maritalrelationships (M 50% versus M 26%), and inchildrens satisfaction with their sibling relation-ships (M 47% versus M 39%). In contrast,actor effects contributed more to the variance insatisfaction in parent-child and child-parentrelationships (M 37% and 60%) than relation-ship effects (M 28% and 18%). Target effectscontributed little to the variance in dyadicrelationship satisfaction (between 0% and 7%).Family effects contributed somewhat more to thevariance in satisfaction (M 23%) than in dis-closure (M 16%), indicating that satisfaction issomewhat more dependent on families as a wholethan is disclosure.

Is relationship satisfaction reciprocal? Similarto disclosure, for relationship satisfaction, dyadicreciprocity was more prevalent in horizontal rela-tionships than in vertical relationships (see Table3, upper panel). In contrast to disclosure, satisfac-tion reciprocity was larger in the marital relation-ship than in the sibling relationship. The onlysignificant dyadic satisfaction reciprocity in ver-tical relationships emerged for the father-olderchild relationship. Thus, married partners andsiblings seem to coordinate their feelings aboutthe relationship. With the exception of father-older child relationships, this is not the case inparent-child and child-parent relationships, whereone person may be satisfied with the relationshipwhereas the other is not.

How Does Relationship Satisfaction in FamiliesVary as a Function of Disclosure?

Are family members who disclose more also moresatisfied with their relationships? The hypoth-esis that family members who disclose more toother family members are also more satisfiedwith their family relationships was tested at

TABLE 3. DYADIC RECIPROCITY FOR RELATIONSHIP-SPECIFIC DISCLOSURE AND SATISFACTION

(N 262 FAMILIES)

Disclosure Satisfaction

FM-MF .38* .63*FO-OF .11* .46*FY-YF .02 .18MO-OM .04 .10MY-YM .05* .07OY-YO .71* .49*

Note: FM disclosure of father to mother (or fatherssatisfaction with relationship with mother); FO disclosureof father to the older child; FY disclosure of father to theyounger child; MF disclosure of mother to father, and soon.*p< .01.


both the individual and the dyadic levels. At theindividual level, the correlations between afamily members actor effect for disclosure andthat family members actor effect for satisfactionwere significant for all four family members(Table 4). Family members who disclose moreto all other family members also appear to bemore satisfied with their relationships with allfamily members.At the dyadic level, the correlations between a

family members relationship effect for disclo-sure and that family members relationship effectfor satisfaction were computed (Table 4). Ourpredictions were partly confirmed. Disclosureand satisfaction were related in horizontal rela-tionships. In vertical relationships where disclo-sure went from parents to children, consistentwith our prediction, disclosure and satisfactionwere weakly related or unrelated. Contrary toour prediction, they were related in vertical rela-tionships where disclosure went from children toparents. Put differently, fathers and mothers whoare more satisfied with their marital relationshipsdisclose more to their spouses than fathers andmothers who are less satisfied. Children who aremore satisfied with specific relationships alsodisclose more in these relationships than do chil-dren who are less satisfied. Parents who are moresatisfied with their relationships with their chil-dren, however, do not seem to disclose more tothem than parents who are less satisfied. Thefamily effect for disclosure is significantly corre-lated with the family effect for relationship satis-

faction (Table 4). This finding suggests thatfamilies who disclose more have more satisfac-tory relationships (and vice versa).Taken together, these results suggest that the

finding that people like those to whom they dis-close extends to families, because family mem-bers who disclose more in unique relationshipsare also more satisfied with these relationships.There is an important exception to this pattern,however: In parents relationships with their chil-dren, disclosure and satisfaction appear to beunrelated.

Is relational disclosure more important forthe recipients relationship satisfaction thandispositional disclosure? Finally, we examinedwhether the finding that people like those whodisclose more extends to families. In addition, wetested whether relational disclosure is relativelymore important than dispositional disclosure forthe recipients relationship satisfactionthat is,whether people like those who disclose to them orwhether they like people who generally disclose.As can be seen in Table 5, with regard to thequestion of whether family members are satisfiedwith relationships with partners who disposition-ally disclose more, only 3 of the 12 correlationsbetween actor effects for disclosure and relation-ship effects for satisfaction were significant.Interestingly, except for the correlation betweenmothers actor effect for disclosure and father-mother relationship effect, these correlationswere negative. This indicates that a family

TABLE 4. CORRELATIONS BETWEEN SOCIAL RELATIONS MODEL COMPONENTS OF DISCLOSURE AND SATISFACTION(N 262 FAMILIES)

Actor Effect Disclosure-Actor Effect Satisfaction

Relationship Effect Disclosure-Relationship Effect Satisfaction

Family Effect Disclosure-Family Effect Satisfaction

F .75* FM .55* .87*M .85* FO .12*O .49* FY .10*Y .50* MF .74*

MO .00MY .08*OF .69*OM .74*OY .77*YF .59*YM .63*YO .81*

Note: F father; Mmother; O older adolescent; Y younger adolescent; FM disclosure of father to mother (orfathers satisfaction with relationship with mother); FO disclosure of father to the older child; FY disclosure of father tothe younger child; MF disclosure of mother to father, and so on.*p< .01.


members disposition to disclose is not or is evenslightly negatively related to the partners rela-tionship satisfaction.In contrast, the correlations between a family

members specific relationship satisfaction andthe partners disclosure in that relationship weresignificant in 8 of the 12 cases. Importantly, theywere strongest in horizontal relationships, indi-cating that the more an individual family memberin a marital or sibling relationship discloses, themore the relationship partner is satisfied with therelationship. For vertical relationships, anotherpicture emerged. Of the eight parent-child andchild-parent relationships, four correlations werenot significant and three correlations were signif-icant but not particularly high. The only substan-tial correlation emerged for the older child-fatherdyad. Thus, in vertical relationships, disclosureby one relationship partner does not seem tostrongly contribute to the other partners relation-ship satisfaction.Finally, we took a closer look at the relative

importance of peoples dispositional versus rela-tional disclosure for others relationship satisfac-tion. In social relations modeling terms, wecompared the correlations between a familymembers relationship satisfaction and (a) thepartners actor effect of disclosure (i.e., one part-ners relationship effect satisfaction, other part-ners dispositional disclosure) and (b) thepartners relationship effect for disclosure (i.e.,one partners relationship effect satisfaction,other partners relational disclosure; see Table 5).

This comparison showed that, in horizontalrelationships, a partners relational disclosure ismore important to the recipients relationshipsatisfaction than is the partners general disposi-tion to disclose. Thus, relationship partners withequal status seem to like those who disclose tothem more than those who disclose more in gen-eral. In vertical relationships, neither disposi-tional nor relational disclosure appears to makemuch of a difference for recipients relationshipsatisfaction. Thus, relationship partners withunequal status seem to like those who do andthose who do not disclose to them. Whether par-ents disclose to their children does not seem tocontribute to the satisfaction children experiencein their relationships with their parents. Similarly,whether children disclose to their parents doesnot seem to contribute to parents satisfactionwith their relationships with their children.

DISCUSSION

The present study applied social relations model-ing (Kenny & La Voie, 1984) to examine dis-closure, relationship satisfaction, and theirassociation in families. It is the first study thatexamined the relative importance of relationalversus dispositional disclosure for relationshipsatisfaction. In addition, it is the first to differ-entiate between horizontal and vertical familyrelationships when examining the three questionsat the heart of the present study, each of which isdiscussed below.

TABLE 5. CORRELATIONS BETWEEN SOCIAL RELATIONS MODEL COMPONENTS OF DISCLOSURE AND SATISFACTION(N 262 FAMILIES)

Actor Effect Disclosure-Relationship Effect Satisfaction

Dyadic ReciprocityDisclosure-Satisfaction

F-MF .03 FM-MF .43*F-OF .07 FO-OF .15*F-YF .22* FY-YF .08*M-FM .27* MF-FM .35*M-OM .17 MO-OM .04M-YM .09 MY-YM .01O-FO .08 OF-FO .35*O-MO .03 OM-MO .04O-YO .11 OY-YO .67*Y-FY .03 YF-FY .11Y-MY .19* YM-MY .13*Y-OY .10 YO-OY .43*

Note: F father; Mmother; O older adolescent; Y younger adolescent; FM disclosure of father to mother (orfathers satisfaction with relationship with mother); FO disclosure of father to the older child; FY disclosure of father tothe younger child; MF disclosure of mother to father, and so on.*p< .01.


The social relations model does not examinemean differences, but differences in variations.When examining mean differences, our sampleof children and parents showed the same genderdifferences that have been well established in theliterature (Buhrmester & Prager, 1995; Dindia &Allen, 1992; Miller & Lane, 1991). That is, chil-dren disclosed more to their mothers than to theirfathers, and girls disclosed more than boys. Inaddition, the disclosure level was highest infemale-female relationships, independent ofwhether disclosure occurred in mother-daughteror sister-sister relationships. In all other relation-ships, independent of the gender constellation,the disclosure level was comparable (Dindia &Allen, 1992). Larger gender differences mayemerge if one were to consider, for example,the emotional content of disclosure (e.g., Barrett,Robin, Pietromonaco, & Eyssell, 1998; Timmers,Fischer, & Manstead, 1998).

How Should Disclosure in FamiliesBe Conceptualized?

Applying social relations modeling, our findingsconfirm that disclosure is a multifaceted conceptthat combines both dispositional and interperson-al processes (Dindia et al., 1997; Miller &Kenny, 1986). Adding to previous findings, dis-closure also varied across families. Hence, dis-closure in families should be conceptualized as asocial phenomenon that combines individual,relational, and group processes.

Disclosure in different types of family relation-ships. Our findings clearly challenge the oftenimplicit assumption that research findings regard-ing one type of relationship generalize to othertypes of relationships. In contrast to findingsfrom adult relationships (Dindia et al., 1997), notarget effects for disclosure in families emerged.This may be due to the type of disclosure thatwas assessed in our study. Indeed, studies assess-ing disclosure behavior find target effects (Dindiaet al.), whereas studies assessing perceptionsof disclosure fail to find target effects (e.g.,Miller & Kenny, 1986). Another explanationmay be more theoretically meaningful. In thesocial relations model, the absence of an effectmay be interpreted in two ways. On the one hand,peoples disposition to elicit disclosure fromothers may be less important in families than itis in other relationships. In families, members arenot free to select the best listener; rather, they

must disclose both to those who are good listen-ers and those who are bad listeners. On the otherhand, the absence of a target effect may alsoindicate that disclosers rate other members lis-tener qualities in a comparable fashion across allfamilies who participated in our study. Forexample, mother may be perceived to be a goodlistener and the younger child to be a bad listenerby all family members in all families. Becausethe social relations model does not allow us todistinguish between these two explanations,research must be designed to compare disclosurein family relationships with disclosure in otherrelationships. Additionally, because our findingsand existing research indicate that, as comparedto men, women elicit more disclosure from bothmales and females (Collins & Miller, 1994),future research should also examine disclosurein same-sex and mixed-sex relationships.In addition, relationship effects were more

important than actor effects for disclosure in hor-izontal relationships and in those vertical rela-tionships where disclosure occurred from aparent to a child. In these types of relationships,the level of disclosure is more likely to be deter-mined by relational processes than by the dispo-sitions of each partner. Further, parents more thanchildren seem to adjust their levels of disclosureto the specific disclosure partner. In our study,mean differences indicated that female siblingsdisclosed more to each other than male siblingsand male-female siblings (cf. Howe et al., 1995).Studies should investigate the extent to whichvariation in sibling disclosure across families isrelated to siblings gender or to other factors suchas age difference or quality of the relationship.In vertical relationships where disclosure

occurred from child to parent, actor effects weremore important than relationship effects. Chil-dren appear to vary in their disposition to discloseto their parents, with some children being moreopen than others. Children who are open are opentoward both parents. Thus, previous findings thatchildren disclose more to their mothers than theirfathers (for an overview, see Buhrmester &Prager, 1995) should be qualified. Although chil-dren in our sample disclosed more to theirmothers than to their fathers (in terms of meandifferences), social relations analyses showedthat children who tend to disclose much to theirmothers also tend to disclose much to theirfathers.Finally, dyadic disclosure reciprocity was

strong in horizontal relationships but absent or


weak in vertical relationships. Partners in hori-zontal relationships (in our case, family memberswith similar family status) match their levels ofdisclosure (Dindia et al., 1997; Jourard, 1971).Partners in vertical relationships (in our case,family members with different or complementarystatus; Hinde, 1979) do not seem to match theirlevels of disclosure. Vertical relationships maynot require a quid pro quo principle for disclo-sure, which may in fact be functional and healthy.High dyadic reciprocity of disclosure in the par-ent-child relationship may be an indication of alack of intergenerational boundaries (Lehman &Silverberg-Koerner, 2002; Minuchin et al., 1978).When a parent and a child engage in highlyreciprocal disclosure and the parent does not dis-close much to more appropriate family members,child adjustment may be negatively affected (seeJacobvitz & Bush, 1996; Lehman & Silverberg-Koerner). Given the important implications ofthis finding for the development of adjustmentproblems in children, future studies shouldexamine whether so-called enmeshed families(Minuchin et al.) with permeable boundariesbetween parents and children show greaterdyadic reciprocity of disclosure between parentsand children than do families with strongerintergenerational boundaries.

How Is Disclosure Linked toRelationship Satisfaction?

The social relations model allowed us to investi-gate the link between disclosure and satisfactionat the individual and the dyadic levels. On theindividual level, family members who disclosedmore were also more satisfied with their familyrelationships. On the dyadic level, children whodisclosed more to a specific family member werealso more satisfied with their relationship withthat specific family member, and spouses whodisclosed more to each other were more satisfiedwith their marital relationship. Our results sug-gest that, in horizontal relationships and in child-parent relationships, partners disclose to the onesthey like and like the ones to whom they disclose.In parent-child relationships, however, this effectdid not emerge. Parents level of satisfaction withtheir relationships with their children was notrelated to their levels of disclosure to their chil-dren. Finally, families who disclosed more hadmore satisfactory relationships. Overall, thesefindings strongly parallel findings on adult rela-tionshipsbetween both intimates and stran-

gersshowing that high disclosure is stronglylinked to high relationship quality (Dindia et al.,1997; Lippert & Prager, 2001). They also shednew light on ongoing questions surrounding themutually transformative nature of disclosure andrelationships (Derlega et al., 1993). Indeed, someauthors argue that disclosure should become aless important determinant of relationship satis-faction over time (Hendrick, 1981). As couplesbuild their relationships, other variables (e.g.,shared activities) become more important, dimin-ishing the impact of disclosure. However, wefound that disclosure remains strongly linked tosatisfaction for both men and women even whenmarriages last for 15 years or more, suggestingthat disclosure and satisfaction remain mutuallytransformative from the beginning to the end of arelationship (Baxter, 1987).Applying social relations modeling also

allowed us to examine the relative importanceof relational and dispositional disclosure for rela-tionship satisfaction. In general, we found littlesupport for a link between dispositional disclo-sure and relationship satisfaction. Apparently,family members who generally disclose are notbetter liked than family members who do not. Incontrast, relational disclosure seemed to make adifference: Family members who disclose to aspecific partner are better liked by that partner(cf. Archer & Burleson, 1980; Taylor et al., 1981;Wortman et al., 1976), above and beyond thedisclosers disposition to disclose. This processseems to operate primarily in horizontal familyrelationships. In relationships between equals, welike the ones who disclose to us. In vertical rela-tionships, this process was weak or not signifi-cant. It is possible that partners in verticalrelationships are more accepting and less condi-tional than partners in horizontal relationships. Ifpartners in horizontal relationships perceive theirpartners to disclose less frequently to them, theymay attribute this lack of disclosure to a poorlyfunctioning marital (Finkenauer & Hazam, 2000)or sibling relationship (Howe et al., 1995). Part-ners in vertical relationships, however, may attri-bute it to implicit relationship rules (Lehman &Silverberg-Koerner, 2002).

Strengths, Limitations, and Implications

To our knowledge, the present study is the first toexplicitly examine disclosure and relationshipsatisfaction in families. The results of this studyare both consistent with existing findings and


new. They confirm findings on disclosure reci-procity and the mutually transformative nature ofdisclosure and relationships. More importantly,they emphasize important differences betweendifferent types of family relationships. At thesame time, there are several limitations of thepresent study.First, our sample consisted of moderately to

highly educated parents and well-adjusted ado-lescent children. Problematic families were rare,restricting the variation in disclosure and satisfac-tion. Second, it is possible that parents discloseother information to their partners than to theirchildren, or that siblings talk about other kinds offeelings or activities than parents do. These pos-sibilities were not included in our measurements,and future studies should investigate whethercontent-specific assessments of disclosure infamilies provide different insights than more gen-eral assessments of disclosure. Third, our studydoes not offer information on the mechanismsand dynamics underlying disclosure, satisfaction,and their relation in families. How does disclo-sure reciprocity in families develop? Which fac-tors play a role in determining who discloseswhat to whom? What are possible mediatorsbetween disclosure and relationship satisfaction(e.g., affection, security, trust)? Does the strengthof the link between disclosure and satisfactionvary over time? Longitudinal and observationalmethods are needed to investigate these questions(e.g., Dindia et al., 1997). Finally, our study didnot examine how and why disclosure, satisfac-tion, and their relation differed across verticaland horizontal family relationships. Feelings ofcommitment, responsibility, power, or closenessshould be investigated to enhance our knowledgeof reasons for the observed differences betweentypes of relationships.We outline two broad implications of the pre-

sent study. First, combining a full family designwith social relations modeling allowed us toexamine interdependent data on disclosure andrelationship satisfaction in families and to ques-tion the widespread, often implicit assumptionthat research findings on adult relationships gen-eralize to all types of relationships. This know-ledge is crucial for our understanding of the waysin which people develop, maintain, and regulaterelationships across different relationship part-ners. Second, our study emphasizes the socialnature of the link between disclosure and rela-tionship satisfaction. Relational processes appearto play a far more important role in this link than

individual differences, and their importanceappears to be determined by the type of familyrelationship. Indeed, they appear much moreimportant in vertical than in horizontal familyrelationships. These results may have importantimplications for the examination of psychosocialproblems in parent-child relationships that awaitfurther investigation.

REFERENCES

Altman, I., & Taylor, D. A. (1973). Social penetration:The development of interpersonal relationships.New York: Holt, Rinehart & Winston.

Archer, R. L., & Burleson, J. A. (1980). The effects oftiming and responsibility of self-disclosure onattraction. Journal of Personality and Social Psy-chology, 38, 120130.

Barrett, L. F., Robin, L., Pietromonaco, P. R., &Eyssell, K. M. (1998). Are women the more emo-tional sex? Evidence from emotional experiences insocial context. Cognition & Emotion, 12, 555578.

Baxter, L. A. (1987). Self-disclosure and relationshipdisengagement. In V. J. Derlega & J. H. Berg (Eds.),Self-disclosure: Theory, research, and therapy(pp. 155174). New York: Plenum Press.

Browne, M. W., & Cudeck, R. (1989). Single samplecross-validation indices for covariance structures.Multivariate Behavioral Research, 24, 445455.

Buhrmester, D., & Prager, K. (1995). Patterns andfunctions of self-disclosure during childhood andadolescence. In K. Rotenberg (Ed.), Disclosureprocesses in children and adolescents (pp. 1056).New York: Cambridge University Press.

Campbell, A., Converse, P. E., & Rodgers, W. L.(1976). The quality of American life: Perceptions,evaluations, and satisfactions. New York: RussellSage Foundation.

Canary,D. J., Stafford, L.,Hause,K. S.,&Wallace, L.A.(1993). An inductive analysis of relational main-tenance strategies: A comparison among lovers,relatives, friends, and others. CommunicationResearch Reports, 10, 514.

Chase, N. D. (Ed) (1999). Burdened children: Theory,research, and treatment of parentification. Thou-sand Oaks, CA: Sage.

Collins, N. L., &Miller, L. C. (1994) The disclosure-likinglink: From meta-analysis toward a dynamic reconcep-tualization. Psychological Bulletin, 116, 457475.

Cook, W. L. (1993). Interdependence and the interper-sonal sense of control: An analysis of family rela-tionships. Journal of Personality and SocialPsychology, 64, 587601.

Cook, W. L. (1994). A structural equation model ofdyadic relationshipswithin the family system. Journalof Consulting and Clinical Psychology, 62, 500509.

Cook, W. L. (2000). Understanding attachment secur-ity in family context. Journal of Personality andSocial Psychology, 78, 285294.

Cook, W. L., & Dreyer, A. S. (1984). The social rela-tions model: A new approach to the analysis offamily-dyadic interaction. Journal of Marriage andthe Family, 46, 679687.


Cook, W. L., Kenny, D. A., & Goldstein, M. J. (1991).Parental affective style risk and the family system:A social relations model analysis. Journal of Abnor-mal Psychology, 100, 492501.

Derlega, V. J., Metts, S., Petronio, S., & Margulis, S. T.(1993). Self-disclosure. London: Sage.

Dindia, K. (1994). The intrapersonal-interpersonal dia-lectical process of self-disclosure. In S. Duck (Ed.),Dynamics of relationships (pp. 2757). London: Sage.

Dindia, K., & Allen, M. (1992). Sex differences in self-disclosure: A meta-analysis. Psychological Bulletin,112, 106124.

Dindia, K., Fitzpatrick, M. A., & Kenny, D. A. (1997).Self-disclosure in spouse and stranger interaction:A social relations analysis. Human CommunicationResearch, 23, 388412.

Dolgin, K. G. (1996). Parents disclosure of their ownconcerns to their adolescent children. PersonalRelationships, 3, 159169.

Dolgin, K. G., & Berndt, N. (1997). Adolescentsperceptions of their parents disclosure to them.Journal of Adolescence, 20, 431441.

Fincham, F. D., & Bradbury, T. (1987). The assess-ment of marital quality: A reevaluation. Journal ofMarriage and the Family, 49, 797809.

Finkenauer, C., & Hazam, H. (2000). Disclosure andsecrecy in marriage: Do both contribute to maritalsatisfaction? Journal of Social and Personal Rela-tionships, 17, 247265.

Finkenauer, C., Engels, R. C. M. E., & Meeus, W.(2002). Keeping secrets from parents: Advantagesand disadvantages of secrecy in adolescence. Jour-nal of Youth and Adolescence, 31, 123136.

Haselager, G. J. T., & Van Aken, M. A. G. (1999).Codebook of the research project Family and Per-sonality: Vol. 1. First measurement wave. Nijmegen,Netherlands: University of Nijmegen, Faculty ofSocial Sciences.

Hendrick, S. S. (1981). Self-disclosure and maritalsatisfaction. Journal of Personality and Social Psy-chology, 40, 980988.

Hinde, R. (1979). Towards understanding relation-ships. London: Academic Press.

Howe, N., Aquan-Assee, J., & Bukowski, W. M. (1995).Self-disclosure and the sibling relationship: What didRomulus tell Remus? In K. Rotenberg (Ed.), Dis-closure processes in children and adolescents(pp. 7899). New York: Cambridge University Press.

Jacobvitz, D. B., & Bush, N. F. (1996). Reconstruc-tions of family relationships: Parent-child alliances,personal distress, and self-esteem. DevelopmentalPsychology, 32, 732743.

Joreskog, K. G., & Sorbom, D. (1993). LISREL 8:Structural equation modeling with the SIMPLIScommand language. Hillsdale, NJ: Erlbaum.

Jourard, S. M. (1971). The transparent self. New York:Van Nostrand-Reinhold.

Kashy, D. A., & Kenny, D. A. (1990). Analysis offamily research designs: A model of interdepen-dence. Communication Research, 17, 462482.

Kenny, D. A., & Kashy, D. A. (1991). Analyzinginterdependence in dyads. In B. M. Montgomery &

S. Duck (Eds.), Studying interpersonal interaction(pp. 275285). New York: Guilford Press.

Kenny, D. A., & La Voie, L. (1984). The social rela-tions model. In L. Berkowitz (Ed.), Advances inexperimental psychology (Vol. 18, pp. 4485).Orlando, FL: Academic Press.

Kenny, D. A., Kashy, D. A., & Bolger, N. (1998). Dataanalysis in social psychology. In D. Gilbert, S.Fiske, & G. Lindzey (Eds.), The handbook of socialpsychology (Vol. 1, 4th ed., pp. 233265). Boston:McGraw-Hill.

Laurenceau, J. P., Feldman Barrett, L., & Pietro-monaco, P. R. (1998). Intimacy as an interpersonalprocess: The importance of self-disclosure, partnerdisclosure, and perceived partner responsiveness ininterpersonal exchanges. Journal of Personality andSocial Psychology, 74, 12381251.

Lehman, S. J., & Silverberg-Koerner, S. (2002).Family financial hardship and adolescent girlsadjustment: The role of maternal disclosure of finan-cial concerns. Merill-Palmer Quarterly, 48, 124.

Lippert, T., & Prager, K. J. (2001). Daily experiencesof intimacy: A study of couples. Personal Relation-ships, 8, 283298.

Locke, H. J., & Wallace, K. M. (1959). Short maritaladjustment and prediction tests: Their reliability andvalidity. Marriage and Family Living, 21, 251255.

Miller, L. C., & Kenny, D. A. (1986). Reciprocity ofself-disclosure at the individual and dyadic levels: Asocial relations analysis. Journal of Personality andSocial Psychology, 50, 713719.

Miller, J. B., & Lane, M. (1991). Relationshipsbetween young adults and their parents. Journal ofAdolescence, 14, 179194.

Miller, L. C., Berg, J. H., & Archer, R. L. (1983).Openers: Individuals who elicit intimate self-disclosure. Journal of Personality and SocialPsychology, 44, 12341244.

Minuchin, S., Rosman, B. L., & Baker, L. (1978).Psychosomatic families: Anorexia nervosa in con-text. Cambridge, MA: Harvard University Press.

Noller, P., & Callan, V. J. (1990). Adolescents per-ceptions of the nature of their communication withparents. Journal of Youth and Adolescence, 19,349361.

Rubin, Z. (1973). Liking and loving. New York: Holt,Rinehart & Winston.

Rubin, Z., Hill, C. T., Peplau, L. A., & Dunkel-Schetter, C. (1980). Self-disclosure in dating couples:Sex roles and the ethic of openness. Journal ofMarriage and the Family, 42, 305317.

Taylor, D. A., Gould, R. J., & Brounstein, P. J. (1981).Effects of personalistic self-disclosure. Personalityand Social Psychology Bulletin, 7, 487492.

Timmers, M., Fischer, A. H., & Manstead, A. S. R.(1998). Gender differences in motives for regulatingemotions. Personality and Social Psychology Bulle-tin, 24, 974985.

Wortman, C. G., Adesman, P., Herman, E., &Greenberg, R. (1976). Self-disclosure: An attribu-tional perspective. Journal of Personality and SocialPsychology, 33, 184191.


2004-disclosure-and-relationship-satisfaction.pdf

Documents