UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The relation of unsupportive actions by the spouse to marital satisfaction Habke, A. Marie 1992

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1992_fall_habke_a_marie.pdf [ 3.38MB ]
JSON: 831-1.0086453.json
JSON-LD: 831-1.0086453-ld.json
RDF/XML (Pretty): 831-1.0086453-rdf.xml
RDF/JSON: 831-1.0086453-rdf.json
Turtle: 831-1.0086453-turtle.txt
N-Triples: 831-1.0086453-rdf-ntriples.txt
Original Record: 831-1.0086453-source.json
Full Text

Full Text

The Relation of Unsupportive Actions by the Spouse toMarital SatisfactionbyA. Marie HabkeB.Sc.N, University of Alberta, 1983B.A., University of Calgary, 1990A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF ARTSinTHE FACULTY OF GRADUATE STUDIESDepartment of PsychologyWe accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAOctober 1992© A. Marie Habke, 1992In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature)Department of ^c-^e7VThe University of British ColumbiaVancouver, CanadaDate ‘ 1(7-i70,g DE-6 (2/88)AbstractDespite the demonstrated benefits of social relationships, there is a growingrecognition that such relationships also contain a significant proportion of negativeinteractions. Even though research has shown that such negative interactions havedetrimental effects on outcomes such as marital satisfaction, little is known abouthow these interactions come to be associated with outcomes. One hundred andthree couples having at least one child from a previous relationship living with themcompleted a telephone interview. As predicted, the results suggest that higherlevels of perceived spousal unhelpfulness was related to lower marital satisfaction.Although spousal criticism was not related to marital satisfaction as hypothesized,reports of a failure of the spouse to provide support was linked to low maritalsatisfaction for husbands. Contrary to predictions, characterological attributionswere not related to marital satisfaction although the hypothesized relation betweenlevels of blame in the attribution and marital satisfaction was supported. However,when controlling for demographics and the type and seriousness of the stressor,extent of blame predicted marital satisfaction only for wives. Wives who expressedmore blame towards their spouse had lower DAS scores than did wives expressingless blame. No significant interactions between behavior and attribution werefound.iiTABLE OF CONTENTSPageAbstract^ iiTable of contents^  iiiList of Tables vAcknowledgments ^ viIntroduction  1Method^ 10Subjects 10Procedure^ 12Measures 14Demographics^ 14Stressors 14Seriousness of the stressor^ 15Spousal behaviors^ 16Attributions 17Dyadic Adjustment Scale^ 19Coding of open-ended items 20Results^ 20Univariate Results^ 21Marital satisfaction and demographics^ 21Most serious family stressor ^ 22Seriousness of the stressor 22Spousal behaviors^ 24Attributions 24i i iBivariate Relations^ 24Marital satisfaction and study variables^ 24Demographic variables and study variables 26Stressor and study variables^ 29Correlations among study variables 32Multivariate prediction of marital satisfaction^ 34Control variables and type of stressor 34Extent of unhelpfulness ^ 36Spousal behaviors 38Type of attribution^ 38Extent of blame in the attribution^ 40Interactions^ 40Discussion^ 41Stressors 42Spousal Behaviors^ 43Attributions 51Interactions between spousal behaviors and attributions^ 54Limitations^ 55Conclusion 57References^ 58Appendix A: Interview Items^ 63Appendix B: Codes for Open-ended Questions^ 64Appendix C: Regression Analyses with Reduced Set of Control Variables ^ 67ivLIST OF TABLESPageTable 1. Sample description^  13Table 2. Spousal differences in reports of stressor, spousal behaviors, andattributions for spousal behaviors^  23Table 3. Correlations between full scale scores of marital satisfaction and studyvariables^  25Table 4. Correlations between demographic variables and study variables -HUSBANDS 27Table 5. Correlations between demographic variables and study variables -WIVES^ 28Table 6. Correlations between type and seriousness of stressor and study variables- HUSBANDS^ 30Table 7. Correlations between type and seriousness of stressor and study variables- WIVES 31Table 8. Correlations between the extent of spousal unhelpfulness and type ofunhelpful spousal behavior, and the type of attribution and extent of blame inthe attribution ^  33Table 9. Regression coefficients and R2 change when regressing DAS on controlvariables^  35Table 10. Regression coefficients and R2 change when regressing DAS on extentof unhelpfulness and type of spousal behavior^ 37Table 11. Regression coefficients and R2 change when regressing DAS on type ofattribution and extent of blame in attribution^  39vACKNOWLEDGEMENTSThe author would like to acknowledge the contribution of Dr. Anita DeLongis tothe completion of this study. I am very thankful for all of her support - tangible,informational, esteem, and emotional. I would also like to acknowledge thesupport of all the family members and friends who have helped me throughout theprocess of completing this research. Finally, it is essential to recognize the hardwork of so many on the Stepfamilies Project that made this study possible.viSpousal Unhelpfulness1INTRODUCTIONRemarried couples with children from previous unions confront a multitudeof stressors and strains in the family reorganization process (Hobart, 1990; Keshet,1990). Although little is known about the adaptation of remarried couples, twoseparate literatures can be brought to bear. First, the research on social support asa stress buffer suggests that perceptions of support or nonsupport are associatedwith outcomes such as depressed mood (Wethington & Kessler, 1986) and maritalsatisfaction (Gray, Lovejoy, Piotrkowski, & Bond, 1990). Second, the maritalliterature has demonstrated that the attributions made for spousal actions areassociated with marital satisfaction (Fincham & Bradbury, 1987; Holtzworth-Monroe & Jacobson, 1985). Both of these literatures will be considered in anattempt to develop a predictive model of marital satisfaction in remarried couples.A substantial amount of research has demonstrated that adaptation tostressful life events is an important predictor of psychological and physical well-being (DeLongis, Folkman, & Lazarus, 1988; Folkman, Lazarus, Gruen, &DeLongis, 1986). The role of social support in determining adjustment to stressand subsequent well-being has been the focus of extensive study (Cohen & Wills,1985; Kessler & McLeod, 1985). Recently, the field has moved from a basicendorsement of the positive effect of social relationships (Dean & Lin, 1977;House, 1981; Lin & McGranahan, 1980) to a recognition that there may also be asignificant proportion of negative interactions within such relationships (Rook,1984; Schuster, Kessler, & Aseltine, 1990). As negative social interactions havebeen shown to have detrimental effects on mood (Pagel, Erdly, & Becker, 1987;Schuster et al., 1990) and life satisfaction (Rook, 1984), the quality of socialSpousal Unhelpfulness2interaction during times of stress may be an important determinant of adaptation.Even so, it is clear that interactions perceived as unhelpful or unsupportive by therecipient occur frequently during stressful life events (Dakof & Taylor, 1990;Lehman & Hemphill, 1991). To date, however, there is little understanding of theprocesses through which unsupportive actions on the part of the spouse might cometo be associated with outcomes such as marital adjustment.Recent studies on failed support attempts have provided useful direction byexamining specific types of unsupportive actions; several kinds of interactionswhich may be perceived as unhelpful by those under stress have been identified(Dakof & Taylor, 1990; Lehman, Ellard, & Wortman, 1986). For example,cancer patients have described various specific actions from potential supportproviders as unhelpful, such as expressing excessive worry or pessimism,minimizing the patient's difficulties, and criticizing the patient's response (Dakof &Taylor, 1990). This 'unhelpful' label was also applied by the cancer patients to theabsence of helpful actions such as expressing concern or providing social contact(Dakof & Taylor, 1990). Among the bereaved, giving advice, encouragingrecovery, minimizing the distress or process of bereavement, forced cheerfulness,and presuming to know the patient's feelings, are all seen as particularly unhelpful(Lehman et al., 1986). These findings are supported by another study that foundthat the bereaved receive considerable negative expressed attitudes and comments,and to a lesser extent unhelpful instrumental behaviors, from their socialrelationships (Malkinson, 1987). Finally, chronically ill patients (those withMultiple Sclerosis) report that minimization or maximization of the seriousness ornature of the disease are unhelpful (Lehman & Hemphill, 1991).Spousal Unhelpfulness3While these studies go beyond the study of negative interactions in generaland provide useful descriptive information on specific behaviors perceived asunhelpful, to date they have failed to address the possible effects of such behaviors.Research on negative behaviors within marriage has demonstrated that fordistressed couples, marital satisfaction is determined, in part, by the occurrence ofdispleasing or negative behavior. For non-distressed couples, satisfaction is relatedto pleasing or positive behavior (Jacobson, Follette, & Waggoner-McDonald, 1982;Jacobson, Waldron, & Moore, 1980). Indeed, the frequency and type of negativebehaviors (including avoidance, verbal confrontation or verbal nonsupport) can beused to discriminate distressed and non-distressed marriages (Fincham & Bradbury,1990). One other type of spousal behavior implicated in the literature isdispleasurable instrumental behaviors surrounding areas of daily living (such ashomemaking, recreation, or finances) that have been linked with lower maritalsatisfaction (Barnett & Nietzel, 1979).There are several types of spousal behaviors commonly associated withdepression that may also be important in determining marital satisfaction,particularly in light of the demonstrated connections between marital satisfactionand depression (Beach, Arias, & O'Leary, 1986; Gotlib & Whiffen, 1989). Ingeneral, the interactions of depressed persons and their spouses have beenconsistently shown to include more negativity (Hautzinger, Linden, & Hoffman,1982; Schuster et al., 1990) and hostility (Kowalik & Gotlib, 1987). Specifically,spousal criticism has been associated with depressive relapse (Hooley & Teasdale,1989) and been linked with specific coping strategies that relate to loweredpsychological well-being in rheumatoid arthritis patients (Manne & Zautra, 1989).Spousal Unhelpfulness4While the studies described above have succeeded in demonstratingconnections among spousal behavior and outcomes, most have been conductedwithin a controlled laboratory setting where ecological validity is uncertain. It isimportant to test these relations in more naturalistic settings. However, without theavailability of an objective observer, it may be difficult to determine whether on nota behavior actually occurred. Further, research has demonstrated considerablediscrepancies in reports of behavior occurring in the context of interpersonalrelationships. For example, Antonucci and Israel (1986) found that across allrelationships, the congruence between recipients' reports of received support andproviders' reports of support given in specific areas was 49-54%. This finding issupported in other studies focusing on the marital relationship that have found, evenfor discrete behaviors, there may be little agreement in terms of whether or not theaction occurred (Booth & Welch, 1978; Fincham & Bradbury, 1990; Jacobson &Moore, 1981).It is not clear whether distressed and non-distressed couples are equallylikely to disagree on the occurrence of negative and positive events. Booth andWelch (1978) found no difference based on the level of distress; yet another studyfound that happily married couples were more accurate in their assessments ofcommunication (Yelsma, 1984).Although these discrepancies in perceptions of behaviors may preclude anobjective assessment of spousal interactions, recent research suggests thatperceptions of support may be more important than actual support received (Kessler& McLeod, 1985; Wethington & Kessler, 1986). For example, working mothersof infants who felt their husbands were supportive were better able to cope withstress even though they continued to do most of the childcare and homemakingSpousal Unhelpfulness5(Gray et al., 1990). Although the majority of the research on negative interactionsutilizes an a priori assumption that particular behaviors will be perceived asnegative, perceived unsupportiveness is well operationalized in studies of supportattempts that fail. Respondents are typically asked to specify any actions from theirsocial network that they felt were unhelpful (Dakof & Taylor, 1990; Lehman &Hemphill, 1991). However, connections between such perceptions ofunsupportiveness and outcome are lacking.One possible link between perceived behavior and outcome is suggested bythe literature on cognitive attributions generated by interpersonal behaviors.Fincham and Bradbury (1990), among others, suggest that it is the cognitiveprocesses stimulated by spousal behavior that define the meaning of the behaviorand thus determine its effect. It has been proposed that spousal action iscognitively assessed along any of several dimensions (e.g. blameworthiness,responsibility, controllability, stability) and that the outcome of that assessment willdetermine how the receiver of the action will respond to it (Bradbury & Fincham,1990). In this way, the same behavior may have different meanings for differentindividuals.The majority of attributional research within the marital literature hasfocused on spousal behavior defined broadly as negative or positive, by assessingseveral aspects of attributions. This typically involves an assessment of thefrequency of attributions, the type of attribution, or both frequency and type. Inthe realm of attributional type, the distinction between causal attributions (is thecause internal or external, stable or unstable, global or specific?) and responsibilityattributions (was it intentional, is the person blameworthy, was it selfishlymotivated?) is common in the literature (Fincham & Bradbury, 1988).Spousal Unhelpfulness6Assessing these different components of attributions, research hasdemonstrated that negative spousal behaviors (those judged as negatively impactingthe recipient) generally produce more attributions (Holtzworth-Monroe & Jacobson,1985; Holtzworth-Monroe & Jacobson, 1988) and specifically, more distress-maintaining attributions than do positive behaviors (Camper, Jacobson, Holtzworth-Monroe, & Schmaling, 1988). This process may vary according to the quality ofthe marital relationship however, as significant attributional differences have beenfound between distressed and nondistressed couples (c.f. Bradbury & Fincham,1990). Men low in marital satisfaction tend to make a greater number ofattributions for their spouse's behavior, and distressed couples in general tend tomake specific attributions which function to maintain the distress (Holtzworth-Monroe & Jacobson, 1985). Further, distressed couples are more likely to viewnegative behavior on the part of their spouse as global and stable (Camper et al.,1988) and internally caused (Jacobson, McDonald, Follette, & Berley, 1985) and toattribute less benign motives to their partners than to themselves (Fincham et al.,1987). Distressed spouses also attribute more blame, negative intent, and selfishmotivation to their partners for negative behavior (Fincham et al., 1987). In areview of the literature, Berley and Jacobson (1984) conclude that dispositionalattributions are positively related to marital distress.Despite the demonstrated relations between negative attributions and maritaldistress, the issue of the direction of causality remains unresolved. Althoughmanipulating attributions for spousal behavior has been demonstrated to altersubsequent behavior and cognitive activity (Fincham & Bradbury, 1988), evidencefor causal links between negative attributions and long term marital satisfaction issparse. There are, however, two important exceptions. One longitudinal studySpousal Unhelpfulness7measuring relationship quality suggests that attributions of responsibility forrelationship maintenance are related to long-term happiness in the relationship(Fletcher, Fincham, Cramer, & Heron, 1987). Further, Fincham and Bradbury(1987) found that causal and responsibility attributions made by wives were relatedto their own marital satisfaction one year later, independent of earlier ratings ofsatisfaction.The research cited thus far has provided important insights into both theimpact of negative spousal behaviors on marital distress and the relation betweenattributions and marital distress. Traditionally, these studies have employedmethodologies that either present hypothetical partner behaviors (Fincham et al.,1989; Holtzworth-Monroe & Jacobson, 1985), stimulate partner behaviors in a labsetting (Camper et al., 1988), or sample a wide range of daily events andinteractions through checklists and ratings (Barnett & Nietzel, 1979; Jacobson etal., 1982). However, these approaches have certain limitations. Perhaps the mostimportant is a lack of attention to the context of the behaviors. It seems unlikelythat any particular behavior has the same impact each time it occurs. Rather, it ismore likely that a particular behavior may be interpreted quite differently indissimilar situational contexts. Consequently, the impact of the behavior will varyas well.Just as the stress and coping literature suggests that coping should beconsidered in the context of a single specific event which is perceived as stressfulby the individual (Compas, Forsythe & Wagner, 1988; Lazarus & Folkman, 1984),it is perhaps advisable to consider the impact of negative behaviors in the context ofa specific stressful event or situation. With the recognition that spousal support intimes of stress is very important to adaptation, unhelpful behaviors that occur in theSpousal Unhelpfulness8context of a particular problem that has been identified as being personally salientmay have considerable impact on outcomes such as marital satisfaction.In summary, the research reviewed above suggests that the perceptions ofunsupportiveness or unhelpful actions performed by the spouse may have asignificant impact on outcomes such as marital satisfaction. Further, there is ademonstrated influence of attributions for spousal behaviors on marital satisfaction.Given the importance of these components, the following basic research questionswere posed: 1) What behaviors on the part of the spouse are viewed as unhelpfulby the recipient? 2) To what extent are spousal behaviors that are perceived to beunhelpful associated with marital satisfaction for married couples? and 3) Do thetypes of attributions made regarding the perceived unhelpful behavior furtherexplain the role that unhelpful spousal behaviors play in determining maritalsatisfaction? In addition, a general purpose of the current study was to identify andexplore the types of problems faced by remarried couples in stepfamilies.This study examined these issues using remarried couples who have at leastone child from a previous union living in the home. This group was chosen forstudy for a number of reasons. The first reason is the increased rate of divorce inWestern society in the last few decades (Visher & Visher, 1985). Although fewerwomen are remarrying after divorce than before (Norton & Moorman, 1987), thereare a large number of families facing integration into new stepfamily units(Anderson & White, 1986). Being part of a stepfamily also generates difficultieswithin the marital relationship. Differences between partners regardingexpectations for the marriage or children (Visher & Visher, 1985), conflict betweenspouses around relationships with the ex-partner (Messinger, 1976) and lowSpousal Unhelpfulness9consensus on attitudes regarding childrearing and stepparenting roles (Keshet,1990), are common sources of conflict.Based on the above, the following hypotheses were developed. 1) First, theextent to which the spouse performs actions perceived to be unhelpful would berelated to marital satisfaction. The more unhelpful the spouse was perceived to be,the lower marital satisfaction was expected to be. 2a) Those spousal behaviors thatwere identified by the respondent as being unhelpful in dealing with a stressfulfamily situation were expected to be associated with lower marital satisfaction.This relation between unhelpful spousal behaviors and marital satisfaction wasexpected to remain significant even after controlling for demographic variables(i.e., family structure, income, and education) and dimensions of the stressfulfamily problem (the respondent's perception of its seriousness and the type ofproblem). b) Further, based on findings in the clinical literature on maritalinteraction, it was expected that those respondents who report that their spousesengaged in criticizing or blaming the respondent for the problem would havesignificantly lower marital satisfaction than other respondents. 3) Certain types ofattributions were expected to have a significant impact on marital satisfaction. a)Specifically, those respondents making characterological attributions for spousalbehaviors were expected to have lower marital satisfaction than other respondents.b) Similarly, the extent of blame in the attribution for spousal behavior wasexpected to be related to marital satisfaction such that higher levels of blame wouldbe associated with lower marital satisfaction. 4) Finally, the impact of spousalbehavior on the respondent's marital satisfaction was expected to vary as a functionof the respondent's attribution for that behavior. a) Specifically, it washypothesized that there would be a synergistic relation between characterologicalSpousal Unhelpfulness10attributions and negative spousal behavior. b) Similarly, the extent of blame in theattribution was expected to interact with the type of spousal behavior such that theimpact of negative spousal behaviors on marital satisfaction would be greater forthose respondents expressing high levels of blame towards their spouse.METHODAs part of a larger study, 103 remarried couples completed a telephoneinterview concerning relationships among family members, issues and problemsfacing them as part of a stepfamily, and basic social support and demographicinformation. Of concern to this study, respondents were asked about the extent towhich their spouse is unhelpful, specific ways in which their spouses are unhelpful,their attributions for spousal behaviors, and their marital satisfaction. Interviewswere conducted from the University of British Columbia and were completed overa period of approximately 18 months from January 1991 through July 1992.SubjectsRespondents were approached through newspaper and radio announcements,and through posters placed in community centers in southern British Columbia.The participant pool was limited to those families having at least one child from aprevious relationship (of either spouse) living in the home for more than threemonths of the year. Further, only respondents who were married or livingcommon-law were included (respondents in common-law relationships arehenceforth referred to here as married). Finally, because of difficulties inadvertising and interviewing in more than one language, participants were limitedto those who were fluent in English. This restriction also served to reducevariability among respondents due to culture. Both partners were requested toSpousal Unhelpfulness11participate. Only couples in which both the husband and the wife completed theinterview were included in the present set of analyses.Interested couples were requested to telephone the Stress and Coping Projectoffice at the University of British Columbia. Of the 189 couples that contacted theproject, 164 met eligibility requirements. Those couples eligible for the study weresent a packet of information including a detailed description of the study. Of these,109 couples agreed to begin the study. In six of these couples, the husband did notcomplete the interview. This resulted in a final sample of 103 couples, 66% ofthose who originally expressed interest in the study. This is somewhat lower thanother studies on marriage that have reported between 74 and 85 percent (Pickersgill& Beasley, 1990; Gaelick, Bodenhausen, & Wyer, 1985) and lower than the 81 to85 percent reported for studies on support attempts (Dakof & Taylor, 1990;Lehman & Hemphill, 1991). It is unclear if this reflects the methodologyemployed in the current study (telephone interviews rather than face-to-faceinterviews, or the greater time demands of this study) or whether this reflectssomething unique about stepfamilies.The mean age for wives in the sample was 38.70 (SD = 5.68) years and forhusbands was 41.67 years (SD = 6.84). Wives were significantly younger thantheir husbands statistically (mean difference = 2.97, SD_ = 6.00; paired 1(102) =4.95, p < .001), although the difference in age between spouses ranged fromhusbands 17 years older than their wives to wives 12 years older than theirhusbands.Seventy-three percent of the wives and 90% of the husbands were employedoutside of the home at the time of the interview. Of the wives, 13.6% weremarried for the first time whereas for 86.4% of the wives, the current marriageSpousal Unhelpfulness12was a second or third marriage (mean number of previous marriages = .98, 5_12 =.51). For 24.3% of the husbands this was a first marriage and for 75.7%, a secondor third (mean number of previous marriages = .87, 512 = .65). The women inthe sample had a mean of 1.37 children from previous relationships while for menthis figure was 1.58. The mean length of time together at the time of the study was4.82 years (511 = 3.69). (See Table 1.)ProcedureInterested couples contacted the Stepfamilies Project by telephone and weremailed a brief summary of the project components and aims, a consent form, andan information sheet on which to record the numbers of children and previousmarriages or unions. Those couples willing to participate in the study returned theinformation and consent forms. Due to time constraints on data collection, duringthe last two months this information was solicited from interested couples over thephone. In these cases, participants did not begin the study until they had receivedthe information sheet and consent form; the respondent verbally consented toparticipate (the consent form was read to them over the phone) and they wererequested to return the signed consent form to the project promptly. 1A telephone interview was completed by trained undergraduates associatedwith the project. Each spouse was assigned to a different interviewer and eachinterviewer was blind to any information received from the other spouse. Allinterviewers were female. Training proceeded according to protocols andconventions outlined in the Institute of Social Research interviewer training manual(Guenzel, Berckmans, & Kannell, 1983). The mean length of the interview was 54minutes. Permission to tape-record the section containing the open-ended1 This change was approved by the University Research Ethics Department.Spousal Unhelpfulness13Table 1Sample DescriptionHUSBANDS^WIVESMean SD Mean SD toAge 41.67 6.84 38.70 5.68 4.95***Number of Previous Marriages .87 .65 .98 .51 -1.35Number of Children fromPrevious Marriages 1.58 1.29 1.37 .97 1.24a Paired t-test***p < .001Spousal Unhelpfulness14questions was obtained from each subject to allow for verbatim transcription of theinterview (none refused). These tapes were also used to ensure that interviewersfollowed standardized protocol.Measures The interview consisted of a number of measures to assess the relationshipswithin the family and with previous spouses, common difficulties that may arisewithin stepfamilies, and demographics. Those portions of the interview used in thisstudy are discussed below. (See Appendix A for Interview items).Demographics. Two demographic variables were included as necessarycontrol variables in this study; socioeconomic status (SES) and family structure.Socioeconomic status was operationalized as the estimated total family incomeprovided by the respondent as well as the years of formal education. Familystructure refers to the nature of the current family unit and was coded into threegroups of respondents for the purposes of the regression analyses; those with onlychildren from the wife's previous relationships living in the home, those with onlychildren from the husband's previous relationships living in their home, and thosewith children both from previous marriages of the husband and the wife living inthe home. 2Stressors. The respondent was asked to describe the one issue or problemthey saw as currently most serious in their family via the following: "I'd like youto take a few moments to consider what you believe the most serious family2 This coding scheme was chosen for entry into the regression analyses over enteringpresence of spouse's children and presence of own children [as suggested by Hobart,(1990)] to provide a clearer measure of structure and to simplify the analyses. Further,coding family structure as suggested by Hobart did not produce different relations withmarital satisfaction.Spousal Unhelpfulness15problem is for you. What would that be?" Following the interview, the responseswere content-coded by two trained coders associated with the project. Codes werethen collapsed into 4 broad categories; a) Issues focused on the marital unit such asa lack of affection between partners, disagreements on raising the children, or otherforms of marital conflict--issues where the primary difficulty is expressed as aproblem that the couple is facing. b) Family difficulties other than those betweenspouses, such as the difficult behavior of a child or stepchild, or the relationshipbetween a child and another member of the family--the problem is labeled asresiding outside of the marital unit and within or between other family members.c) Broad stepfamily issues such as integration of the family into a single unit, ordifficulties with ex-partners over such things as custody or visitation issues--problems generally unique to stepfamilies. And finally, d) Miscellaneous problemsor issues such as difficulties in the allocation of tangible resources such as moneyor time, and other types of problems not classified elsewhere--primarily problemsthat impact the family as a whole but that are not unique to stepfamilies. 3 (SeeAppendix B for codes to this and subsequent open-ended questions). Inter-raterreliability was calculated at .90 using Cohen's kappa indicating high interrateragreement. These codes were entered into the regression analyses as controlvariables, to address the possibility that the type of stressor represents a thirdvariable in the relation between spousal unhelpfulness and marital satisfaction.Seriousness of the stressor. Following the assessment of the type ofproblem, a Liked scale was used to assess the perceived seriousness of the problemby asking "From your perspective, how serious would you say this problem is?3 This variable was dummy coded for entry into the regression analyses.Spousal Unhelpfulness16One means not at all serious and 5 means very serious." This rating was used as acontrol variable in the regression analyses in order to control for the possibility thatthe perceived seriousness of the problem might enter into the relation betweenspousal unhelpfulness and marital satisfaction as a third variable.Spousal Behaviors. A Likert scale was used to assess the extent to which therespondent perceived their spouse as engaging in unhelpful behaviors in regards tothe specified problem ("To what extent does your husband/wife say or do thingsthat are unhelpful, whether or not he/she intends to be unhelpful?"). Respondentswere presented with a rating scale of spouse is not at all unhelpful (1) to spouse isunhelpful to a large extent (5).Continuing to focus on the problem they described as most serious,respondents were then asked about the unhelpful behaviors their spouse engages invia the open-ended question "What does your husband/wife say or do that isunhelpful, whether or not they intend to be unhelpful?" Respondents were probedfor clarity, and were asked to mention all of the things their spouse did that theyperceived to be unhelpful in regards to this problem. Eighty-nine percent of wivesmentioned at least one unhelpful behavior; 16% mentioned two or more behaviors.The figures for husbands were 76% and 13 % respectively.Following the same procedure as mentioned above, the responses werecontent coded and then collapsed into the following broad categories: a) criticism orblame expressed by the spouse towards the respondent, b) failure to support thespouse in decision making or in efforts to cope or deal with the physical demandsof the situation such as family maintenance or childrearing, or a failure to talkabout issues or feelings, c) other unhelpful behaviors directed toward the spouseSpousal Unhelpfulness17such as being unrealistic, blowing the problem out of proportion, refusing toacknowledge the problem or any responsibility for it, or getting angry about theproblem, and d) unhelpful behaviors directed toward others such as beingoverprotective or overindulgent with the children, being critical of a child, ordealing with ex-partners inappropriately. 4 Inter-rater reliability was high (Cohen'skappa= .87). Three husbands and one wife were dropped from the analysesregarding the specific behaviors as they were unable to specify what their spousedid that was unhelpful even though the spouse was unhelpful to some extent.Attributions. After describing the types of unhelpful behaviors their spouseengaged in, the respondent was asked "Why do you think he/she might say or dothat?" Although this question does not allow for a fine-grained analysis of allattributional dimensions, it has been used to elicit explanations for interpersonalbehavior in other studies (e.g. Camper et al., 1988). Further, necessary limitationson the length of the interview precluded a lengthier measure of attributions.As with the behavior question, respondents were probed for as many reasonsfor their spouse's behavior as they felt should be mentioned. Twenty-four percentof wives and 18 % of husbands mentioned more than one attribution while 89% ofwives and 76% of husbands provided at least one attribution.The attribution categories used in this analysis were as follows: a)Characterological causes such as personality, a lack of understanding regarding theneed for help, or unrealistic expectations b) Situational causes such as the influenceof other people in the situation or the demands implicit in being part of a4 In order to deal with multiple mentions, the behaviors were coded into dichotomiesrepresenting the presence (1) or absence (0) of the particular behavior. For the purposesof the regression analyses, each behavior was treated as an independent predictor.Spousal Unhelpfulness18stepfamily, and c) Benign/Positive causes such as good intentions, demands fortime or energy from sources extraneous to the problem, or past history. 5 The firsttwo categories follow the attribution literature in their distinction between externaland internal or situational and dispositional causes of behavior (Fincham, Beach, &Nelson, 1987; Kyle & Falbo, 1985). However, studies on support attempts thatfail have also identified attributions of a generally benign nature such as attributingthe behavior to good intentions or to something about the individual that is 'nottheir fault' such as their upbringing or experience. As these attributions do notclearly fit into the categories described above and given that they may in factrepresent a unique factor, it was decided to include the third category. Cohen'skappa based on the content coding was .79 indicating good inter-rater agreement.One woman was dropped from the analyses because she was not able to identifyany reason for her spouse's behavior.Finally, the attribution question was coded for the extent of blame in theattributions made by the respondent in order to assess the attributional componentof responsibility. This rating was on a scale from 9 being very negative/blaming to1 being very positive/no blame. An example of an attribution rated 1 is "She onlywants the best for us and she sees this as the best way to make things happen.""He won't give me anything of my own because he doesn't want to give up any ofhis power over me" would be rated as very blaming and would be assigned a value5 As with the behaviors, attributions were coded into dichotomies representing thepresence (1) or absence (0) of the particular attributions, and entered into the regressionanalyses as independent predictors.Spousal Unhelpfulness19of 9. The Pearson correlation between the two coders was .77 for men and .69 forwomen. The average of the two values assigned by the coders was used in theanalyses.Dyadic Adjustment Scale. The Spanier Dyadic Adjustment Scale (DAS;Spanier, 1976) was included in the interview as a measure of marital satisfaction.Slight modifications were necessary to make it more appropriate for administrationduring a telephone interview. The modifications included minor rewording ofseveral questions, and a change in the scale options from a six option scale ofalways agree to always disagree in the original to a five point scale from neverdisagree to always disagree in the current version. This range (from 1 to 5) wasused for all items. Finally, two items from the original scale were dropped. Theitem "How often do you disagree about matters of recreation" was seen as toosimilar to "How often do you disagree about leisure time interests and activities"and was deleted to avoid confusion. The question "Do you and your mate engagein outside interests together" (with a scale from all of them to none of them) wasdropped to minimize the number of different response scales presented to therespondent and because respondents found it confusing. The final score on theDAS was represented by the mean of all items.The reliability coefficient for the full scale has been reported to be .96(Corcoran & Fischer, 1987). In the current study, the alpha for both women andmen was .91. This scale has been used in numerous studies of marital interactionto discriminate between distressed and non-distressed couples (Camper et al., 1988;Long & Andrews, 1990). The DAS functions to assess both event occurrences andsubjective quality of the relationship which have been proposed asSpousal Unhelpfulness20necessary to assess the complex nature of marital satisfaction (Fincham &Bradbury, 1990). The DAS has also been shown to correlate with other measuresof marital adjustment (Spanier, 1976).Coding of Open-ended ItemsThe development of the coding categories for the open-ended questionsproceeded as follows. A random sample of responses was reviewed and an initialset of codes was devised. Next, a sample of responses were coded byundergraduate research assistants associated with the project using this preliminarycoding scheme. Codes were adjusted to increase the clarity of each category and toinclude additional categories as necessary. Upon completion of the final set ofcodes, two undergraduate coders were trained to code each question. These coderswere all women between the ages of 22 and 45 years. After reaching an agreementof 80%, coders proceeded independently although the majority of responses werecoded by both. Agreement between coders was assessed on a regular basis anddiscrepancies were discussed in order to maintain an agreement between coders ofat least 80%. In total, 68 % of responses on the stressor question, 79% on thespousal behavior question, and 81 % on the question regarding attributions werecoded by both. Disagreements between the coders were resolved by the principalinvestigator.RESULTSThe analyses are presented in three sections. First, univariate statistics arereported for marital satisfaction, demographic variables, types of stressors, extentof spousal unhelpfulness, and types of behaviors and attributions. In the secondSpousal Unhelpfulness21section, bivariate relations between variables are presented and finally, regressionanalyses using marital satisfaction as a dependent variable are presented in the thirdsection. The data for husbands and wives were analyzed separately because,having chosen specifically couples where both spouses would participate, thegroups cannot be considered independent.Univariate Results Marital satisfaction and demographics. Full scale DAS scores were used asthe measure of marital satisfaction as suggested by Spanier (1976). However,because Fincham and Bradbury (1990) suggest that using full scale scores cannotcompletely rule out the possibility of overlap between measures of dyadicadjustment and those of attribution, the analyses including attributions wererepeated using the single DAS item measuring overall happiness in the marriage.This second set of regressions was done in those cases where a significant effectwas found for attributions in the analyses with the full scale DAS in order toexamine the possibility that the effect was only a reflection of shared contentbetween the attribution measure and DAS scores.On the full DAS scale, the wives obtained a mean score of 4.06 ( ^= .52)whereas for husbands this score was 4.07 (512 = .51; paired 1 (102) = -.17, p >.10). The mean on the DAS item measuring overall happiness with the relationshipwas 4.22 (SD = .85) for husbands and 4.22 (SD_ = .80) for wives. Happiness wascorrelated .72 (p < .001) with the overall scale score for wives and .74 (p < .001)for husbands.Wives reported a mean of 14.12 years of formal education (02 = 2.09) whiletheir husbands reported a mean of 14.16 years (SD_ = 2.60; paired 1(102) = -1.00,p > .10). The mean family income reported by wives was $75,310 (SD =Spousal Unhelpfulness22$41,385) while the mean for husbands was $79,970 (112 = $52,770; paired 1(102)= -1.97, p = .06). It is impossible to determine if these differences in reportedincome are the result of some respondents inflating or minimizing estimates ofincome, perhaps due to social desirability, or due to spouses reporting differentcomponents of family income in response to the question.Twenty-six percent of couples had only children from the wife's previousrelationships living in the home whereas 20% had only children from the husband'sprevious relationships living in the home. Fifty-four percent of couples hadchildren from previous relationships of both the husband and the wife living in thehome.Most serious family stressor. The problems most commonly mentioned bywives as the most serious were problems between or within individuals in thefamily other than the marital couple (coded in the category "family problems - notmarital;" 42%). (See Table 2.) Husbands most often stated that problems facingthe couple were the most serious (29%) although family problems (28 %) and broadstep-family issues (28%) were almost as frequently mentioned. The miscellaneouscategory consisted primarily of difficulties in managing tangible resources such astime or money; this was the least frequently mentioned category for both husbandsand wives. Comparisons between husbands and wives revealed that wives (n = 43)were significantly more likely than husbands (n = 29) to indicate that the mostdifficult problem was a family problem (/2 = 4.18, p < .05).Seriousness of the stressor. The mean rating of the severity of the problemor issue was 4.04 (SD = .91) for wives and 3.83 (S_D_ = 1.04) for husbands.Husbands rated the stressor as significantly less serious than did their wives (paired1 (102) = 1.99, p < .05).Spousal Unhelpfulness23Table 2Spousal Differences in Reports of Stressors. Spousal Behaviors, and Attributionsfor Spousal BehaviorsCategories of Most Serious StressorHUSBANDS WIVESFreq % Freq % Chi-sqaMarital^ 30 29.1 24 23.3 .90Family (not marital) 29 28.2 43 41.7 4.18*Broad Stepfamily Issues 29 28.2 21 20.4 1.69Other Issues 15 14.6 15 14.6 .00Types of Unhelpful Spousal BehaviorCriticism 22 21.4 24 23.3 .10Failure to Support 11 10.7 26 25.2 4.97*Other Behaviors 26 25.2 21 20.4 2.33Directed at Others 32 31.1 40 38.8 .10None 25 24.3 11 10.7 7.74**Attributions made for Unhelpful Spousal BehaviorSituation^ 28^27.2 29 28.2 .31Spouse's Character 32 31.1 49 47.6 2.77Benign/Positive 37 35.9 41 39.8 .08Note: Percentages represent the percent of all respondents reporting that category;for types of behaviors and attributions these add to more than 100% due to multiplementions.a With the exception of the category of "no unhelpful behaviors," Chi squareanalyses for Behaviors and Attributions include only couples where both spousesmentioned at least one unhelpful behavior (therefore, n = 70 pairs).*p < .05, **p < .01Spousal Unhelpfulness24Spousal behaviors. The mean rating of the extent of spousal unhelpfulnesswas 2.98 (02 = 1.13) for wives and 2.57 (SD = 1.16) for husbands. Wives ratedtheir spouses to be significantly more unhelpful than their husbands reported(paired 1 (102) = 2.73, p < .01).Twenty-five husbands (24.3%) stated that their spouse does nothingunhelpful while only 11 wives (9.6%) said the same about their husbands, 9C 2 =7.74, p < .01). For those respondents mentioning at least one unhelpful behavior,the spousal behaviors most frequently mentioned by both wives and husbands wereactions directed towards other individuals (39% and 31% respectively; ns). Inaddition, 26 wives and 11 husbands (25 % and 11 % of the total sample respectively;ns) mentioned that their spouses failed to support them and 24 wives and 22husbands (23 % and 21 % respectively; ns) stated that their spouses were critical ofthem. Wives were significantly more likely to indicate that their husbands failed tosupport them than were husbands (?( 2 = 4.97, p < .05).Attributions. The most frequently mentioned type of attribution for bothhusbands was benign or positive attributions (40%); 36% of wives made this typeof attribution. There were no significant differences between the frequencies withwhich husbands and wives made specific attributions (see Table 2). The meanextent of blame in the attribution for wives was 4.67 (SD = 2.3) and was 5.65(02= 2.9) for husbands. This difference was not significant (paired I (77) = 1.76, p< .09).Bivariate Relations Marital satisfaction and study variables. The correlations between DAS scoresand the study variables of interest are provided in Table 3. With one exception, thepattern of relations of the unhelpful and attributional variables to DAS scores wereSpousal Unhelpfulness25Table 3Correlations between Full Scale Scores of Marital Satisfaction and Study VariablesHUSBANDS WIVESExtent of Unhelpfulness -.43*** -.32**Specific Unhelpful BehaviorsCriticism -.12 -.18Fails to Support Spouse -.26* -.10Other Behaviors Directed at Spouse -.09 .08Behaviors Directed at Others .18 .02Type of AttributionSituation .06 -.06Spouse's Character -.09 -.18Benign/Positive .13 .05Extent of Blame in Attribution -.26* -.28**Note: Coefficients for specific behaviors and attributions include only thoserespondents that reported at least one codeable unhelpful spousal behavior andattribution (n = 78 husbands, n = 90 wives).*p<.05,**p<.01,***p<.001Spousal Unhelpfulness26similar for husbands and wives. The greater the perceived extent of spousalunhelpfulness, the lower the scores on the DAS (wives r = -.32, p < .01;husbands r = -.43, p < .001), and the more positive the tone (the less blaming) ofthe attribution, the higher the DAS score (wives r = -.28, p < .01; husbands r = -.26, p < .02). For husbands, reports that their wife failed to support them inregards to the problem was associated with lower marital satisfaction (r = -.26, p< .02). There were no significant associations between other types of behaviorand DAS scores for husbands. None of the unhelpful behaviors were significantlyassociated with the DAS for wives. Further, there were also no significantrelations between the type of attribution and DAS scores for husbands or wives.Demographic variables and study variables. Correlations between thedemographic variables and the study variables are reported in Tables 4 (Husbands)and 5 (Wives). There were few significant relations among any of these variablesfor either husbands or wives (4 of 36 relations for husbands and 5 of 36 for wives),making those that did reach significance difficult to interpret. For husbands,having only children from their spouse's previous relationships living in the homewas associated with reporting that their spouse failed to support them (r = .25, p< .05) and making characterological attributions was related to higher familyincome (r = .23, p < .05). The extent of blame in the attribution wassignificantly correlated with family structure such that having only children fromthe spouse's previous relationships living in the home was associated with moreblame (1. = .41, p < .001) and having only children from their own previousrelationships was associated with lower levels of blame (r =-.24, p < .05). Forwives, family structure was related to the type of attribution made; having onlychildren from the spouse's previous relationships living in the home was associatedSpousal Unhelpfulness27Table 4Correlations between Demographic Variables and Study Variables - HUSBANDS1 2 3 4 5Extent of Unhelpfulness -.03 -.07 .16 -.05 -.09Type of Unhelpful BehaviorCriticism .08 -.10 .00 .13 -.10Failure to Support -.03 -.03 .25* -.10 -.13Other Types of Behaviors .04 .06 -.06 .07 -.02Directed at Others -.11 .02 -.15 -.08 .14Type of AttributionSituation -.20 -.18 -.09 -.03 .10Spouse's Character .07 .23* .08 -.01 -.06Benign/Positive .18 -.05 -.16 .20 -.02Extent of Blame -.08 -.16 .41 ***-.24* .18Marital Satisfaction (DAS score) -.07 .00 -.07 .04 .041 = Education, 2 = Income, 3 = Only spouse's children in the home, 4 = OnlyRespondent's children in the home, 5 = Children from previous marriages of bothpartners.Note: Coefficients for specific behaviors and attributions include only thoserespondents that reported at least one unhelpful spousal behavior (n = 78).* p < .05, **p < .01, ***p < .001Spousal Unhelpfulness28Table 5Correlations between Demographic Variables and Study Variables - WIVES1 2 3 4 5Extent of Unhelpfulness -.19 -.01 -.12 .09 .03Type of Unhelpful BehaviorCriticism .28** .17 .14 .11 -.02Fails to Support -.09 -.06 -.13 .08 .02Other Actions -.02 .02 .06 -.19 .11Directed at Others .26* .16 -.05 .05 -.01Type of AttributionSituation -.01 -.15 .14 -.23* .08Spouse's Character .06 .13 -.25* .14 .07Benign/Positive -.09 .00 .06 -.05 -.02Extent of Blame -.01 -.03 -.13 .12 .00Marital Satisfaction (DAS score) .23** -.04 -.01 -.01 .011 = Education, 2 = Income, 3 = Only spouse's children in the home, 4 = Onlyrespondent's children in the home, 5 = Children from previous marriages of bothpartners.Note: Coefficients for specific behaviors and attributions include only thoserespondents that reported at least one codeable unhelpful spousal behavior andattribution (n = 90).*p < .05, **p < .01Spousal Unhelpfulness29with a lower frequency of making characterological attributions (r = -.25, p <.05). Further, those wives having only children from their own previousrelationships were less likely to make situational attributions (r = -.23, p < .05).Family income was positively related to wives' perceptions of spousal criticism (r= .28, p < .01) and to reports that their husbands engaged in behaviors directed atindividuals other than the spouse (r = .26, p < .05). Finally, higher maritalsatisfaction for wives was associated with higher levels of education (r = .23, p <.01). That the pattern of relations differed for husbands and wives makesinterpretation of these findings more difficult. In general, it appears that thesedemographic factors have little influence, if any, on the variables of interest in thisstudy. However, in order to err on the conservative side, these variables wereretained as controls for the regression equations, as had been originally proposed.Stressor and study variables. The type of stressor and the seriousness of thestressor were correlated with the study variables. There were few significantrelations between the stressor described (neither the type of stressor reported northe respondent's rating of its seriousness) and the variables of interest in the presentstudy [see Table 6 (Husbands) and Table 7 (Wives)]. For both husbands andwives, those reporting that their most serious stressor was a family problem wereless likely to report that their spouse was critical (husbands r = -.22, p < .05;wives r = -.26, p < .01). For wives, indications that the most serious problemwas in the realm of family problems outside of the marital relationship were morelikely to be associated with unhelpful spousal actions aimed at individuals otherthan the spouse (r = .36, p < .001). This type of behavior was less likely tooccur when the problem was a broad stepfamily issue (r = -.30, p < .001) andwhen the problem was primarily one of allocation of tangible resources (r = -.31,Spousal Unhelpfulness30Table 6Correlations between Type and Seriousness of Stressor and Study Variables -HUSBANDS 1 2 3 4 5Extent of Unhelpfulness .19 .17 .02 -.13 -.06Type of Unhelpful BehaviorCriticism .14 .14 -.22* -.03 .13Fails to Support .09 -.11 -.06 .06 .13Other Actions .00 .12 -.15 -.04 .08Directed at Others -.14 -.10 .19 .09 -.21Type of AttributionSituation -.08 -.04 .14 -.02 -.10Spouse's Character .03 .07 -.05 -.14 .15Benign/Positive -.03 .00 -.05 .13 -.11Extent of Blame .16 .05 .01 -.02 -.06Marital Satisfaction (DAS score) -.32** -.34 .02 .03 .121 = Seriousness of the Stressor, 2 = Marital Issues, 3 = Family Problems - notmarital, 4 = Broad Stepfamily Issues, 5 = Miscellaneous Issues.Note: Coefficients for specific behaviors and attributions include only thoserespondents that reported at least one codeable unhelpful spousal behavior (_n =78).*p < .05, **p < .01Spousal Unhelpfulness31Table 7Correlations between Type and Seriousness of Stressor and Study Variables -WIVES 1 2 3 4 5Extent of Unhelpfulness .13 .14 -.19 .03 .05Type of Unhelpful BehaviorCriticism .10 .02 -.26** .16 .16Fails to Support .06 -.07 .05 .01 .00Other Actions -.01 -.06 -.20* .14 .20*Directed at Others -.04 .13 .36*** -.30** -.31**Type of AttributionSituation .17 -.05 .03 -.14 .17Spouse's Character -.03 .01 .01 .05 -.09Benign/Positive .11 -.03 .09 .09 -.19Extent of Blame -.20 .09 .02 -.20 .08Marital Satisfaction (DAS score) -.34*** -.04 .10 -.04 -.031 = Seriousness of the Stressor, 2 = Marital Issues, 3 = Family Problems - notmarital, 4 = Broad Stepfamily Issues, 5 = Miscellaneous Issues.Note: Coefficients for specific behaviors and attributions include only thoserespondents that reported at least one codeable unhelpful spousal behavior and oneattribution (n=90).*p < .05, **p < .01, ***p < .001Spousal Unhelpfulness32p < .01). For both husbands and wives, the more serious they perceived theirproblem to be, the less satisfied they were in their marriage (husbands r = -.32, p< .01; wives r = -.34, 12 < .001).Correlations among study variables. The relations among the independentvariables were also assessed using correlations in order to examine for potentialmulticollinearity problems effecting the regression analyses. The type of spousalbehavior was unrelated to the type of attributions made by the wives; however, theextent of spousal unhelpfulness was positively related to characterologicalattributions (r = .24, p < .05). Further, the relation between having a spouse thatfailed to offer support and the extent of blame in the attribution was significant (r= .21, p < .05) such that those wives who had spouses that failed to support themexpressed higher levels of blame in their attributions. For husbands, however,mentioning behaviors directed towards others was significantly related to makingboth situational attributions (r = .30, p < .01) and characterological attributions (r= -.22, p < .05). Further, the extent of blame in the attribution was less when thebehavior was directed towards others (r =-.35, p< .01), and the attribution wasmore blaming when the wife failed to offer support (r = .32, p < .01). (See Table8.)Finally, the specific types of unhelpful spousal behaviors were correlatedwith the extent to which the partner was perceived to be unhelpful to determine ifsome behaviors were more likely to be seen as unhelpful than other behaviors. Forthose respondents mentioning at least one unhelpful action by their spouse, the typeof behavior did not significantly correlate with the extent of unhelpfulness for wivesor husbands. Similarly, the types of attributions were correlated with the extent ofblame in the attribution to determine if some attributions were more likely to beSpousal Unhelpfulness33Table 8Correlations Between the Extent of Spousal Unhelpfulness and Type of UnhelpfulSpousal Behavior, and the Type of Attribution and Extent of Blame in theAttributionATTRIBUTIONSHusbandsaUNHELPFUL SPOUSAL BEHAVIOR1^2^3^4^5Situation .04 .11 .15 .08 .30**Spouse's Character .08 .11 .19 .13 -.22*Benign/Positive .06 .12 .01 .11 .12Extent of Blame -.17 -.20 .32** -.05 -.35**WivesbSituation -.06 .13 -.06 .19 .02Spouse's Character .24* -.09 .15 -.11 .03Benign/Positive .05 .08 .03 -.16 -.02Extent of Blame .17 .05 .13 -.07 .021 =Extent of spousal unhelpfulness, 2 =Criticism, 3 =Fails to support, 4 =Otherbehaviors directed towards the spouse, 5 =Behaviors directed towards others.an=  78 , b n = 90*p < .05, ** p < .01Spousal Unhelpfulness34associated with blame than other attributions. For both husbands and wives,characterological attributions were related to higher blame ratings (r = .39, p <.001 and r = .30, p < .001 respectively) and benign attributions were associatedwith lower ratings of blame (r = -.35, p < .01 and r = -.28, p < .01respectively). Making situational attributions was unrelated to extent of blame forboth husbands and wives.Multivariate Prediction of Marital SatisfactionHierarchical multiple regressions were performed between maritalsatisfaction scores as the dependent variable and the predictor variables asdescribed below. In order to address concerns about low power, it was decided torepeat the analyses with a reduced number of control variables. Three variableswere chosen for these analyses. Education was retained because of its relation toDAS scores for wives and seriousness because of the independent contribution tomarital satisfaction for both husbands and wives. The presence or absence ofmarital issues (the dichotomous vector used as a part of the set of dummy codes forthe type of problem) was retained for two reasons; it had the largest beta weight forboth husbands and wives, and secondly, whether or not the issues focused on theneed for spouses to work together would seem particularly germane to anassessment of marital satisfaction. The results of these "reduced" analyses areavailable in Appendix C.Control variables and type of stressor. Table 9 displays the results ofregressing the DAS score on the control variables. For wives, the four controlvariables -- education, family income, seriousness of the problem, and familystructure (two dummy variables indicating the presence of only the spouse'schildren living in the home and the presence of only the respondent's childrenSpousal Unhelpfulness35Table 9Regression Coefficients and R2 Change when Regressing DAS on ControlsHUSBANDS^WIVESVariable Beta^R2change Beta R2changeSTEP 1 - DEMOGRAPHICS AND SERIOUSNESS RATINGSeriousness of Stressor -.30** -.32**Education .09 .22*Income -.07 -.16Only Spouse's Children -.10 -.06Only Own Children .04 .09 -.08 .17**STEP 2 - TYPE OF PROBLEMBroad Stepfamily Issues .03 -.02Marital Issues -.13 -.13Family (not marital) .07 .03 .11 .04*TOTAL R2 .12 .21***p < .05, **p< .01Spousal Unhelpfulness36living in the home) -- predicted a significant amount of the variance in DAS scores(R2 = .17, p < .01). Higher seriousness ratings predicted lower maritalsatisfaction (beta = -.32, p < .01) and the greater the years of education, thehigher the DAS scores (beta = .22, p < .05). For husbands, the beta foreducation failed to reach significance. However, the ratings of the seriousness ofthe stressor did contribute independently to the explained variance (beta = -.30, p< .01) such that husbands rating the stressor as more serious had lower maritalsatisfaction than did those husbands who perceived the problem as less serious.Even so, the total amount of variance explained by the set of control variables wasnot significant (R2 = .09, p > .10) for husbands. The analysis using the reducedset of control variables resulted in the same findings in terms of which predictorswere significant (see Appendix C - Table 1).In an attempt to determine if there was any unique contribution of the type ofproblem to overall marital satisfaction, this set of variables was entered next. Forhusbands, there was no significant increase in the prediction of marital satisfactionover and above that explained by the control variables. For wives, the addition ofthe type of problem as a predictor resulted in a significant R 2 change of .04.Extent of unhelpfulness. After entering the control variables as Step 1, theextent to which the spouse was perceived to be unhelpful was entered into theequation as Step 2. As hypothesized, this variable significantly improvedprediction of marital satisfaction for both wives (R2 change = .05, beta = -.24 p< .01) and husbands (R2 change = .15, beta = -.40, p < .001). Higher ratingsof spousal unhelpfulness were related to lower marital satisfaction. These resultsare described in Table 10. Results were similar in the reduced variable analysisSpousal Unhelpfulness37Table 10Regression Coefficients and R2 change when Regressing DAS on Extent ofUnhelpfulness and Type of Spousal BehaviorHUSBANDS WIVESVariable Beta R2change Beta R2changeSTEP 1 - CONTROL VARIABLES-.29**.03-.36**.22Seriousness of StressorEducationIncome -.09 -.17Only Spouse's Children -.14 -.05Only Own Children .00 -.10Broad Stepfamily Issues .01 -.02Marital Issues -.13 -.13Family (not marital) .07 .11 .12 .21 **STEP 2 - EXTENT OF UNHELPFULNESSExtent^ -.40*** .15*** -.24** .05**TOTAL R2 .27*** .26***STEP 2 - TYPE OF UNHELPFUL SPOUSAL BEHAVIORSaCriticism -.20 -.17Fails to Support -.32** -.17Other Behaviors -.23 .00Towards Others -.14 -.19None .22 .26*** -.06 .03TOTAL R2 .37*** .24*a n = 100 husbands and 102 wives due to uncodeable responses.*p < .05, **p < .01, *** p > .001Spousal Unhelpfulness38(See Appendix C - Table 2).Spousal behaviors. An additional set of regression equations were run inwhich Step 1 was again the control variables but with the type of spousal behaviorsas Step 2. Table 10 summarizes these results. For husbands, the resulting R 2change of .26 was significant at p < .001. However, only a failure to providesupport was independently associated with marital satisfaction (beta = -.32, p <.01); those husbands reporting this type of spousal behavior had lower maritalsatisfaction than husbands not reporting this behavior. 6 For wives, there was nosignificant association between having a spouse that failed to be supportive andmarital satisfaction (beta = -.17, p > .10). 7 Contrary to predictions, the beta forcriticism failed to reach significance for husbands (beta = -.20, p > .10) or wives(beta = -.17, p > .10). The results of the reduced analysis supported thoseobtained using the full set of control variables. (See Appendix C, Table 2.)Type of attribution. The next set of regressions involved enteringattributions for spousal behaviors following the control variables. (See Table 11.)Contrary to predictions, characterological attributions did not add significantly tothe prediction of satisfaction (beta = ..03, p > .10) for husbands. For wives,characterological attributions were negatively related to marital satisfaction (beta =6 Given the correlation between this type of unhelpful spousal behavior and the extent ofspousal unhelpfulness for husbands (r = -.39, R < .02), a second regression was doneentering the spouse's failure to offer support after first partialling out extent of spousalunhelpfulness. This analysis indicated that a spouse's failure to provide support was stillnegatively related to marital satisfaction scores (beta = -.28, p < .01).7 The difference between husbands and wives in the regression coefficients for a spouse'sfailure to support was tested and not found to be significantly different ft (166) = -1.12, p> .10).Spousal Unhelpfulness39Table 11Regression Coefficients and R2  change when Regressing DAS on Type ofAttribution and Extent of Blame in AttributionHUSBANDS WIVESVariable Beta R2change Beta R2changeSTEP 1 - CONTROL VARIABLESSeriousness of Stressor -.35** -.33**Education .13 .22Income -.18 -.13Only Spouse's Children -.13 -.05Only Own Children .00 -.08Broad Stepfamily Issues .04 -.03Marital Issues -.15 -.08Family (not marital) .00 .17 .09 .17*STEP 2 - TYPE OF ATTRIBUTIONSituation .05 -.18Spouse's Character .03 -.28*Benign/Positive .11 .01 -.01 .06TOTAL R2 .17 .23**STEP 2-EXTENT OF BLAME IN ATTRIBUTIONBlame -.21 .03 -.36*** . 12***TOTAL R2 .20 .29***Note: Analyses include only those respondents mentioning at least one type ofunhelpful behavior and one codeable attribution (n = 78 husbands and n = 90wives).*p < .05, **p < .01, ***12 < .001Spousal Unhelpfulness40-.28, p < .05) as predicted. 8 '9 However, it should be noted that the relationbetween these two variables was not significant at the bivariate level.Extent of blame in the attribution. As predicted, using the ratings of theextent of blame in the attribution resulted in a significant increase in varianceexplained. However, this was true only for wives (R2change = .12, p < .001,beta for the extent of blame = -.36; see Table 11). Wives expressing high levelsof blame in their attributions for spousal behaviors were more likely to have lowermarital satisfaction than were wives expressing less blame towards their spouse.For husbands, the extent of blame did not explain a significant amount of variancein marital satisfaction. 1 ° In order to rule out the possibility that the significantresult for wives was due to content overlap between measures of blame and DAS,the analysis was repeated using overall happiness in the marriage as the dependentvariable. The pattern of results was similar (R 2change = .05, p < .01, beta forextent of blame = -.24). The results using the reduced set of control variableswere also similar (see Appendix C, Table 3).Interactions. Finally, a set of two regression equations were constructed foreach unhelpful spousal behavior. Control variables were entered as Step 1, theunhelpful spousal behavior and characterological attributions as Step 2, and theinteraction between the behavior and characterological attributions in Step 3. Thesecond equation was similar, using the extent of blame in the attribution in the8 Again, the difference between the regression coefficients for husbands and wives wastested for characterological attributions and not found to be significant (1 (166) = 1.84, p‘? .10).Y This regression was replicated with the overall happiness in the marriage as thepredicted variable and the results were similar (beta for characterological attributions =-26 p < .05).lu The difference between the regression coefficients for the extent of blame for husbandsand wives was tested and not found to be significant (1(166) = 1.25, p > .10).Spousal Unhelpfulness41place of characterological attributions. For those subjects reporting more than onetype of behavior or more than one attribution, interaction terms were constructedby matching each behavior with the attribution made for that behavior. In any casewhere it was unclear whether or not the respondent was linking a particularattribution with a behavior, no interaction was assigned. For all of theseregressions, the interaction terms failed to explain a significant amount of thevariance in DAS scores over and above that explained by the control variables, thetype of spousal behavior, and the attribution for the behavior and are not reportedhere. The reduced variable analyses replicated these results.DISCUSSIONThis study provides evidence for the impact of unsupportive behaviors fromthe spouse on marital satisfaction when such behavior occurs in the context of aparticularly important issue or problem. As hypothesized, the analyses demonstratea significant relation between the perceived extent of spousal unhelpfulness andmarital satisfaction for both wives and husbands; for both, greater unhelpfulnesswas related to lower marital satisfaction. However, the expected relation betweenunhelpful behaviors and marital satisfaction was found only for husbands; forwives, considering the type of spousal behavior did not significantly improve theprediction of marital satisfaction over that explained by the control variables. Theresults do not support the hypothesized effect of perceived spousal criticism, butrather, point to the importance of husbands' perceptions of their wife's failure tosupport them in regards to this problem. Husbands who report their spouse'sbehavior as a failure to support them had significantly lower DAS scores than thosehusbands who did not report this type of behavior. Although the specific types ofSpousal Unhelpfulness42attributions did not reliably contribute to the prediction of DAS scores ashypothesized, the ratings of the extent of blame in the attribution did contribute, asexpected, to predictions of marital satisfaction; more blame was associated withgreater marital distress. After controlling for possible third variables, however,this relation no longer held for men. Finally, the findings did not support thehypothesized interaction between the type of behavior and the type of attribution.The results will be discussed in terms of the stressors faced by this sample, theunhelpful spousal behaviors experienced, the attributions made for spousalbehaviors, and the relations between behavior and attribution.Stressors The problems mentioned by respondents as currently the most serious weresimilar to those produced in prior research on stepfamilies (Keshet, 1990;Messinger, 1976; Visher & Visher, 1985). Problems with relationships within thefamily, difficulties with children and stepchildren, and conflicts with ex-partnerswere common themes. This suggests that the respondents in this study are fairlyrepresentative of the larger population of stepfamilies.For the most part, husbands and wives did not differ significantly in theirreports of the types of stressor currently facing their families; of the fourcategories, husbands and wives differed only on the frequency with which theymentioned problems that resided within or between family members outside of themarital relationship. It seems most probable that wives' more frequent reports ofthis stressor reflect gender differences in roles within the family. Kessler andMcLeod (1984) point to the wife's greater involvement in the daily interactionsbetween family members and the accumulated effect of frequent interpersonalhassles to explain the greater impact of stress on women. This is consistent withSpousal Unhelpfulness43the finding in the current study that women tended to rate the family problems thatthey were experiencing as more serious than did their husbands.The type of stressor was not significantly related to marital satisfaction. It ispossible that the codes used were too broad to capture important dimensions in thedifficulties facing these couples. Alternatively, the types of problems may havebeen too similar for there to be differences in effect in that they all involvedproblems that the family was facing. However, current models of stress and coping(Lazarus & DeLongis, 1983) suggest that structural dimensions of stressors (suchas marital vs. family) are distal variables that should not be expected to predictstress outcomes. More proximal variables, such as the perceived seriousness of thestressor, would be expected to be more strongly related to stress outcomes.Consistent with prior research on stress and coping that suggests that the cognitiveappraisal of a stressor is an important factor in determining the outcome of stress(e.g. Folkman, Lazarus, Dunkel-Schetter, DeLongis, & Gruen, 1986), theperceived seriousness of the stressor did predict low marital satisfaction in thecurrent study.Spousal BehaviorsThe extent to which the spouse was perceived to be unhelpful differedsignificantly between husbands and wives; wives less frequently reported that theirspouse did nothing unhelpful and in general indicated that their spouse was moreunhelpful than their husbands reported. It is difficult to determine if this representsa gender difference in perceptions of spousal support or a veridical representationof the nature of relationship between spouses. Previous research has suggested,however, that husbands are less likely to engage in supportive actions (Gray et al.,1990; Yogev, 1986). It is important to remember that this comparison is post hocSpousal Unhelpfulness44and therefore must be interpreted cautiously; clearly further research on genderdifferences in perceptions of spousal unsupportiveness is necessary.The types of unhelpful behaviors mentioned by respondents in this studywere both similar to and different from those found by other researchers. Similarto the findings of research on support attempts that fail, respondents identifiedcriticism of their response to the problem by their spouse, a failure to talk about theproblem or generally offer support, a failure to help with instrumental tasks, andother behaviors such as minimizing or maximizing the problem, as unhelpful(Dakof & Taylor, 1990; Lehman et al., 1986). Thus it would seem that thebereaved (Lehman et al., 1986), cancer patients (Dakof & Taylor, 1990) andmultiple sclerosis patients (Lehman & Hemphill, 1990) experience unhelpfulbehaviors similar to those experienced by remarried couples.Along with similarities in types of unsupportive behaviors, the frequencieswith which each behavior was mentioned in the current sample were comparable toprevious studies. Dakof and Taylor (1990) report that 20% of cancer patients intheir sample experienced criticism from their spouse, and 11 % indicated that theirspouse failed to support them. A further 26% indicated that their spouse engagedin behaviors classifiable as "other" in the current study. These proportions arequite similar to those of the current study (21 %, 11 %, and 25% for husbands and23%, 25%, and 20% for wives, respectively).One major difference between the current study and previous research is inthe frequency of reporting minimization (stating that the problem is less serious orimportant than it seems) and maximization (stating that the problem is more seriousor important than it seems, or expressing excessive concern) as unhelpfulbehaviors. In the Dakof and Taylor (1990) study, 13% of respondents indicatedSpousal Unhelpfulness45that their spouse engaged in each of these types of behaviors. Prior research hasalso found that these two types of behaviors are frequently experienced by multiplesclerosis patients (Lehman & Hemphill, 1990 - although this study did notdifferentiate the spouse from other support givers). However, in the current study,these behaviors were included under the code "other unhelpful behaviors aimed atthe spouse" as they were only occasionally mentioned (a mean across spouses of8.5% for minimization and 3.5% for maximization). Also unlike previousresearch, respondents in this study regularly described actions that their spousedirected towards others (children, stepchildren, ex-partners, etc.) as unhelpful.These differences make sense for several reasons. First of all, the types ofproblems faced by an individual who is chronically ill, such as physical limitationsor emotional stress, are different from those faced by remarried couples. It is thuslikely that they have different needs in terms of support and therefore experience alack of support in different ways. However, the types of problems are importantfor support providers as well as the individual experiencing the problem. It hasbeen suggested that support providers may be uncomfortable in the face of chronicillness or bereavement and that they may attempt to minimize problems in thesesituations to manage their own emotional reaction to it (Wortman & Lehman,1985). Similarly, as support providers are also impacted by such stressors, theymay maximize some aspect of the situation as an expression of their own distress.In contrast, it is less likely that support providers would be uncomfortable withremarried couples, or would experience a great deal of distress around remarriage.On the other hand, remarried couples may be more prone to experience problemssurrounding interactions with others because of the difficult roles inherent inSpousal Unhelpfulness46stepfamilies, while such interpersonal demands may be reduced for the chronicallyill or bereaved.It is also possible that beyond the frequency of experiencing differentunhelpful behaviors, different behaviors are more salient for these two populations.Chronically ill individuals are perhaps more vulnerable to having their physicalconditions exaggerated or their difficulties passed over and are therefore moreaware of when support providers engage in these behaviors. Similarly, individualsin stepfamilies may be sensitized to comments regarding their natural children orparticularly aware of spousal behaviors surrounding ex-partners.Further, it should be noted that the way the question was posed torespondents in this study differed from that used in previous studies. The currentstudy asked subjects about behaviors that their spouse engaged in that wereunhelpful, whether or not the spouse intended them to be unhelpful. Previousresearch asked about behaviors that support providers engaged in that the recipientfelt were unhelpful even though the action was intended to be helpful. It is possiblethat these two prompts elicit responses that cannot be expected to coincide. Thefirst would seem to elicit primarily behaviors that were blatantly negative, while thesecond would be unlikely to produce such negative behaviors under the rubric of"attempts to be helpful."Finally, the current study focused on a family problem rather than placingthe question in the context of personal physical disability or bereavement. Inparticular, this would seem to make it more likely that respondents would considera wide range of spousal behaviors, including those involving other familymembers, rather than focusing only on those directed at themselves. Conversely,Spousal Unhelpfulness47asking about unhelpful actions in the context of a personal stressor might make itless likely that behaviors directed toward others would be mentioned.Despite the focus on a stressor involving the family however, the frequencywith which respondents spontaneously mentioned actions directed towardindividuals other than themselves as unhelpful is a provocative finding that hasimplications for the standard typologies of social support. These classificationstypically fail to recognize actions directed toward someone outside of therelationship as an important component of that relationship. Spousal behaviorsdirected toward other family members may be particularly important in the contextof stepfamilies where a spouse's actions aimed at children that are not their ownmay be threatening to the natural parent. Similarly, unhelpful spousal actionsaimed at ex-partners may be particularly detrimental given the interpersonaldifficulties inherent in divorce, remarriage, and reformulation of the family. Thisfinding may also have implications for marital therapy with remarried couples as itpoints to the necessity of considering spousal behaviors directed toward otherfamily members as potential difficulties in the relationship.Before concluding that spousal behaviors directed at family members otherthan the spouse are meaningful however, it must be clear that even though thesebehaviors are perceived to be unhelpful, they were not related to maritalsatisfaction in the current study. One possibility that should be considered is thatthese behaviors may be seen as unhelpful at the time they occur but that over timethey lose their "sting" and do not influence outcomes such as marital satisfaction.It is also possible that the importance of this group of behaviors is expressedthrough outcomes other than low marital satisfaction, such as depression or familyconflict. However, it is difficult to make conclusions based on the current studySpousal Unhelpfulness48given the heterogeneity of responses (behaviors directed at children, step-children,ex-partners, etc.) included in this category. A finer analysis of this group ofbehaviors is warranted but was not possible with the current sample givenlimitations in sample size.Several other interesting results arose from the analyses of the contributionof spousal unhelpfulness and specific unhelpful actions by the spouse to maritalsatisfaction. First of all, as hypothesized, the perceived extent of spousalunhelpfulness occurring in the context of a single stressor faced by the family wasrelated to marital satisfaction. This is in line with the previous literature onmarriage that has connected marital happiness with levels of conflict in therelationship (Jacobson et a., 1980) but extends this research to behaviors perceivedto be unhelpful that occur around a family stressor. This suggests that evendissatisfaction with support received for a single family problem affects maritalsatisfaction.It cannot be ruled out however, that spouses unhelpful in one situation aregenerally unhelpful in other situations as well. In other words, if their spouse'sbehavior is similar across situations, it is possible that asking the respondent tofocus on one problem generates data that is no different from what would begenerated by asking how their spouse is unhelpful generally. The number ofrespondents that qualified their descriptions of their spouse's behavior with "... butgenerally (he/she) is pretty good" suggests that this is not universally the case.Further research is necessary to determine the pattern of unhelpfulness acrosssituations and to differentiate the impact of spousal behaviors around a majorstressor from the impact of behaviors occurring in a variety of less seriouscontexts.Spousal Unhelpfulness49Contrary to expectations, unhelpful spousal behavior was unrelated to wives'marital satisfaction. Further, despite hypothesizing that spousal criticism would bemost strongly related to marital satisfaction, for husbands only perceptions of theirwife's failure to provide support were found to be related to low maritalsatisfaction. Although this result must be interpreted cautiously as anyconsideration is post hoc, the importance of this general class of behaviors is alsoconsistent with the literature on the effects of a lack of support on other outcomessuch as health and mood (DeLongis et al., 1988). This also extends previousresearch on helpful and unhelpful actions that has suggested that a lack of supportwas the most unhelpful behavior encountered (Dakof & Taylor, 1990) bydemonstrating that it also had the strongest relation to marital adjustment. Clearly,further research on this type of unhelpful behavior is warranted.Although the effect of perceptions of a spouse's failure to support wasdemonstrated only for husbands, the lack of a significant difference between theregression coefficients of husbands and wives suggests that this behavior impactsboth spouses in similar ways. However, it is also possible that there are genderdifferences in the response to this spousal behavior but the current study lackedsufficient power to identify those differences. Further research is necessary todetermine if men and women differ in their response to a spouse's failure tosupport them.The hypothesis regarding the effect of spousal criticism was not supported.This is somewhat surprising given the extent to which this type of action has beenfound to be related to negative outcomes in the literature. It would seem that thereare several possible explanations. One possibility that cannot be ruled out is thatremarried couples are unique in their response to criticism. Perhaps individualsSpousal Unhelpfulness50that have been through a divorce are less sensitive to criticism, having learned todeal with such behavior from a spouse in a previous relationship. However, thispresumes that all previous relationships involved criticism, and it seems equallyplausible that exposure to criticism in a prior marriage might increase reactivity tocriticism rather than decrease it.A second explanation that should be recognized is that the effect of criticismmay be on outcome variables other than marital satisfaction. For example, studiesusing mood as an outcome measure have demonstrated that some degree of upsetwithin a relationship is related to more negative mood following the conflict (Pagelet al., 1988). This also suggests that the effect of criticism may be more immediatethan would be detectable by measures of marital satisfaction, such as the DAS, thatfocus on a period of at least six months. Possibly, it is only in the long term, or inthe presence of repeated criticism, that marital satisfaction would be undermined(perhaps through a relationship with depressed mood). Thus, behaviors perceivedas criticism now (elicited in this study by a focus on a current problem), may onlyimpact marital satisfaction in the future. Unfortunately, it is not possible to assessthis given the present data set. An assessment of the prevalence and chronicity ofthe behavior, preferably through the collection of longitudinal data, is necessary.A third reason for the failure to replicate the effect of criticism found inother studies that should be considered, lies in differences between the currentstudy and previous research in the definition and measurement of criticism.Behavioral measures of the frequency of critical remarks from the spouse (Manne& Zautra, 1989; Vaughn & Leff, 1976) may not map onto criticism perceived bythe recipient. It may be that such measures are more sensitive to spousal criticismthan self-report measures.Spousal Unhelpfulness51The most likely explanation, however, may be that the lack of an effect forcriticism is related to the statistical limitations of using a dichotomous predictor.An assessment of the presence or absence of criticism is not as sensitive as a ratingof the extent of criticism, and thus restriction of range in the classification ofcriticism may be obscuring the effect of this variable. This is supported by thestrong relation of ratings of the extent of unhelpfulness to marital satisfaction. Thissuggests that a rating of spousal criticism that allows for a gradation in the extent towhich the spouse is critical would be a more effective method for assessing thisbehavior. Indeed, previous research using rating scales of perceived criticism havedemonstrated significant relations between these ratings and negative outcomes(Hooley & Teasdale, 1989). The influence of restriction of range may also explainthe lack of effect for the other behaviors included in this study.AttributionsSeveral findings from the analyses regarding attributions deserve discussion.These include the frequency of benign attributions for unhelpful behavior, therelation of the extent of blame to marital satisfaction for wives, and the interactionsbetween attributions and spousal behavior. Each of these will be discussed in turn.Prior research on support attempts that fail has identified a high proportionof benign or positive attributions made for unhelpful actions (Lehman & Hemphill,1991). However, research in this area typically has asked respondents to reportbehaviors that their support providers intended to be helpful; this would seem tonaturally elicit attributions of a positive nature. Having addressed the possibilitythat benign attributions are due to the nature of the question by asking aboutspousal actions which were unhelpful regardless of whether the actor intended themto be unhelpful, it is interesting that the current study also elicited a high proportionSpousal Unhelpfulness52of such attributions. This suggests that the findings of earlier studies were not anartifact of the methods used.Explanations in the literature for the frequency of benign attributions fornegatively impacting behaviors have primarily involved the reluctance ofindividuals to criticize their support providers (Lehman & Hemphill, 1991).However, it is possible that this is a function of using the chronically ill or thebereaved as a sample population; in these circumstances, individuals may beparticularly dependent on support providers and thus less willing to do anything(such as criticize or blame) that might compromise the relationship. This might beextended further to a reluctance to jeopardize feelings of security that a caregiver orsupport provider can give by labeling them as blameworthy. The frequency ofbenign attributions in stepfamilies, a sample that is perhaps less openly dependenton their social relationships than a chronically ill sample, suggests that support ismeaningful when provided in a variety of circumstances and that people dealingwith a wide spectrum of stressors are loathe to jeopardize their relationships withtheir support providers.There are at least two alternative explanations that must be considered,however. First of all, the respondents may have been influenced by the perceivedsocial desirability of particular responses. Making a positive attribution fornegative behavior may be seen as reflecting well on the individual by demonstratingmagnanimity or benevolence in light of negative spousal behavior, or it mayfunction to avoid negative judgements from the listener that could result frompublicly blaming the spouse. The distinction between attributions made public andattributions kept private has been made by other authors (Fincham & Bradbury,1988) and it may be that these two types are not congruent for these and otherSpousal Unhelpfulness53reasons. Unfortunately, it is difficult to assess an attribution made by theindividual (the "real" attribution) but not made public. Assurances of anonymityand confidentiality were used during the interview in this study to minimize the biastowards responses perceived as socially desirable.It is also true, however, that this study assessed only a subset of spousalbehaviors by asking about just those behaviors that occurred in the context of theproblem identified as the most stressful. It is likely that attributions for thesebehaviors will be affected by the quality of the relationship as a whole. Perhapsrespondents attribute behaviors to benign reasons because, on the whole therelationship is good or because the behaviors only occur in this particular situationand are therefore judged as atypical. This interpretation would lead one to expectbenign attributions to be associated with high marital satisfaction, however. Thiswas not demonstrated to be the case.Unlike the specific types of attributions, the extent of blame in theattributions for spousal behaviors did significantly relate to DAS scores at thebivariate level for both husbands and wives; this finding is consistent with previousresearch (Fincham et al., 1987; Fincham & Bradbury, 1988). Respondentsexpressing more blame in their attributions for spousal behavior had lower maritalsatisfaction than those respondents expressing less blame; after controlling fordemographic variables and the type and seriousness of the stressor, however, thisrelation no longer held for men. The contrast between husbands and wives in theeffect of blame is difficult to interpret given the lack of a significant differencebetween the regression coefficients for this variable.It should be recognized that information on the extent of blame is notnecessarily independent of that received from assessing the type of attribution, as itSpousal Unhelpfulness54is likely that attributions of responsibility map onto a characterological/noncharacterological distinction. (If a cause of a behavior is believed to be due tosomething within the spouse, it is more likely that the spouse will be blamed forthat behavior). This is supported by the significant correlations betweencharacterological attributions and the extent of blame. Still, given that thecorrelations between the types of attributions and the extent of blame were onlymoderate (although statistically significant), knowing the extent to which the spouseis blamed for their behavior is important supplemental information on the impact ofattributions.The fact that the extent of blame predicts DAS scores while the type ofattribution does not predict marital satisfaction, may again be due to thepsychometric limitations associated with restriction in range. As with the type ofbehavior, using a dichotomous predictor for attributions would seem to restrict thevariability in responses that could have been achieved through the use of ratingscales. Thus it would appear that it is important that future research on supportattempts and unhelpful behaviors moves from content coding of attributions (andbehaviors) to the use of rating scales similar to those used in current research onattributions.Interactions between Spousal Behaviors and Attributions The results failed to support the hypothesis that there is a synergistic effectbetween the type of attribution or the extent of blame in the attribution and the typeof behavior. There are at least two possible explanations for this finding. The firstand most obvious is that spousal actions and attributions are actually unrelated andare independent in their contribution to marital satisfaction. However, a morelikely explanation is that the lack of an effect for the interaction betweenSpousal Unhelpfulness55attributions and spousal behaviors may be the result of the statistical limitations ofsample size. Given the relatively small number of cases available for eachinteraction term, it is perhaps not surprising that the interactions failed to reachstatistical significance. 11 It is clear that more research using larger samples isneeded to address this limitation.Limitations There are several inherent limitations to a study of this nature. Perhaps themost serious is that cross-sectional data such as those described here preclude anyfinal assessment of the direction of causality between attributions or behaviors andmarital satisfaction. It is impossible to say whether perceptions of unhelpfulnesscause lower marital satisfaction or whether low marital satisfaction increases thechance that the spouse will be perceived as unhelpful. Similarly, it is not possibleto tease apart the causal relations between spousal attributions of blame and lowmarital satisfaction. However, although these difficulties may restrict conclusions,cross-sectional data are also valuable, particularly with initial data collection in newareas of inquiry such as the one under investigation here.The self-report nature of the data also creates difficulties to the extent towhich recall is biased and not representative of all interactions surrounding anissue. Spouses may be biased by recent events or by particularly salient behavioron the part of the spouse, and their answers may not be reflective of theirinteractions around this problem. If a wife's failure to support her husband isparticularly salient to him, he may recall such incidents when asked about unhelpfulspousal behavior even though they occur infrequently. However, it can be argued11 Power for the current sample size with an estimated R 2 change of .04 for theinteraction, was estimated between .49-.51 for husbands and .54-.56 for wives.Spousal Unhelpfulness56that bias in the recall of a behavior is useful as an indication of the impact ofperceptions of behaviors. If an individual differentially recalls the distressingbehaviors of a spouse, it is likely that those behaviors exert a more powerful impacton the recipient. As argued earlier, it is the perception of the behavior that may bemore powerful than its veridicality. However, the relatively low frequency withwhich men reported that their spouses failed to support them, argues against theinterpretation that husbands are impacted more by a spouse's failure to supportbecause such a failure is recalled more frequently than other behaviors.Two other sources of bias cannot be ruled out. It is possible that therespondents were affected by the extent to which they perceived a particularresponse as being "better" in terms of what the researchers might want or in termsof the way the response might appear to the interviewer. However, assurances ofanonymity and confidentiality, and standardization of interviewer technique, shouldhave minimized this effect. Second, it should be recognized that respondents mayhave been biased by the fact that all the interviewers were female. This may haveserved to alter responses, perhaps by making women more comfortable and thusmore open about their experiences or by making men less likely to respond in waysthat might make them look bad. However, it has been found by a number ofinvestigators (e.g., Veroff, Kulka, & Douvan, 1981) that both women and mentend to feel most comfortable confiding in women.Finally, the generalizability of research results is always restricted by theextent to which the sample is not representative of the larger population. It must beacknowledged that couples consenting to participate in this study may have beensignificantly different from those who did not, and couples in which both partnersagreed to participate may have been significantly different from those in which oneSpousal Unhelpfulness57spouse refused. (This may be a partial explanation for the relatively high levelsand restricted range of marital satisfaction found in the current study). Also, theresults may not be applicable to non-White, non-English speaking remarriedcouples. Further, it is clear that the sample used in this study was more affluentthan many families, given that the average family income and the years ofeducation of respondents were above national averages (reported as $40,646 and12.2 respectively in 1991; Statistics Canada, 1992).ConclusionTo conclude, this study offers preliminary evidence for the inclusion ofunhelpful spousal behaviors and attributions for those behaviors in a model ofmarital satisfaction. In particular, it would seem that perceptions of a failure on thepart of the spouse to provide support may have important links to maritalsatisfaction. Further, responsibility attributions for unhelpful spousal behavior maybe particularly important in understanding marital distress. However, these resultsare far from unequivocal; differences between husbands and wives in the effect ofthese variables clearly warrants more research. Additional research is alsonecessary to delineate the role of spousal behaviors and attributions on bothproximal measures of distress as well as more long-term measures of relationshipsatisfaction. Finally, further research is needed to clarify the relations betweenunhelpfulness and helpfulness in the context of particularly important issues andmore generally in the relationship. It would seem necessary to move beyonddescriptive assessments of supportive behaviors and move from the reliance oncontent-coding responses to the inclusion of rating scales for spousal behaviors.The current study provides useful guidelines to this effect for future studies onunhelpful spousal behaviors and attributions for spousal behavior.Spousal Unhelpfulness58REFERENCESAnderson, J., & White, G.D. (1986). An empirical investigation of interaction andrelationship patterns in functional and dysfunctional nuclear families andstepfamilies. Family Process,  25, 407-422.Antonucci, T. C., & Israel, B. (1986). Veridicality of social support: Acomparison of principal and network members' responses. Journal ofConsulting and Clinical Psychology, 54, 432-437.Barnett, L. R. & Nietzel, M. T. (1979). Relationship of instrumental andaffectional behaviors and self esteem to marital satisfaction in distressed andnondistressed couples. Journal of Consulting and Clinical Psychology, 47,946-957.Beach, S.R.H., Arias, I., & O'Leary, K.D. (1986). The relationshi of maritalsatisfaction and social support to depressive symptomology.Psychopathology and Behavioral Assessment, 8, 305-316.Berley, R. A., & Jacobson, N. S. (1984). Causal attributions in intimaterelationships: Toward a model of cognitive-behavioral marital therapy. InP. Kendall (Ed.), Advances in cognitive-behavioral research and therapy(Vol. 3, pp. 1-60). New York: Academic.Booth, A., & Welch, S. (1978). Spousal consensus and its correlates: Areassessment. Journal of Marriage and the Family, 0, 23-32.Bradbury, T. N. & Fincham, F. D. (1990). Attributions in marriage: Review andcritique. Psychological Bulletin, 107, 3-33.Camper, P., Jacobson, N., Holtzworth-Monroe, A., & Schmaling, K. (1988).Causal attributions for interactional behaviors in married couples. CognitiveTherapy and Research, 12, 195-209.Cohen, S. & Wills, T. (1985). Stress, social support, and the buffering hypothesis.Psychological Bulletin, 98, 310-357.Compas, B.E., Forsythe, C.J., & Wagner, B.M. (1988). Consistency andvariability in causal attributions and coping with stress. Cognitive Therapyand Research, 12, 305-320.Corcoran, K., & Fischer, J. (1987). Measures for clinical practice. New York:The Free Press.Dakof, G., & Taylor, S. (1990). Victims' perceptions of social support: What ishelpful from whom? Journal of Personality and Social Psychology,  5, 80-89.Spousal Unhelpfulness59Dean, A., & Lin, N. (1977). The stress-buffering role of social support. Journalof Nervous and Mental Disease, 165, 403-417.DeLongis, A., Folkman, S., & Lazarus, R.S. (1988). The impact of daily stresson health and mood: Psychological and social resources as mediators.Journal of Personality and Social Psychology, 54, 486-495.Fincham, F. D., Beach, A., & Nelson, G. (1987). Attribution processes indistressed and nondistressed couples: 3. Causal and responsibilityattributions for spouse behavior. Cognitive Therapy and Research, 11, 71-86.Fincham, F., & Bradbury, T. (1990). Social support in marriage: The role ofsocial cognition. Journal of Social and Clinical Psychology,  2, 31-42.Fincham, F., & Bradbury, T. (1988). The impact of attributions in marriage:Empirical and conceptual foundations. -British Journal of ClinicalPsychology, 27, 77-90.Fincham, F. D. & Bradbury, T. N. (1987). The impact of attributions in marriage:A longitudinal analysis. Journal of Personality and Social Psychology, I,510-517.Fletcher, G. J., Fincham, F. D., Cramer, L., & Heron, N. (1987). The role ofattributions in the development of dating relationships. Journal ofPersonality and Social Psychology, 5, 481-489.Folkman, S., Lazarus, R.S., Dunkel-Schetter, C., DeLongis, A., Gruen, R.J.(1986). Dynamics of a stressful encounter: Cognitive appraisal, coping andencounter outcomes. Journal of Personality and Social Psychology, U, 992-1003.Folkman, S., Lazarus, R., Gruen, R., & DeLongis, A. (1986). Appraisal, coping,health status, and psychological symptoms. Journal of Personality and Social Psychology, IQ, 571-579.Gaelick, L., Bodenhausen, G.V., Wyer, R.S. (1985). Emotional communicationin close relationships. Journal of Personality And Social Psychology, 49,1246-1265.Gotlib, I. H. & Whiffen, V. E. (1989). Depression and marital functioning: Anexamination of specificity and gender differences. Journal of AbnormalPsychology, 98, 23-30.Gray, E., Lovejoy, M., Piotrkowski, C., & Bond, J. (1990). Husbandsupportiveness and the well-being of employed mothers of infants. Familiesin Society: The Journal of Contemporary Human Services, 71, 332-341.Spousal Unhelpfulness60Guenzel, P.J., Berckmans, T.R., & Kannell, C.S. (1983). General InterviewingTechniques. Ann Arbor, Michigan: University of Michigan, SurveyResearch Centre of the Institute of Social Research.Hautzinger, M., Linden, M., & Hoffman, N. (1982). Distressed couples with andwithout a depressed partner: An analysis of their verbal interaction. Journalof Behavior Therapy and Experimental Psychiatry, 13, 307-314.Hobart, C. (1990). Experiences of remarried families. Journal of Divorce,  la,121-144.Holtzworth-Monroe, A., & Jacobson, N. (1985). Causal attributions of marriedcouples: When do they search for causes? What do they conclude when theydo? Journal of Personality and Social Psychology, 4/1, 1398-1412.Holtzworth-Monroe, A., & Jacobson, N. (1988). Toward a methodology forcoding spontaneous causal attributions: Preliminary results with marriedcouples. Journal of Social and Clinical Psychology, 7, 101-112.Hooley, J. M., & Teasdale, J. D. (1989). Predictors of relapse in unipolardepressives: Expressed emotion, marital distress, and perceived criticism.Journal of Abnormal Psychology, 98, 229-235.House, J. S. (1981). Work stress and social support. Reading, Mass.: Addison-Wesley.Jacobson, N., Follette, W., & Waggoner-McDonald, D. (1982). Reactivity topositive and negative behavior in distressed and nondistressed marriedcouples. Journal of Consulting and Clinical Psychology, 5Q, 706-714.Jacobson, N., Waldron, H., & Moore, D. (1980). Toward a behavioral profile ofmarital distress. Journal of Consulting and Clinical Psychology, 48, 696-703.Jacobson, N. S., McDonald, D. W., Follette, W. C., & Berley, R. A. (1985).Attributional processes in distressed and nondistressed married couples.Cognitive Therapy and Research, 9, 35-50.Jacobson, N. S., & Moore, D. (1981). Spouses as observers of the events in theirrelationship. Journal of Consulting and Clinical Psychology, 49, 269-277.Keshet, J. (1990). Cognitive remodeling of the family: How remarried peopleview stepfamilies. American Journal of Orthopsychiatry,^196-203.Kessler, R., & McLeod, J. (1984). Sex differences in vulnerability to undesirablelife events. American Sociological Review, 49, 620-631.Kessler, R., & McLeod, J. (1985). Social support and mental health in communitysamples. In S. Cohen & S. Syme (Eds.) Social support and health (pp. 219-240). New York: Academic Press.Spousal Unhelpfulness61Kowalik, D.L., & Gotlib, I.H. (1987). Depression and marital interaction:Concordance between intent and perception of communication. Journal ofAbnormal Psychology, 21 127-134.Kyle, S.O., & Falbo, T. (1985). Relationships between marital stress andattributional preferences for own and spouse behavior. Journal of Social andClinical Psychology, 3, 339-351.Lazarus, R.S., & DeLongis, A. (1983). Psychological stress and coping in aging.American Psychologist, la, 245-253.Lazarus, R.S., & Folkman, S. (1984). Stress. appraisal. and coping. New York:Springer Publishing Company.Lehman, D., Ellard, J., & Wortman, C. (1986). Social support for the bereaved:Recipients' and providers' perspectives on what is helpful. Journal ofConsulting and Clinical Psychology, 54, 438-446.Lehman, D., & Hemphill, K. (1991). Recipients' perceptions of support attemptsand attributions for support attempts that fail. Journal of Social and PersonalRelationships, 7, 563-574.Lin, J., & McGranahan, D. (1980). Personal disruptions, social integration,subjective well-being, and predisposition toward the use of counsellingservices. American Journal of Community Psychology, 8, 87-100.Long, E., & Andrews, D. (1990). Perspective taking as a predictor of maritaladjustment. Journal of Personality and Social Psychology, 52, 126-131.Malkinson, R. (1987). Helping and being helped: The support paradox. DeathStudies, 11, 205-219.Manne, S. L., & Zautra, A. J. (1989). Spouse criticism and support: Theirassociation with coping and psychological adjustment among women withrheumatoid arthritis. Journal of Personality and Social Psychology, 5&, 608-617.Messinger, L. (1976). Remarriage between divorced people with children fromprevious marriages: A proposal for preparation for remarriage. Journal ofMarriage and Family Counseling, 2, 193-200.Norton, A. J. & Moorman, J. E. (1987). Current trends in marriage and divorceamong American women. Journal of Marriage and the Family, 42, 3-14.Pagel, M., Erdly, W., & Becker, J. (1987). Social networks: We get by with (andin spite of) a little help from our friends. Journal of Personality and SocialPsychology, 53, 793-804.Spousal Unhelpfulness62Pickersgill, M.J., & Beasley, C.I. (1990). Spouses' perceptions of life eventswithin marriage. Personal and Individual Differences, 11, 169-175.Primomo, 2 J., Yates, B.C., & Woods, N.F. (1990). Social support for womenduring chronic illness: The relationship among sources and types toadjustment. Research in Nursing & Health, la, 153-161.Roberts, T. W., & Price, S. J. (1989). Adjustment in remarriage: Communication,cohesion, marital and parental roles. Journal of Divorce, a 17-43.Rook, K. (1984). The negative side of social interaction: Impact on psychologicalwell-being. Journal of Personality and Social Psychology, 46, 1097-1108.Schuster, T. L.,. Kessler, R. C., & Aseltine, R. H. (1990). Supportive interactions,negative interactions, and depressed mood. American Journal ofCommunity Psychology, la, 423-438.Spanier, G. (1976). Measuring dyadic adjustment: New scales for assessing thequality of marriage and similar dyads. Journal of Marriage and the Family,2a, 15-28.Statistics Canada (1992). Canada Yearbook. Ottawa, Ontario: Author.Vaughn, C., & Leff, J. (1976). The measurement of expressed emotion in thefamilies of psychiatric patients. British Journal of Clinical Psychology,  15,157-165.Veroff, J., Kulka, R., & Douvan, E. (1981). Mental Health in America: Patternsof Help-seeking from 1957 to 1976. New York: Basic Books.Visher, E., & Visher, J. (1985). Stepfamilies are different. Journal of FamilyTherapy, 7, 9-18.Wethington, E., & Kessler, R. (1986). Perceived support, received support, andadjustment to stressful life events. Journal of Health and Social Behavior,27, 78-89.Wortman, C., & Lehman, D. (1985). Reactions to victims of life crises: Supportattempts that fail. In I.G. Sarason & B.R. Sarason (Eds.), Social support: Theory. research and applications (pp. 463-489). Dordrecht, TheNetherlands: Martinus Niijhoff.Yelsma, P. (1984). Marital communication, adjustment and perceptual differencesbetween "happy" and "counseling" couples. The American Journal ofFamily Therapy, 12, 26-36.Yogev, S. (1986). Relationships between stress and marital satisfaction amongdual-earner couples. Women and Therapy,  5, 313-330.Spousal Unhelpfulness63Appendix AInterview ItemsTYPE OF STRESSORFor the next 15 minutes or so we are going to focus on the one problem orchallenge that you currently see as the most serious. Fm going to ask youquite a few questions about this problem, so I'd like you to take a fewmoments to consider what you believe the most serious family problem is foryou. This may be a problem we've already mentioned or it might besomething different. What would that problem be?SERIOUSNESS OF THE STRESSORFrom your perspective, how serious would you say this problem is? (1means not at all serious and 5 means very serious).EXTENT OF UNHELPFULNESS With this problem, to what extent does your (husband/wife) say or do thingsthat are unhelpful, whether or not (he/she) intends to be unhelpful? (1 meansnot at all and 5 means to a large extent).UNHELPFUL BEHAVIORIF NOT 1: What does your (husband/wife) say or do (that is unhelpful)?(AO? Are there any other things....)ATTRIBUTIONSWhy do you think (he/she) might say or do that? (AO? Are there any otherreasons...)Spousal Unhelpfulness64Appendix BCodes for Open-ended Questions WHAT DO YOU SEE AS THE MOST SERIOUS FAMILY PROBLEM?I. Marital issues. Examples include: difficulties in raising the children,differential treatment of children or stepchildren, marital conflict(unspecified as focused on other issues), concern for the future, and lack ofaffection/warmth between partners.II. Family problems other than those between the respondent and thespouse. Examples include: Isolation of a specific individual, difficulties inspecific relationships with child or stepchild, misbehavior of child orstepchild, acceptance of an individual into family or extended family.III. Broad stepfamily issues. Examples include: Integration of the family as aunit, adjustments in family roles necessitated by remarriage, and difficultieswith ex-partner of the respondent or their spouse.IV. Miscellaneous problems not classifiable elsewhere. Examples include:financial difficulties, inadequate time or difficulty coordinating time, divisionof labor within the household.WHAT DOES YOUR SPOUSE DO THAT IS UNHELPFUL?I. Criticizes or blames respondent for the problem. Examples include: iscritical of respondent's response to problem, takes the child's side inarguments, says that it is the respondent's fault that the problem continues.II. Fails to engage in supportive behaviors. Examples include: Lack ofsupport in decision making/efforts to cope, lack of tangible help in familymaintenance/ childrearing, ignores spouse's need for help, doesn't talk aboutSpousal Unhelpfulness65his/her feelings or Respondent's feelings or listen when R wants to discussfeelings or issues.III. Other unhelpful behaviors directed towards respondent. Examplesinclude: Offers unrealistic solutions or suggestions, minimizes or maximizesthe problem, denies responsibility, is stubborn/inflexible, is insensitive ornegative, gets impatient/frustrated, gets angry about the problem, or yells.IV. Unhelpful behaviors directed towards people other than the spouse.Examples include: Is overprotective or restrictive of towards child orstepchild, gives preferential treatment to one child, is overly lenient orindulgent or is inconsistent in discipline or interaction with child orstepchild, has expectations of child(ren) that are too high or too low,criticizes or lays blame on child, or offers unsolicited advice to the child, orinteracts inappropriately with ex-partners (is not firm enough with them, orspouse gets upset during interactions).WHY DO YOU THINK THEY MIGHT SAY/DO THOSE THINGS THAT AREUNHELPFUL?I. Situational Influences cause the spouse to act that way. Examples include:something about the situation or the interaction between the spouse and thesituation such as feelings of frustration or guilt because of..., my daughter isdifficult, etc. There is no indication that this is due to something that ispermanent or pervasive about the spouse.II. Constitutional or Characterological Factors cause the spouse to act thatway. It is something that is a part of who the spouse is, or something thatholds true across situations. Examples include: Lacks understanding ofSpousal Unhelpfulness66problem or need for help, personal hangups or expectations, differencesbetween spouse and any other member of the family, is afraid ofjeopardizing a relationship, is not motivated to solve problem for any reason.III. Benign or positive attributions; factors other than the situation or thespouse's character cause the behavior. Examples include: Limitedresources (does or doesn't do things because of demands from othersituations), good intentions cause partner to act that way (partner intends orwants to be helpful), unaware of what to do to be helpful or that what theyare doing is not helpful, or spouse's past history (their upbringing or lack ofexperience or something which clearly happened prior to current family) iscausing their actions.Spousal Unhelpfulness67Appendix CRegression Analyses using Reduced Set of Control VariablesTable C-1Regression Coefficients and R2 Change when Regressing DAS on DemographicsAnd Type and Seriousness of the StressorHUSBANDS^WIVESVariable^ Beta^R2change Beta^R2changeSTEP 1 - EDUCATION AND SERIOUSNESS OF THE STRESSORSeriousness -.26** -.30**Education -.09 .08 .17* .14***STEP 2 - TYPE OF STRESSORMarital Issues -.14 .02 -.17 .03TOTAL R2 .10* .17****p < .05, **p < .01Spousal Unhelpfulness68Table C-2Regression Coefficients when Regressing DAS on Extent of Unhelpfulness andType of Spousal ActionHUSBANDS^WIVESVariable^ Beta R2change Beta R2changeSTEP 1 - CONTROL VARIABLES.10*.16***.26***-.32***.17-.17***-.26**.17***.06**.23***Seriousness^ -.27**Education -.07Marital Issues -.14STEP 2 - EXTENT OF UNHELPFULNESSExtent^ -.40***TOTAL R2STEP 2 - TYPE OF UNSUPPORTIVE SPOUSAL BEHAVIORS4Criticism -.16 -.19Fails to Support -.31** -.17Other Behaviors -.21 -.01Towards Others -.12 -.15None .24 .27*** -.07 .03TOTAL R2 .35*** .20**a n = 100 husbands and 102 wives due to uncodeable responses.*p < .05, **p < .01, ***p < .001Spousal Unhelpfulness69Table C-3Regression Coefficients when Regressing DAS on Type of Attribution and Extentof Blame in AttributionHUSBANDSa^WIVESbVariable^ Beta R2change Beta R2changeSTEP 1 - CONTROL VARIABLES.13-.31**.18*-.11 .14**Seriousness^-.32**Education .05Marital Issues^-.15STEP 2 - TYPE OF ATTRIBUTIONSituational^.09 -.13Characterological^.01 -.26*Benign/Positive .15 .02 .01 .05TOTAL R2 .15 .19**STEP 2 - EXTENT OF BLAME OF ATTRIBUTIONAffect^-.21^.04 -.34*** .11***TOTAL R2 .17** .25***an=  78 , b n = 90*p < .05, **p < .01, p < .001


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items