Those men who nominated disagreements over stance,that the larger the family the greater the children as a cause of breakdown were particular- housing difficulties.The mean values show,how- ly likely to describe strongly disapproving parents ever,that it is among three-child families that (p =.036).For both men and women,parental these complaints are most common,followed by approval was associated with higher SES(r =.17, one-child families,larger families(4+),and lastly p =.02)and with a drop in living standards two-child families. following the separation (r =.17,p =.02;see below). Standard of Living After Divorce Respondents'evaluations of their standards of Parental Marital Stability living following separation or divorce were signifi- Twenty-one percent of the respondents'parents cantly associated with causes (F=1.49,df= and 24 of their spouses'parents were reported 16,320,p =.02).Those people who reported to have been divorced or separated.This is some- financial (p<.000)and housing (p =.03)prob- what higher than the incidence of 18.9(18.3 lems considered that their standards of living were for women and 19%%for men)in a sample of still now higher than during their marriages,while married Australian couples interviewed at a simi- those who reported disagreements over children lar time (Antill and Cunningham,personal com- (p =.02)evaluated their living standards as munication)but in line with the general finding lower.The relationship with husband's cruelty from the U.S.A.that there is a modest amount of (p =.03)was curvilinear;i.e.,these women's liv- intergenerational transmission of marital break- ing standards were now described as either higher down (Kitson and Raschke,1981).Nearly one- or much lower.A greater drop in standard of liv- third (30%)of those respondents who came from ing was reported by those of higher SES(r =.27, divorced or separated homes had married spouses p=.001). from a similar background.(It was not possible to estimate the likelihood of such homogamy in the Duration of Marriage general population,given the broad age range of As the Figure shows,many couples continued the sample and the absence of population statis- to live together for considerable periods after the tics that would indicate the likelihood of persons time at which they felt their marriages had begun within this range coming from dissolved mar- to break down-26 years in one case.The pa- riages.) tience of wives is particularly notable;although Marital status of the respondents'parents was half of the women considered the breakdown to not significantly associated with any of the have occurred within two years of marriage,the marital complaints,but that of the spouses' great majority stayed with their husbands for a parents was (F 1.88,df 16,314,p<.001). further 5 to 20 years.Perhaps because of this dis- The relevant causes are an "other man''(p< crepancy between attitudes and behavior,dura- .000)and disagreements over children (p =.02). tion of marriage did not emerge as a strong The direction of the association was unexpected: predictor,although there is a seemingly logical in each case it is the widowed parents who stand association between length of marriage and wife's out from all others. loss of interest in the marriage (p =.025).Dura- Acquaintance Before Marriage tion of marriage is associated,however,with spouse's religion (F=3.31,df 2,309,p =.02). There is no overall association with length of Religiously homogamous marriages have the acquaintance before marriage,but the univariate longest duration,with the most longstanding be- analyses showed that those who had known their ing those between two nonbelievers.The briefest spouses for the shortest period of time (0-3 marriages were those between a Protestant and a months)were more likely to nominate husband's Roman Catholic and between a Protestant and a cruelty as a cause.Age at marriage and premarital nonbeliever.When the data were analyzed sepa- acquaintance period were not correlated (r rately for men and women,it emerged that wives 06). whose ex-husbands were Roman Catholics had ex- Number of Children perienced the shortest marriages.The husbands in these cases were significantly less likely than all Again,the multivariate Fis not significant,but others to maintain contact with children after the friction with relatives (p =.022)and housing separation (F=4.45,df 2,309,p =.01). problems (p =.033)both show a trend towards association with number of children.Interesting- Associations Among Perceived Causes ly,this is not a strictly linear function,as might The analyses reported above suggested that cer- perhaps be expected;it would seem logical,for in- tain marital complaints tended to cluster together. 558 JOURNAL OF MARRIAGE AND THE FAMILY August 1984 This content downloaded from 211.80.94.134 on Mon,19 Dec 2016 05:27:55 UTC All use subject to http://about.jstor.org/termsThose men who nominated disagreements over children as a cause of breakdown were particular- ly likely to describe strongly disapproving parents (p = .036). For both men and women, parental approval was associated with higher SES (r = .17, p = .02) and with a drop in living standards following the separation (r = .17, p = .02; see below). Parental Marital Stability Twenty-one percent of the respondents' parents and 24%o of their spouses' parents were reported to have been divorced or separated. This is some- what higher than the incidence of 18.9% (18.3% for women and 19% for men) in a sample of still married Australian couples interviewed at a simi- lar time (Antill and Cunningham, personal com- munication) but in line with the general finding from the U.S.A. that there is a modest amount of intergenerational transmission of marital break- down (Kitson and Raschke, 1981). Nearly one- third (30%) of those respondents who came from divorced or separated homes had married spouses from a similar background. (It was not possible to estimate the likelihood of such homogamy in the general population, given the broad age range of the sample and the absence of population statis- tics that would indicate the likelihood of persons within this range coming from dissolved mar- riages.) Marital status of the respondents' parents was not significantly associated with any of the marital complaints, but that of the spouses' parents was (F = 1.88, df = 16,3i4, p < .001). The relevant causes are an "other man" (p < .000) and disagreements over children (p = .02). The direction of the association was unexpected: in each case it is the widowed parents who stand out from all others. Acquaintance Before Marriage There is no overall association with length of acquaintance before marriage, but the univariate analyses showed that those who had known their spouses for the shortest period of time (0-3 months) were more likely to nominate husband's cruelty as a cause. Age at marriage and premarital acquaintance period were not correlated (r = .06). Number of Children Again, the multivariate F is not significant, but friction with relatives (p = .022) and housing problems (p = .033) both show a trend towards association with number of children. Interesting- ly, this is not a strictly linear function, as might perhaps be expected; it would seem logical, for in- stance, that the larger the family the greater the housing difficulties. The mean values show, how- ever, that it is among three-child families that these complaints are most common, followed by one-child families, larger families (4 +), and lastly two-child families. Standard of Living After Divorce Respondents' evaluations of their standards of living following separation or divorce were signifi- cantly associated with causes (F = 1.49, df = 16,320, p = .02). Those people who reported financial (p < .000) and housing (p = .03) prob- lems considered that their standards of living were now higher than during their marriages, while those who reported disagreements over children (p = .02) evaluated their living standards as lower. The relationship with husband's cruelty (p = .03) was curvilinear; i.e., these women's liv- ing standards were now described as either higher or much lower. A greater drop in standard of liv- ing was reported by those of higher SES (r = .27, p = .001). Duration of Marriage As the Figure shows, many couples continued to live together for considerable periods after the time at which they felt their marriages had begun to break down-26 years in one case. The pa- tience of wives is particularly notable; although half of the women considered the breakdown to have occurred within two years of marriage, the great majority stayed with their husbands for a further 5 to 20 years. Perhaps because of this dis- crepancy between attitudes and behavior, dura- tion of marriage did not emerge as a strong predictor, although there is a seemingly logical association between length of marriage and wife's loss of interest in the marriage (p = .025). Dura- tion of marriage is associated, however, with spouse's religion (F = 3.31, df = 2,309, p = .02). Religiously homogamous marriages have the longest duration, with the most longstanding be- ing those between two nonbelievers. The briefest marriages were those between a Protestant and a Roman Catholic and between a Protestant and a nonbeliever. When the data were analyzed sepa- rately for men and women, it emerged that wives whose ex-husbands were Roman Catholics had ex- perienced the shortest marriages. The husbands in these cases were significantly less likely than all others to maintain contact with children after the separation (F = 4.45, df = 2,309, p = .01). Associations Among Perceived Causes The analyses reported above suggested that cer- tain marital complaints tended to cluster together. 558 JOURNAL OF MARRIAGE AND THE FAMILY August 1984 This content downloaded from 211.80.94.134 on Mon, 19 Dec 2016 05:27:55 UTC All use subject to http://about.jstor.org/terms