Marital Processes Predictive of Later Dissolution Behavior, Physiology, and Health John M. Gottman Department of Psychology University of Washington Robert W. Levenson Department of Psychology University of California, Berkeley ABSTRACT Seventy-three married couples were studied in 1983 and 1987. To identify marital processes associated with dissolution, a balance theory of marriage was used to generate 1 variable for dividing couples into regulated and nonregulated groups. For studying the precursors of divorce, a "cascade" model of marital dissolution, which forms a Guttman-like scale, received preliminary support. Compared with regulated couples, nonregulated couples had (a) marital problems rated as more severe (Time 1); (b) lower marital satisfaction (Time 1 and Time 2); (c) poorer health (Time 2); (d) smaller finger pulse amplitudes (wives); (e) more negative ratings for interactions; (f) more negative emotional expression; (g) less positive emotional expression; (h) more stubbornness and withdrawal from interaction; (i) greater defensiveness; and (j) greater risk for marital dissolution (lower marital satisfaction and higher incidence of consideration of dissolution and of actual separation). We wish to acknowledge National Institute of Mental Health (NIMH) Grant ROI PHS MH42722 and Research Scientist Development Award, and Research Scientist Award 1K02MH00257 to John M. Gottman and NIMH Grant MH39895and National Institute on Aging Grant AG07476 to Robert W. Levenson. Correspondence may be addressed to John M. Gottman, Department of Psychology, University of Washington, Guthrie Hall, NI-25, Seattle, Washington, 98195. Received: October 15, 1990 Revised: March 23, 1992 Accepted: April 3, 1992 There are currently over one million divorces a year in the United States, with estimates that almost 50% of marriages will ultimately end in divorce ( Cherlin, 1981 ). Marital dissolution is a serious social issue in terms of its negative consequences for the mental and physical health of spouses ( Levinger & Moles, 1979 ) and their children ( Emery, 1988 ). Previous Studies Despite the importance of marital dissolution, empirical research has not been very successful at predicting which married couples will separate or divorce and which married couples will stay together. Attempts at prediction have usually been epidemiological, designating cohorts and demographic groups that are thought to be at the greatest risk for marital dissolution ( Bennett, Blanc, & Bloom, 1988 ; Cherlin, 1981 ). Lamentably, studies attempting to identify marital processes that are antecedents of Journal of Personality and Social Psychology © 1992 by the American Psychological Association August 1992 Vol. 63, No. 2, 221-233 For personal use only--not for distribution. Page 1 of 1911/27/2000http://spider.apa.org/ftdocs/psp/1992/august/psp632221.htmlmarital dissolution have been quite rare (for a review, see Newcomb & Bentler, 1981 ). Our current lack of knowledge concerning which patterns of marital interaction lead to marital dissolution stems partly from the fact that, in most studies, divorce and separation have been viewed as independent rather than dependent variables (e.g., Hetherington, Cox, & Cox, 1978 , 1982 ). Thus, these studies have been primarily concerned with the effects of marital dissolution on other variables and on the adjustment of spouses and children to marital dissolution. Of the nearly 1,200 published studies to date with the terms marital separation or divorce in their titles, we know of only four prospective longitudinal studies that have attempted to predict future separation and divorce ( Bentler & Newcomb, 1978 ; Block, Block, & Morrison, 1981 ; Constantine & Bahr, 1980 ; Kelly & Conley, 1987 ). 1 In the Block et al. study of 57 families with children who were 3.5 years old, parental disagreement about child-rearing practices discriminated between the intact and divorced groups 10 years later. Constantine and Bahr, in a 6-year longitudinal study, found that the group of men who either divorced or separated had a greater "internal orientation" on the leadership subscale of a measure of locus of control than did men who remained married. Bentler and Newcomb (1978) found that couples who remained married were more similar in age, interest in art, and attractiveness than couples who separated or divorced. Men who separated or divorced described themselves as more extraverted, more invulnerable, and more orderly than men who stayed married. Women who separated or divorced described themselves as less clothes conscious and less congenial than women who stayed married. Kelly and Conley (1987) , using acquaintance ratings of personality in a prospective 35-year longitudinal study of marital stability, reported that the men who remained married were more conventional and less neurotic, and had greater impulse control than those who divorced. A similar pattern was found for women, with the additional finding that women who stayed married were judged as higher in emotional closeness and lower in tension in their families of origin. Aggregating findings from these four studies does not provide a coherent theoretical picture of couples or individuals at risk for marital dissolution. Furthermore, effect sizes in these studies were not particularly large. Nonetheless, that relations did obtain is encouraging for additional efforts at longitudinal prediction using the same and other methods. From our perspective, an important methodological improvement would be the addition of direct observation of marital behavior, which could provide greater descriptive clarity in prospective longitudinal research and might account for greater amounts of variance in marital dissolution. The Problem of Low Base Rates of Divorce in Short-Term Longitudinal Studies Ironically, although many marriages will ultimately end in divorce, attempts to predict marital dissolution over short, 3- to 5-year periods are often plagued by low base rates of divorce. In part, this problem simply reflects that it can take many years for an unsatisfying marriage to formally dissolve, but it also may reflect sampling issues (e.g., couples who are willing to participate in these kinds of research projects may be those who are least likely to divorce). Examples of low base rates for divorce in short-term longitudinal studies are common. In Kelly and Conley's (1987) study of 278 couples who were married in 1935, the
View Full Document
Unlocking...