Australian Couples in Millenium Three  

The major determinants of strong couple relationships can be thought of as falling into two broad categories: generic and specific determinants. Generic determinants refer to variables that influence relationship outcomes across a broad range of life stages and circumstances. For example, the capacity to negotiate conflict seems to impact upon the relationship satisfaction and stability of couples across a wide variety of ages and circumstances (Gottman, 1998). Specific relationship determinants are variables that exert particular influence under certain circumstances. For example, negotiating the parenting role of a stepparent is particularly difficult for stepfamilies during the process of family formation (Visher & Visher, 1982).

Generic determinants of positive relationship outcomes

The most important information on the influences on relationship outcomes comes from longitudinal studies of the course of relationships. There are over 120 published studies assessing psychological variables and the longitudinal course of couple relationship satisfaction and stability (Karney & Bradbury, 1995). There also are a large number of studies that have examined sociodemographic variables and their relationship to the satisfaction and stability of couple relationships (Glenn, 1998; Larson & Holman, 1994). Bradbury (1995) adapted the stressvulnerability- coping model to offer a heuristic model by which this comprehensive literature usefully can be summarized. He suggested that there are three broad classes of variables that impact upon the etiology of relationship problems: adaptive processes within the couple system, stressful events impinging upon the couple, and enduring individual vulnerabilities of the partners. I believe this model is very useful, but it does focus upon the determinants of relationship distress and divorce.

Three modifications make Bradbury's (1995) model even more useful when considering how to promote strong, mutually satisfying and stable couple relationships (Halford, in press). Bradbury's concept of individual vulnerabilities refers to individual differences that make relationship problems more likely. I prefer the term individual characteristics rather than individual vulnerabilities, as some individual characteristics have positive effects on relationships (e.g. gender role flexibility). Similarly, I prefer life events to stressful life events, as the latter refers only to presumed damage that negative events have on relationships. The reports of long-married satisfied couples (Scott, Halford, & Ward, 1999; Gagnon, Hersen, Kabacoff, & van Hasselt, 1999), highlight that shared history of positive events, and supporting each other through difficult times, contribute to couple relationship satisfaction and commitment. A third modification to Bradbury's model is that I add a fourth class of factors that influence relationship outcomes: contextual variables. Contextual variables refer to the cultural and social circumstances within which couple relationships exist.

[ top ]

The context of couple relationships

Couple relationships occur within broader contexts that can serve to promote relationship satisfaction and stability, or can serve to undermine relationship functioning. Many approaches to relationship education do not include specific attention to these contextual variables, but understanding these factors may be crucial to helping couples sustain strong relationships.

The socio-cultural context of marriage

Couple relationships occur within a cultural context that defines how marriage and other couple relationships are supposed to be. Whilst there are certain general assumptions shared across western cultures, there also are important variations between those cultures. For example, German couples without relationship problems engage in similar level of verbal negativity as Australian distressed couples (Halford et al. 1990), suggesting that greater levels of negativity are more acceptable and less dysfunctional in the German cultural context than in Australia. Even within one country there is great diversity in acceptable relationship behavior. Winkler and Doherty (1983) found that verbal conflict was reported as more common in New York couples who were born in Israel than in Anglo couples living in New York. However, verbal conflict was less often associated with physical aggression or relationship distress in the Israeli-born couples than the Anglo couples. Thus, the cultural appropriateness and functional impact of behavior varies considerably even within western cultures.

It can be important to assess the cultural context within which relationship standards develop and may be reinforced. Partners who differ in their ethnic, racial or cultural background often differ in their expectations and beliefs about relationships (Jones & Chao, 1997). This diversity in partner assumptions and beliefs can be a source of great strength for a relationship when the partners are able to draw on the wisdom and strengths of different cultural traditions. At the same time, substantial differences in expectations can be a significant source of conflict between the partners (Jones & Chao, 1997), and marriages in which partners have very different cultural backgrounds tend to break down at somewhat higher rates than other marriages (Birchnall & Kennard, 1984; Kurdek, 1991). The magnitude of effects of ethnic dissimilarity on relationship satisfaction and stability generally are small (White, 1990).

Other relationships and roles

Whilst the partner role is central to most adults in couple relationships, this is not the only relationship or role that the partners have. Other relationships and roles of each partner are part of the context in which couple interaction occurs, and these other relationships and roles can impact in a positive or negative manner on the couple relationship. For example, work often provides extra stimulation and ideas to enrich the relationship, but work demands also can compete for time with the partner (Thompson, 1997). Friends may provide support and shared activities that complement the relationship, and reduce the chance of excessive dependence upon the spouse. However, friendships also can take away time from the partner. Parenting, sports, hobbies, and community service activities all have the capacity to enrich or erode relationship quality.

There are consistent findings that approval of one's spouse and relationship by friends and extended family are predictive of better relationship satisfaction and stability (Booth & Johnson, 1988; Cate, Hustom, & Nesselroade, 1986, Kurdek, 1991). At the same time there also is evidence that excessive intrusion by family on selection of dating partners and subsequent mate selection may predict relationship problems (Benson, Larson, Wilson, & Demo, 1993).

[ top ]

Adaptive couple processes

Adaptive processes refer to the cognitive, behavioral and affective processes that occur during couple interaction. Certain deficits in these adaptive processes seem to predispose couples to relationship problems. More specifically, deficits in communication and conflict management behaviors observed in engaged couples prospectively predict divorce and relationship dissatisfaction over the first years of marriage (Gottman, Coan, Carrere, & Swanson, 1998; Markman, 1981; Markman & Hahlweg, 1993). Dysfunctional communication in engaged couples also predicts the development of relationship verbal and physical aggression in the first few years of marriage (Murphy & O'Leary, 1989; O'Leary et al. 1989), at least for mild to moderate severity aggression. Relationship aggression often is established early in the relationship, and usually continues and escalates once established (Murphy & O'Leary, 1989; O'Leary et al. 1989).

It is noteworthy that the communication deficits observed in some engaged couples do not correlate with their reported relationship satisfaction at the time (Markman & Hahlweg, 1993; Sanders et al. 1999). It seems that these communication difficulties do not stop couples from forming committed relationships, but the difficulties may predispose couples to develop relationship problems later (Pasch & Bradbury, 1998). In couples who have been married for some time, these same communication difficulties predict deterioration in relationship satisfaction, and decreased relationship stability (Gottman, 1993b; Gottman, 1994).

The beliefs and expectations individuals have when entering into relationships and marriage predict the risk of divorce in the first few years of marriage (Olsen & Fowers, 1986; Olsen & Larsen, 1989). Couples characterized by unrealistic expectations and beliefs in areas such as importance of communication, appropriate methods of conflict resolution, importance of family and friends, and gender roles, have higher rates of erosion in relationship satisfaction than couples not so characterized. Negative attributions, in which partners ascribe blame for relationship problems to stable, negative characteristics of their spouse also prospectively predict deterioration in relationship satisfaction (Fincham & Bradbury, 1991).

Finally, certain patterns of emotional expression are predictive of relationship problems. Showing contempt, disgust, fear or emotional withdrawal toward partners during interaction is predictive of relationship deterioration and taking steps toward separation (Gottman, 1994). Thus, certain behavioral, cognitive and affective characteristics of the couple's adaptive processes predate, and prospectively predict, relationship problems.

Life events

Life events refer to the developmental transitions, and acute and chronic circumstances that impinge upon the couple or individual partners. Relationship problems often are argued to be more likely to develop during periods of high rates of change and stressful events (Karney & Bradbury, 1995). For example, the transition to parenthood sometimes is associated with decline in couple relationship satisfaction (Cowan & Cowan, 1992). However, many couples report that the transition to parenthood enhances relationship satisfaction and commitment. Similarly, partners who successfully support each other through stressful events such as severe illness in one partner often report the experience brings them closer together (Halford, Scott, & Smythe, in press). Significant life events have the potential both to increase or decrease relationship satisfaction.

Couples with less robust adaptive processes are believed to be particularly vulnerable to the negative effects of a range of stressful life events (Markman, Halford, & Cordova, 1997). In particular, couples who lack communication skills, or who have inflexible or unrealistic expectations of relationships, find it hard to negotiate the changes required to adapt to major life transitions. For example, couples in which the woman was recently diagnosed with breast or gynecological cancer and who display poor communication and ineffective mutual support, show deterioration in their relationships and poor individual coping with the cancer (Scott, Halford, & Ward, 1999). In contrast, couples with good communication and mutual support often report having been brought closer together emotionally, by the experience of supporting each other through adversity (Scott et al. 1999).

Individual characteristics

Individual characteristics refer to the stable historical, personal, and experiential factors which each partners brings to a relationship (Bradbury, 1995). High levels of education, high income, and high status occupation each are associated with increased chances of relationship satisfaction and stability (Glick, 1984; Kurdek, 1991, 1993; Martin & Bumpass, 1989; Mott & Moore, 1979). The reasons for these effects are not entirely clear. The relationships of poor and less educated people do have greater cumulative exposure to stress over the early years of marriage, such as financial and health problems, and this exposure may mediate greater relationship problems (Kurdek, 1993).

[ top ]

Relationship history variables

There are a number of aspects of relationship history that couples bring to their current relationships that are predictive of relationship satisfaction and stability. Relationship history includes family of origin experiences, prior relationships and the development of the current relationship, each of which predicts relationship satisfaction and stability (Larson & Holman, 1994).

Negative family-of-origin experiences increase the chance of relationship problems. In particular, parental divorce is associated with greater marital problems in the offspring when they become adults. In the United States and Europe there are substantially higher rates of divorce in adult offspring of divorce than for people with no family history of divorce (DeGraaf, 1991; Glen & Kramer, 1987; Glenn & Shelton, 1983; Pope & Mueller, 1976). Parental divorce seems to have a particularly strong impact upon women. Women experiencing parental divorce have a 60% higher divorce rate than women without such a history, while men with a history of parental divorce have a 35% higher divorce rate than men without such a history (Glenn & Shelton, 1983).

Another well established risk indicator for couple relationships is violence in the family of origin (e.g., Burgess, Hartman, & McCormack, 1987; Mihalic & Elliot, 1997; Stith & Farley, 1993; Stets & Straus, 1990; Straus et al. 1980; Widom, 1989). More specifically, men who report witnessing violence between their parents have a substantially higher risk of being violent themselves (Hotaling & Sugarman, 1986; Mihalic & Elliot, 1997; Riggs, O'Leary & Breslin, 1990; Stith & Farley, 1993; Stets & Straus, 1990; Straus et al. 1980; Widom, 1989). Parent-to-child violence sometimes has been found to increase the risk of men being violent (Stets & Straus, 1990; Straus et al. 1980), though several studies have not found this association (e.g., Hotaling & Sugarman, 1986). Separating the effects of parent-toparent versus parent-to-child physical aggression is difficult, as there is substantial overlap in occurrence of inter-partner and parent-child violence. Rosenbaum and O'Leary (1981) reported 82% of men who reported witnessing violence between their parents also reported being victims of violence by their parents. In the study with the largest sample used to date, family of origin inter-parental violence was found to elevate risk of adult male relationship violence substantially more than parent-to-child violence (Kalmuss, 1984).

The mechanism by which exposure to parental divorce or aggression may impact upon subsequent adult relationships is becoming clearer. Exposure to parental divorce is associated with more negative expectations of marriage (Black & Sprenkle, 1991; Gibardi & Rosen, 1991), and with observable deficits in communication and conflict management in couples prior to marriage (Sanders et al. 1999). Adult offspring of parents who were aggressive also show deficits in communication and conflict management skills in dating and marital relationships (Halford, Sanders, & Behrens, in press; Skuja & Halford, 1999). Negative expectations and communication deficits may well be learned from the parents' relationship and subsequently this learned behavior impacts negatively upon the adult relationships of the offspring. The argument that communication difficulties may be acquired through observation and interaction with parents is supported by the finding that couple communication style assessed at the beginning of adult relationships predicts subsequent communication style when the partners become parents and are interacting with their children (Howes & Markman, 1989).

The longer and better couples know each other before marriage, the greater their reported relationship satisfaction after marriage (Birchnell & Kennard, 1984; Grover, Russel, Schumm, & Paff-Bergen, 1985; Kurdek, 1991, 1993). It has been speculated that shorter dating periods before entering commitment may not allow people to screen out potentially incompatible partners (Grover et al. 1985). Whilst this may be true, there are other variables operating that might explain this association. For example, those who choose to marry quickly may differ in attachment style from those who delay marriage. Perhaps the attachment style predicts both rapid marriage and high risk for relationship problems.

Cohabitation before marriage consistently is associated with increased risk of relationship distress and separation (Balakkrishnan, Rao, Lapierre-Adamcyk, & Krotski, 1987; Janus & Janus, 1993; Trussel, & Rao, 1987). Choosing to cohabit is associated with a variety of other factors, such as low religiosity, uncertainty about committing to the relationship, and negative perceptions of marriage. Any of these variables might account for the high risk of relationship breakdown of married couples who cohabited before marriage.

The association between personality variables and relationship problems has been widely studied. Most normal personality variations do not seem to contribute much variance to relationship satisfaction (Gottman, 1994; Karney & Bradbury, 1995). One exception is poor negative affect regulation, (high neuroticism), which consistently has been found to predict higher risk for relationship problems and divorce (Karney & Bradbury, 1995). Poor negative affect regulation refers to an inability to respond constructively to negative feelings such as anger, sadness, or frustration.

A second personality characteristic related to relationship satisfaction and stability is attachment style (Feeney, 1998). Attachment style refers to a general way of thinking and responding emotionally in close relationships. It is believed that attachment style is learned early in life based upon early relationship experiences with parents. A general style is argues to develop of people having secure or insecure attachment styles. Insecure attachment styles are characterized by high discomfort with emotional closeness or anxiety over being abandoned. How neuroticism or attachment impact upon the development of relationship problems is not yet well understood.

[ top ]

Psychological disorder

A major risk indicator for relationship distress and divorce is past or present history of psychological disorder. Higher rates of relationship problems and divorce consistently have been reported in populations with severe psychiatric disorder (Halford, 1995), and in people with depression, alcohol abuse and some anxiety disorders (Emmelkamp, De Haan & Hoogduin, 1990; Halford, Sanders & Behrens, 1999; Halford & Osgarby, 1993; O'Farrell & Birchler, 1987; Reich & Thompson, 1985; Ruscher & Gotlib, 1988; Weissman, 1987). As described earlier in this chapter, relationship problems and individual problems both can exacerbate each other (Halford et al. 1999). In addition certain personal vulnerabilities may dispose people to both psychological disorders and relationship problems. For example, deficits in interpersonal communication and negative affect regulation are risk factors that predict the onset of both alcohol abuse (Block, Block, & Keyes, 1988), and relationship problems (Markman & Hahlweg, 1993). This common risk factor might be part of the explanation for the common co-occurrence of relationship and alcohol problems.

Gender

There are important differences between how men and women function within relationships. For example, relative to men, women are more likely to report dissatisfaction with a lack of emotional closeness in their marriages (Clements & Markman, 1996; Julien, Arellano & Tugeon, 1997); to be more emotionally expressive when discussing relationship issues (Weiss & Heyman, 1997); to report greater conflict between their work and family roles (Thompson, 1997); and to initiate divorce (Wolcott & Glazer, 1989). There also is evidence that men and women experience intimacy somewhat differently. Women are more likely to experience self disclosure of feelings as high in intimacy, whereas men are more likely to experience shared activity as intimacy (Markman & Kraft, 1989). Marriage and relationship education programs need to provide information to participants on gender differences, and to assist couples to develop ways to meet couple needs for both male and female partners.

Specific influences on relationships at major transition points

The establishment of a committed relationship

The first few years of marriage or cohabitation are associated with substantial change for couples. Most couples find that the initial overwhelming attraction to their spouse moderates, that new relationship roles and routines need to be developed, and means of negotiating conflict evolve (Huston, McHale, & Crouter, 1986; Veroff, Douvan, & Hatchett, 1995). Across numerous studies the mean quality of relationship satisfaction declines in an approximately linear fashion over the first few years of marriage (Huston et al. 1986; Karney & Bradbury, 1997; Kurdek, 1998; Veroff et al. 1995). It has been speculated that some decline in satisfaction is inevitable, given that many couples entering relationships have extreme, and probably unrealistic, positive views about their relationship and partner (Kudek, 1998). It is striking that separations in the first 4 years of marriage account for about one third of all divorces (Clark, 1995). Clearly many couples have a great struggle to adapt to cohabiting in ongoing committed relationships.

There are a number of well-replicated predictors of deterioration in relationship satisfaction across these early crucial years of the relationship. Most of them are the generic risk factors identified above. For example, couple communication and conflict management skills predict relationship satisfaction (Gottman et al. 1998; Markman, 1981). In addition, there are some factors that seem particularly important during the transition into a committed relationship. Relationship aggression is a predictor of deteriorating satisfaction and higher risk of separation that seems particularly important in the early years of marriage (O'Leary, Barling, et al. 1989; Rogge & Bradbury, 1999). A significant minority of young couples need special assistance to resolve conflict without physical aggression, with estimates of 15 to 15% of young couple reporting physical aggression occurring in their relationship in the last year (O'Leary et al. 1989).

[ top ]

The transition to parenthood

Ninety percent of married couples have children (Houseknecht, 1987), and so along with moving in together, having children can be regarded as one of the most universal transitions couples experience. The transition to parenthood universally is reported to bring a wide range of changes in the partners' individual functioning, their relationship with each other, and their relationships with extended family (Belsky & Kelly, 1994; Belsky & Pensky, 1988; Cox, 1985; Cowan & Cowan, 1992; Palkovitz & Sussman, 1988).

It is well established that the transition to parenthood is associated with declines in mean marital satisfaction for couples, though the source of these mean changes is a source of some debate. There are over 20 studies that examine marital satisfaction and marital conflict from early pregnancy across the first months or years of parenthood (Cowan & Cowan, 1995). Almost all of these studies report declines in mean relationship satisfaction and increases in conflict across this time (e.g., Belsky, 1985; Belsky, Lang, & Rovine, 1985; Cowan et al, 1985; Cowan, Cowan, Heming, & Miller, 1991; Wallace & Gotlib, 1990), which has been interpreted to mean the transition to parenthood itself has a negative impact upon relationship satisfaction. However, long term studies of relationship satisfaction show mean relationship satisfaction is highest pre-maritally, that it declines over time, and that couples with and without children show similar downward trajectories (Huston et al. 1986; McHale & Huston, 1985; Duncan & Markman, 1988; Markman, Clements, & Wright, 1991). Hence it may be more accurate to say that couples are experiencing mean declines in relationship satisfaction over time that continue across the transition to parenthood (Clements & Markman, 1996).

Perhaps even more important than the mean changes in satisfaction across the transition to parenthood is the variability between couples in the changes in their relationship as they become parents. In one series of studies 12% of all new parents showed a severe decline, 30% showed moderate decline, 30% showed no change, and about 20% showed definite improvement in their marital relationship satisfaction (Belsky & Rovine, 1990; Belsky & Kelly, 1994). Thus, while the transition to parenthood is associated with a mean decline in relationship satisfaction across couples, for many couples becoming parents is associated with enhanced relationship and life satisfaction.

There are two broad classes of variables that predict changes in relationship functioning across the transition to parenthood: generic relationship risk factors, and parenthood specific factors. Variables such as negative family-of-origin experiences, marital problems before the birth of the child, pre-existing psychological disorders in either partner, and poor couple communication each predict poor individual and couple adjustment across the transition to parenthood (Belsky & Rovine, 1990; Cowan & Cowan, 1992; Cox, Owen, Lewis, & Henderson, 1989). As these same variables also predict deterioration in couple relationship satisfaction among all couples, they are generic relationship risk factors.

One parenthood specific factor predicting deteriorating couple relationship satisfaction and functioning is maternal depression (Cowan & Cowan, 1995; Belsky & Kelly, 1994; Fleming, Ruble, Flett, & Shaul, 1988; Antonucci & Mikus, 1988; Cutrona, 1982). For women, the peak onset of depression is during the childbearing years, with the most widely accepted incidence rate being around 10% to 20% of childbearing women (Romito, 1989). Ten percent of postpartum women develop depression serious enough to interfere with their daily functioning (Campbell, Cohn, Flanagan, Proper, & Myers, 1992). For men, there also appears to be an increased risk of depression during the transition to fatherhood, but much less is known about psychological and psychiatric disorders in new fathers compared with new mothers. There are no epidemiological data documenting the incidence of depression in new fathers, but recent studies have documented higher rates of anxiety, depression and relationship problems in the postnatal period for partners of depressed women (Pope, Evans, McLean, & Michael, 1998).

Parenthood brings with it substantial changes in roles for couples (Sanders et al. 1997). For some couples additional roles that are involved in parenting provide extra sources of satisfaction and are associated with greater reported general life satisfaction (Luckey & Bain, 1970). However, role strain resulting from the buildup of competing demands associated with their parenting roles can add stress for each of the spouses (Rollins & Galligan, 1978). This seems often to have most impact on women (Belsky, 1990). In particular, the extent to which male partners meet the women's expectations for support and sharing parenting responsibilities is a major predictor of the women's relationship satisfaction (Belsky & Rovine, 1990). Gender roles often become more traditional in couples after the birth of a child, and often remain that way until children reach adolescence (Cowan & Cowan, 1990; Belsky et al. 1985). If this gender traditionalism is not consistent with the women's expectations, this can be a major source of dissatisfaction, which may account for why decreases in marital satisfaction are more reliably found for women than for men (e.g. Belsky, 1990).

When children have medical and emotional problems, couples relationships are at increased risk for distress (Klaus & Kennell; 1976; Wikler, 1986). In the early stages of the transition to parenthood factors like prematurity of the child (Pope et al. 1998), and child problems with sleep, eating and other health problems all predict increased risk of stress in the parents, and increased risk of couple relationship problems (Cowan & Cowan, 1992). Moreover, when marriages are distressed before the birth of the child, couples often cope more poorly with such child problems and may exacerbate the severity of these childhood problems (Fincham, Grych, & Osbourne, 1994, Cowan & Cowan, 1992; 1995). In other words, couple relationship problems and early childhood problems reciprocally influence each other.

[ top ]

Separation and re-partnering

Separation and divorce are reported by adults as extremely stressful (Bloom et al. 1978), and it is likely that all adults experiencing separation have at least some adjustment difficulties. For many adults the first six to 12 months after separation are the most difficult, and their psychological well being returns to normal ranges over a period of months to a year or two (Bloom et al. 1978). However, for some people maladjustment is severe and persists for many years after separation (Braver, Whitley, & Ng, 1993; Jordan, 1985; Kincaid & Calwell, 1995). The long term effects of separation can have major negative effects on individuals, the relationships they have with their children (Funder, 1991; Furstenburg, Morgan & Allison, 1987), and on the outcome of subsequent relationships they may enter (Booth & Edwards, 1992; Smith-Barnett, 1990).

Whilst defining what constitutes better adjustment to separation is a matter of some controversy, there are some responses that almost universally would be acknowledged as poor adjustment and coping. Some indicators of poor adjustment that are common are individuals becoming chronically socially withdrawn, depressed, or abusing alcohol and other drugs (Bruce and Kim, 1992; Miller, Smerglia, Gaudet, & Kitson, 1998; Stack & Banowski, 1994). Many separated people report thinking and being preoccupied about the ex-partner, and this preoccupation can range in severity from mild discomfort from unwanted memories and thoughts, to extreme distress and obsessional rumination (Kitson & Morgan, 1990). For some people their distress is so severe that they contemplate, or attempt, or actually commit suicide (Cantor & Slater, 1994).

Parents' relationships with children can be severely affected by separation, particularly for parents who do not have primary contact with their children. Whilst most non-custodial parents initially declare a high level of interest in their children, over time the frequency of contact and the quality of the relationship often greatly diminish (Amato, 1997; Depner & Bray, 1993). An Australian Bureau of Statistics report (April, 1997) found 36% of children who most often reside with the primary care-giver are visited by their other parent from less than once a year to never. In a five-year longitudinal study, 23% of non-custodial fathers had no contact with their children during the previous five years (Furstenberg et al. 1987). The active involvement of both parents in children's lives after separation predicts better adjustment of the children (Hines, 1997), particularly when the noncustodial parent engages in what has been called authoritative parenting (Amato, 1997). Authoritative parenting refers to when the parent is actively involved in the setting of rules for children, discipline, and promoting engagement in schooling and peer social activities for the child.

[ top ]

Influences on adjustment of adults after separation

The stress-vulnerability-coping model adapted by Karney and Bradbury (1995) to explain relationship outcomes can also be adapted to explain the adjustment of expartners to separation. That is, one can see adjustment as a function of the life events occurring around the separation, individual partner characteristics, and interactions between the ex-partners.

There are a number of stressful life events associated with separation, and the more severe these stresses, the poorer adjustment people tend to make. Separation often follows a period of declining relationship satisfaction, and often increasing conflict (Gottman, 1994). In Australia, the most common reasons given by ex-partners for separation are communication problems (23% of women and 33% of men), incompatibility (20% of women and 23 % of men), an affair by either party (20% of both men and women), physical abuse in the relationship (10% of women and 3% of men), or alcohol or drug abuse (11% of women and 3% of men) (Wolcott & Hughes, 1999). In all these reported reasons a common element is dissatisfaction with the partner, and such dissatisfaction usually is associated with anger and conflict with the ex-partner both before and after separation. The evidence that ongoing conflict erodes individual adjustment was reviewed earlier, and highlights that separation often follows a period of severe distress.

Up to 70% of separations are reported as being initiated by women (Braver et al. 1993; Wolcott & Hughes, 1999). Many men often report shock and surprise at the decision of their partner to separate from them, whilst the women are more likely to report they have perceived relationship problems, and have contemplated separation for years before initiating the separation (Kincaid & Caldwell, 1995; Braver et al. 1993). These differences may explain the consistent finding that men exhibit more short term adjustment problems to separation than women (e.g. Helgeson, 1994).

There also are severe stresses after separation. Most ex-partners report a substantial decrease in their disposable income and standard of living after separation, and in people with low initial incomes separation can be associated with extreme financial hardship (Hanson, McLanahan & Thomson, 1998; Duncan, 1994). Involvement in current or impending legal proceedings often is viewed as very stressful, and the longer and more complex the legal proceedings the poorer people's adjustment (Funder, 1991). Many people who experience extended legal action through the Family Law system in Australia perceive the courts as unfair, and as adding to their problems after separation. For example, many men report a view that the system is biased against them (Funder, 1991). Whatever the merits of this opinion, it is clear that the perception of unfairness by the Family Law court is associated with poorer adjustment after separation.

Given that poor conflict management and communications skills predict increased risk for separation and divorce, and that separation often follows prolonged periods of conflict, it is not surprising that many separated couples report ongoing conflict with their ex-partner (Johnson, Gonzales & Campbell, 1987). Severe conflict has been reported to continue for five or more years after separation in 25% of separated couples (Johnson et al. 1988). Difficulties in sharing and coordination of child rearing are major problems for many separated couples (Barris & Garrity, 1997; Funder, 1991), and the extent of conflict between ex-partners has a major impact upon the partners themselves, and their children (Grych & Fincham, 1990).

The relationship with an ex-partner can impact on subsequent couple relationships, and may place them at risk (Lawton & Sanders, 1994). Either high levels of ongoing conflict with the ex-spouse, or ongoing close attachment to the ex-partner, are associated with lower relationship satisfaction in subsequent relationships (Buunk & Mutsaers, 1999).

Individual characteristics of the ex-partners that are associated with adjustment to separation have not been extensively studied. Neuroticism, which can be seen as the ability to regulate and respond to negative feelings, is a relatively stable personality characteristic likely to effect adjustment. Individuals with a history of affect problems such as depression, anxiety and anger, seem most likely to have problems after separation. In addition, ex-partners who have an avoidant or anxious attachment style report more distress after separation than those with secure attachment style (Birnbaum, Orr, Mikulincer & Florian, 1997).

The interpretations ex-partners make of the process and reasons for separation affect adjustment. People who attribute the causes of the break up entirely to their ex-partner often report anger, and this probably is unhelpful in the longer term (Smith-Barnett, 1990). In particular, a partner-blaming perspective is likely to prevent the person examining their own communication skills, ability to manage conflict and so forth. To the extent that these processes put future relationships at risk, the person is missing the chance to develop their ability to have a better relationship in the future. On the other hand, attributing the separation entirely to oneself often is unhelpful, particularly if this is associated with thoughts such as being unlovable, or having faults that one is unable to change.

[ top ]

Stepfamilies

Stepfamilies are an increasing proportion of all families in Australia. The high divorce rate combined with the fact that the majority of separated people repartner contributes to the high rate of step family formation (Pasley & Ihinger- Tallman, 1987). Unfortunately stepfamilies break up at particularly high rates, substantially higher rates than for either first relationships (Fergusson, Horwood, & Dimond, 1985; Pasley & Ihinger-Tallman, 1987), or second marriages without children from prior relationships (Messinger & Walker, 1981; Pasley & Ihinger- Tallman, 1982; White & Booth, 1985; Whitsett & Land, 1992). Furthermore, in stepfamilies where both partners have children from previous marriages (complex families) there is a lower level of marital satisfaction compared to stepfamilies with children from only one partner (Clingempeel, 1981; Clingempeel & Brand, 1985).

The high rates of break up of stepfamilies result in many children experiencing multiple breakdowns of families. About 50% of children whose parents divorce will experience a second relationship break up of one of their parents (Bumpass & Sweet, 1985). Repeated breakdowns of families is associated with substantial long-term adjustment problems for children (Amato, 1997).

A high percentage of remarried couples divorce within the first four years of remarriage (Booth & Edwards, 1992). The satisfaction with couple relationships (Booth & Edwards, 1992), stepparent-child relationships (Bray & Berger, 1993), parent-child relationships (Hetherington, Cox, & Cox, 1982), and family cohesion (Bray & Berger, 1993) often decline soon after remarriage. Those couples who remain together as stepfamilies for about seven years then show substantially higher levels of relationship stability (Papernow, 1984). Given the extremely high rates of early break down of stepfamilies, relationship education programs need to come very early in the stepfamily formation process.

The stepparent-stepchild relationship is often the most crucial to determining stepfamily outcomes (Brown, Green, & Druckman, 1990). The quality of this relationship strongly predicts the child's adjustment (Crosbie-Burnett & Giles-Sims, 1994), and whether couples remain together (Bray & Berger, 1993; Clingempeel, Brand, & Ievoli, 1984; Crosbie-Burnett, 1984). Many writers and researchers in the field argue that the stepparent-stepchild relationship is problematic because society has not established a set of clear and realistic guidelines on how to fulfil the roles of step-family members (Cherlin, 1978). As a result, stepfamily members often start their life together unsure about the roles and boundaries within the new family. This is particularly hard for those stepparents who lack prior parenting experience (Sanders et al. 1997). Furthermore, adolescent stepchildren have the highest rates of difficulty in adjusting to stepfamily life (Ihinger- Tallman,1988; Hughes & Schroeder, 1997). Given that the assertion of independence from family life is socially normative in adolescence, this process may be further complicated in step families by negotiating new family roles.

Couples in stepfamilies have substantially more conflict over parenting practices than other couples (Cissna, Cox & Bochner, 1990), and the role of the stepparent is often central to that conflict. In particular, developing a caring relationship with stepchildren can be problematic. It is often assumed that stepparents and stepchildren will quickly develop a loving relationship, but this is often not the case (Lawton & Sanders, 1994). Issues such as resentment of the new stepparent, unresolved hope for reconciliation between the biological parents, and lack of shared history together make the stepparent-stepchild relationship difficult to forge. When stepparents are rejected by their stepchildren, they may withdraw from the children and their relationship may become more and more negative over time (Hetherington & Clingempeel, 1992), which in turn predicts increasing negativity in the couple relationship (Clingempeel et al. 1984). Negotiating a mutually acceptable level of intimacy, such as the role of a friend or housemate, early in stepfamily formation can reduce the negativity and conflict between stepparents and stepchildren (Visher & Visher, 1991).

The role of stepparents in disciplining children often is a source of conflict in families. Stepparents who assume a disciplinarian role for children too early in the family formation are often resented by children, and the biological parent often finds themselves torn between loyalty to a new partner and loyalty to the child (Hetherington, Cox, & Cox, 1985; Visher & Visher, 1982). On the other hand, lack of child discipline, or inconsistent, irritable or coercive parenting predicts child behavior problems and maladjustment (Buchanan, Maccoby, & Dornbusch 1996). These parenting problems often occur during transition from the first relationship to separation, and then into the new couple relationship (Hetherington, 1988).

The biological parent-child relationship also is important to the health of the new stepfamily. The bond between biological parent and child often is very strong and can interfere with the development of a stable couple bond and the development of the stepparent-stepchild relationship (Hetherington, Bridges, & Insabella 1998). Stepfamilies often have unhealthy biological parent-child coalitions and this may make it difficult for a stepparent to join with the family (Anderson & White, 1986). When stepfamilies first form, parent-child bonds are likely to deteriorate as the parent spends time strengthening the couple relationship at the expense of the parent-child relationship (Bray, 1988). Children may resent the time and love given to their new stepparent and may feel abandoned by their parent (Visher & Visher, 1991). This may result in problem behaviours or withdrawal (Hetherington et al. 1998).

On average, stepfamilies are less competent at problem solving and communicating effectively than first marriage families (Bray, 1988). The source of these problems is unclear; it may be due to the fact that they are attempting to manage particularly difficult issues. There are a number of guidelines that have been offered for formation of stepfamilies that seem most likely to develop a strong stepfamily (Cissna et al. 1990). Couples first need to establish a strong couple relationship, as partners who cohabit soon after meeting are less likely to have stable stepfamily relationships than other stepfamilies. The biological parent needs to introduce the new partner gradually to children, and to affirm the importance of the new relationship to the children jointly with the new partner. The couple need to negotiate a clear and mutually acceptable means of asserting parental authority to resolve family conflicts, to communicate that understanding to the children, and the partners need to avoid conflict about parenting practices occurring in front of the children.

After problems coping with children, stepfamilies report finances as their second biggest problem area (Albrecht, Barr & Goodman, 1983; Hartin, 1990; Messinger, 1976; Pasley & Ihinger-Tallman, 1982). This includes paying and receiving child support, financial burdens incurred from a previous marriage, and management issues related to independence, security and control (Hetherington et al. 1998). There are a number other significant strains in stepfamily life that differentiate stepfamilies from other families. These issues include: loyalty issues with past and present family members, and children's difficulty in adjusting to their biological parent loving another person who is not their parent (Cherlin, 1978; Hartin, 1990; Visher & Visher, 1978; Visher & Visher, 1989). In addition the roles of grandparents, non-custodial parents, and long-time friends in the new stepfamily have to be negotiated.

[ top ]

Older couples

The relationship needs and satisfaction of older couples has received limited research attention (Gagnon et al. 1999). Australia, like most western countries has an aging population and the number and length of marriages continuing into older age is increasing (McDonald, 1995). The divorce rate in older couples is lower than for younger couples, though divorce rates in older couples are increasing internationally (Hammond & Muller, 1992).

The correlates of relationship satisfaction in older couples have many commonalties with the correlates of relationship satisfaction in younger couples, but there also are some important distinctive factors in older couples. Like younger couples, older couples who are satisfied negotiate life transitions more successfully than dissatisfied couples (Gagnon et al. 1999). In particular, dissatisfied couples often report the onset of problems around developmental transitions, such as children leaving the parental home, and retirement (Booth & Johnson, 1994). Many couples respond to the reduced time spent in child rearing or employment by increasing shared positive activities, and this often is associated with reports of enhanced relationship satisfaction (Guilford & Bengston, 1979). However, developing a mutually satisfying balance of independent and couple activities can present challenges for some older couples. Effective communication and conflict management that help to negotiate transitions correlate with relationship satisfaction in older couples, as they do with younger couples, but there is substantially less overt conflict in distressed older couples than in distressed younger couples (Levenson, Carstensen, & Gottman, 1993). Sometimes the lack of obvious conflict might lead outsiders to assume the relationship is a mutually satisfying one. However, there is an important minority of older couples who report long-term relationship problems that are not recognized by others (Levenson et al. 1993), and these problems are predictive of relationship breakdown at times of transition.

Support for the spouse is a factor reported by older couples to be particularly important, and low levels of support during crises are a major source of relationship dissatisfaction in older couples (Cutrona, 1986). There is a reciprocal relationship between sustaining good health and relationship satisfaction in older couples. Declining health is associated with increased relationship problems, particularly amongst couples who provide each other with low levels of mutual support (Melton, Hersen, VanSickle, & Van Hasselt, 1995). Conversely, couples who offer high levels of support and have high satisfaction have better health (Melton et al. 1995). The role of mutually satisfying relationships in promoting the health and well being of older couples has great social significance. In severe health problems of the elderly, such as Alzheimer's disease, a strong couple relationship with good spouse support predict less deterioration in the sufferer, and enhanced ability to support the sufferer in the couple's home rather than in nursing home care (Sanders, 1995).

Implications of evidence for relationship education

The variables that predict relationship satisfaction and stability usefully can be conceptualized as falling into two categories: risk indicators that are static, and risk factors that are dynamic. Static risk indicators allow identification of couples at high risk for relationship problems, but in isolation tells us little about what we can do to enhance their relationships. For example, parental divorce is a predictor of risk for relationship problems, but this historical variable cannot be changed. Dynamic risk factors can be changed, and often are the targets of relationship education. For example, communication skills predict sustained relationship satisfaction (Markman, 1981), and these skills can be taught to couples who have low levels of such skills (Halford & Behrens, 1996).

Many static risk indicators are easily and reliably assessed, for example parental divorce, whether either partner is under 21 years of age at the time of marriage, and education level all are easily assessed. In contrast, some dynamic risk factors are more difficult to assess. For example, behavioral observation of communication skills detects subtle deficits that predict future relationship problems (Markman & Hahlweg, 1993; Sanders et al. 1999). However, couple self reports of communication are not as sensitive to these deficits. Thus, direct observation of couple communication can be very important in research to establish risk factors.

There is an association between some established static risk indicators and dynamic risk factors. For example, parental divorce and parental aggression in the family of origin is associated with couples having poor conflict management skills at the time of marriage (Halford et al. in press; Sanders et al. 1999). Most people learn intimate communication and conflict management skills within the family of origin (O'Leary, 1988), and having parents who divorced or who were violent is associated with a lesser chance of developing adequate skills. The easily measured risk indicator of parental divorce can be used to identify couples at high risk of relationship problems. Moreover, assessment of this easily assessed risk indicator may be a better means of detecting couples with communication problems than self-reports by the couple about their communication. In other words, the difficult-to-measure risk factor of poor communication can be the target of the relationship education program, but its detection might be through the easily-measured risk indicator of parental divorce status.

Table 1 is a summary of the research reviewed previously, presenting the generic predictors of relationship satisfaction and stability in categories of context, individual partner characteristics, life events, and couple interaction. It is clear that existing evidence provides a substantial ability to predict couples at high risk for relationship dissatisfaction and instability, particularly in the early years of marriage. Marriage and relationship education programs need to be developed and provided in such a manner that all couples, including high-risk couples, are attracted to participate in marriage and relationship education programs. Second, the identified dynamic factors that promote relationship satisfaction and stability should be the key educational goals of marriage and relationship education.

[ top ]

1. The variables identified as risks are correlates of relationship satisfaction, in most cases the causal relationship(s) are unclear, the implications are educated guesses based on likely causal relationships.

Relationship education programs need to include teaching of skills that have been shown to be generic predictors of relationship satisfaction and stability across multiple life circumstances and developmental transitions. In addition, specific programs need to include the specific risk factors that have been shown to be relevant to particular life transitions. Hence, the development of appropriate expectations and conflict management skills seem to be particularly important for couples making the transition into committed relationships. The transition into parenthood will require greater attention to the likely impact of being a parent on the couple's communication, time together, and sexual relationship. In addition, skills would need to be developed which help couples to renegotiate gender roles, sharing of parenting and household responsibilities, and adapting to changing financial circumstances. In the case of step-parenting, particular attention needs to be provided to giving information on the stepparent/stepchild relationship, and helping the partners to negotiate mutually acceptable parenting arrangements. Thus, across a number of different examples relationship education programs need to incorporate some generic factors (e.g., communication skills would probably be an important component of nearly all programs), and some specific skills for coping with particular life transitions.