The relationship between family-of-origin and marital satisfaction

1 ABSTRACT The aim of this article is to explore the relationship between marital satisfaction and family-oforigin factors amongst couples with children. Locke and Wallace’s Marital Adjustment Test and the McMaster Family Assessment Device were used to measure marital satisfaction and familyof-origin factors respectively. Family-of-origin factors, such as problem solving, communication, roles, affective involvement and behavioural control were investigated. The sample consisted of 47 married couples. A signifi cant relationship was found between roles and affective responsiveness as family-of-origin factors and marital satisfaction, while roles as a family-of-origin factor played an important role in the wife as well as her husbands’ marital satisfaction. The fi ndings emphasise the importance of functioning in the family-of-origin as a potential determinant of future marital satisfaction.


INTRODUCTION
It is cause for concern that approximately half of all fi rst marriages end in divorce (Bradbury, Fincham & Beach 2000:964) and that the level of satisfaction in established marriages has been declining continuously since the seventies (Rogers & Amato 1997:1098).Snyder and Lopez (2005:429) asserts that even in happy marriages there is a gradual decline in satisfaction within the fi rst decade of matrimony, and thereafter marital satisfaction continually decreases.The extent of marital discord and confl ict is also refl ected in alarmingly high divorce statistics in South Africa that reveal that almost two out of three marriages end in long-term separation or divorce (South African Statistics 2004).'Marriages exist in highly complex, multifaceted environments, of course, and a full understanding of how these environments interact and impinge upon marriage is just beginning to develop' (Bradbury et al. 2000:969).Combinations of interpersonal factors (such as marital problem-solving and affective expression) and broader contextual processes (including the dynamics of the family-of-origin and family life cycle and economic and work-related stressors) have been identifi ed as potential determinants of the quality of marital relationships (Halford, Kelly & Markman 1997:4).However, most of the existing studies are conceptualised within a pathogenic paradigm emphasising factors that erode marital satisfaction.As it is widely recognised that a fulfi lling marriage is not merely characterised by the absence of marital dissatisfaction (Bradbury et al. 2000:973), more information is needed on the unique qualities that contribute to marital satisfaction.The study on which this article is based differed from existing studies on family-of-origin and marital satisfaction with respect to the fortogenic focus of the study.A fortogenic focus highlights human strengths, resiliency factors and optimal human experiences such as happy marriages opposed to the pathogenic focus that focuses on dysfunctions, risk factors and human suffering such as marital confl ict and divorce.In the current article, the authors integrated the fortogenic perspective with the multigenerational perspective of adjustment in the family-of-origin and how this adjustment affects the experience of marital satisfaction of couples.Therefore, this article focuses on factors that promote marital satisfaction.Jacobson and Addis (1993:8) are of the opinion that interventions promoting marital satisfaction must occur when the potential for confl ict is greater.Every phase in the life cycle presents various demands and accompanying confl icts to couples (Language 1998:14).However, marital satisfaction tends to decline over time, and as most couples with children will have been married for some time, this age group can provide better information than a newly-wed sample.
Numerous studies have shown the infl uence of family-of-origin on various aspects of peoples' lives, including relationships (Conger et al. 2000;Whiston & Keller 2004:493).Thus, family-of-origin variables may be a fruitful area to explore with regard to marital satisfaction.In this regard, the McMaster Family Assessment Device will be useful, as it is a pragmatic, all-inclusive model of family assessment.This article explores the relationship between spouses' marital satisfaction and their appraisal of the quality of the relationship of their family-of-origin with regard to McMaster's six dimensions of family structure and family interaction (Epstein, Baldwin & Bishop 1983:171), including communication, problem solving, roles, affective responsiveness, affective involvement and behavioural control.The cross-effects of the impact of the family-of-origin, that is the influence of the wife's family-of-origin experiences on her husband and vice versa, will also be determined.

Marital satisfaction in couples with children
Marital satisfaction is the degree of happiness, everything considered, in one's marriage (Locke & Wallace 1959:252).Investigations that gather information pertaining to the factors that influence marital satisfaction have become prevalent since the nineties (Bradbury et al. 2000:970).Marital satisfaction is considered an important determinant of an individual's general well-being (Larson & Holman 1994:231).According to Snyder and Lopez (2005:67), it appears that happily married couples experience better well-being and report less feelings of depression.Therefore, in the quest for well-being and happiness it is important to determine the factors that influence marital satisfaction.Goldenberg and Goldenberg (2002:10) report that married couples are divorcing earlier than ever before with 38% of couples divorcing within four years and 50% within seven years.Couples with young children therefore seem to be at increasingly greater risk for marital dissatisfaction.It appears that, while children increase the stability of a marriage, they also decrease marital satisfaction (Waite & Lillard 1991:950) because couples with children have less time to spend on their marriages (Cavanaugh 1993:345).Productivity in their careers and involvement in the family or community are important tasks for couples with children, and the approach used by each spouse to achieve this productivity may lead to conflict (Goldenberg & Goldenberg 2004:27).Given that true intimacy is established only after the age of 30 (Goldenberg & Goldenberg 2004:27), it would appear that, during this time, relationships experience either a gradual drifting apart or an increase in intimacy (Glick & Kessler 1980:47).

The influence of family-of-origin on marital satisfaction
It is widely accepted that people are shaped by experiences within their families of origin (Conger et al. 2000:3;Goldenberg & Goldenberg 2004:455).According to Hovestadt et al. (1985:289) the family-of-origin is seen as the unit in which the individual has his or her physiological and emotional beginnings and in which the person spent most of his or her childhood years.When a couple experiences a lack of marital satisfaction, it often means that one or both of the marriage partners still struggle to deal with issues stemming from the family-of-origin (cf.Sabatelli & Bartle-Haring 2003:161).The marriage system is by definition not a closed system and it can be expected that functioning within this system will be influenced by all the other systems to which the individual belongs.
The relationship between marriage partners is not determined only by the functionality of the family as a whole (Freeman 1992), but various studies have indicated that marital problems between parents also influence their children's future relationships (Doucet & Aseltine 2003:819;Sabatelli & Bartle-Haring 2003:160). De Wet (1985) and Naude (1990:18) reported South African studies confirming the negative impact of the family-of-origin on the ability to deal effectively with the roles and responsibilities in intimate adult relationships.Naude (1990:30) studied the marriages of South African couples and mentioned that transference from childhood can have a negative influence on marital satisfaction.Research by Larson and Holman (1994:230) confirms that the processes and interactions in the family-of-origin predict the quality of a marriage more accurately than a single incident, for example divorce.Factors that influence the quality of a marriage include frequent conflict in the family-of-origin (Doucet & Aseltine 2003:835), the extent to which communication skills are acquired in the family-of-origin (Sanders, Halford & Behrens 1999:10), and the extent to which the backgrounds of the respective families of origin differ, resulting in greater adjustments that both partners have to make (Du Rand 1991:71).Sabatelli and Bartle-Haring (2003:159) were of the first researchers to study the cross-effects of family-of-origin factors.They found that the wife's family-of-origin experiences influence both her and her husband's marital adjustment.However, the husband's family-of-origin influences only his adjustment.Further, it became apparent from the research by these authors that the influence of the family-of-origin on marital adjustment extends into middle age.
The majority of the above-mentioned studies have been conceptualised within a pathogenic paradigm.These studies frequently focus on parental divorce (Amato 1996:628;Sanders et al. 1999:3), dysfunctional attachment (Klohnen & Bera 1998:219), lack of support from parents and parents-in-law, and dysfunction within the family (Larson & Holman 1994:230).

McMaster model of family functioning and family-of-origin
The present study differs from the aforementioned studies on marital satisfaction in its fortogenic approach as well as the use of a different theoretical approach to study family-of-origin factors, namely the McMaster model of family functioning (Epstein et al. 1983:445).Several studies have been conducted to investigate the relationship between roles and marital satisfaction, but other researchers rarely define roles in the same way as the McMaster model does.The McMaster model of family functioning is a pragmatic, all-inclusive model of family assessment and treatment that was developed over a period of 30 years (Miller et al. 2000:169).The model has six dimensions that are used to describe family structure, organisation and transactional patterns (Epstein et al. 1983:448).These dimensions include problem solving, communication, roles, affective involvement and behavioural control.Epstein and Bishop (1991:448) contend that healthy families do not necessarily have fewer problems than dysfunctional families, but they maintain that healthy families are more adept at solving their problems.Freeman (1992) supports this viewpoint and emphasises that love, care and commitment are not entirely absent in unhappy marriages.
Research that makes use of the McMaster model has dealt mostly with mood disorders, medical conditions, children and therapeutic treatment (Miller et al. 2000:181).The six dimensions of McMaster's model have not yet been studied together in relation to marital satisfaction.Furthermore, studies on aspects found in McMaster's model, such as behavioural control, affective involvement or affective responsiveness and marital satisfaction, could not be traced.This is understandable, as these terms are generally not used outside of the McMaster's model.Therefore, comparisons between different research studies should be approached with caution.
Although the McMaster model is based on a problem-centred systems approach, it also emphasises the optimal development of the individual (Epstein & Bishop 1991:447) and deems it important to understand both problems and strengths of the family system.A fortogenic approach would thus not be in conflict with the underlying principles of this model.The current article attempts to determine which of the six dimensions (problem solving, communication, roles, affective responsiveness, affective involvement and behavioural control) influence marital satisfaction.There are no recent South African studies that approach marital satisfaction in terms of familyof-origin factors, according to a search conducted on the SA Studies, Academic Search Premier and PsycINFO databases on August 8, 2008.However, Sabatelli and Bartle-Haring (2003:163) have studied these factors among American middle-aged couples.While this article aims to determine the cross-effects as determined in the above-mentioned study, different research instruments will be employed.Regardless of this, it remains meaningful and important to compare the results of the two studies.
The purpose of this study is firstly to explore the applicability of the McMaster model of family functioning in determining a relationship between couples' level of marital satisfaction and their own family-of-origin factors.The above-mentioned relationship will be explored for the group as a whole and for males and females separately.Secondly, the influence of the relationship of the husband's family-of-origin on the wife's marital satisfaction, as well as the relationship of the wife's family-of-origin factors on the husband's marital satisfaction, will be investigated.The target group of the study on which this article is based comprised white couples with children.

METHODS
A non-experimental research design was employed during the study reported here (Huysamen 1993:26).In order to determine the relationship between the variables, the following was done: Firstly, the relationship between the total group's level of marital satisfaction (criterion variable) and their own family-of-origin factors (predictor variable) was investigated.Following that, these relationships were investigated independently for males and females.Secondly, the relationship between the husband's family-of-origin factors and his wife's marital satisfaction, as well as the relationship between the wife's family-of-origin and her husband's marital satisfaction, was investigated.

Measuring instruments
A biographical questionnaire was used to obtain demographic information such as age, gender, marital status of parents and number of siblings as well as information about current family size, number of previous marriages and number of years married.Locke and Wallace's (1959) 15-item Marital Adjustment Test was used to measure marital satisfaction (Fincham & Bradbury 1993:444).An individual's score is obtained by totalling the raw scores.A high score indicates better adjustment, with a maximum score of 158.The items include a variety of dimensions of marital functioning, among others management of finances, conflict, relationship with one's in-laws, leisure time, friendship, sexual relationship, life philosophy and trust (Schutte & Molouff 1995).According to Fincham and Bradbury (1993:444), the instrument has a split-half reliability of 0.90 and is considered a widely used, reliable and valid measure in spite of the age of the measuring scale.
The McMaster Family Assessment Device was used to measure family functioning in the following six areas (Miller et al. 2000 The questionnaire consists of 60 items.An average score is determined after the raw scores of the items have been totalled.
A minimum score of 1 and a maximum score of 4 is possible for each scale.A higher score is indicative of greater dysfunction.
In an American sample, alpha-coefficients ranging from 0.72 to 0.92 for the various sub-scales have been found by Epstein et al. (1983:176).
The internal consistency of all the subscales was subsequently investigated for the current sample, and Cronbach's α-coefficients (SPSS Incorporated, 2001) are reported in Table 1.
The ά-coefficients show that the majority of the variables have high levels of internal consistency.Only affective involvement (0.60) and behavioural control (0.649) show relatively low levels of consistency, with coefficients lower than 0.7.However, according to Nunnally and Bernstein (1994:251) ά-coefficients higher than 0.6 are acceptable for non-cognitive constructs.

Sample
A snowball sample (Huysamen 1993:46) of 47 married couples was obtained with the assistance of pre-primary school principals, ministers and business managers in Bloemfontein, a South African city with approximately 300 000 inhabitants.The respondents were selected from a homogeneous group of white, married couples (with both partners between the ages of 29 and 40) who had at least one child.Married couples with children were targeted as statistics and proved these couples to be at risk for marital dissatisfaction.
The couples had been married for an average number of nine years, with a standard deviation of approximately 3.5 years.They had an average of two children and an average of 2.5 siblings (family-of-origin).The average age was 34.57years (men) and 33.28 years (women).The group consisted primarily of highly educated people, with 71 individuals (75.5% of the total group) coming from the professional, semi-professional, technical, managerial and executive occupational group.

Data collection and statistical analysis
The questionnaires were administered in the home language of participants.The respondents were ensured of the confidentiality of all information.Both measures were self-report questionnaires and were completed individually by respondents after they had received detailed instructions from the test administrator.Furthermore, each couple was afforded the opportunity to receive feedback on their questionnaire.Questionnaires were distributed to 90 couples, but only 47 were returned.The response rate was 52%.
Pearson's product moment correlations were calculated to investigate the relationship between the variables.Furthermore, to determine whether there is a statistically significant difference in this relationship in terms of gender, Fisher's r-to Z-transformation was used (Howell 2007:681).Besides determining the simple relationship between the family-oforigin factors and marital satisfaction, a hierarchical regression analysis technique was used to determine the amount of variance in the criterion variable (marital satisfaction) accounted for by the predictor variables (family-of-origin factors).

RESULTS AND DISCUSSION
The descriptive statistics (means and standard deviations) for both the criterion and the predictor variables are indicated in Table 1.
The average marital satisfaction scores for men and women indicate relatively high levels of marital satisfaction within the sample (the maximum is 158).In addition, it appears that functioning with respect to most family-of-origin factors is good, given that the scores are closer to 1 (healthy functioning) than to 4 (dysfunction).For both males and females, functioning with respect to problem solving was the weakest and behavioural control was the best.Moreover, the scores of both groups on the general functioning scale indicate moderate adjustment.

Relationship between marital satisfaction and family-of-origin factors
Using Pearson's product moment correlation to indicate the relationship between marital satisfaction and family-of-origin factors, the following was found: As can be inferred from Table 2, significant correlations between roles and marital satisfaction (on the 1% level) were found for the total group, as well as between affective responsiveness and marital satisfaction (on the 5% level).Both coefficients were negative; therefore, the more clearly the roles within the family-of-origin were defined, and the more apt their emotional responsiveness was, and the higher the level of marital satisfaction the person experienced.When the group was examined in terms of gender, no statistically significant relationships between family-of-origin factors and marital satisfaction were found for men.In the case of women, the roles scale is significantly correlated to marital satisfaction (1% level).As was the case with the total group, this is a negative relationship; hence, the more clearly the roles within the familyof-origin were defined, the higher the level of marital satisfaction the wife experienced.On examination of the effect sizes of these correlations, it appears that they have medium to large effect sizes and are thus of moderate practical significance.According to the calculated z-values, there are no significant differences between men and women in terms of the relationship between family-of-origin factors and marital satisfaction.
In order to determine the extent to which family-of-origin factors (predictors) accurately predict marital satisfaction, a multiple regression analysis was performed for the total group and thereafter by gender.The results are shown in Table 3.
The R² value for the total group was 0.1562 and is significant on the 5% level (F7;85 = 2.25).Thus, it can be deduced that 15.62% of the variance in marital satisfaction for the total group is accounted for by the predictors.The multiple regression analysis by gender indicates that for women a significant percentage (F7;39 = 2.64) of variance in marital satisfaction is accounted for by family-of-origin factors.According to the R² value, 32.12% of the variance is accounted for.For men, family-of-origin factors only accounted for 17.31% of the variance in marital satisfaction and the result was not statistically significant.For the total group and for the women, the results were further investigated by a hierarchical regression analysis.The results for the total group appear in Table 4.
It is clear that only one family-of-origin factor, namely roles, is significant on the 1% level.Taking the guidelines for effect sizes into account (see f²), it is clear that this result has a medium effect, indicating that it is of moderate practical significance.Therefore, it appears that when the family-of-origin defines its roles clearly, later in life it makes a positive contribution to marital satisfaction of children who grew up within that family.
A hierarchical regression analysis was preformed for the women.
The results appear in Table 5.
On examining the contribution of the individual variables to R² for women, it is clear that two family-of-origin factors, namely roles (on the 1% level) and affective involvement (on the 5% level) contribute significantly to the percentage of variance accounted for in marital satisfaction.Taking the guidelines for effect sizes into account, it is clear that affective involvement has a small to medium effect, whereas roles have a large effect.The extent to which the roles within the family-of-origin are clearly defined thus plays a prominent role in the marital satisfaction of women.Moreover, it appears that affective involvement (the extent to which the family-of-origin was genuinely interested in the woman's life) makes a positive contribution to her marital satisfaction.Pearson's product moment correlation coefficient was calculated to investigate the cross-effects of the relationship between the wife's family-of-origin factors and her husband's marital satisfaction, as well as the husband's family-of-origin factors and the wife's marital satisfaction.The results appear in Table 6.
From Table 6, it appears that the correlation between roles as the wife's family-of-origin factor and the marital satisfaction of her husband is significant on the 5% level.Approximately 9% of the variance in the husband's marital satisfaction is accounted for by roles as his wife's family-of-origin factor.The coefficient is negative, indicating that the better defined the roles in the wife's family-of-origin were, the greater the level of satisfaction her husband will experience in their marriage.The corresponding effect of this correlation was of medium size; therefore, the result is of moderate practical value.No significant relationship was found between any of the husband's family-of-origin factors and his wife's marital satisfaction.
A multiple regression analysis was performed to explore the extent to which the wife's collective family-of-origin factors could successfully predict the marital satisfaction of her husband, and also the extent to which the husband's collective family-of-origin variables could successfully predict the wife's marital satisfaction (see Table 7).However, no significant crosseffects could be found for the husband or wife's family-of-origin factors and the marital satisfaction of their partners.Given that no significant cross-effects were found, no further analysis was conducted.

CONCLUSION
In this study, it was found that there is a relationship between certain family-of-origin factors and marital satisfaction.Sabatelli and Bartle-Haring (2003:160) answered the question regarding  whether family-of-origin factors can affect adjustment in marriage, even into middle-age.Consistent with their findings, it was found that the patterns of interaction established in the family in which the individual grew up could play a role in marital satisfaction well into adulthood.
A relationship was found between roles and marital satisfaction.Furthermore, roles accounts for a significant amount of the variance in marital satisfaction in middle age.The more clearly the roles in the family-of-origin were, the higher an individual's marital satisfaction is likely to be (Epstein et al. 1983:172).These roles include the provision of resources, support and care, the sexual satisfaction of the parents, the management of systems and the acquisition of life skills (Epstein & Bishop 1991:460).
It may be that children in such families experience a secure environment (i.e.their needs are met) and that the skills they acquire in their families enable them to choose good/appropriate marriage partners.They may also learn at an early age that each family member has certain responsibilities and functions to fulfil.Therefore, they are able to meet the needs of their partner in a marriage and assume responsibility for establishing marital happiness.Furthermore, there is a relationship between affective responsiveness and marital satisfaction.If the individual learns to express a full range of emotions in an appropriate manner in his or her family-of-origin, his or her marital satisfaction will also be higher (Epstein et al. 1983:173).The appropriate expression of emotion encourages communication, couples continually reaffirm their love and appreciation for each other, misunderstandings are cleared up and problems are solved.
It is surprising that no significant relationships were found between any of the other family-of-origin factors and marital satisfaction, given that various researchers have already shown that both problem solving and communication skills play an important role in the success of a long-term relationship (Conger & Conger 2002:372;Kaslow & Robison 1996:155;Sanders et al. 1999:6).Although these researchers did not necessarily direct their studies to family-of-origin factors, the skills that they investigated are learnt in the family in which the individual grew up (Du Rand 1991:71).The use of a small homogenous sample could have affected the results in that a more diverse sample would have evidenced more family-of-origin factors playing a role in marital satisfaction.
If further appears that there is a relationship between roles as the wife's family-of-origin factor and her marital satisfaction.Both roles and affective involvement account for a significant percentage of the variance in a wife's marital satisfaction.The extent to which roles were clearly defined in her original family, and the extent to which her family was genuinely interested in her activities contribute to her marital satisfaction.According to Bartle-Haring and Sabatelli (1998:909), there is a direct relationship between interaction within the family-of-origin and marital adjustment.It may be that their genuine interest in her enhanced her sense of self-worth and that she therefore has more self-confidence in her ability to make her own marriage a success.Affective involvement possibly plays a role in marital satisfaction because the wife's parents or family still supports her and remains involved in her life.Her family-of-origin is thus an important social support system that can contribute to her satisfaction in marriage.A wife's family-of-origin accounts for a significant percentage of variance in her marital satisfaction.
The results of the current research differ from those of Sabatelli and Bartle-Haring (2003:166) in the sense that no significant relationship was found between the husband's family-oforigin and his marital satisfaction.Sabatelli and Bartle-Haring's sample also consisted of educated white people.It was an American group, however, and their average age was 49 years (approximately 15 years older than the current sample).
It was found that a correlation exists between roles as the wife's family-of-origin factor and the husband's marital satisfaction.
The fact the wife learnt in her family-of-origin to fulfil family functions effectively may contribute to her being able to meet the needs of her husband, thereby increasing his satisfaction.
The implications for practice are twofold: On the one hand, interventions with respect to the family-of-origin factors, more specifically the defining and assigning of roles, increase both the husband's and the wife's satisfaction.On the other hand, parents should be informed about the contribution they can make to ensure the future marital satisfaction of their children by ensuring that healthy family functioning is promoted.Aside from roles, the expression of appropriate emotions and genuine interest is also important and should be encouraged in family therapy.
In the light of the number of divorces in South Africa, it is important to work both preventatively (especially in critical periods such as within the first seven years of marriage) and to ensure effective interventions for couples who undergo marital therapy.Much research regarding the marriages of couples with children is still required, and factors that contribute to effective interventions may become clearer.
The study contributed to research in that factors that promote marital satisfaction and well-being were identified.The importance of the family-of-origin, especially in the marital satisfaction of the wife, has been highlighted.There is a lacuna in recent research dealing with family-of-origin factors in South African couples, despite the fact that research overseas indicates that such factors can no longer be ignored (Doucet & Aseltine 2003:819;Larson & Holman 1994:228).
Limitations of the current study include the use of a small snowball sample, the use of self-report measures and the crosssectional design.Not only do these factors pose a threat to internal validity, but also the sample may not be representative of the cohort (Huysamen 1993:102).Due to the small sample the results cannot be generalised, but at least the model is confirmed within the South African context in the absence of recent local studies concerning family-of-origin.However, it cannot be denied that more comprehensive studies would be invaluable in this field.
Future studies can replicate and expand on the findings of the current research.The sample consisted of highly educated and predominantly professional people.According to Rogers andAmato (1997:1098), well-qualified people will experience greater marital satisfaction.The influence of family-of-origin factors in middle-and lower-class populations can be investigated in the future.It will also be important to conduct cross-cultural research because the effect the family-of-origin has on the individual may differ from culture to culture.

TABLE 1
Marital satisfaction and family-of-origin factors for the total sample and by gender

TABLE 3
Results of the multiple regression analysis for the total group and by gender

TABLE 4
Contributions of the individual variables to R² for the total group

TABLE 5
Contributions of the individual variables to R² satisfaction for women