Universidade Federal do Espírito Santo

This study examined associations between family income, family structure and relationship quality among members of Brazilian families. Participants (n = 77) were mothers with at least one child within the age of zero to nine years-old living in two adjacent cities in the State of Espirito Santo, Brazil. Face-to-face interviews were conducted during home visits. Quality of family relationship was evaluated in four dimensions: general quality of family life, marital conflict, mother-children conflict, and mother-children proximity. All dimensions of family relationship were significantly correlated. Quality of family life was affected by family income, co-residence structure and presence of  siblings. We concluded that the mothers' perception of the relationships among family members is interdependent, and their perception on general quality of life is affected by socio-economic variables.


Palavras-chave: relacionamento intra-familiar; vida familiar; mães; contexto
The major transactive role of the family is to present its members to the social world around them (Francis, 2007) and in this task there is a great complexity of factors involved.
Contemporary studies point toward the multiple roles that parents take and indicate that the expression of heritable traits depends on experience, including specific parental behaviors, as well as predispositions and age-related factors in the child (Collins, Maccoby, Steinberg, Hetherington & Bornstein, 2000).The intra-familiar relationship and the parental care are accompanied and affected by a series of genetics, structural, ecological and other variables.
However, some researchers have given an especial attention to the social variables that may influence the family dynamic (Farver, Xu, Eppe, Fernandez & Schwartz, 2005).Concerning the family structure, for example, the conception of family as composed by two genitors and their offspring is now viewed through more complex lens.The family arrangement known as "traditional model" composed by the couple and their children, with the woman working out of home or not, predominated during a long time in the history of Brazil and it is still strongly maintained.In the last decades, nevertheless, this family structure has been changing and slowly substituted by other models (Goetz & Vieira, 2008).Silva and Tokumaru (2008) categorized family structures in "nuclear family" (composed by father, mother and siblings), "extended family" (father, mother, siblings and other people), "paternal mono-parental family" (cohabitation only with father), "maternal mono-parental family" (cohabitation only with mother), "extended paternal mono-parental family" (cohabitation with father and other people) and "extended maternal mono-parental family" (cohabitation with mother and other people).Aiming to investigate the parents' and other adults' participation in the supply of care to children in two socio-economic contexts, the authors found that the nuclear family was the more frequent structure in their sample, followed by the extended family and by the extended maternal mono-parental family.These data indicate a gradual change in family structure, but it still shows the maintenance of the traditional model in Brazil.Mothers were also found to be the primary caregivers.Tudge et al. (2000), in a cross-cultural work concerning parent's participation in cultural practices with their children, also found the same trend.According to the study, the children in the five countries were more involved in play than in lessons, work, or conversation, and this was unaffected by the presence of either parent.However, parents were relatively less likely to be involved in their children's play than in the other activities.Their findings indicate that mothers were more likely to be found in the same setting as their children and, proportionally, were more likely to be involved with their children than were fathers.Hamilton (2005), with the purpose of examining the association between adolescent's well-being and the presence of non-parental adults in the household, categorized the family structure in: "two-parent", "stepparent", "one-parent", "grandparents" and "other adults" (e.g.aunts, uncle).However, no significant association was found between family structure and adolescent well-being.Instead, the author found that adolescents residing with other adults and an above average number of siblings reported less deviant behavior than their counterparts residing with a below average numbers of siblings.According to Hamilton (2005), a possible explanation may be that in large sibling families, other adults serve a more parental function whether by providing extra attention to children that would otherwise not receive it, or providing economic or social support to parents and/or children.Turagabeci, Nakamura, Kizuki, and Takano (2007) performed a study to analyze the associations between family structure and health.The following definitions of family structures were used: "alone" (single person household); "couple" (with no children or siblings sharing the same household); "nuclear family" (conventional family of parent(s) and child(ren)); and "extended family" (grandparent(s), parent(s) and child(ren) [3 or more generations present]).Some of the results were that subjects living alone were more likely to be ill in comparison to those in extended families.Another one was that subjects living only with spouse or in nuclear family had higher probabilities of mental illness in the absence than in the presence of people showing concern for their well-being.
Those studies illustrate the importance of studying the structure of family cohabitation, the interpersonal relationship quality and the unfolding of these factors to other areas of people's life, as health.The marital relationship has been pointed out as a preponderant factor to the families' quality of life, particularly regarding the relations that fathers and mothers keep with their children.The conjugal adjustment, the communication methods and the strategies used by the couple to solve conflicts influence the development of children's care and the quality of relations among genitors and their children (Emery, Fincham & Cummings, 1992;Kitzmann, 2000).Some studies identified several direct and indirect damages, both for the spouses and for the children, provoked by an unsatisfactory matrimonial relationship.In a study developed by Emery (1982), it was found that children from broken or intact homes characterized by interparental conflict are at a greater risk than children from broken or intact homes that are relatively harmonious.With the same direction, Gottman and Katz (1989) found that children in highly stressed families may have a reduced ability to play with a peer.As a result, these children can never get a chance to rehearse a whole set of social skills with their friends.In contrast, children in low stressed families that present higher levels of peer interaction will have the chance to further develop social skills.
For Grych and Fincham (1990), the impact of a particular episode of marital conflict depends in part on the broader context in which it occurs.Contextual factors are potentially relevant because they provide a backdrop against which episodes of conflict are perceived, and are therefore likely to affect the child's understanding of and response to conflict.In a similar sense, Emery et al. (1992) consider that parenting must be viewed within the larger family and social context just as children's behavior must be viewed within the context of parenting.In their concern about the interpersonal determinants of problematic child behavior, there is a tendency to neglect the social determination of parenting problems.
Using meta-analysis procedures Erel and Burman (1995), Buehler et al. (1997), and Krishnakumar & Buehler (2000) concluded that regardless of causality, positive parent-child relations are less likely to exist when marital relationship is troubled.This proposition suggests that the emotions and tensions aroused during negative marital interactions are carried over into parent-child interactions.Kitzmann's (2000) study, however, highlights the importance of three types of context in understanding conflict-related disruptions in parenting: first, families have their own histories of experience with conflict that influence their reactions to conflict and strategies for dealing with it.Second, interactions in one family subsystem (i.e., marital, parental) impact subsequent interactions in other subsystems; and third, each family member influences others in the current context.
Like most studies, a recent Brazilian research (Stutzman, Miller, Hollist & Falceto, 2009) obtained results that provide evidence that marital conflict has an impact on early child development.They contrast their results to the hypothesis that in Latino populations marital conflict did not affect child outcomes because of characteristics of family organization, as extended family structure.Although the authors did not accessed family structure of cohabitation they suggest that the importance of family cohesion in Latino cultures could amplify the impact of marital conflict over children.
According to Sarrazin and Cyr (2007), children are sometimes not merely the witnesses to their parents' conflicts: they also get trapped into a damaging dynamic in which they are forced to take sides or to bring some support to one parent in detriment of the other.
The authors emphasize that these conflicts make children more susceptible to psychological problems.The studies performed by Sturge-Apple, Davies, Winter, Cummings and Schermerhorn (2008), and Bradford, Vaughn and Barber (2008) showed that these psychological problems are also manifested as learning difficulties and through depressive and anti-social behaviors.Both Micheli and Formigoni (2004), investigating adolescent students from public schools in a Brazilian city, and Carlini-Cotrim and Barbosa (1993), using a national sample of Brazilian students, found associations between drug use and poor family relationships.Nevertheless, drug use may result in behavior alterations which, in turn, can lead to relationship difficulties between young and family (Tavares, Béria & Lima, 2004).
The research of Gottman & Katz (1989) showed that maritally distressed and physiologically under aroused couples have a parenting style that is cold, unresponsive, angry, and low in limit setting and structuring, and this interaction style may relate to anger and noncompliance in their children as well as high levels of stress-related hormones.
According to these authors, children from such homes tend to play less with peers, display more negative peer interactions and have worse health.Brody, Pellegrini and Sigel (1986) performed a study to examine the association between marital relationship quality and the interaction of genitors with their scholar-aged children.The results of this research indicated that the couples who offered reciprocal support and whose matrimonial relations were satisfactory showed higher sensibility in their parental role then the couples whose relations were unsatisfactory.The genitors, therefore, also kept satisfactory interactions with their children and these were equally responsive to the father and to the mother.However, the unsatisfied mothers tended to compensate their children, being more responsive and demanding more from the young.In opposition to this, the satisfied couples tended to show more coherence among themselves and in relation to their children.
In other similar work about the relevance of marital relationship quality to the parental behaviors and to the development of small children in families with two genitors, Goldberg and Easterbrooks (1984) showed that secure child-mother and child-father attachments were most likely to occur in families in which husbands and wives were highly satisfied with their marriages, whereas insecure child-parent attachments were most likely to occur when marital adjustment was poor.In addition, the security of the child-father attachment tended to be related positively to marital harmony.Marital harmony was also related to the children's task competence in a parent-assisted problem-solving situation with the father present (in this case, the extent of interactive harmony (husband-wife) related significantly to positive task affect and good task orientation).The authors demonstrated that both the direct effects (for example, the conjugal adjustment of the husband influencing his attitudes as a father) as the indirect ones (for example, the marital satisfaction of the wife and its influence on her husband's attitudes as a father) were associated to the marital relationship, and to the parental behaviors.
In a paper that reviews the literature concerning parental conflicts and their effects on children, Sarrazin and Cyr (2007) found many studies about divorce and children's well-being.According to this review, many researchers agree that parental conflicts as well as child-parent relationships have more important effects on the child's adjustment than the divorce itself.Moreover, a child's adjustment does not appear to be linked to the type of custody put into place following the divorce, but rather to the quality of the relationships between the parents.The review also shows that parental separation and divorce are not significantly correlated to childhood illnesses.However, childhood illnesses are significantly associated with the level of hostility present in the parents' relationship.Therefore, it is clear that parental conflicts may have a much more devastating effect on children than the marital status of their parents itself.Braz, Dessen and Silva (2008) performed a study that discusses aspects of parental and marital relations' quality of Brazilian families from middle and low classes.Among the results concerning parental care they showed that while families coming from middle class emphasize the transmission of values linked to sociability and affectivity, families coming from low class emphasize the formal education and the transmission of moral values to their children.
The ecological perspective not only emphasizes the potential significance of extrafamilial influences on the child's development, but also stresses the interactive and synergistic, rather than additive and competitive, nature of the links between the family and other influences (Collins et al., 2000).It is important to emphasize that the family does not assume a passive role in relation to the contextual events in which it is inserted, but family interacts with the context, changing it and being changed by it.
An important study performed by Conger, Ge, Elder, Lorenz, and Simons (1994) have demonstrated that many of the deleterious effects of poverty on children's development and family relationship are mediated through the effect of poverty on parenting; economic stress while Serra (the neighbor city) has 0,761.These data are from the year 2000.All participants signed the informed consent to participate in the study.Face-to-face interviews were conducted by three trained interviewers during home visits.

Instrument
The instrument was a structured questionnaire that contained questions about monthly family income; whether parents lived together or apart; duration of the present relationship; age of the family members; number of children; educational level of the parents; whether the mother spent some time out of home daily (studying/working); who take care of the children when mother is out of home; family co-residence structure (described below), the age and sex of the youngest child and a Likert scale to measure the quality of family relationship.
Family co-residence structures were classified in: 1) nuclear family: mother, father and children (other relatives or non-relatives could co-reside); 2) mono-parental family: mother or father and children (other relatives or non-relatives could co-reside); 3) other family: children and relatives and/or other non-related adults (no parents co-residing).
Analysis included, initially, a descriptive and frequency statistics.As data did not present normal distribution, non-parametric statistics was used, including Spearman Correlations for continuous variables, Mann-Whitney tests for comparisons between two independent groups of mothers and Kruskal-Wallis tests for three independent groups of mothers.

Results
We interviewed 77 mothers, 38 living in Serra and 39 living in Vitória.Monthly family income varied from R$102,00 to R$21.800,00 and did not differ significantly between families living in the two cities (Mann-Whitney test, Z = -0,69, p > 0,05).Fifty five couples lived together and twenty two were separated.Duration of the present relationship varied from 6 months to 27 years, with a mean of 10,46 years.The age of the mothers varied from 17 to 51 years old (mean = 32,5), the age of the fathers varied from 19 to 59 years old (mean = 36,4) and the age of the youngest child varied from 0 to 9 years old (mean = 3,7).The number of children varied from 1 to 9 although most mothers (89,6%) had from 1 to 3 children.
Thirty seven of the 77 mothers reported to spend some time working or studying out of home daily.Sixty mothers reported to leave their children with other people when they go out.From these, 33 mothers left the children with relatives and 27 left them with nonrelatives.
The most frequent family co-residence structure in our sample was the nuclear tumultuous family life was perceived by the mothers.Mothers living in mono-parental families perceived the quality of family life as more tumultuous (M = 4,0) than mothers living in nuclear families (M = 2,93; Wilcoxon's Z = 2,09, p < 0,05).Mothers whose youngest child had siblings also perceived the quality of family life as more tumultuous (M = 3,56) than those whose child did not have siblings (M = 2,57; Wilcoxon's Z = 2,03, p < 0,05).
We also analyzed the mother-child conflict and proximity for mothers that had two (N = 25) and three (N = 16) children.There were no significant differences in the relationship of the mothers with their first and their second child (Wilcoxon test, conflict Z = -1.33,proximity Z = -1.22),or in the relationship of the mothers with their first, second or third child (Friedman test, conflict χ 2 = 3,75, proximity χ 2 = 2,0).

Discussion
This study examined the associations between family income, family structure and the relationship quality among members of Brazilian families.The four dimensions of family relationship measured were significantly correlated.We interpret this result as an indication of the validity of the questions used to measure the dimensions of family relationship.Some of these findings are consistent with previous studies, for example: the positive correlation between the quality of marital relationship and of parent-children relationship (Bradford et al., 2008;Buehler et al., 1997;Emery, 1982;Emery et al., 1992;Erel & Burman, 1995;Gottman & Katz, 1989;Grych & Fincham, 1990;Kitzmann, 2000;Krishnakumar & Buehler, 2000;Sarrazin & Cyr, 2007;Sturge-Apple et al., 2008).
Our results expand these findings.Parent-child relationship was here measured in relation to conflict and proximity.The correlation between these two measures in the expected direction and of each one with the general quality of life and with marital conflict, also in the expected directions, indicate that these measures should be considered as dimensions of parent-children relationship and be evaluated separately in future studies.
Although correlated between them and with the other variables of family relationship, these dimensions were not affected by the other variables considered.This was also true for the marital conflict dimension.Although related to the other dimensions of family relationship it was not affected by the other variables considered.
The only dimension of family relationship affected by some of the other variables considered was the general quality of family life.One possible interpretation of these results is that although the dimensions measured are interrelated, the dimensions of mother-children conflict and proximity and marital conflict were perceived as more specific measures of family relationship than the general quality of family life.This dimension seems to have been interpreted as a more global measure related both to the dimensions of family relationship and to other contextual variables.We are suggesting that one's consideration of his/her general quality of family life includes both the evaluation of the relationship among family members and the evaluation of other contextual variables.
In the present study, three contextual variables showed impact over the general quality of family life: family income, co-residence structure and the presence of siblings.
The negative correlation between the general quality of life and family income can lead to the conclusion that shortage of resources is linked to the perception of a more tumultuous life.Nonetheless, this conclusion should be seen with caution as it can be the result of other relations.Income has been showed to be related to other contextual factors as educational level, place of residence, exposure to violence and access to health care (Gawryszewski & Costa, 2005;Giatti & Barreto, 2006;Macedo, Paim, Silva & Costa, 2001, for example).These variables were not analyzed in the present study and their effects over the perception of the general quality of life cannot be ruled out.
Mothers living in mono-parental families perceived the quality of family life as more tumultuous than mothers living in nuclear families.This result may be related the availability of social support for the mother.Some studies demonstrated that among the relatives the husband was pointed out by the mother as the main source of support while maternal grandmothers occupied a secondary place (Belsky, 1981;Bronfenbrenner, 1986;Levitt, Weber & Clark, 1986).
The number of siblings also correlated with the quality of family life.Mothers whose youngest child had siblings also perceived the quality of family life as more tumultuous than those whose child did not have siblings.The hypothesis for the understanding of this result can also be related to the network of support.A Brazilian study performed by Dessen & Braz (2000) indicated the permanence of values related to the gender roles within the family dynamics.According to these authors, the majority of women did not work out of home and the responsibility of providing and sustaining the family was assigned to men while the women took the responsibility for the caring and rising of the children and the organization of the house.Although in the present study the number of mothers who worked out or studied (37) was not far from the number of mothers who remained home with the children (40) the correlation between the presence of siblings and the perceptions of a more tumultuous life may indicate an overload in parental responsibilities to the mother.The results of Tudge et al. (2000) also showed that fathers spent less time in activities with their children than mothers.
This may be an important variable to the understanding of the perception of the mothers about the general quality of family life.Interestingly, the number of children did not correlate with the other dimensions of family relationship, a result that support our view that mother-children conflict and proximity are dimensions of relationship while the general quality of life involves the perception of other variables.

Conclusion
Our results showed that the mothers' perceptions of the relationships among family members are interdependent.Conflict with the children and the partner are positively related while both measures are negatively related to children proximity.We also showed that their perception of the general quality of life is not restricted to their perception of the relationships among family members.Other contextual variables as family income, co-residence structure and number of siblings compose their perception of their general quality of life.We suggest that future researches investigate further these relations isolating each one of these variables and searching for the contribution they have to the mothers' perception of their general quality of life.