Children’s socioemotional competence, the ability to engage appropriately with interactional partners, self-regulate, process social information, and understand emotions, is fundamental to their concurrent and prospective outcomes in multiple domains of functioning (Bohlin & Hagekull, 2009). Socioemotionally competent children tend to experience more positive relationships, feelings of connectedness, higher academic achievement, and better mental health outcomes (Blair, 2002; Eisenberg et al., 2005; Graziano et al., 2007). Crucially, socioemotional competence from as early as kindergarten has been linked with outcomes in education, mental health, employment, substance use and criminal behaviour 13 to 19 years later (Jones et al., 2015). In contrast, inadequate or delayed socioemotional development may result in poor inter-personal relationships, low academic achievement, school dropout (Eisenberg et al., 2005; Hill & Craft, 2003; Nelson et al., 1999) and poorer mental health in adolescence and adulthood (Bornstein et al., 2010).
Thus, an important research goal is to identify determinants of children’s socioemotional development during early childhood with the aim of devising interventions that can promote healthy child outcomes. A focus on early childhood is important as this is a time during which children’s social skills undergo rapid development and are most susceptible to environmental influence (Eisenberg et al., 2010). According to Bronfenbrenner’s ecological model of development (Bronfenbrenner (1979)), early in life, the environmental layer that is most salient to the child is usually their nuclear family. Thus, it is essential to uncover family-level factors that impact child socioemotional outcomes.
Child Socioemotional Development within the Family Context
According to the tripartite model of the family’s role in child socioemotional development (Morris et al., 2007), children develop socioemotional competencies through three mechanisms. First, children can develop such competencies through exposure to parenting practices. For instance, children’s socioemotional development benefits when parents display positive affect, respond to their children’s bids for attention and actively regulate children’s emotions (e.g., Eisenberg et al., 1999; Morris et al., 2007; Rutherford et al., 2015). Second, children may learn through observation. That is, parents may promote children’s socioemotional development by modelling appropriate responses to emotional situations and adequate self-regulation strategies (Blair et al., 2004). However, there is a third, relatively overlooked, mechanism through which parents may influence child development. According to family systems theory (Minuchin, 1974), family relationships are interdependent, and each dyadic relationship influences all other relationships in the family. In line with this theory, the emotional climate of the family such as the quality of the interparental relationship may represent a third mechanism through which children’s socioemotional competencies develop. That is, relationship dissatisfaction within the spousal subsystem is likely to spill-over into the parental subsystem and impede a parent’s responsiveness and attunement to their child, which are critical to optimal child development (e.g., Fletcher, 2009; Kwok et al., 2015; Salo et al., 2022). Parents who are preoccupied with their spousal relationship are less equipped to attend to their child’s needs (Joyce, 2016). Conversely, parents who share harmonious relationships are also more likely to function effectively as co-parents and thereby foster child development. Equally possible is that children develop maladaptive coping strategies against negative internal states provoked by parental conflict and hence their social development derails (Cummings & Davies, 2010). Therefore, understanding how the positivity and negativity that parents experience in relation to each other can ‘spill over’ into the parent-child relationship may broaden our understanding of child socioemotional development within the family context.
Quality of Interparental Relationship and Child Socioemotional Development
The interparental relationship is an important subsystem of the family unit (Cox & Paley, 1997). The quality of the interparental relationship has implications not only for the mother and father, but also for all other members of the family including children. Interparental conflict has been shown to exert direct and indirect effects on children’s psychological and behavioural wellbeing (Davies et al., 2015; Harold & Sellers, 2018; Linville et al., 2010; Zhou et al., 2017). For instance, Salo and colleagues (2022) found that low maternal and paternal relationship satisfaction from as early as pregnancy is linked with child behavioural problems at two-years-old (Salo et al., 2022). Similarly, Katz and Gottman (1993) reported that observed parental mutual hostility when children were aged five years was linked with externalising problems in children at eight years (Katz & Gottman, 1993). In particular, fathers who demonstrated anger and withdrawal towards mothers at child age five years had children with greater internalising problems at eight years (Katz & Gottman, 1993). Furthermore, in families of six- to eight-year-old children, couple negativity has been linked with family negativity following both pleasant and conflictual discussions (Kitzmann, 2000). Crucially, mothers’ and fathers’ (dis)satisfaction early in life can exert effects on child outcomes that persist as children grow. For instance, Lindblom and colleagues (2017) found that the quality of interparental relationship satisfaction during the perinatal period (birth to 12 months) predicts children’s anxiety and depression at seven- to eight-years-old (Lindblom et al., 2017). Panula and colleagues (2020) similarly found that mothers’ relationship dissatisfaction during the 12th week of pregnancy is associated with lower social competence (increased impulsivity and disruptiveness) in three-year-olds. As family systems theory suggests, it is likely that the hostility and conflict experienced in the spousal subsystem extents into the parent-child subsystem by impinging upon effective parenting such as a parents’ emotional availability and responsiveness (Sturge-Apple et al., 2006). In fact, recent evidence suggests that marital conflict is associated with poorer social skills among five- to six-year-olds through negative parenting (Hosokawa & Katsura, 2017). These findings demonstrate the spill-over effects that the couple relationship may have on the entire family system and highlight the relevance of interventions aimed at targeting the interparental relationship.
The evidence highlighting the role of the interparental relationship for child development is convincing. Nevertheless, our understanding of how parent relationship satisfaction relates to child socioemotional development remains incomplete. First, although previous studies have focused on interparental conflict and its negative impact on child outcomes, far fewer studies have examined interparental satisfaction and its beneficial role for child development. Yet the little existing evidence suggests that marital harmony is linked with better emotion regulation in toddlers (Porter et al., 2003) and relationship satisfaction is linked with fewer child behavioural problems, controlling for parenting behaviour (Linville et al., 2010). Presumably, relationship satisfaction, characterised by dyadic adjustment, displays of positive affect, and constructive behaviours (Spanier, 1976) are conducive to a positive emotional climate, and consequently to children’s socioemotional development (Morris et al., 2007). Moreover, children who have the opportunity to observe their parents engage in positive behaviour with each other, and adequately regulate their emotions, may adopt strategies for self-regulation and learn to articulate their internal states.
Second, research has mainly focused on one parent’s report of relationship satisfaction (e.g., Graham et al., 2010; Linville et al., 2010). The limited number of studies that have included both mothers and fathers have conceptualised relationship satisfaction at the family-level by either summing or averaging couples’ scores (Sturge-Apple et al., 2004) or loading mothers’ and fathers’ scores onto a single latent construct of couple satisfaction (Kaczynski et al., 2006). Yet, mothers and fathers may have dissimilar perceptions of the quality of their couple relationship (Kamp Dush et al., 2008). Including both parents is imperative especially given that each parent’s relationship satisfaction may differentially influence children’s outcomes (Korja et al., 2016). For instance, Korja and colleagues (2016) found that mothers’ high relationship satisfaction during the prenatal period was associated with better overall family relationships during infancy whereas no such association was found for fathers. However, Salo and colleagues (2022) found that mothers’ relationship satisfaction during pregnancy is associated with child behaviour problems at 12 months through parental reflective functioning whereas fathers’ relationship satisfaction is directly linked with social competency and behaviour problems. Moreover, although the quality of the inter-parental relationship has been found to influence both mothers’ and fathers’ parenting behaviour, associations are stronger for fathers (Belsky et al., 1991; Goeke-Morey & Cummings, 2007; Pedro et al., 2012). Similarly, paternal, but not maternal, marital satisfaction has been linked with greater social referencing towards fathers in infants (Dickstein & Parke, 1988).
Third, although longitudinal studies exist (e.g., Katz & Gottman, 1993), most have conceptualised interparental satisfaction as a static construct. However, parental perception of relationship quality is dynamic and likely to fluctuate over time, although the direction of change remains unclear (for a review see Karney & Bradbury, 2020). On the one hand, it has been suggested that couple relationship satisfaction tends to decrease over time (Karney & Bradbury, 1995). However, other evidence suggests that only a fraction of couples experience this decline, while the majority undergo minimal or no change (e.g., Lavner & Bradbury, 2010). Among parents, findings have been more consistent. Meta-analytic findings suggest that compared to non-parents, parents experience less relationship satisfaction (Twenge et al., 2003). Similarly, in a recent meta-analysis of 165 studies with a combined sample size of 165,039 participants Bühler et al. (2021) found that relationship satisfaction tends to decrease across time and the presence of children is a significant moderator. The transition to parenthood appears to be a particularly vulnerable period for declines in couple relationship satisfaction (see Mitnick et al., 2009 for a meta-analysis) and this trend has been replicated among parents in the United States (Crohan, 1996), Asia (Luo, 2006) and Europe (e.g., Lorensen et al., 2004; Salmela-Aro et al., 2006).
While decline in relationship satisfaction over time may be universal, the onset and rate of deterioration may differ between mothers and fathers. For example, some findings indicate that following childbirth mothers experience abrupt declines in relationship satisfaction whereas fathers’ declines occur more gradually and only become evident from 6 to 15 months postpartum (e.g., Belsky & Hsieh, 1998; Grote & Clark, 2001). However, it remains unknown whether these declines persist, accelerate or attenuate over time. Yet, from a transactional perspective (Sameroff, 2009) children are likely to respond differentially to changing levels of interparental satisfaction and as a result their socioemotional development may vary as a function of change. Clinically, understanding how change or stability in interparental relationships may influence children’s socioemotional functioning bears practical implications for couples at risk of low relationship satisfaction and by conjecture poor socioemotional competence in children.
The Present Study
In this study we aim to contribute to the research base on child socioemotional development within the family system through two research goals. First, we aim to characterise and compare the change in mothers’ and fathers’ relationship satisfaction from nine months until nine years. As such, we capture the entire childhood period during which children’s socioemotional development may be susceptible to parental influence. Second, we aim to examine how change in mothers’ and fathers’ relationship satisfaction from nine months until nine years may influence children’s socioemotional outcomes at nine years. We examine associations with outcomes in male and female children separately given evidence that males and females may show differential susceptibility to parenting influence (e.g., Barnett and Scaramella (2013); van Eldik et al., 2020). Based on existing research, we hypothesise that both mothers and fathers will experience a decline in their relationship satisfaction from nine months to nine years. We further hypothesise that smaller declines in maternal and paternal relationship satisfaction will be associated with better socioemotional outcomes in both male and female children, regardless of baseline levels of relationship satisfaction.
Method
Participants
Data in the present study come from the Growing Up in Ireland (GUI) study, a large representative cohort study of children in Ireland. Ethical approval for this study was obtained from the Research Ethics Committee convened by the Department of Children and Youth Affairs. Full details regarding the study design and data collection have been described elsewhere (Quail et al., 2019). Briefly, 11,134 children born between December 2007 and May 2008, along with their primary and secondary caregivers were recruited from the Republic of Ireland Child Benefit Register (Quail et al., 2019). Child benefit is a universal monthly social welfare payment to families with children in Ireland. This administrative database was in electronic form and families were recruited through a systematic selection that entailed a random start and constant sampling fraction. The 11,134 families that agreed to participate represent a 65% response rate of all families approached.
Data were collected when children were nine months, three years (n = 9073), five years (n = 9001), seven to eight years (n = 5344; postal survey only) and nine years (n = 8032) through interview, direct assessment, and questionnaires. The present study uses data that were collected when children were nine months and nine years respectively given that during these timepoints parents provided data pertaining to their relationship satisfaction. Only families whose children had no siblings at the nine-month timepoint were included to exclude the effect that an additional child may exert on the couple relationship. Similarly, only couples who shared an interparental relationship at nine years with the same partner as they did at nine months were included. Data were weighted to statistically adjust for systematic non-response and attrition. The final sample consisted of 2074 mothers, fathers, and their child. All mothers in the present study were the child’s primary caregiver.
Mothers ranged in age from 18 to 40 years (M = 31.06, SD = 4.70) and fathers ranged in age from 18 to 60 years (M = 33.50, SD = 5.54). Child sex was roughly equally split with 55.1% male and 44.9% female children. Families were well educated with 27.8% of mothers and 18% of fathers holding a university degree or higher. Over 50% of the sample had managerial and professional careers. Participant demographic information is presented in Table 1.
Table 1
Descriptive Statistics of Demographic Information at Nine Months
Mothers (n = 2074) | Fathers (n = 2074) | |||
|---|---|---|---|---|
Mean (SD) or % | Min–Max | Mean (SD) or % | Min– Max | |
Parental age | ||||
Years | 31.06 (4.70) | 18–40 | 33.50 (5.54) | 18–60 |
Child sex | ||||
Male | 55.1% | - | - | - |
Female | 44.9% | - | - | - |
Parental education | ||||
No qualification | 28% | - | 32.2% | - |
Technical/vocational qualification | 8.5% | - | 14.6% | - |
Non-degree | 35.3% | - | 29.3% | - |
University degree | 12.6% | - | 7.2% | - |
Postgraduate | 14.2% | - | 10.1% | - |
Doctorate | 1% | - | 0.7% | - |
Family socioeconomic status | ||||
Professional | 16.8% | - | - | - |
Managerial | 44.7% | - | - | - |
Non-manual | 18.9% | - | - | - |
Skilled/unskilled manual | 18% | - | - | - |
Never worked/unknown | 1.7% | - | - | - |
Measures
Exposure: Interparental couple satisfaction
Mothers and fathers completed the abbreviated form of the Dyadic Adjustment Scale (DAS-7; Sharpley & Rogers, 1984) when children were aged nine months and subsequently when children were aged nine years. Couples indicated their dyadic satisfaction across seven items on a five-point scale ranging from ‘low satisfaction’ to ‘high satisfaction’. Final scores range from 0 to 36, with higher scores indicating greater relationship satisfaction. Scores lower than 21 reflect low relationship satisfaction whereas scores higher than 21 reflect high relationship satisfaction (Hunsley et al., 2001). Cronbach’s alphas (α) were 0.63 and 0.60 for mothers and fathers respectively at nine years. Reliabilities at nine months have not been disclosed by the GUI team.
Outcome: Socioemotional difficulties
Children’s socioemotional difficulties were assessed via teacher report on the Strengths and Difficulties Questionnaire (SDQ; Goodman, 1997) at nine years. Teachers responded to questions regarding the child’s behaviour over the past six months on a three-point scale ranging from 0 (‘not true’) to 2 (‘certainly true’). Higher scores indicate greater levels of the behaviour of interest. Four subscales of the SDQ, namely Emotional Problems (α = 0.78), Conduct Problems (α = 0.73), Hyperactivity (α = 0.85) and Peer Problems (α = 0.69) were combined to yield an overall score of Total Difficulties (α = 0.87).
Control variables
Demographic variables
We controlled for maternal and paternal age, education, family socioeconomic status and number of people living in the house.
Child baseline socioemotional competence
We also controlled for baseline levels of child socioemotional competence as measured by parent report of the ‘Personal-Social’ subscale of the Ages and Stages Questionnaire (ASQ, Squires et al., 2002) at nine months. The ASQ has excellent psychometric properties with 92% test-retest reliability, 87.4% sensitivity and 95.7% specificity (e.g., Singh et al., 2017).
Analytic strategy
First, we performed a series of descriptive and correlational analyses to describe the study variables. Second, we stratified participants into parents who reported high levels of relationship satisfaction (i.e., scores > 21; Hunsley et al., 2001) and parents who reported low levels of relationship satisfaction (i.e., scores < 21; Hunsley et al., 2001) at nine months. Third, we performed residualised change analysis to describe the change in mothers’ and fathers’ relationship satisfaction from nine months to nine years. Residualised change analysis refers to the statistical approach that considers early and later assessments of a construct to analyse longitudinal change (Allison, 1990). Next, we performed multiple linear regressions whereby we regressed child socioemotional outcomes at nine years onto mothers’ and fathers’ longitudinal patterns of change in relationship satisfaction. All models adjusted for control variables as well as the alternative parent’s baseline levels of relationship satisfaction.
Missing data
As is common in longitudinal research, there was some missing data due to attrition and non-response. Participants whose data were missing were less educated (t (7499) = −19.80, p < 0.001 for mothers; t (5643) = −8.70, p < 0.001 for fathers), reported lower relationship satisfaction levels at nine months (t (6141) = −5.30, p < 0.001 for mothers; t (5643) = −8.70, p < 0.001), had a larger number of people living in the house (t (7507) = 45.78, p < 0.001), were of lower socioeconomic status (t (7507) = 25.23, p < 0.001) and had children with poorer socioemotional behaviour at nine months (t (7393) = −7.61, p < 0.001). In addition, mothers with missing data were older (t (7507) = 3.84, p < 0.001) whereas fathers with missing data did not differ significantly from fathers with complete data in terms of age (t (7086) = 0.013, p = 0.99). Data were analysed for missingness patterns and Little’s MCAR test indicated that data were missing completely at random, χ2(160) = 182.17, p = 0.11.
Results
Correlations among study variables are presented in Tables 1 to 4 in Appendix A. Results indicate that at nine months mothers’ and fathers’ levels of relationship satisfaction were not statistically different, t (4146) = −1.96, p = 0.05. Although both mothers’ and fathers’ relationship satisfaction declined from nine months to nine years, fathers experienced a significantly greater decline than mothers, t (4146) = 3.74, p < 0.001. Finally, at nine years, fathers’ relationship satisfaction was significantly lower than that of mothers, t (4146) = 2.85, p = 0.004. It is noteworthy that parents of male and female children showed similar levels of relationship satisfaction at nine months (mothers: t (2022) = −0.24, p = 0.81; fathers: t (1864) = 0.47, p = 0.64), at nine years (mothers: t (2037) = 0.78, p = 0.81; fathers: t (1660) = 0.26, p = 0.79) and experienced similar levels of change in satisfaction (mothers: t (1990) = 0.97, p = 0.33; fathers: t (1536) = 0.04, p = 0.97).
Once participants were stratified according to level of relationship satisfaction at nine months results showed that mothers with high levels of relationship satisfaction at nine months experienced significantly greater declines than mothers with low relationship satisfaction at nine months, t (2072) = 28.81, p < 0.001. Similarly, fathers with high levels of relationship satisfaction at nine months experienced significantly greater declines than fathers with low relationship satisfaction at nine months t (2072) = 26.34, p < 0.001 (see Table 2).
Table 2
Descriptive Statistics of Mothers and Fathers’ Interparental Relationship Satisfaction at Nine Months, Nine Years and Longitudinal Change in Relationship Satisfaction
RSBaseline Mean (SD) | RSNine years Mean (SD) | RSchange Mean (SD) | t | p | Cohen’s d | |
|---|---|---|---|---|---|---|
Mothers High RSbaseline | 26.83 (3.61) | 17.08 (2.55) | −9.75 (4.03) | 99.89 | <0.001 | 4.03 |
Mothers Low RSbaseline | 18.19 (1.95) | 15.72 (3.28) | −2.46 (3.69) | 11.24 | <0.001 | 3.69 |
Fathers High RSbaseline | 26.99 (7.71) | 16.85 (2.92) | −10.15 (4.19) | 88.87 | <0.001 | 4.19 |
Fathers Low RSbaseline | 18.30 (1.81) | 15.39 (3.01) | −2.91 (3.28) | 12.23 | <0.001 | 3.28 |
In addition, male children showed significantly greater levels of overall total difficulties than female children at nine years t (1896) = 7.93, p < 0.001.
Mothers’ and fathers’ Interparental Relationship Satisfaction and Child Socioemotional Outcomes
Among parents who reported high relationship satisfaction at nine months, after adjusting for covariates neither mothers’ nor fathers’ declines in relationship satisfaction until nine years were significantly linked with socioemotional outcomes in either male or female children (see Table 3).
Table 3
Multiple Regression Models for the Prediction of Child Socioemotional Outcomes by Parents’ Change in Relationship Satisfaction from Nine Months to Nine Years Among Parents with High Relationship Satisfaction at Nine Months
Total Difficulties | ||||||||
|---|---|---|---|---|---|---|---|---|
Female Children | Male children | |||||||
B | β | p | Model Summary | B | β | p | Model Summary | |
Mothers’ RSchange | −0.01 | −0.01 | 0.75 | R2 = 0.03, F = 2.16, f2 = 0.03, p = 0.02 | −0.02 | −0.02 | 0.59 | R2 = 0.03, F = 3.14, f2 = 0.03, p = 0.001 |
Mothers’ age | −0.02 | −0.03 | 0.63 | −0.14 | −0.12 | 0.02 | ||
Fathers’ age | 0.08 | 0.09 | 0.06 | 0.08 | 0.08 | 0.09 | ||
Mothers’ education | −0.06 | −0.03 | 0.45 | −0.03 | −0.02 | 0.65 | ||
Fathers’ education | −0.10 | −0.06 | 0.16 | −0.01 | −0.01 | 0.89 | ||
Family social class | 0.30 | 0.08 | 0.06 | 0.40 | 0.10 | 0.01 | ||
Fathers’ RSbaseline | −0.06 | −0.06 | 0.25 | −0.11 | −0.09 | 0.06 | ||
Fathers’ RSchange | −0.04 | −0.03 | 0.49 | −0.04 | −0.03 | 0.54 | ||
Number of people in the house | 0.52 | 0.05 | 0.17 | −0.91 | −0.06 | 0.06 | ||
Child baseline socioemotional skills | −0.02 | −0.06 | 0.11 | −0.01 | −0.04 | 0.23 | ||
Fathers’ RSchange | −0.004 | −0.004 | 0.92 | R2 = 0.04, F = 2.55, f2 = 0.04, p = 0.01 | 0.02 | 0.01 | 0.75 | R2 = 0.03, F = 2.98, f2 = 0.03, p = 0.001 |
Mothers’ age | −0.09 | −0.08 | 0.11 | −0.11 | −0.09 | 0.05 | ||
Fathers’ age | 0.11 | 0.12 | 0.01 | 0.08 | 0.08 | 0.08 | ||
Mothers’ education | −0.03 | −0.02 | 0.69 | −0.05 | −0.03 | 0.51 | ||
Fathers’ education | −0.09 | −0.05 | 0.21 | −0.03 | −0.02 | 0.65 | ||
Family social class | 0.30 | 0.07 | 0.08 | 0.45 | 0.11 | 0.01 | ||
Mothers’ RSbaseline | −0.12 | −0.11 | 0.09 | −0.04 | −0.04 | 0.57 | ||
Mothers’ RSchange | −0.11 | −0.11 | 0.09 | −0.08 | −0.07 | 0.26 | ||
Number of people in the house | 0.53 | 0.05 | 0.18 | −1.08 | −0.07 | 0.03 | ||
Child baseline socioemotional skills | −0.02 | −0.06 | 0.11 | −0.02 | −0.05 | 0.15 | ||
In contrast, among parents who reported low relationship satisfaction at nine months, only mothers’ declines in relationship satisfaction until nine years were significantly negatively linked with total socioemotional difficulties in female but not in male children (β = −0.22, p = 0.01; R2 = 0.15, F = 2.31, p = 0.02). Thus, mothers who reported low levels of relationship satisfaction at nine months and who then experienced further declines until nine years had female children with more total socioemotional difficulties at nine years. However, among fathers with low relationship satisfaction at nine months, change in relationship satisfaction until nine years was not significantly linked with any socioemotional outcomes in neither male nor female children (see Table 4).
Table 4
Multiple Regression Models for the Prediction of Child Socioemotional Outcomes by Parents’ Change in Relationship Satisfaction from Nine Months to Nine Years Among Parents with Low Relationship Satisfaction at Nine Months
Total Difficulties | ||||||||
|---|---|---|---|---|---|---|---|---|
Female Children | Male children | |||||||
B | β | p | Model Summary | B | β | p | Model Summary | |
Mothers’ RSchange | −0.33 | −0.22 | 0.01 | R2 = 0.15, F = 2.31, f2 = 0.18, p = 0.02 | −0.09 | −0.07 | 0.43 | R2 = 0.09, F = 1.33, f2 = 0.10, p = 0.22 |
Mothers’ age | −0.17 | −0.15 | 0.19 | 0.13 | 0.14 | 0.23 | ||
Fathers’ age | 0.11 | 0.12 | 0.28 | −0.12 | −0.14 | 0.19 | ||
Mothers’ education | 0.32 | 0.15 | 0.13 | −0.38 | −0.20 | 0.04 | ||
Fathers’ education | −0.01 | −0.01 | 0.95 | −0.12 | −0.07 | 0.52 | ||
Family social class | 1.15 | 0.26 | 0.01 | 0.14 | 0.04 | 0.68 | ||
Fathers’ RSbaseline | 0.04 | 0.03 | 0.79 | −0.03 | −0.03 | 0.84 | ||
Fathers’ RSchange | −0.06 | −0.04 | 0.73 | −0.02 | −0.02 | 0.90 | ||
Number of people in the house | 1.44 | 0.15 | 0.11 | −0.04 | −0.001 | 0.99 | ||
Child baseline socioemotional skills | 0.04 | 0.11 | 0.21 | −0.04 | −0.13 | 0.13 | ||
Fathers’ RSchange | 0.12 | 0.06 | 0.52 | R2 = 0.14, F = 1.64, f2 = 0.16, p = 0.11 | −0.21 | −0.11 | 0.24 | R2 = 0.09, F = 1.28, f2 = 0.10, p = 0.25 |
Mothers’ age | 0.07 | 0.07 | 0.64 | 0.11 | 0.10 | 0.55 | ||
Fathers’ age | 0.09 | 0.11 | 0.41 | −0.03 | −0.03 | 0.84 | ||
Mothers’ education | 0.20 | 0.10 | 0.33 | −0.62 | −0.31 | 0.01 | ||
Fathers’ education | 0.13 | 0.06 | 0.53 | −0.05 | −0.03 | 0.83 | ||
Family social class | 1.26 | 0.29 | 0.01 | −0.36 | −0.08 | 0.45 | ||
Mothers’ RSbaseline | −0.29 | −0.29 | 0.08 | 0.07 | 0.05 | 0.73 | ||
Mothers’ RSchange | −0.38 | −0.34 | 0.03 | −0.06 | −0.05 | 0.76 | ||
Number of people in the house | 1.03 | 0.10 | 0.31 | 0.09 | 0.01 | 0.94 | ||
Child baseline socioemotional skills | 0.02 | 0.04 | 0.67 | −0.05 | −0.13 | 0.15 | ||
Discussion
In this study we aimed to characterise and compare the change in mothers’ and fathers’ relationship satisfaction from nine months until nine years and to examine how this change may influence children’s socioemotional outcomes at nine years. We found that at nine months both mothers and fathers reported similar levels of relationship satisfaction. However, compared to mothers, fathers experienced greater declines in relationship satisfaction from nine months until nine years and reported lower satisfaction at nine years.
The decrease in relationship satisfaction over time identified in this study is in line with previous literature and suggests that some decline in relationship satisfaction may be a universal experience for the average parenting couple (e.g., Lorensen et al., 2004; Salmela-Aro et al., 2006). However, whether and how such declines can influence children’s socioemotional development over time remains unclear. Static measures of relationship satisfaction suggest that the quality of interparental relationship satisfaction during the perinatal period is linked with child social competence at three (Panula et al., 2020) and seven to eight years of age (Lindblom et al., 2017). In this study however we found that change in relationship satisfaction over time was related to child socioemotional difficulties as a function of parents’ satisfaction during infancy. Both mothers and fathers with high levels of relationship satisfaction at nine months experienced greater declines in satisfaction compared to mothers and fathers who reported low relationship satisfaction at nine months. Nevertheless, we found no evidence for a significant link between either mothers’ or fathers’ decline in relationship satisfaction over time with child socioemotional functioning if parents had high levels of satisfaction during their child’s first months of life. This finding suggests that given high levels of relationship satisfaction in infancy, the decline that follows in the subsequent years is not necessarily detrimental to children’s socioemotional outcomes.
In contrast, mothers who reported low levels of relationship satisfaction at nine months and experienced further declines until nine years had female, but not male, children with more total socioemotional difficulties at nine years. This finding accords with recent meta-analytic evidence which suggests that females may be more likely to develop internalizing problems following interparental hostility (van Eldik et al., 2020). Arguably, females place greater importance on interpersonal relationships and as a result are more sensitive to interparental discord (Davies & Lindsay, 2001; Davies et al., 2016). Alternatively, given that mothers are known to use more emotion words with female children than they do with male children (Aznar and Tenenbaum (2015)), and given that emotion labelling is one mechanism through which children develop social skills (e.g., Tang et al., 2021), this parenting behaviour may be compromised among mothers preoccupied with poor relationship quality and as a result negatively impact children’s social adjustment. This hypothesis aligns with family systems theory which postulates that experiences from one subsystem (e.g., relationship satisfaction) may spill-over and influence another subsystem (e.g., impaired parenting). Considering the transactional nature of family systems, it is possible that ineffective parenting may further increase parent conflict and result in persistent relationship dissatisfaction. Thus, promoting relationship satisfaction during the perinatal period bears important implications for parenting competence throughout childhood and ultimately for child development. In fact, evidence suggests that interventions that exclusively target parenting skills are most likely to fail in the presence of conflictual marriages (e.g., Webster-Stratton, 1994), which further warrants continuous efforts at promoting couple relationship satisfaction (e.g., Cowan & Cowan, 2002).
In contrast, among fathers with low relationship satisfaction at nine months, change in relationship satisfaction over time was not significantly linked with children’s socioemotional outcomes. This finding aligns with previous research which similarly found a link between mothers’ relationship satisfaction during the perinatal period and overall family relationships in infancy but found no significant association for fathers (Korja et al., 2016). However, the null result contradicts other research which found relationship satisfaction to be a better predictor of paternal than maternal parenting behaviour (Belsky et al., 1991). One potential explanation for the lack of statistical significance is that all fathers in the present study were the child’s secondary caregiver. It would therefore be useful for future studies to also include fathers who are children’s primary caregiver as associations between parental behaviours and child outcomes may vary not as a function of parent gender but rather the roles that parents undertake within the family unit. It is also possible that maternal and paternal levels of relationship satisfaction may influence children in different ways. For example, there is now a considerable body of research documenting fathers’ unique roles in children’s socioemotional development (e.g., Flanders et al., 2010; Peterson & Flanders, 2005; Stgeorge et al., 2017). Thus, it is possible that certain aspects of fathering behaviour may buffer against socioemotional difficulties despite fathers’ low and declining interparental relationship satisfaction.
Overall, our findings suggest that compared to parents with low relationship satisfaction, parents with higher relationship satisfaction at nine months experienced significantly larger declines in satisfaction over time. However, despite the larger magnitude, this decline appears to be a common experience for parents (e.g., Lorensen et al., 2004; Salmela-Aro et al., 2006) and may not pose risk to children’s socioemotional development. In contrast, our findings suggest that, in the context of low relationship satisfaction in infancy, even small declines over time can exert a negative influence on female children’s socioemotional development at nine years. This may be particularly the case for primary caregivers, all of whom were mothers in this study. These findings warrant the development of programs for parents at risk for low relationship satisfaction as early as the perinatal period. Crucially, although we found no statistical evidence for a link between fathers low and declining relationship satisfaction and children’s socioemotional difficulties, relationship-promoting interventions should be inclusive of both mothers and fathers. That is, while mothers may develop effective interparental problem-solving strategies, they may have difficulty implementing them in the presence of unsupportive and conflictual partners. Henceforth, integrative interventions on early relationship satisfaction are strongly encouraged to target both mothers and fathers with the aim of promoting positive parenting.
Limitations and Future Directions
The present study is not without limitations. First, interparental relationship quality was assessed via parent report and therefore only represents mothers’ and fathers’ perceived relationship quality. Future studies should aim to include observed interparental relationship quality which would provide a more comprehensive view of how the parent couple relationship may influence child outcomes. In addition to observation, including child report of interparental relationship can also provide valuable information, and greater insight regarding the mechanisms through which the quality of the spousal subsystem influences the parental subsystem and ultimately children’s development. Such approaches can offset biases and limitations associated with participant report and the internal consistency of measures as evinced by the modest Cronbach’s alphas for the DAS in this study.
Second, although the study captures change in interparental relationship quality spanning a period of nine years, it does not reflect fluctuations in change over time. Despite this significant limitation, our findings provide incentive for future research to assess interparental relationship quality at frequent intervals. Such an approach would yield a more representative profile of relationship satisfaction change. In addition, it would allow examination of the onset and duration of either growth or decline in relationship satisfaction. It may be that change in relationship satisfaction that occurs earlier in life may have more pronounced effects on children’s outcomes. Equally possible is that gradual change may exert a different effect from sudden change. However, these hypotheses remain to be tested.
Third, although we adjusted for relevant covariates, we did not include any measure of parenting behaviour. Yet the interparental relationship does not exist independent of parenting behaviour and it would be relevant to examine how mothering and fathering interact with relationship satisfaction to influence child development. Such future directions can clarify whether parental behaviours compound the effect of relationship (dis)satisfaction on child outcomes or whether they work in a compensatory way.
Relatedly, from a family systems theory lens, future research should also explore how children’s individual differences such as temperament influence the mother-father relationship. Such approaches can effectively inform interventions which may need to target the whole family in some instances whereas in others it may be sufficient to focus on the parent dyad.
Finally, despite the large sample size and longitudinal design causality cannot be inferred. Thus, the assumptions regarding the direction of effects remain speculative.
Implications and Conclusion
Despite limitations, our findings suggest that following the arrival of a new baby it is common for couples to experience some decline in their interparental relationship satisfaction. In families where initial levels of couple satisfaction are high, the decline over time does not adversely affect children’s socioemotional outcomes. However, in couples where mothers’ initial levels of relationship satisfaction are low, the experience of further decline may have a negative influence on socioemotional competencies in female children. This finding indicates that supporting the interparental relationship during the perinatal period bears preventive clinical value.
While the parental dyad has long been recognised as important to the whole family unit (Minuchin, 1974), the evidence base from interventions studies targeting parental relationship satisfaction and child outcomes is sparse (see Harold & Sellers, 2018). Hence, our findings bear both research and clinical implications as they highlight a need for carefully tailored interventions to target and promote couple relationship satisfaction during the transition to parenthood such that the negative impact on children of any decline experienced over time can be mitigated. Such interventions should involve both mothers and fathers given the well documented systemic nature of families and the transactional functioning of family subsystems. Such an exercise bears implications not only for children’s immediate socioemotional development but may also longitudinally reduce internalising and externalising problems (Basten et al., 2016), anxiety and depression (Egger & Angold, 2006) during preschool, behavioural problems during early childhood (de Lijster et al., 2019) and promote academic achievement in adolescence (Buehler et al., 2007).
Additionally, there is a need for increased effort for general practitioners and perinatal specialists to identify couples at risk for low relationship satisfaction and to provide them with resources and education on coping strategies, conflict-resolution, and communication techniques. In addition, families should be encouraged to develop an appreciation of the importance of the interparental relationship to the whole family system as this can increase understanding and encourage collective work on those parts of the system that are amenable to change.
Supplementary information
The online version contains supplementary material available at https://doi.org/10.1007/s10826-023-02601-4.
Compliance with ethical standards
Conflict of interest
The authors declare no competing interests.
Ethical approval
Ethical approval was obtained from the Research Ethics Committee convened by the Department of Children and Youth Affairs.
Informed consent
Children’s parent (s)/guardian(s) provided written consent prior to the start of the study.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.