Mothers' work-family conflict and enrichment: associations with parenting quality and couple relationship.

BACKGROUND: Employment participation of mothers of young children has steadily increased in developed nations. Combining work and family roles can create conflicts with family life, but can also bring enrichment. Work-family conflict and enrichment experienced by mothers may also impact children's home environments via parenting behaviour and the couple relationship, particularly in the early years of parenting when the care demands for young children is high.
METHODS: In order to examine these associations, while adjusting for a wide range of known covariates of parenting and relationship quality, regression models using survey data from 2151 working mothers of 4- to 5-year-old children are reported. RESULTS/
CONCLUSION: Results provided partial support for the predicted independent relationships between work-family conflict, enrichment and indicators of the quality of parenting and the couple relationship.

Introduction

The rise in employment participation of women over recent decades, particularly mothers of young children, has been well documented (Australian Bureau of Statistics 2011; Huerta et al. 2011). Dual-earner families are now the norm in industrialized countries, and sole parents participate in employment at an increasing rate (Australian Bureau of Statistics 2008; OECD 2011). Raising children occurs in a context where both parents are employed, but in employment that is increasingly insecure, requiring long hours and is likely to encroach onto family life (Jacobs & Gerson 2001; Skinner & Pocock 2010). The established consequences of work-to-family conflict include poorer physical and mental health, poorer quality of life, lower job satisfaction and higher turnover (Allen et al. 2000; Eby et al. 2005; Nomaguchi et al. 2005). However, participation in employment can also be rewarding and enriching, associated with a range of mental health indicators (Cooksey et al. 1997; Grzywacz & Bass 2003; Bianchi & Milkie 2010).

While a vast literature investigating the negative effects of work–family conflict exists, knowledge about the effects of workplace experiences on maternal parenting behaviours specifically is scarce (Eby et al. 2005). This paper is one of few examining the relationship between work–family conflict and rewards and parenting behaviours known to influence children's outcomes during a key phase of childhood – the year prior to formal schooling. The aim of the present study is to investigate associations between work–family conflict, work–family enrichment, and parenting behaviours and perceptions of the couple relationship, for mothers of pre-school aged children.

Our focus here is on mothers, rather than parents, in dual earner families. While paid work continues to be a key normalizing principal for fathers (Marsiglio et al. 2004; Ranson 2012), mothers' employment is assumed to be supplementary to that of the primary ‘breadwinners’, fathers. Women continue to perform a substantive, disproportionate amount of unpaid work compared with men, particularly in Australia (Craig & Sawrikar 2009; Bianchi & Milkie 2010; Craig et al. 2010), and mothers curtail their involvement in the paid labour force in response to family demands more frequently than fathers (Kaufman & Uhlenberg 2000; Sayer 2005; Craig & Sawrikar 2009; Bianchi & Milkie 2010). Work–family conflict is reported more often for mothers compared with fathers (Frone et al. 1992; Duxbury & Higgins 2001), even when mothers participate in part-time employment (Gronlund 2007). Examining the work–family interface is particularly salient for mothers' interactions with their partners and children; ultimately these determine the quality of children's home environments (Parke 2004).

Work–family conflict and enrichment

In the present study, we use the complementary constructs of work–family conflict and work–family enrichment (Marshall & Barnett 1993). Work–family conflict is based on the scarcity hypothesis that poses limits to individuals' time and energy expenditure (Goode 1960) such that competing roles (family, employment) produce inevitable tension, resulting in conflict (Froberg et al. 1986). Work–family enrichment stems from the alternative concept that accumulating roles enhances social support, interaction and skill-building, enacting an overall net benefit, despite time constraints (Sieber 1974; Marshall & Barnett 1993). We use these two constructs simultaneously. It is well accepted that work–family conflict and enrichment can exact independent concurrent influences on employees (Marshall & Barnett 1993; Barnett 1998; Grzywacz & Marks 2000; Grzywacz & Bass 2003). While there is also evidence to suggest that work–family conflict and enrichment can interact, with one mitigating the effects of the other (Greenhaus & Powell 2006; Gronlund & Oun 2010), this patterning appears to be related to the particular outcomes of interest. For the couple relationship and parenting behaviours, evidence suggests that an independent ‘additive’ approach is most appropriate (Gareis et al. 2009; Cooklin et al. in press).

For mothers, there has been much research describing the negative outcomes of combining employment with family care, for themselves, for their partners and for children's development (Scarr et al. 1989; Voydanoff 1989; Menaghan & Parcel 1990; Parcel & Menaghan 1990; Brooks-Gunn et al. 2002). Consequently, the benefits and positive inter-dependencies potentially conveyed by employment have been under-investigated (Eby et al. 2005; Greenhaus & Powell 2006).

Work–family conflict and enrichment: parenting practices and behaviours

Ecological perspectives of parenting and children's development propose that while parenting occurs within the family, parents' behaviours are influenced by interaction with broader social, structural and institutional environments that exert direct and indirect effects on parent–child interactions (Parke 2004; Bronfenbrenner & Morris 2006). It is plausible that for mothers, the incompatibility between paid work and care of young children acts as one such contextual stressor. There is evidence of associations between work–family conflict and family relationships measured via broad-level constructs including parenting satisfaction, family functioning or family stress (Repetti 1994; Galambos et al. 1995; Kinnunen & Mauno 1998; Galinsky 1999; Crouter & Bumpus 2001). Work–family conflict is associated with reduced time with children, and poorer child mental health (Allen et al. 2000; Crouter & Bumpus 2001; Grzywacz & Bass 2003; Chandola et al. 2004; Butterworth et al. 2011; Strazdins et al. 2013). Parenting behaviours may explain these associations (Rose 1993; Stewart & Barling 1996). However, the role of employment– as a putative stress, but also a potential support – for parenting has not been understood (Allen et al. 2000; Eby et al. 2005).

Three parenting domains have an established relationship with children's social, emotional, cognitive and behavioural outcomes. Parenting warmth is an aspect of parental responsiveness, the degree to which parents intentionally support individuality and self-regulation via sensitivity to the child's needs (Baumrind 1991). Warm, affectionate parenting behaviours are associated with optimal child outcomes (Bradley & Caldwell 2006; Berk 2001; Chao & Willms 2002; Dallaire & Weinraub 2005; Landry et al. 2006). Conversely, parenting irritability including frequent parental rejection, harsh or irritable responses is associated with increased externalizing and internalizing behaviour, poorer academic achievement, less pro-social behaviour and worse physical health (Repetti et al. 2002; Low & Stocker 2005; Romano et al. 2005; Waylen & Stewart-Brown 2010). Parental consistency is persistence in parental discipline and behaviour management. Consistency contributes to fewer behaviour problems in children and promotes pro-social conduct (Webster-Stratton et al. 2011; Cussen et al. 2012). The first aim of the present study therefore was to examine the relationship between mothers' work–family conflict, work–family enrichment and parenting warmth, irritability and consistency. We make the following specific hypotheses:

Higher work–family conflict will be associated with lower parenting warmth, lower consistency and higher irritability in mothers;

Higher work–family enrichment will be associated with higher warmth, higher consistency and lower irritability in mothers.

Work–family conflict, enrichment and the couple relationship

Couple functioning, including any conflict present, is inextricably linked to parent mental health, parent–child interactions and children's outcomes (Repetti et al. 2002; Low & Stocker 2005). For all adults, work–family conflict is associated with poorer marital quality (Allen et al. 2000). For mothers in particular, inequitable division of household work is associated not only with work–family conflict (Duxbury & Higgins 2001; Alexander & Baxter 2005; Stevens et al. 2007), but with reduced marital satisfaction and perceived unfairness within the household (Wilkie et al. 1998).

Conversely, it is also possible that the rewards and enrichment conveyed by employment participation enhance the quality of the intimate relationship for mothers (Marshall & Barnett 1993; Rothbard 2001). Complex, rewarding jobs are likely to improve optimism, affect and self-efficacy, enhancing inter-personal interactions (Rothbard 2001; Greenhaus & Powell 2006). However, given the research interest in work–family conflict, particularly for mothers', positive effects of employment enrichment on marital quality for mothers have been under-investigated. Our second aim is to investigate the relationship between work–family conflict and enrichment on couple relationship quality and couple conflict for mothers of pre-school aged children. We make the following specific hypotheses:

Higher work–family conflict will be associated with a poorer quality couple relationship, including more frequent relationship conflict;

Higher work–family enrichment will be associated with a higher-quality couple relationship, and less frequent relationship conflict.

The current study

The aim of the current study was to investigate the relationship between work–family conflict and enrichment on parenting practices and the couple relationship for mothers of 4- to 5-year-old children. Our sample was a diverse, nationally representative cohort of Australian mothers. The number of hours of employment participation per week, child characteristics including health care needs, and family socio-economic position are all likely to confound the relationship between employment participation and parenting outcomes. Child temperament is an established influence on parenting (Sanson et al. 2004), and exerts additional strain on mothers' employment functioning (Shibley-Hyde et al. 2004). Similarly, mental health is particularly important, with psychologically distressed individuals reporting adverse parenting practices and poorer relationship quality (Low & Stocker 2005; Romano et al. 2005). Adjusting for mental health in our analyses allows us to focus on our primary main relationships of interest, to test associations that may exist outside of mediation through maternal mental health. Accordingly we present the independent contributions of work–family conflict and enrichment to parenting and couple outcomes when maternal demographic, maternal mental health and child characteristics and temperament are adjusted for in multivariate models.

Method

Participants

Data for the current study were taken from the Kindergarten (K) cohort of the Longitudinal Study of Australian Children (LSAC) (for full sample details see Soloff et al. 2005b; Misson & Sipthorp 2007). LSAC was approved by the Australian Institute of Family Studies Ethics Committee (Gray & Sanson 2005; Soloff et al. 2005b). Study design and methods are described in detail elsewhere (Soloff et al. 2005a). LSAC used a two-stage cluster sampling, using Australian postcodes and the Medicare Australia database. This sample was broadly representative of all Australian children. Data were collected in 2004 via face-to-face interview and a self-report questionnaire by the parent who knew the child best (98.6% were the child's biological mother) (Department of Family and Community Services 2004; Soloff et al. 2005a). Data are from Wave 1 (collected in 2004) when children were aged 4–5 years. Of the contactable children selected and residing in the sampled postcode, 4983 took part in LSAC (59% response rate).

Inclusion and exclusion criteria

Mothers were included in the present study if they were the child's biological mother, employed at the time of data collection, with complete data on all variables included in the (a) parenting and (b) couple analyses. To retain single mothers in parenting analyses (a), the sample was limited to participants with full data across all variables in the final parenting model. Then, single mothers were excluded from the couple relationship models only (b), assuring maximum data for all analyses.

Measures

Work–family conflict

A four-item adaptation of the scales authored by Marshall and Barnett (1993) assessed employment-related constraints on family life and parenting (‘Because of my work responsibilities …. I have missed out on home or family activities that I would like to take part in … My family time is less enjoyable and more pressured’) and constraints from family responsibilities that affect employment (e.g. ‘Because of my family responsibilities …. My work time is less enjoyable and more pressured …. I have to turn down work activities or opportunities that I would prefer to take on’). Respondents indicated agreement using a scale ranging from 1 = strongly disagree to 5 = strongly agree (Cronbach's α mothers 0.71), creating a continuous scale, less than one missing item, with higher scores indicating more work–family conflict.

Work–family enrichment

A six-item adaptation of Marshall and Barnett's (1993) scales assessed the benefits of combining employment with parenting (Cronbach's α = 0.84). Three items assessed parents' views of the benefits for children (e.g. ‘My working has a positive effect on my children, Working helps me to better appreciate the time that I spend with my children’) and themselves (e.g. ‘Having both work and family responsibilities … Makes me a more well-rounded person …. Makes me feel more competent’). Respondents indicated their agreement using a scale ranging from 1 = strongly disagree to 5 = strongly agree, to create a continuous scale, with no more than one missing item, with higher scores indicating more enrichment. Both of these indicators of the work–family interface have been used in recent Australian research (Strazdins et al. 2013; Cooklin et al. in press).

Irritable parenting was measured with five items (10-point scale, 1 ‘not at all’ to 10 ‘all of the time’) assessing the frequency of hostile behaviours and feelings towards the child (Statistics Canada 2000). Example items include ‘I have been angry with this child; I have lost my temper with this child’.

Parenting warmth was measured using six items (five-point scale, 1 ‘never/almost never’ to 5 ‘always/almost always’) assessing frequency of positive and affectionate verbal and physical behaviours and feelings towards the child, e.g. ‘how often do express affection by hugging kissing or holding this child?; how often do you have warm close times together with this child?’

Parenting consistency was measured using six items (five-point scale, 1 ‘never/almost never’ to 5 ‘always/almost always’) assessing consistency in parenting behaviours (e.g. ‘How often does this child get away with things that you feel should have been punished?’) Scores for irritable parenting, parenting warmth and parenting consistency were the mean of individual items (with no more than two missing items), creating a continuous variable with higher scores indicating more irritability, higher warmth and higher consistency.

Couple argumentativeness was assessed using four items (five-point scale, 1 ‘Never’ to 5 ‘Always’) from the Quality of Co-parental Interaction Scale (Ahrons 1981), assessing the frequency of arguments and conflict with their partner. Items included ‘How often do you argue’; ‘How often is there hostility between you'? Final scores were the mean of individual items with higher scores indicating more frequent conflict.

Couple relationship quality was assessed using six items from the Relationship Assessment Scale (Hendrick 1988) measuring mothers' report of quality of the partner relationship. Items include ‘How well does your partner meet your needs’; and ‘How good is your relationship compared with most’. Rating scales vary but all are rated on a five-point response scale. Mean scores for item responses are calculated with higher scores indicating a higher-quality intimate relationship.

Maternal psychological distress was assessed using the Kessler-6 (K6), a brief assessment tool, widely used in population studies, which measures the frequency of symptoms of psychological distress over the previous 4 weeks (Furukawa et al. 2003). Responses to the six items (five-point rating scale) were summed to give a total score 0–30, with higher scores indicating greater psychological distress.

Child temperament was assessed using the Sociability, Persistence and Reactivity subscales of the Short Temperament Scale for Children (Sanson et al. 1994; Smart & Sanson 2005), which each contain four items. The Sociability scale assesses how comfortable the child is in new situations or with unfamiliar children or adults. The Persistence scale assesses the child's capacity to sustain tasks to completion. The Reactivity scale assesses child volatility. Items from each subscale were averaged for participants with no more than one missing item, to create continuous scales, with high scores indicating more sociability, persistence and reactivity.

Demographic characteristics including maternal age at time of data collection (<25 years = 1; ≥25 years = 0), the primary language at home (primary language at home English = 0; primary language not English = 1), family type (two parent household = 0; single mother household = 1) and the number of children in the household (1 or no other children in family = 0; 2 or more children in family = 1) were collected. Child age was recorded in months. Any child special health care needs was recorded (none = 0; one or more = 1). Maternal employment hours performed each week were recorded as a single, continuous variable. Socio-economic position (SEP) was rated using a continuous, composite variable, ranking each family's relative socio-economic position based on parental income, education and occupational prestige (Blakemore et al. 2009). Families with a standardized score at or below the 25th percentile were classified as ‘low’ SEP, those above the 75th percentile were classified as ‘high’ SEP, and the remainder were classified as ‘medium’ SEP, as recommended by the authors of this measure. A single-item recorded whether the child had any special health care needs (no = 0; yes = 1).

Statistical analyses

Data were analysed in Stata 11.0. LSAC sample weights and adjustment for complex sampling design were applied. Analyses were weighted for non-response accounting for unequal probability of selection into the sample. First-order Taylor linearization was used to obtain estimates of standard error, accounting for the multi-stage, clustered sampling design. Means, standard errors, range and Cronbach's alpha (if relevant) were calculated for all variables. Parenting and couple variables were the primary dependent variables and work–family conflict and enrichment the main independent variables. To test for association between dependent and independent variables, unadjusted linear regression analyses were conducted. Multiple linear regression analyses were performed, with parenting and couple relationship variables as the dependent variables, and work–family conflict and enrichment as the main predictor variables. Maternal demographic (employment hours, socio-economic position, family type, number of children, main language), child characteristics (child age, health, temperament) and maternal mental health were entered as controls into the multivariate models. Interaction terms (Work–family conflict × Work family enrichment) were tested in parenting models to compare independent main (vs. interaction) effects. For the couple relationship variables, the family structure variable (single mother household versus other) was removed, as these models were restricted to couple mothers only. Associations between work–family conflict, enrichment, maternal employment hours and family socio-economic position were tested using unadjusted linear regression analyses.

Results

Of the 4983 mothers of 4- to 5-year-olds in the data set, a final sample of n = 2151 were eligible for inclusion in the parenting models (43% of total cohort), and a final sample of n = 1878 (38% of cohort) were eligible for inclusion in the analyses with primary couple outcomes. Compared with those included in parenting analyses (n = 2151), those who were eligible but excluded because of missing data were more likely to be of low socio-economic position (38% versus 14%, P < 0.001) and employed in an unskilled occupation (28 versus 14%, P < 0.001); a single parent (18% versus 11%, P < 0.001) and to have a language other than English as primary (22% versus 10%, P < 0.001). Sample characteristics are presented in Table 1. Mothers were employed for a mean (SD) of 23.5 (14.3) hours per week.

Table 1

Sample characteristics for mothers (n = 2151) 1920

Variables	M/%	SD	Range	α
Work–family enrichment	3.67	0.69	1–5	0.85
Work–family conflict	2.65	0.87	1–5	0.74
Social/demographic factors
Socio-economic position (quartiles)
Most disadvantaged families	0.15	NA	0–4
Most advantaged families	0.29	NA	0–4
Family structure – single mother family	0.11	NA	0–1
2 or more children in family	0.88	NA	0–1
Age of study child (months)	56.98	2.57	51–67
Mother's age (years)	35.09	4.93	20–52
Mother's work hours per week	23.50	14.27	0–1
Language other than English spoken at home	0.11	NA	0–1
Current maternal psychological distress	3.76	3.33	0–24	0.81
Child temperament
Sociability	3.87	1.20	1–6	0.82
Persistence	4.00	0.92	1–6	0.79
Reactivity	2.67	0.89	1–6	0.65
Parenting and couple relationship
Parenting warmth	4.46	0.43	1–5	0.82
Parenting irritability	2.36	0.56	1–4	0.71
Parenting consistency	4.14	0.63	2–5	0.70
Couple relationship quality (n = 1920)	4.28	0.74	1–5	0.90
Couple argumentativeness (n = 1920)	2.02	0.52	1–5	0.80

Unadjusted analyses

Unadjusted linear regression analyses provided initial support for significant associations between work–family conflict and lower parenting warmth (β = −0.06, P < 0.001); higher irritability (β = 0.09, P < 0.001); and less consistency (β = −0.08, P < 0.001). Work–family enrichment was associated with higher warmth (β = 0.06, P < 0.001); less irritability (β = −0.06, P = 0.003) and more consistency (β = 0.10, P < 0.001). Conflict was also associated with a poor quality intimate relationship (β = −0.18, P < 0.001), and frequent conflict within the relationship (β = 0.13, P < 0.001). Work–family enrichment was significantly associated with a high-quality relationship (β = 0.20, P < 0.001), and infrequent argumentativeness (β = −0.11, P < 0.001).

Higher work–family conflict (but not enrichment) was associated with longer employment hours (β = 0.02, P < 0.001). Women of high and medium socio-economic position reported similar work–family conflict and enrichment; however, both higher conflict [mean (SD) 2.8 (0.89) versus 2.6 (0.85), P < 0.001], and higher enrichment [mean (SD) 3.8 (0.68) versus 3.6 (0.69), P < 0.001] were reported by women of low socio-economic position, compared with those of medium socio-economic position.

Work–family conflict and enrichment and parenting: adjusted analyses

Results of the multiple linear regression analyses, identifying the relationship between conflict and enrichment parenting warmth, irritability and consistency, adjusted for control variables are presented in Table 2. Higher work–family conflict was significantly associated with less warm parent–child interactions, and more irritable parent–child interactions, but not with parenting consistency. Work–family enrichment was significantly associated with higher parenting consistency and parenting warmth (but not parenting irritability), in adjusted analyses (Table 2) even when maternal mental health and child temperament variables were included in analyses. There was no evidence of interaction between work–family conflict and work–family enrichment.

Table 2

Summary of adjusted multiple regression models, predicting parenting outcomes

	Parenting warmth		Parenting irritability		Parenting consistency
Variables	B	SE. B	B	SE. B	B	SE. B
High work/family enrichment	0.03*	0.01	0.01	0.02	0.05*	0.02
High work/family conflict	−0.04**	0.01	0.04*	0.01	−0.02	0.02
Increasing employment hours	0.01	0.001	−0.02*	0.001	−0.01	0.001
Most advantaged SEP	−0.71	0.02	0.03	0.03	0.11***	0.03
Most disadvantaged SEP	−0.05	0.03	0.02	0.04	−0.16***	0.04
Couple family	0.03	0.04	−0.11*	0.04	0.03	0.05
2 or more children in family	−0.12***	0.03	0.03	0.04	0.03	0.04
Language other than English	0.04	0.03	−0.03	0.04	−0.22	0.04
Child age (months)	−0.01	0.004	0.01	0.004	−0.01	0.005
Child special health needs	0.05	0.03	−0.01	0.03	−0.01	0.04
Current maternal psychological distress	−0.01**	0.003	0.03***	0.004	−0.02***	0.004
Child temperament: Sociability	0.01	0.01	0.02**	0.01	0.02	0.01
Child temperament: Persistence	0.05***	0.01	−0.09***	0.01	0.06**	0.02
Child temperament: Reactivity	−0.04**	0.01	0.19***	0.01	−0.16***	0.01
n	2151		2151		2151
R2	0.07		0.21		0.15
F	7.72		43.28		27.77

P < 0.05.

P < 0.01.

P < 0.001.

Work–family conflict, work–family enrichment and the couple relationship: adjusted analyses

Results of the multiple linear regression analyses, identifying the relationship between conflict, enrichment and assessments of the intimate relationship, adjusted for sociodemographic, maternal mental health and child temperament are presented in Table 3. Work–family conflict was independently associated with poorer couple relationship on both outcomes in fully adjusted models. Work–family enrichment was significantly associated with reporting a higher-quality intimate relationship in adjusted analyses (Table 3). However, no protective association between work–family enrichment and couple conflict was observed in adjusted analyses.

Table 3

Summary of adjusted multiple regression models, predicting the couple relationship

		Couple relationship quality		Couple argumentativeness
Model	Variables	B	SE. B	B	SE. B
2	High work/family enrichment	0.09**	0.03	−0.03	0.02
	High work/family conflict	−0.06*	0.02	0.05**	0.02
	Increasing employment hours	−0.01***	0.001	0.01	0.001
	Most advantaged SEP	0.05	0.03	−0.03	0.02
	Most disadvantaged SEP	−0.12*	0.05	0.06	0.04
	2 or more children in family	0.24**	0.07	−0.09*	0.04
	Language other than English	−0.07	0.06	0.06	0.05
	Child age (months)	−0.01	0.006	0.01	0.04
	Child special health needs	−0.03	0.05	−0.03	0.04
	Current maternal psychological distress	−0.07***	0.01	0.05***	0.01
	Child temperament: Sociability	−0.01	0.01	−0.003	0.01
	Child temperament: Persistence	0.01	0.02	−0.005	0.01
	Child temperament: Reactivity	−0.01	0.02	0.06***	0.01
	n	1920		1920
	R2	0.17		0.17
	F	15.30		20.87

P < 0.05.

P < 0.01.

P < 0.001.

Discussion

While a vast literature investigating the negative effects of work–family conflict exists, few have examined the effect of adverse workplace experiences on maternal parenting behaviours specifically. Similarly, evidence about the potential benefits to mothers of employment participation while their children are young has been slow to emerge. This paper is one of very few examining the relationship between work–family conflict and rewards and specific parenting behaviours known to influence children's outcomes. We extended our analyses to include investigation of the work–family interface on the quality of the partner relationship for partnered mothers, another important predictor of outcomes for young children.

A striking finding is that higher work–family conflict was independently associated with less warm, affectionate responding towards the child and more negative irritable interactions even when maternal sociodemographic characteristics, maternal mental health and child temperament were controlled. Such parenting adversely affects children's optimal social, educational, behavioural and physical development (Bradley & Caldwell 2006; Berk 2001; Chao & Willms 2002; Dallaire & Weinraub 2005; Waylen et al. 2008). Mothers' capacity to parent optimally may be compromised when work–family conflict is high. Univariate associations between high work–family conflict and parenting inconsistency were not sustained in multivariate models.

Other notable findings include the independent association between work–family enrichment, and parenting warmth and consistency. These associations persisted even when all other potential explanatory variables were included. Historically, mothers' employment participation has been assumed to be a risk – for women themselves, for their partners and families (Barnett & Hyde 2001; Gilbert & Rader 2001). It is now established that employment participation is protective against common physical and mental health problems (Lee & Powers 2002) and increases opportunities for social support, income, skill-building and identity (Cooksey et al. 1997; Rothbard 2001; Grzywacz & Bass 2003; Greenhaus & Powell 2006; Bianchi & Milkie 2010). Our findings show that satisfaction and efficacy gained in the workplace might strengthen mothers' interactions with children.

This effect was net of the effect that work–family enrichment might impart via women's mental health. Our findings suggest a main effect between work–family enrichment, and more warm and consistent parenting. It may be that mothers revalue the time that they do spend with their children when time is constrained by employment participation, and prioritize affection and consistency for their children. For some families, increased income might mean that some mundane, routine domestic tasks can be outsourced, alleviating stress. Alternatively, a satisfying, high-quality job conveys competence, optimism, motivation and self-esteem. These are all supportive qualities for promoting parenting confidence and competence.

Higher work–family conflict was associated with reporting a poorer quality couple relationship, and frequent couple conflict. Previous research has demonstrated that low job quality, work–family conflict and stress are associated with relationship problems such as withdrawal, conflict and inequitable sharing of domestic work (Allen et al. 2000). Our findings provide further evidence that negative experiences in the workplace adversely affect the quality of emotional exchanges within the intimate partner relationship. Notably, we report an independent association between work–family enrichment and a higher-quality partner relationship, Optimally, employment participation can provide satisfaction, optimism and efficacy that flow-on to promote high-quality emotional exchanges between parents. Alternatively, it may be that positive experiences in the workplace improve mothers' inter-personal skills and tolerance and that these are brought to bear in interactions with an intimate partner (Greenhaus & Powell 2006).

Work–family conflict was associated with longer employment hours, and socio-economic disadvantage. This suggests that women who are occupied in low-skilled, poorly remunerated jobs, or jobs characterized by long, inflexible working hours experience high strains. Higher work–family enrichment was associated with lower socio-economic position. This warrants further exploration. However, income and access to resources provided by employment participation may, for women who are vulnerable to financial hardship, relieve financial stress in a way that supports them to focus on parenting.

There are several notable strengths to our study. Our sample was a diverse, nationally representative cohort of Australian mothers. All analyses were adjusted for maternal demographic characteristics, in particular, those likely to confound the relationship between employment participation and parenting outcomes. Maternal mental health was adjusted for in all analyses to ensure that any effect on parenting and couple outcomes were not those of mothers psychological functioning, but rather of the conflicts and rewards inherent in combining paid work with parenting.

We acknowledge some limitations to our approach. These data are cross-sectional, so causal relationships cannot be established. Potential reverse causality of the variables cannot be examined (e.g. poor couple relationship predicting work–family conflict). We used mothers' self-reports to assess our criterion variables, which may be subject to social desirability and other forms of reporting bias. Women excluded from the analyses because of missing data were more likely to be of low socio-economic position, in a low-skilled occupation, a single parent, and have a primary language other than English, compared with those included. Employment for these women is likely to be in occupations that are unsatisfying, menial and stressful, and unlikely to convey multiple rewards. Hence, the relationships observed have limited generalizability to those most vulnerable in the paid labour force. Our sample of Australian mothers are predominantly employed part-time, as is common in Australia, so results may not be generalizable to mothers' resuming employment full-time, more common in other settings. Our analyses were limited to employed mothers only; comparisons with non-employed women on our outcomes of interest are not presented here.

Finally, we acknowledge that the relative strength of the relationship between work–family conflict, enrichment, parenting and couple outcomes reported here, although significant, was small in multivariate models. This is likely a reflection of the necessarily brief measures used in LSAC that are perhaps somewhat insensitive and not likely to detect all differences across populations. Furthermore, employment on a part-time basis is likely to minimize both work–family conflict and enrichment, explaining the small associations. Our purpose was not to explain maximum variance in our outcomes. Rather, it was to identify whether work–family conflict and enrichment have an independent association with parenting, and the couple relationship, reflected in our findings here.

Implications of study findings

It is well established that high-quality jobs are those associated with higher levels of work–family enrichment and lower levels of work family conflict (Allen et al. 2000; Eby et al. 2005; Greenhaus & Powell 2006). Our findings provide further impetus for the provision of optimal employment conditions. High-quality jobs, those that optimize opportunities for work–family enrichment, ultimately improving family environments are characterized by manageable workloads, reasonable hours, flexibility, optimal supervisor support, job security and a sense of control/autonomy over work (Gronlund 2007; Strazdins et al. 2010). Furthermore, for parents of young children, quality employment would also include provision of paid family-related leave, flexible start and finish times, work–family specific supervisor support, and a range of genuine options for rewarding part-time roles. This is particularly relevant for women of low socio-economic position, employed in low-skilled, or menial occupations, who are likely to experience high work–family conflict brought about by inflexible workloads or working hours.

Conclusion

Using a nationally representative sample, and adjustment for a wide range of control variables, this study demonstrated associations between work–family rewards and enrichment and key indicators of children's family environments – parenting and relationship quality. These findings show the potential for mothers' experiences of employment to ultimately impact children both positively and negatively. They provide further impetus for workplaces and public policy to provide optimal employment conditions to parents of young children, given the flow-on implications work has to families, family relationships, parent–child interactions, and ultimately, children's development.

Key messages

Work–family conflict and enrichment have the potential to impact on children's family environments. However, the specific relationship between work–family conflict, enrichment, and parenting behaviours and couple functioning has not been widely investigated for mothers of young children.

We report a significant relationship between mothers' work–family experiences and their family relationships. Work–family conflict is associated with less optimal parenting behaviours (lower warmth, higher irritability) and higher conflict within the couple relationship; work–family enrichment appears to promote optimal, warm and consistent parent–child interactions, and is associated with a high-quality couple relationship.

These findings provide the impetus for the provision of optimal employment conditions for mothers' of young children. Reducing work–family conflict and enhancing work–family enrichment is protective for family relationships, family environments and ultimately, for children's outcomes.