Parental work characteristics and time with children: The moderating effects of parent’s gender and children’s age

Abstract

Spanish

French

This article investigates the association between work characteristics and parent–child interaction time. In addition to studying the commonly considered working hours, the authors investigated the effects of job demands and resources such as job insecurity, autonomy and non-standard hours. Moreover, they analysed whether these associations were different for fathers than for mothers and for parents with young vs adolescent children. The authors analysed self-collected data on 2593 Dutch parents and found that parents participated more in parent–child activities when they worked shorter hours, experienced more autonomy, could be reached by their children at work and worked during non-standard hours. Nevertheless, the work characteristics had little explanatory value additional to the working hours. The association between working hours and parent–child time was weaker for mothers and for parents with young children. Moreover, the beneficial effect of non-standard hours was stronger for mothers and autonomy more relevant for parents with adolescent children.

Keywords

gender roles parenting the Netherlands working conditions working parents

Introduction

It is a general concern that families spend increasingly less time together because female labour force participation and work demands have increased (e.g. Bianchi et al., 2006; Gauthier et al., 2004; SCP, 2006). However, notwithstanding the fact that time is a fixed resource, research has found surprisingly little empirical evidence for the expectation that parents who work more hours spend substantially less time with their children (e.g. Bianchi, 2000; Bianchi et al., 2006; Crouter et al., 2001; Nock and Kingston, 1988; Roeters et al., 2009). The presumption that work demands cut into parent–child time is grounded in the conflict approach (e.g. Eby et al., 2005; Greenhaus and Beutell, 1985), arguing that work demands reduce parents’ available time and energy. In this study we re-examine the association between parental work characteristics and parent–child time by investigating whether this association is stronger when we consider a wider range of work characteristics and specify relevant conditions that previous studies have not yet looked into.

The aim of this study is two-fold. First, although most studies have focused on work hours, work is more than spending time away from home. Work demands absorb energy and attention as well as time and previous research showed that work demands reduce parental and child well-being (e.g. Baxter et al., 2007; Strazdins et al., 2006; Thompson et al., 1999). Work characteristics other than work hours received relatively little attention, however, and the studies that did address work characteristics tended to retain a focus on time-related dimensions, such as non-standard work schedules (Baxter, 2009; Estes, 2004; Noonan et al., 2007). We extend this research by studying a wider range of work demands and resources, including organizational culture and supervisor support, as possible antecedents of parent–child time.

Second, previous research has considered parents as a homogeneous group, both theoretically and in the analysis of the data. Nevertheless, fathers and mothers as well as parents of young and adolescent children may respond differently to work characteristics. Bielby (1992) claimed that men allow, and are allowed, to let paid work reduce their participation in family life, whereas this applies less to women. Moreover, the impact of work characteristics on parent–child time is likely to be limited when children are unavailable to interact with (Bianchi, 2000) and children who enter adolescence become less available because adolescents spend more time in school and with peers (Crouter et al., 2001).

Direct comparisons of work-to-family effects for fathers and mothers are scarce, although there are some notable exceptions (Bianchi et al., 2006; Galambos et al., 1995; Nock and Kingston, 1988). Nevertheless, none of these studies actually tested whether gender differences were significant. With regard to age, most studies consider children of all ages (e.g. Bianchi et al., 2006; Zick and Bryant, 1996), whereas some focused either on young children (Greenberger et al., 1994) or adolescents (Crouter et al., 2001). Differences in findings may therefore have resulted from sample differences. Studying whether the strength of the association between work characteristics and parent–child time is conditioned by the parent’s gender and the children’s age group is the second contribution of this study.

The current study is set in the Netherlands, which is characterized by a high prevalence of part-time work: in 2005, 61 percent of employed women held a part-time job (30 hours per week or less), which is the highest percentage in Europe, followed by Germany and the UK with part-time rates of 39 percent (SCP, 2006: 145). In the US, 18 percent of the women in salaried employment held a part-time job in 2005 (SCP, 2006). Moreover, when women work part-time in the Netherlands they usually have small part-time jobs of fewer than 20 hours per week (SCP, 2006). Part-time work has become increasingly popular in the Netherlands among men as well: in 2005, 15 percent of Dutch men had a part-time job (SCP, 2008). Dutch parents strongly feel that a child is harmed when both parents work full-time (SCP, 2006, 2009) and have wide access to part-time jobs as Dutch employees are entitled to reduce their work hours unless employers can prove that the company suffers substantially (SCP, 2006).

Theoretical framework

The theoretical framework first sets out the expectations regarding the main effects of the work characteristics on parent–child time. Subsequently, we argue why the impact of work characteristics on parent–child time may be stronger for some groups of parents than for others.

Work characteristics and parent–child time

The conflict approach is the most widely used approach in studying the impact of parental paid work on family life (Eby et al., 2005). Greenhaus and Beutell (1985) set out three sources of conflict between work and family roles: time-, strain- and behavioural-based conflict. Participation in the work role requires time, can produce strain and can require behaviour that is incompatible with parenting behaviour. Hence, higher work demands are argued to decrease the amount of parent–child time.

Parents’ available time, energy and attention are likely to be limited not only by work hours, but by other work demands as well. Plausible candidates are job insecurity and an organizational culture that is unsupportive towards employees’ family demands. Because job insecurity reduces individual well-being (Van der Lippe, 2007), parents with insecure jobs may have less energy to actively take up activities with their children. Moreover, these parents may be inclined to increase their work involvement in order to safeguard their job or career. The organizational culture may also affect the amount of time, energy and attention that is available for the family (Glass and Estes, 1997). Employees may be reluctant to make use of family-friendly policies when they fear that their career will suffer when they do. Moreover, when the amount of face-time at work is considered to be an indicator of organizational commitment and employees receive little family support from their managers, parents are discouraged to prioritize family over work responsibilities (Thompson et al., 1999), hindering parental involvement. For example, a parent may be less inclined to take a day off when a child is sick.

The empirical literature indeed suggests that paid work hours, job insecurity and a family-unfriendly organizational culture restrict parents’ involvement in the family. Employed mothers spend less time with their children than their non-employed counterparts (Bianchi, 2000; Bianchi et al., 2006; Sandberg and Hofferth, 2001) and longer work hours decrease parent–child time (e.g. Bianchi et al., 2006; Zick and Bryant, 1996). These effects have not always been found, however, and the effects that were found were small (Bianchi, 2000; Crouter et al., 2001; Nock and Kingston, 1988). Prior research showed that job insecurity increases stress and decreases individual well-being, family satisfaction and functioning (Mauno and Kinunnen, 1999). Similarly, parents who work in ‘family-unfriendly’ organizational cultures feel less entitled to make use of work–family benefits and report more feelings of work–family conflict (e.g. Grandey et al., 2007; Thompson et al., 1999). Our first hypothesis thus reads: More work demands (as indicated by the number of work hours, job insecurity and a less family-friendly culture) are associated with less parent–child time (H1: Work Demands Hypothesis).

Job resources

The enrichment approach (Greenhaus and Powell, 2006) proposes that performance in one role can benefit from involvement in another role. Skills, abilities and values that are acquired in the work domain can be applied in the family domain and positive experiences at work may increase a parent’s general well-being, improving interactions at home (Greenhaus and Powell, 2006). Furthermore, the job demands–resources model (Demerouti et al., 2001) argues that work performance and well-being are enhanced when a job entails more resources, such as supervisor support.

Flexibility in scheduling accommodates parental involvement as it helps parents to fine-tune work and family demands. Parents can plan their work efficiently and around child-related activities when they are able to plan and organize their work tasks flexibly (e.g. leaving work early to pick up children from school). Moreover, job autonomy enables employees to finish their work more efficiently (Hackman and Oldham, 1976), freeing time for one’s children. One particular component of job flexibility that we explore here, and that previous research has not yet looked into, is parents’ availability at work. When children are able to reach their parents at work, parents can more easily respond to family emergencies or general needs, such as a bad experience at school. Thomas and Ganster (1995) identified supervisor support as an additional relevant component of the family-supportive work environment. Supportive supervisors are likely to accommodate employees’ family needs and be more lenient when family responsibilities interfere with work responsibilities. Moreover, supportive colleagues can assist in meeting the direct work demands and compensate for parents’ absence. Finally, on the basis of the enrichment approach it can be expected that the beneficial effects of job resources on individual well-being create energy, stimulating parents to actively take up and participate in child-related activities (Glass and Estes, 1997; Greenhaus and Powell, 2006).

Previous studies demonstrated that parents with flexible schedules experience less work–family conflict, are healthier and are more satisfied with family life (e.g. Butler et al., 2009; Thomas and Ganster, 1995). Autonomy increases productivity (Hackman and Oldham, 1976) and organizational commitment and decreases burn-out (e.g. Bakker et al., 2003; Demerouti et al., 2001). Moreover, autonomy buffers the detrimental effects of work demands on individual well-being (Karasek and Theorell, 1990). With regard to social contacts at work, empirical studies showed that when these are of a higher quality, employees’ are mentally and physically more healthy, the quality of parenting is higher and work–family conflict is lower (e.g. Bakker et al., 2003; Greenberger et al., 1989; Thomas and Ganster, 1995). We thus expect: More job resources (as indicated by the level of flexibility and social support from supervisors and colleagues) are associated with more parent–child time (H2: Job Resources Hypothesis).

Non-standard work hours

Non-standard scheduling is likely to be relevant for the organization of family life, but is difficult to interpret as either a work demand or resource. Non-standard work hours provide specific challenges for family life but provide flexibility at the same time (Mills and Täht, 2010). On the one hand, work hours outside regular office hours make it more difficult to separate work and family life and are likely to cut into family life severely, as most family events take place during evenings and weekends (Lesnard, 2008; Mills and Täht, 2010; Presser, 1999). On the other hand, non-standard hours decrease the need for alternative childcare arrangements (Mills and Täht, 2010) and come with more flexibility. For example, morning shifts enable parents to pick up their children from school. Working outside office hours may also reflect a strategy of freeing time from work when children are available (e.g. after school) and catching up with work when children are asleep or away. Barnett and Gareis (2007) found no differences in the mother–child time of mothers who worked during days versus evenings, but Lesnard (2008) demonstrated that although off-scheduling in the evening reduced family time with both parents, it increased one-on-one father–child time when the father finished work directly after school hours.

Because previous research is inconclusive, we formulated two competing hypotheses: Working non-standard hours is associated with more parent–child time (H3: Non-standard Hours Hypothesis I) and working non-standard hours is associated with less parent–child time (H4: Non-standard Hours Hypothesis II). A drawback of formulating competing hypotheses is that a lack of findings either reflects a lack of effects or that the two effects outweigh each other. One way of disentangling this is to formulate expectations on the conditions under which one of the two effects dominates, as we do in the following section.

The moderating effects of gender and age group

Gender

Bielby (1992) argued that societal norms on work and family commitment are gender specific. Whereas men’s main responsibility is to provide for the family, women are expected to prioritize the family domain over the work domain. Subsequently, men let work interfere with the family more easily than women (e.g. working late and skipping a family meal) and women more easily let family interfere with work (e.g. skipping a meeting to pick up a sick child).

Previous empirical results are mixed. Some studies found that work characteristics had a similar impact on men and women (e.g. Galambos et al., 1995), but others indeed found that the effects of work characteristics on family life were stronger for men (e.g. Bumpus et al., 1999; Menaghan and Parcel, 1990; Nock and Kingston, 1988). Also, mothers are more likely to scale back and adjust their work to accommodate family demands when children are young (Becker and Moen, 1999; England and Farkas, 1986). We therefore hypothesize that the negative effects of work demands are weaker for mothers than for fathers (H5: Gender/Work Demands Hypothesis).

Bielby’s (1992) discussion of gender differences relates to the effects of work demands but could be extended to job resources, predicting that women protect family life from the influences of paid work altogether. However, Bielby’s arguments also make it likely that women are more motivated than men to use job resources for the benefit of the family. Thompson et al. (1999) as well as others (e.g. Grandey et al., 2007; Hochschild, 1997) showed that although family-friendly policies are gender neutral, organizational norms are more accepting towards utilization among women. We therefore expect that the positive effects of job resources on parent–child time are stronger for mothers than for fathers (H6: Gender/Job Resources Hypothesis). Finally, we hypothesize that non-standard hours are less detrimental for parent–child time for mothers than for fathers (H7: Gender/Non-standard Hours Hypothesis) because women are more likely to plan their work hours strategically to fit family demands (Becker and Moen, 1999).

Young vs adolescent children

Bianchi (2000) argued that when children are at school, in extracurricular activities, or spend time with friends, it is irrelevant whether their mother works during this time. In the current study we assume that the availability of children decreases as children grow older. Once they enter high school (in the Netherlands children do so around the age of 12), their school hours increase, they are required to do more homework and the peer group increases in significance (Crouter et al., 2001; Giordano, 2003). We therefore expect differences in work characteristics to be less salient during adolescence than during the early childhood stages. A second reason why paid work may become less salient is that parent–child time becomes scarcer during the child’s adolescence. Crouter et al. (2001) found that father–adolescent time was unaffected by paternal work overload and argued that an additional work hour of the father made no difference because joint activities were rare. We extend prior research by directly comparing parents with young children and adolescent children and hypothesize that the effects of work demands, resources and non-standard hours on parent–child time are weaker in households with adolescent children than in households with young children (H8: Adolescent/Work Characteristics Hypothesis).

Method

Data and sample

The analyses are based on self-collected household data in the Netherlands. The sample was drawn from the Taylor Nelson Sofres – Netherlands Institute for Public Opinion (TNS–NIPO) Household Panel comprising 200,000 households. The information available on the panel members allowed us to approach those households relevant to our study. The sample was representative for the Dutch population in terms of earner types, educational level and region, although ethnic minorities were underrepresented. A total of 2620 households (equalling 4912 parents) were approached and 1874 households (2816 parents, 57 percent) returned at least one questionnaire. We selected those parents who were in paid employment and had at least one child aged 17 or younger. This resulted in an effective sample size of 2593 parents. Because 862 of the two-parent households returned both of the questionnaires, the data are partly nested.

Measures

Parent–child time was measured by asking parents to rate the frequency of a list of child-related activities, such as having meals and watching television together, in the week preceding the survey. This scale was based on Bianchi et al.’s (2006: 79) ‘estimated daily activities with children’ measure and contained 18 one-on-one parent–child activities (without the partner) and 10 family activities (in which the partner participated as well). The response categories ranged from 0 = never to 6 = more than three times per day. We took the mean (α: .88 for fathers and .89 for mothers), resulting in a scale ranging from 0 (low frequency) to 6 (high frequency). We minimized the skewness of the dependent variable by performing a zero-skewness log transformation in Stata.

Paid work hours were measured by asking parents how many hours they worked in the week preceding the survey (including overtime). The scale measuring job insecurity was based on Crompton et al. (2007) and consisted of five items (α: .82, e.g. ‘I am worried that I will lose my job’, 1 = totally disagree to 5 = totally agree). We took the average score. The family friendliness of the organizational culture was measured through a selection of 12 items from Thompson et al.’s (1999) ‘Family-friendliness’ scale (e.g. ‘To get ahead at this organization, employees are expected to work overtime on a regular basis’, α: .89), ranging from 1 = totally disagree to 5 = totally agree. We recoded the items so that higher values corresponded with a less family-friendly organizational culture and took the mean score.

Schedule flexibility was measured by taking the mean score of the respondent’s answers on two questions: ‘To what extent do you determine when you start and end work?’ (1 = others fully determine this to 5 = I fully determine this) and ‘When something unexpected happens, is it possible for you to take time off or work from home?’ (1 = this is impossible to 5 = this is very well possible). Job autonomy was based on the task authority dimension of Karasek and Theorell’s (1990) control scale, and consisted of three items (e.g. ‘Can you decide how to perform your work?’, 1 = never to 5 = always, α: .83). We measured availability at work with a single self-designed item, asking respondents to what extent they agreed with the statement ‘When necessary, my children can reach me at any moment of the day’ (1 = fully disagree to 5 = fully agree). The measure for informal supervisor support was based on Greenberger et al. (1989). Items related to general socio-emotional support (seven items, e.g. ‘My supervisor is willing to listen to personal problems’) and support with regard to family demands (two items, e.g. ‘My supervisor is understanding when I receive phone calls from home at work’, 1 = fully disagree to 5 = fully agree). We took the mean over the nine items (α: .90). Social support from colleagues is based on a scale from the time competition survey (Van der Lippe and Glebbeek, 2003) and concerned emotional (five items, e.g. ‘I feel at home among my colleagues at work’, 1 = fully disagree to 5 = fully agree) and instrumental support (three items, e.g. ‘Can you ask your colleagues for help?’, 1 = never to 5 = always). All eight items loaded on a single factor (α: .87) so we took the mean.

In order to separate work outside office hours as a result of the parent’s own planning from fixed work hours (determined by the parent’s supervisor), we constructed two measures for non-standard work hours. First, those parents who reported working during evenings or weekends on a regular basis, but who did not work in permanent or rotating shifts, were assigned a 1 on the dummy variable non-fixed standard hours. Second, parents who reported working during evenings or weekend on a regular basis and worked in fixed or rotating shifts, scored a 1 on the dummy variable fixed non-standard hours.

To test for moderating effects, two dummies were included, indicating whether the parent was female (0 = male, 1 = female) and whether the youngest child was an adolescent (0 = no, 1 = yes). We controlled for the number of children and the parent’s years of education. We also included a dummy indicating whether the respondent was self-employed (0 = no, 1 = yes). Self-employment differs from wage jobs in multiple aspects. Self-employment results in more work–family conflict, but also enhances autonomy and flexibility (Parasuraman and Simmers, 2001). Self-employed respondents were not presented with the questions relating to schedule flexibility, the organizational culture and social support. Because we wanted to preserve the full sample, self-employed respondents were assigned the mean score of the respondents in paid employment on these specific variables. This is a common procedure, used by Poortman and Kalmijn (2002: 184) among others. Descriptive statistics are presented in Table 1.

Table 1.

Descriptive statistics (N = 2593)

	M	SD	Range	Correlation with parent–child time (transformed)
Parent–child time (untransformed)	1.98	.55	1–5.36	1.00
Work demands
Work hours	31.02	14.76	1–100	−.27***
Job insecurity	2.34	.77	1–5	−.04
Restrictive organizational culture ^a	2.61	.64	1–5	−.04
Job resources
Flexibility	2.89	1.01	1–5	−.01
Autonomy	3.26	.98	1–5	.01
Availability	4.28	.90	1–5	.12***
Supervisor support ^a	3.58	.73	1–5	.01
Colleague support ^a	3.53	.65	1.38–5	.04*
Non-standard hours
Non-fixed (1 = yes)	.20	.40	0–1	.02
Fixed (1 = yes)	1.18	.38	0–1	.01
Control variables
Female (1 = yes)	.48	.50	0–1	.25***
Number of children	1.83	.80	0–5	.12***
Age youngest child	7.58	5.44	0–17	−.39***
Adolescent child (1 = yes)	.31	.46	0–1	−.36***
Years of education	12.59	2.49	5–20	.04^†
Self-employment (1 = yes)	.09	.29	0–1	−.01

N = 2350 because this variable was only measured for the respondents in paid employment.

†

= p < .10; * = p < .05; ** = p < .01; *** = p < .001.

Method of analysis

The hypotheses were tested using OLS regression with a cluster correction, as implemented in Stata. The cluster correction was applied because the individual parents were nested in households. We estimated multiple models. First, we ran a model that included only the control variables (Model 1). In the second and third step we added the work hours (Model 2) and the other work characteristics (Model 3). Fourth, we ran the models with the interaction effects. In order to avoid entering too many variables in the model, we included one interaction term at a time, while keeping the control variables and work characteristics in the model.

Results

Main effects

Table 2 presents the results of the models with the main effects. To examine how work characteristics contribute to the explanation of parent–child time, we first estimated a model that included the controls only (Model 1). As expected, mothers participated more in child-related activities than fathers and parent–child time was higher when children were younger and the number of children was higher. Two variables had a marginally significant effect: parents spent less time with their children when they had an adolescent child and, unexpectedly, when they were higher educated.

Table 2.

The main effects of the work characteristics on parent–child time (N_individuals = 2593, N_households = 1784)

	Hyp.	Model 1		Model 2		Model 3
		B	SD	B	SD	B	SD
Work demands
Work hours	–			−.004***	.001	−.005***	.001
Job insecurity	–					−.004	.011
Restrictive organizational culture	–					−.008	.015
Job resources
Flexibility	+					−.008	.008
Autonomy	+					.023**	.008
Availability	+					.061***	.009
Supervisor support	+					−.033*	.014
Colleague support	+					.022^†	.013
Non-standard hours
Non-fixed (1 = yes)	−/+					.050**	.019
Fixed (1 = yes)	−/+					.036^†	.020
Control variables
Female (1 = yes)		.218***	.012	.146***	.017	.134***	.017
Adolescent child (1 = yes)		−.057^†	.030	−.061*	.029	−.058*	.029
Age youngest child		−.026***	.003	−.025***	.003	−.027***	.003
Number of children		.056***	.012	.053***	.012	.052***	.012
Years of education		−.005^†	.003	−.003	.003	−.002	.003
Self-employed (1 = yes)						−.014	.026
R²		.243		.259		.285

†

= p < .10; * = p < .05; *** = p < .001.

Model 2 included work hours. Parents who worked longer hours reported significantly less parent–child time, but the coefficient was small (–.004) and the variable added little to the explained variance. Adding the other work demands to the model (Model 3) also resulted in only a small increase in R². Nevertheless, six out of the 10 work characteristics reached (marginal) significance. Hypothesis 1 received partial support. Parents spent less time with their children when they worked longer hours, but parent–child time did not depend on the level of job insecurity or the organizational culture. As predicted by Hypothesis 2, parent–child time increased when parents reported more job autonomy, a higher availability and more support from colleagues (marginally significant). The association with flexibility was non-significant however, and supervisor support decreased rather than increased parent–child time. The results further suggest that parent–child time was higher among parents who worked during non-standard hours, both when these hours were voluntary (i.e. non-shift and non-fixed) and enforced (marginally significant), supporting Hypothesis 3, but contradicting Hypothesis 4. Additional analyses (results not shown) indicated that the effects of non-standard hours were mostly attributable to the effects of working during the weekend and that the negative effect of supervisor support only applied to the higher educated respondents.

Moderating effects

In the subsequent analyses, two additional models were estimated for each of the 10 work characteristics: one model including an interaction term with the dummy ‘female’ and one model including an interaction term with the dummy ‘adolescent child’. Each model included the other work characteristics and the controls. Table 3 presents the coefficients of the main and interaction effects for each model.

Table 3.

The moderating effects of parent’s gender and children’s age on the association between work characteristics and parent–child time

	Gender of parent			Age group of children
	Female (1 = yes)	Work characteristic	Interaction	Adolescent (1 = yes)	Work characteristic	Interaction
Work demands hypothesis
	+	–	+	–	–	+
Work hours	.064	−.006***	.002^†	−.201***	−.006***	.005***
Job insecurity	.177***	.005	−.019	−.086	−.008	.152
Restrictive organizational culture	.029	−.027	.040^†	−.037	−.006	−.008
Job resources hypothesis
	+	+	+	–	+	–
Flexibility	.139**	−.008	−.002	−.110*	−.014	.019
Autonomy	.061	.016	.023	−.256***	.010	.064**
Availability	.073	.055***	.014	.034	.065	−.021
Supervisor support	.203*	−.024	−.019	−.019	−.030^†	−.011
Colleague support	.132	.022	.000	−.051	.023	−.002
Non-standard hours hypothesis
	+	+/–	+	–	+/–	−/+
Non-fixed (1 = yes)	.118***	.007	.092**	−.070*	.031	.069^†
Fixed (1 = yes)	.137***	.043	−.015	−.048	−.053*	−.051

Note: The cells in the table present the unstandardized coefficients in the different models that tested the interaction effects. For example, in the model that tested the interaction effect between gender and working hours the coefficient of the main effects of gender and working hours were respectively .064 and –.006. The coefficient of the interaction effect between gender and working hours was .002.

†

= p < .10; * = p < .05; ** = p < .01; *** = p < .001.

Gender

The parent’s gender moderated few of the associations between the work demands and parent–child time. In line with Hypothesis 5, the interaction term with work hours was marginally significant, suggesting that the negative impact of work hours was weaker for mothers than for fathers. The interaction between gender and the family-friendliness of the culture was marginally significant as well. However, when combined, the results suggested that when the organizational culture was less family-friendly, father–child time was lower, but mother–child time was higher, which contradicts our expectation that work demands reduce parent–child time (H1). Because models with interaction effects cannot be conclusive with regard to the direction of the effects, it is also possible that the positive effect of being female was stronger in organizations with less family-friendly organizational cultures. None of the resources interacted with the gender of the parent, rejecting Hypothesis 6. Finally, the parent’s gender interacted with the dummy variable indicating regular, non-fixed non-shift work during non-standard hours. The coefficients suggest that the positive association between working non-standard hours and parent–child time was stronger for mothers than for fathers, supporting Hypothesis 7.

Adolescent children

In line with Hypothesis 8 the negative association between parental work hours and parent–child time was weaker for parents with adolescent children than for parents with younger children. Moreover, the beneficial effect of job autonomy was stronger, rather than weaker, for parents with adolescent children as compared to parents with younger children. Similarly, the effect of non-fixed/non-shift non-standard hours interacted (marginally significant) with the age group of the children, but the effect was stronger, rather than weaker for parents with adolescent children. The children’s age group did not moderate the effects of the other work characteristics, although the effects for job insecurity and supervisor support had the predicted signs. Thus, Hypothesis 8 received partial support.

Finally, we estimated additional models in which fathers and mothers were analysed separately to validate our findings (results not shown). Consistent with the previous results, the size effect of work hours was larger in the model with fathers than in the model with the mothers. The size effects of the other work demands and resources were larger for the mothers than for the fathers, but the results did not show a substantially different pattern than the pattern that emerged from the models with the gender interactions. Finally, the effect of organizational culture did not yield a significant effect in the model with mothers only, which warrants further caution when interpreting the unexpected interaction effect between gender and organizational culture.

The interactions between work characteristics and the presence of adolescent children yielded similar effects for fathers and mothers. The interaction effect with autonomy only reached significance for mothers, suggesting that autonomy is a more salient resource for mothers. Although we expected a similar effect for non-fixed non-standard hours, the models yielded a non-significant interaction effect between non-standard hours and the presence of an adolescent for both fathers and mothers. Furthermore, we found that the presence of an adolescent buffered the association between work hours and parent–child time for both fathers and mothers.

Finally, we looked at gender differences in the effects of the controls. Prior research showed that women are more responsive to family demands than men (e.g. Mattingly and Bianchi, 2003; Sanchez and Thomson, 1997) and more often use self-employment as a strategy to reconcile work and family responsibilities (Budig, 2006). Also, women’s educational levels have a stronger effect than men’s (Sayer et al., 2004), and gender differences in the division of labour decrease when children enter adolescence (Craig and Sawrikar, 2009). In line with our expectations, the separate models for fathers and mothers suggested that all controls had a stronger impact for the mothers than for the fathers.

Discussion

The aim of this study was two-fold. First, we analysed whether the time parents spent with their children depended on their work demands and resources. In addition to the commonly examined work hours, we considered job insecurity, the family-friendliness of the organizational culture, flexibility, autonomy, availability, social support and the prevalence of non-standard hours. Second, we examined whether the association between these work characteristics and parent–child time was different for fathers and mothers and for parents of adolescent children and parents of younger children.

In line with the conflict approach, we found that parents who worked longer hours participated less in parent–child activities. None of the other work demands appeared to be relevant however, and differentiating between fathers and mothers and adolescent and younger children did not change this. With regard to the job resources, parent–child time increased when parents had more autonomy at work and when their children could reach them at work. We also found that parents who regularly worked non-standard hours, but did not work fixed non-standard hours or in shifts, spent more time with their children. This could indicate that parents plan their time strategically, leaving work when children are available and catching up on work in the evenings or weekends when children are sleeping or away. The results concerning social support were remarkable: although support from colleagues increased parent–child time as expected, support from supervisors reduced parent–child time. These results mainly applied to the higher educated parents and tentatively suggest that supervisor support increases work commitment, withdrawing energy and attention from the family as a consequence. Prior research indeed showed that supervisor support increases work commitment (e.g. Thompson et al., 1999) and the risk of eroding boundaries between the work and family domain may be higher for higher educated parents.

Our expectation that women would protect family life more than men held true for some of the work characteristics. Work hours reduced parent–child time less for mothers than for fathers and the beneficial effects of autonomy and working non-standard hours applied more to mothers. Because previous research on gender differences in work-to-family effects focused on work demands (e.g. Bumpus et al., 1999; Nock and Kingston, 1988), it is interesting to note that non-standard hours had a stronger positive effect for women than for men. This suggests that women plan their work more strategically around the availability of the children. We also found that the family-friendliness of the organizational culture interacted with the parent’s gender (the effect was marginally significant). The direction of the effects can be interpreted in two ways: they could imply that mother–child time increases when a culture is less family-friendly, whereas the father–child time is reduced, but our findings could also indicate that the gender differences in parent–child time are larger in less family-friendly organizational cultures. The second explanation is most likely.

In response to Bianchi’s (2000) claim that research on the association between work and family life should take into account that children are often unavailable, we compared the impact of work characteristics on parents with young and adolescent children. Work hours appeared to matter less when children had reached adolescence, which could be attributed to the fact that there is less joint time to be affected. It should be noted, however, that our theoretical assumption that children are less available when they are older might be problematic. Although adolescents will spend more time with peers and doing homework, their bedtime is also likely to be postponed, providing more opportunity for interaction. For autonomy and non-standard hours, we found that these mattered more during adolescence. Craig (2006) suggested that parents of adolescent children attach increasing importance to autonomy and flexibility and apparently autonomy also becomes more relevant for the actual time parents spend with their children. Autonomy and non-standard hours may become more salient resources that accommodate paid work to the unpredictable family demands.

While interesting findings emerged from the current study, effects of work characteristics remained small and limited. Examining a wider range of work characteristics and specifying conditions under which the effects are more likely to occur did not change this. This may be explained in two ways: the effects may simply be minimal or they may not have been detected because the theoretical approach and data did not go into enough detail. Because the combination of paid work and family responsibilities remains a pressing societal and scientific issue (e.g. Moen, 2003), it is worth the effort to further consider the second possibility. Future research needs to exclude the possibility that paid work impacts on family life in ways that have not been covered here. For example, it is likely that ‘qualitative’ aspects of family life, such as the extent to which a joint activity is enjoyed by parents and children, are affected more strongly by paid work, as behaviour may be less elastic than feelings. Prior research focusing on more qualitative outcomes, such as parenting quality, indeed found more effects of paid work (e.g. Baxter et al., 2007; Bumpus et al., 1999; Crouter et al., 1999; Strazdins et al., 2006). Hence, a study providing a joint analysis of the effects on quantitative and qualitative family outcomes would be useful. Also, research could further distinguish between different types of activities, for example between parent–child activities that do and do not include the other parent. Future research could also provide more insight into the exact strategies parents use to minimize the detrimental impact of work. Whereas the conflict approach assumes that parents are passively reacting to the work demands that are imposed upon them, previous research showed that parents find ways to reconcile work and family demands or select work hours to match the time they want to spend with their children (e.g. Becker and Moen, 1999; Moen and Wethington, 1992). Mothers in particular are likely to accommodate their work to their family circumstances (e.g. England and Farkas, 1986). If work characteristics are indeed endogenous, rather than exogenous, this would imply that we overestimated the effects of paid work. Panel data could provide more insight into the causality of such processes. The implications of such strategies could also be studied by means of a cross-national comparison of association between work hours and parent–child time. If the notion of strategic action held true, the association between parent–child time and work hours would differ between countries, depending on how difficult it is to reconcile work and family life and to select one’s own work hours. This would also provide insight into the generalizability of the results. Although Van den Broek (2006) claimed that the determinants of parent–child time are similar in the Netherlands and the US, differences in the country contexts may condition the effects of work. The current study was set in the Netherlands where the wide accessibility of part-time work could have buffered work encroachments, as Bianchi et al. (2006) suggested. Part-time work provides parents with more flexibility (e.g. Becker and Moen, 1999; Gornick and Meyers, 2003; Jacobs and Gerson, 2004). This could facilitate the reconciliation of work and family demands for Dutch parents. Yet, while this would account for mothers, the explanation applies much less to fathers, who still predominantly work full-time in the Netherlands. Alternatively, it could also be that parent–child time is affected more strongly by family characteristics than by work characteristics. Note that future research could further explore the modest effects through a cross-national comparison, such as, for example, Gornick and Meyers (2003) did.

With regard to the data, we may have underestimated the effects of work characteristics on parent–child time because the parents who experience most time pressure are less likely to participate in the panel from which we derived our sample. Furthermore, asking parents to estimate the frequency of activities with their children, as we did, is common in the literature (e.g. Bianchi et al., 2006: 79), but results in less detailed information than time diary data and may prompt socially desirable answers.

In conclusion, this study showed that the time parents spent with their children partly depended on the characteristics of their job, but that these effects were small. Parents spent less time with their children when they worked longer hours, and parent–child time also appeared to be related to the content and organization of their work. Moreover, the finding that the gender of the parent and the age group of the children condition the effects of work hours may partly account for the inconsistent findings of previous research on the association between work hours and time with children.

Footnotes

Acknowledgements

The authors would like to thank Lieke ten Brummelhuis, Elianne van Steenbergen and Gerdientje Danner-Vlaardingerbroek for their useful comments on a previous version of the manuscript.

Funding

This research was supported by a Utrecht University High Potential grant awarded to Tanja van der Lippe and Esther Kluwer (2004).

Biographical notes

Anne Roeters is assistant professor at the Department of Sociology, Utrecht University/Interuniversity Centre for Social Science Theory and Methodology (ICS). The current study is part of her PhD thesis on the impact of parental work characteristics on family time.

Tanja van der Lippe is Professor of Sociology of Households and Employment Relations and head of the Department of Sociology at Utrecht University/ICS. Her main research interests are family sociology, organizational sociology, gender inequality and cross-national research.

Werner Raub is Professor of Theoretical Sociology at the Department of Sociology, Utrecht University/ICS. The main themes in his research are theoretical sociology, economic sociology and mathematical sociology, with a focus on the use of game-theoretic models in social science.

Esther Kluwer is an associate professor at the Department of Social and Organizational Psychology, Utrecht University. Her research focuses mostly on close relationships, and the themes of revenge and forgiveness, post-divorce relationships, transition to parenthood and work–family balance in particular.

References

Bakker

Demerouti

De Boer

Schaufeli

(2003) Job demands and job resources as predictors of absence duration and frequency. Journal of Vocational Behavior 62(2): 341–356.

Barnett

Gareis

(2007) Shift work, parenting behaviors, and children’s socioemotional well-being: A within-family study. Journal of Family Issues 28(6): 727–748.

Baxter

(2009) Parental time with children: Do job characteristics make a difference? Research Paper No. 44. Australian Institute of Family Studies, Melbourne.

Baxter

Gray

Alexander

Strazdins

Bittman

(2007) Mothers and fathers with young children: Paid employment, caring and wellbeing. Social Policy Research Paper No. 30. Department of Families, Community Services and Indigenous Affairs.

Becker

Moen

(1999) Scaling back: Dual-earner couples’ work–family strategies. Journal of Marriage and the Family 61(4): 995–1007.

Bianchi

(2000) Maternal employment and time with children: Dramatic change or surprising continuity? Demography 31(4): 401–414.

Bianchi

Robinson

Milkie

(2006) Changing Rhythms of American Family Life. New York: Russell Sage Foundation.

Bielby

(1992) Commitment to work and family. Annual Review of Sociology 18: 281–302.

Budig

(2006) Gender, self-employment, and earnings: The interlocking structures of family and professional status. Gender and Society 20(6): 725–753.

10.

Bumpus

Crouter

McHale

(1999) Work demands of dual-earner couples: Implications for parents’ knowledge about children’s daily lives in middle childhood. Journal of Marriage and the Family 61(2): 465–475.

11.

Butler

Grzywacz

Ettner

Liu

(2009) Workplace flexibility, self-reported health, and health care utilization. Work and Stress 23(1): 45–59.

12.

Craig

(2006) Work and balance: The adolescent years. Newsletter Social Policy Research Centre 94: 1–7.

13.

Craig

Sawrikar

(2009) Work and family: How does the (gender) balance change as children grow? Gender, Work and Organisation 16(6): 684–709.

14.

Crompton

Lewis

Lyonette

(eds) (2007) Women, Men, Work and Family in Europe. New York: Palgrave Macmillan.

15.

Crouter

Bumpus

Maguire

McHale

(1999) Linking parents’ work pressure and adolescents’ well-being: Insights into dynamics in dual-earner families. Developmental Psychology 35(2): 1435–1461.

16.

Crouter

Bumpus

Head

McHale

(2001) Implications of overwork and overload for the quality of men’s family relationships. Journal of Marriage and Family 63(2): 404–416.

17.

Demerouti

Bakker

Nachreiner

Schaufeli

(2001) The job demands-resources model of burnout. Journal of Applied Psychology 86(3): 499–512.

18.

Eby

Casper

Lockwood

Bordeaux

Brinley

(2005) Work and family research in IO/OB: Content analysis and review of the literature (1980–2002). Journal of Vocational Behavior 66(1): 124–197.

19.

England

Farkas

(1986) Households, Employment, and Gender: A Social, Economic, and Demographic View. New York: Aldine.

20.

Estes

(2004) How are family-responsive workplace arrangements family friendly? The Sociological Quarterly 45(4): 637–661.

21.

Galambos

Sears

Almeida

Kolaric

(1995) Parents’ work overload and problem behavior in young adolescents. Journal of Research on Adolescence 5(2): 201–223.

22.

Gauthier

Smeeding

Furstenberg

(2004) Are parents investing less time in children: Trends in selected industrialized countries. Population and Development Review 30(4): 647–671.

23.

Giordano

(2003) Relationships in adolescence. Annual Review of Sociology 29: 268–281.

24.

Glass

Estes

(1997) The family responsive workplace. Annual Review of Sociology 23: 289–313.

25.

Gornick

Meyers

(2003) Families that Work: Policies for Reconciling Parenthood and Employment. New York: Russell Sage Foundation.

26.

Grandey

Cordeiro

Michael

(2007) Work–family supportiveness organizational perceptions: Important for the well-being of male blue-collar hourly workers? Journal of Vocational Behavior 71(3): 460–478.

27.

Greenberger

Goldberg

Hamiil

O’Neil

Payne

(1989) Contributions of a supportive work environment to parents’ well-being and orientation to work. American Journal of Community Psychology 17(6): 755–783.

28.

Greenberger

O’Neil

Nagel

(1994) Linking workplace and homeplace: Relations between the nature of adults’ work and their parenting behaviors. Developmental Psychology 30(6): 990–1002.

29.

Greenhaus

Beutell

(1985) Sources of conflict between work and family roles. Academy of Management Review 10(1): 76–88.

30.

Greenhaus

Powell

(2006) When work and family are allies: A theory of work–family enrichment. Academy of Management Review 31(2): 72–92.

31.

Hackman

Oldham

(1976) Motivation through the design of work: Test of a theory. Organizational Behavior and Human Performance 16(2): 250–279.

32.

Hochschild

(1997) The Time Bind: When Work Becomes Home and Home Becomes Work. New York: Metropolitan.

33.

Jacobs

Gerson

(2004). The Time Divide: Work, Family, and Gender Inequality. Cambridge, MA: Harvard University Press.

34.

Karasek

Theorell

(1990). Healthy Work: Stress, Productivity, and the Reconstruction of Working Life. New York: Basic Books.

35.

Lesnard

(2008) Off-scheduling within dual-earner couples: An unequal and negative externality for family time. American Journal of Sociology 114(2): 447–490.

36.

Mattingly

Bianchi

(2003) Gender differences in the quantity and quality of free time: The U.S. experience. Social Forces 81: 999–1030.

37.

Mauno

Kinnunen

(1999) The effects of job stressors on marital satisfaction in Finnish dual-earner couples. Journal of Organizational Behavior 20(6): 879–895.

38.

Menaghan

Parcel

(1990) Parental employment and family life: Research in the 1980s. Journal of Marriage and the Family 52(4): 1079–1098.

39.

Mills

Täht

(2010) Nonstandard work schedules and partnership quality: Quantitative and qualitative finding. Journal of Marriage and Family 72(4): 860–875.

40.

Moen

(ed.) (2003) It’s about Time: Couple and Careers. Ithaca, NY: Cornell University Press.

41.

Moen

Wethington

(1992) The concept of family adaptive strategies. Annual Review of Sociology 18: 233–251.

42.

Nock

Kingston

(1988) Time with children: The impact of couples’ work-time commitments. Social Forces 67(1): 59–85.

43.

Noonan

Estes

Glass

(2007) Do workplace flexibility policies influence time spent in domestic labor? Journal of Family Issues 28(2): 263–288.

44.

Parasuraman

Simmers

(2001) Type of employment, work–family conflict and well-being: A comparative study. Journal of Organizational Behavior 22(5): 551–568.

45.

Peters

Van der Lippe

(2007) The time-pressure reducing potential of telehomeworking: The Dutch case. International Journal of Human Resource Management 18(3): 430–447.

46.

Poortman

Kalmijn

(2002) Women’s labour market position and divorce in the Netherlands: Evaluating economic interpretations of the work effect. European Journal of Population 18(3): 175–202.

47.

Presser

(1999) Toward a 24-hour economy. Science 284(5421): 1778–1779.

48.

Robinson

Godbey

(1999) Time for Life: The Surprising Ways Americans Use their Time. University Park, PA: Pennsylvania State University Press.

49.

Roeters

Van der Lippe

Kluwer

(2009) Parental work demands and the frequency of child-related routine and interactive activities. Journal of Marriage and Family 71: 1193–1204.

50.

Sanchez

Thomson

(1997). Becoming mothers and fathers: Parenthood, gender, and the division of labor. Gender and Society 11: 747–772.

51.

Sandberg

Hofferth

(2001) Changes in children’s time with parents: United States, 1981–1997. Demography 38(3): 423–436.

52.

Sayer

Gauthier

Furstenberg

(2004) Educational differences in parents’ time with children: Cross-national variations. Journal of Marriage and Family 66(5): 1152–1169.

53.

SCP (2006) Emancipatiemonitor 2006 [Emancipation monitor 2006]. Den Haag: Sociaal en Cultureel Planbureau.

54.

SCP (2008) Nederland deeltijdland. Vrouwen en deeltijdwerk [The Netherlands, part-time country: Women and part-time work]. Den Haag: Sociaal en Cultureel Planbureau.

55.

SCP> (2009) Emancipatiemonitor 2008 [Emancipation monitor 2008]. Den Haag: Sociaal en Cultureel Planbureau.

56.

Strazdins

Clements

Korda

Broom

D’Souza

(2006) Unsociable work? Nonstandard work schedules, family relationships, and children’s well-being. Journal of Marriage and Family 68(2): 394–410

57.

Thomas

Ganster

(1995) Impact of family-supportive work variables on work–family conflict and strain: A control perspective. Journal of Applied Psychology 80(1): 6–15.

58.

Thompson

Beauvais

Lyness

(1999) When work–family benefits are not enough: The influence of work–family culture on benefit utilization, organizational attachment, and work–family conflict. Journal of Vocational Behavior 54(3): 392–415.

59.

Van den Broek

(2006) Parental trends and time use in the Netherlands: 1975 to 2000. In: Bianchi

Robinson

Milkie

(eds) Changing Rhythms of American Family Life. New York: Russell Sage Foundation, 193–196.

60.

Van der Lippe

(2007) Dutch workers and time pressure: Household and workplace characteristics. Work, Employment and Society 21(4):693–711.

61.

Van der Lippe

Glebbeek

(2003) Codebook Time Competition. Utrecht: Department of Sociology, Utrecht University.

62.

Zick

Bryant

(1996) A new look at parents’ time spent in child routine: Primary and secondary time use. Social Science Research 25(3): 260–280.