Flexibility or Constraint? The Implications of Mothers’ and Fathers’ Nonstandard Schedules for Children’s Behavioral Outcomes

Abstract

Approximately 17.7% of the U.S. workforce is employed in a nonstandard schedule. Research thus far indicates that these schedules negatively influence children’s behavioral development. However, few studies examine the roles of the child’s gender and age. To broaden understanding of the relationships between nonstandard schedules and child behavior, and how these relationships may depend on the gender and age of the child, I analyze data from the National Longitudinal Survey of Youth-1979 and its Child Supplement from 1992 to 2006. My findings show that some types of parental nonstandard shifts, such as evening and night shifts, are associated with fewer behavioral problems among children, though these results depend on the gender and age of the child. In contrast, parents’ rotating and split shifts are associated with more behavior problems among children, indicating that it is relatively unstable and unpredictable work schedules that may have the most harmful associations with children’s outcomes.

Keywords

child behavior work and family gender and family nonstandard schedules age effects

Introduction

In the past few decades, technological breakthroughs have enabled societal advances that would have been unthinkable for the vast majority of human history. One of those breakthroughs is the ability to connect with almost any person at any time and to have 24/7 access to a world of knowledge accessible at the click of a button. The expectation that information can now be demanded at any time has reinforced a relatively recent phenomenon: the demand for services at all hours (McMenamin, 2007; Presser, 2003). Whether the 24/7 economy is viewed through the lens of a night-shift nurse, a stockbroker awake in the early hours to confer on the Japanese stock exchange, or an employee of a fast-food restaurant chain open at 2 a.m., accommodating the desire for on-demand service has become a reality for a significant percentage of employees in the American economy and across the world (Lesnard, 2008; Presser, 2003; Strazdins, Clements, Korda, Broom, & D’Souza, 2004).

According to the latest published statistics on the size of the U.S. nonstandard workforce, 17.7% of the workforce is employed in a nonstandard work schedule such as an evening, night, rotating, split, or irregular shift, and the prevalence of such schedules is only expected to grow (McMenamin, 2007). Over the next 10 years, the fastest growing occupations are projected to be in sectors of the economy in which nonstandard schedules are very common, such as in food, retail, and health services (Bureau of Labor Statistics, 2013). In the face of these changes, it is important to understand how nonstandard schedules affect the individuals working them, and the families that must respond to them.

Indeed, it is not just adults who face the consequences of their nonstandard schedules, but their children as well. Children may spend very little time with their mothers or fathers if their parents are away during evening shifts; the quality of time children spend with parents may decline if parents are exhausted because they work night shifts, and the instability created by rotating or split shifts may preclude engaging in extracurricular activities or making plans to spend time together if parents do not reliably know when they will be home from work each week. However, some parents may specifically choose nonstandard schedules in order to spend more time with their children, picking a night shift to be present during their children’s waking hours, for example, and sleeping while children are at school.

Parental involvement and time spent with children is important to children’s development: it enhances bonds between parents and children, it establishes a safe and stable environment, and it provides space for parents to model good behaviors and discipline bad behaviors, among a multitude of other benefits (Amato & Rivera, 1999; Cripps & Zyromski, 2009; Miedel & Reynolds, 1999; Waldfogel et al., 2010). Therefore, it is important to understand how nonstandard schedules, which affect the amount and structure of time families spend together, influence child well-being.

Additionally, the relationships between nonstandard schedules, parental involvement, and child outcomes may be stronger when children are young and in their early developmental stages, when parental involvement is likely particularly important (Englund, Luckner, Whaley, & Egeland, 2004; NICHD Early Child Care Research, 2005). Alternatively, these relationships may be stronger among adolescents if nonstandard schedules alter parents’ abilities to monitor their children and prevent them from engaging in problem or risky behaviors, such as smoking and drinking alcohol (Han, Miller, & Waldfogel, 2010; Han & Waldfogel, 2007). Nonstandard schedules may offer families flexibility for addressing these issues, or may impose constraints. The extent to which they do one or the other has important implications for family functioning and children’s well-being as they develop.

The study of children’s behavioral problems is particularly informative for understanding how nonstandard schedules influence children’s well-being across the lifespan. Behavioral problems such as depression, aggression, and hyperactivity are associated with life satisfaction and subjective well-being (Gilman & Huebner, 2003; Huebner & Alderman, 1993). Behavioral problems may also influence children’s later-life outcomes by impinging on their academic achievement or their development of peer relations, for example. Behavioral problems therefore represent a “canary in the coal mine,” providing a signal that something is wrong in a child’s life.

Consequently, this study aims to examine the relationship between parental nonstandard schedules and children’s behavioral problems by analyzing five central questions:

1. Are parental nonstandard schedules associated with children’s behavioral problems?

2/3. Do these associations depend on the gender and/or age of the child?

4. Do these associations differ by the type of nonstandard shift schedule?

5. Do these associations depend on the gender of the parent working the nonstandard shift?

By investigating these questions, this study attempts to improve on previous research by modeling explicitly the relationships among nonstandard work and child behavioral outcomes for both fathers and mothers, by accounting for whether these relationships depend on the gender and age of the child, and by using longitudinal data that is capable of following children as they age from young children into teenagers. I use panel data from the National Longitudinal Survey of Youth-1979 (NLSY79) main file and Mother–Child Supplement for the years 1992 to 2006 to seek answers to these research questions.

Nonstandard Schedules and Child Outcomes

Nonstandard schedules create both flexibility and constraints. On the one hand, parents and caregivers, particularly women, may choose to work nonstandard schedules in order to negotiate the demands of work and caring for children or, in rarer cases, relatives (Hosking & Western, 2008; Presser, 2003). Approximately 30% to 40% of mothers cite better child care arrangements as a primary reason for working a nonstandard schedule (Presser, 2003). On the other hand, most employees, including 50% of single-mothers and 40% to 46% of married mothers, say they work these shifts because they had no other choice or because it was “the nature of [their] job” (Presser, 2003). Thus, for some families, nonstandard schedules offer flexibility in negotiating child care responsibilities between family members, particularly spouses, while for other families, nonstandard schedules may be imposed and seen as inconvenient and problematic.

Indeed, nonstandard schedules have been linked to a variety of negative outcomes, including acute and chronic health problems (Kantermann, Juda, Vetter, & Roenneberg, 2010; Maume Sebastian, & Bardo, 2010; Wight, Raley, & Bianchi, 2008), separation and divorce (Kalil, Ziol-Gues, & Epstein, 2010; Presser, 2003), and parenting practices that are less sensitive to children’s needs and more aggressive (Joshi & Bogen, 2007; Strazdins et al., 2006). Nonstandard schedules also place severe time constraints on families, often clashing with the hours during which child care facilities and schools operate.

Consistent with these findings, previous studies indicate that nonstandard schedules may correspond to negative outcomes for children. For example, toddlers of mothers who worked a nonstandard schedule at some point during the first 3 years of the child’s life had significantly lower scores in terms of various developmental outcomes such as expressive language and motor skills, sensory perception, learning, and memory (Han, 2005). Likewise, Joshi and Bogen (2007) find that maternal nonstandard schedules are associated with increased behavioral problems among 2- to 4-year-old children in low-income families.¹ Among 23 studies examining nonstandard schedules and child outcomes, 21 studies found that parental nonstandard schedules were associated with worse behavioral, cognitive, or health outcomes among children (Li et al., 2014). Of these 23 studies, the most consistently negative associations between parental nonstandard schedules and child outcomes were found for infants, toddlers, and adolescents. Though, of the 16 of these studies examining behavioral outcomes, only one was longitudinal, utilized a nationally representative sample, and examined behavioral outcomes in children of a relatively wide range of ages (4- to 10-year olds in this particular case), hindering the ability to systematically study the role of age in these relationships (Han, 2008).

Additionally, nonstandard schedules tend to have larger negative associations with children’s behavioral problems in single-parent families (Dockery, Li, & Kendall, 2009; Han, 2008; Han & Waldfogel, 2007). Han (2008), finds, for example, that children whose mothers are single and employed in nonstandard schedules exhibit significantly higher levels of behavioral problems than children whose mothers are married and employed in nonstandard schedules. Likewise, Dockery et al. (2009) demonstrate that the relationship between parental nonstandard schedules and child mental health is significantly larger and more harmful for children in single-parent families than for children in two-parent families. It is therefore reasonably well-established that nonstandard schedules are negatively associated with child outcomes among families who might find nonstandard schedules especially constraining, such as single-parent families.

It is possible, however, that nonstandard schedules may benefit children from two-parent families if two-parent families are more able to utilize nonstandard schedules as opportunities for flexibility. Indeed, many dual-earner couples choose to off-set their work schedules in order to allow a parent to always be present for their children (Hattery, 2001; Lindsay, Maher, & Bardoel, 2009) and some parents attest that they prefer nonstandard schedules because they allow them to feel like stay-at-home mothers or fathers (Hattery, 2001; Lindsay et al., 2009). In some cases, nonstandard schedules may therefore enhance parental involvement. For example, Brayfield (1995) found that fathers spent more time with their children when their wives worked evening or night shifts. Children also rate their fathers’ parenting skills and parental knowledge higher when their mothers work nonstandard schedules, with higher parental effectiveness ratings corresponding to lower levels of risk-taking behavior and negative outcomes among youths (Barnett & Gareis, 2007). It may therefore be the case that nonstandard schedules enable parents, particularly fathers, to spend more time with their children. Because increased father closeness tends to be associated with fewer behavior problems and better socioemotional adjustment among children (Barnett & Gareis, 2007; Brayfield, 1995; Harris, Furstenberg, & Marmer, 1998), this increased time could correspond to higher overall child well-being. Some types of nonstandard schedules may also allow the parents working those schedules to be at home for a larger proportion of children’s nonsleeping, nonschool hours than day shift schedules.

The Role of Gender and Age

Parental nonstandard schedules therefore have a multitude of associations with family functioning, influencing the relationships between parents and children, as well as the well-being of both. However, there are a number of unanswered questions concerning how nonstandard schedules are associated with children’s behavioral problems. In particular, the consequences of nonstandard schedules may depend on the gender of the child and the gender of the parent employed in the nonstandard schedule because children’s relationships with their parents and the effects of those relationships tend to be gendered. Indeed, fathers are more likely to remain in the household after a son is born (Lundberg, McLanahan, & Rose, 2007) and tend to be more involved in their sons’ lives (Harris et al., 1998; Mammen, 2011; Raley & Bianchi, 2006; Williams & Kelly, 2005). This higher level of involvement includes exhibiting greater closeness and warmth toward sons (Harris et al., 1998), spending more time in collaborative, leisure, and achievement-related activities (reading, helping with homework, teaching, talking, and playing) (Harris et al., 1998; Mammen, 2011; Raley & Bianchi, 2006); spending more time engaging in primary care for sons (Mammen, 2011; Raley & Bianchi, 2006), and engaging in more one-on-one time with sons (Raley & Bianchi, 2006). It may therefore be the case that fathers take greater steps to mitigate any negative effects or enhance any positive effects associated with their or their wives’ nonstandard schedules when their children are boys, by maintaining high levels of involvement, for example. In contrast, mothers’ involvement, including the warmth and closeness they feel toward their child and the level of time they spend with their child in various activities, is not influenced by children’s genders (Raley & Bianchi, 2006; Starrels, 1994).

Nonstandard schedules may also have associations with gender if one gender is particularly vulnerable to changes in maternal or paternal employment. For example, girls may be more vulnerable to changes in their mothers’ employment status if girls are more likely to model their mothers’ behavior and if mothers’ mental health is affected by their nonstandard employment (Bussey & Bandura, 1999; Deater-Deckard & Dodge, 1997). It is therefore important to separately examine girls’ and boys’ behavioral adjustment in response to both mothers’ and fathers’ shift schedules because the effect of nonstandard schedules and parental employment more broadly may be gendered. Consequently, the relationships between nonstandard schedules and child behavior problems may depend on the gender of both the child and the parent employed in the nonstandard schedule. The extent to which these relationships operate differently for girls and boys, fathers and mothers, has important implications for understanding how and why parental employment influences children, how to go about addressing any negative associations between parental employment and child behavioral outcomes, and how to go about enhancing any positive associations between employment and children’s and families’ well-beings.

Yet few studies have examined whether the associations between nonstandard schedules and child outcomes differ for girls and boys, limiting the ability to hypothesize about these associations. Among the few that have examined the gendered nature of these relationships, results have been mixed. One study found that mothers’ night shifts increase the number of delinquent behaviors exhibited by adolescent boys from nonprofessional, low-income families (Han et al., 2010). In contrast, Joshi and Bogen (2007) find that maternal nonstandard schedules increase internalizing and externalizing behaviors among 2- to 4-year-old girls to a marginally greater extent than among 2- to 4-year-old boys. Additionally, Barton, Aldridge, and Smith (1998) find that 8- to 11-year-old girls’ feelings of competence in school are more negatively associated with fathers’ nonstandard shift schedules than boys’ feelings of competence. There is therefore suggestive evidence that the relationships between nonstandard schedules and child outcomes may be gendered, though these findings would be strengthened by examining a wider range of ages, utilizing longitudinal data, and comparing different articulations of mothers’ and fathers’ nonstandard schedules, such as night and rotating shifts, rather than operationalizing nonstandard schedules as a binary indicator. Moreover, many studies, including those by Han et al. (2010) and Joshi and Bogen (2007), focus on low-income families. While these are very important investigations, it is also integral to understand how nonstandard schedules influence the outcomes of children from a wider range of families.

These relationships may also depend on children’s ages. Indeed, the research on nonstandard schedules has largely focused on the influence of these schedules on very young children, often finding that nonstandard schedules have negative associations with infants’ and toddlers’ behavioral development (Daniel et al., 2009; Joshi & Bogen, 2007; Li et al., 2014), or on adolescents and their engagement in risky behaviors, such as smoking and drinking alcohol (Han et al., 2010; Han & Waldfogel, 2007). It is unclear, however, whether these relationships operate similarly for all ages or show changes in their associations for young children, children in middle childhood, and adolescents. Utilizing longitudinal data and studying a wider range of ages may therefore provide greater insights into the influence of nonstandard schedules on child behavior problems.

Nonstandard schedules therefore have consequences for children, though it is not clear whether these consequences are positive or negative once the characteristics of children and their families are accounted for. Understanding for whom nonstandard schedules have the worst and most beneficial associations offers important insights into how employment structures family time and dynamics and whether the benefits of certain schedules should be emulated more broadly and whether the costs of certain schedules should be mitigated through intervention.

Theoretical Background and Hypotheses

This study is informed by the conceptual resources framework (Brooks-Gunn, 1995), which is integral for framing why parental nonstandard schedules may negatively or positively influence children. Parental shift schedules are thought to influence children’s behavioral outcomes through changes in parental stress and changes in the quality and/or quantity of time parents spend with children in a variety of activities (Han, 2008). These mechanisms correspond with the conceptual resources framework (Brooks-Gunn, 1995), which argues that parents provide four types of resources for their children: income, time, human capital (such as reading ability and training), and psychological capital (e.g., modeling mental health and positive social relationships), and these resources correspond with children’s well-being. Nonstandard work schedules may change parents’ abilities to provide this capital, either in negative ways by elevating parents’ stress and straining their ability to model positive relationships and/or engage in developmental activities such as reading to children, for example, or in positive ways by increasing the amount of time parents have with their children, thereby influencing child well-being. These relationships may be particularly strong for work schedules that are unusually unpredictable and therefore stressful, such as rotating and split shifts, because research has found that unpredictable and varied work schedules tend to be associated with higher levels of stress (Fenwick & Tausig, 2001) and mental health problems than other types of shift schedules (Bara & Arber, 2009). Alternatively, these relationships may be strongest for shift schedules that alter the amount and/or structure of time parents spend with children, such as evening or night shifts. Shift schedules that alter the amount of time parents spend with children may also alter the amount of human capital parents are able to provide, by changing the amount of time parents are able to spend reading to children for example, thereby influencing parents’ provision of conceptual resources in multiple, interacting ways. The conceptual resources framework therefore helps illustrate why nonstandard schedules may influence child development and why it is important to study different articulations of shift schedules, given the different mechanisms each type of shift schedule might trigger.

Moreover, the other parent (who is likely not employed in a nonstandard schedule) may also change their provision of conceptual resources in response to changes in their partner’s provision of conceptual resources, suggesting that this is a dynamic and interactive process. Because father involvement tends to be gendered, his provision of conceptual resources is also likely to be gendered, potentially corresponding to differences in the relationships between nonstandard schedules and behavioral outcomes for girls and boys.

The fourth type of conceptual resource—psychological capital—could also explain potential gender differences in the relationships between parental nonstandard schedules and child behavior problems. For example, research has found that parent mental health influences children’s mental health (Mattejat & Remschimidt, 2008) and that children may be more likely to model the behavior and mental health of their same-gender parent (Bussey & Bandura, 1999; Deater-Deckard & Dodge, 1997). Therefore, changes in parent mental health as a result of employment in nonstandard work schedules may prompt changes in children’s mental health and these changes may be particularly large for children who are of the same gender as the parent employed in the nonstandard schedule. Consequently, if parental nonstandard schedules primarily influence child behavioral outcomes through the provision of psychological capital, then children who share the same gender as the parent employed in the nonstandard schedule should exhibit the largest changes in behavioral problems.

These theories lead to the following hypotheses: (a) I hypothesize that consistent with prior studies, children whose mothers or fathers work nonstandard schedules will exhibit more behavior problems relative to children whose mothers or fathers work day shifts. (b) I hypothesize that these relationships will differ depending on the type of shift schedule, because different articulations of shift schedules have different implications for family relationships, time, and parents’ resulting ability to provide conceptual resources to their children. In particular, I expect rotating/split shifts to exhibit more positive (harmful) associations with behavioral problems, because of their instability and because of the association between instability and stress aforementioned. The hypothesized direction of these relationships for night/evening shifts is ambiguous because, on the one hand, they may be associated with increased time with children leading to negative relationships with behavioral problems. On the other hand, they may be associated with increased stress, worse parent mental health, and higher levels of behavioral problems in children. (c) I expect the relationships between nonstandard schedules and behavioral outcomes to differ by the gender of the child and the gender of the parent employed in the nonstandard schedule because parents’ provisions of conceptual resources and the relationship between nonstandard schedules and the provision of conceptual resources likely differs by child and parent gender, though it is unclear whether girls or boys will be affected more by nonstandard schedules.

Data and Research Design

Sample Selection

This analysis utilizes the NLSY79 and its Child Supplement (NLSY79-CS) for the years 1992 to 2006. Later years are not included because the number of children in the sample dropped precipitously after 2006. Restricting the analysis to these years does, however, limit the generalizability of these findings to children who were 5 to 15 years between 1992 and 2006. The NLSY79 is a probabilistic sample composed of 12,686 men and women interviewed annually between 1979 and 1994, and interviewed biennially thereafter. In 1986, the NLSY79-CS began to interview the children of the women interviewed in the NLSY79, with children continuously added to the supplement as they are born.

The NLSY79-CS children and their mothers provide the data for this analysis. This study focuses on children aged 5 to 15 years in order to follow children throughout a significant period of their childhood and adolescence (each child is observed for up to 10 years or 5 observation periods, with an average of 2.5 observations per child), and to capitalize on the availability of the behavior problems index (BPI), which is only available for children aged 5 to 15 years. The sample is restricted to 4,180 children—1,950 girls belonging to 1,420 unique families and 2,044 boys belonging to 1,488 unique families—for whom the father is in the household at all-time points and who have valid BPI scores and valid responses for all covariates. Children who have experienced changes in their family compositions—either by fathers entering or leaving the household—during the study period are excluded from this analysis. Only 7% of the original sample was excluded on the basis of this criterion, however, and the results are largely similar when this group is included and excluded. A central goal of this study is to understand whether the association between parental nonstandard schedules and children’s behavior problems depends on the gender of the child and/or the parent employed in the nonstandard shift. These gender dynamics are likely substantially different for families in which the biological father is not present. Likewise, partners, stepfathers, and fathers who are not in the household likely have systematically different relationships with their or their partner’s children relative to fathers who are in the household. A Chow test for different estimates for the sample with the father in the household compared with estimates for children whose fathers live elsewhere suggested that this is, in fact, the case (F = 8.01, p < .01). Two-parent families are therefore solely examined to reduce the potential bias associated with these relationships.

Behavioral Problems Index

The outcome variable is represented by the BPI, an index of 28 questions answered by the mother that assess various behavioral difficulties exhibited by their children. The index is normalized to have a mean of 100 and a standard deviation of 15, with higher scores indicating more behavior problems. Additionally, the index is composed of six subscales (abbreviated examples of the questions asked for each subscale are provided in parentheses): antisocial (e.g., cheats or tells lies, bullies others, is disobedient at school); anxious/depressed (fearful or anxious, feels worthless or inferior, is unhappy, sad or depressed); dependent (clingy, cries too much, dependent on others); headstrong (argues too much, is disobedient at home, has strong temper and loses it easily); hyperactive (has difficulty concentrating or paying attention, is impulsive, is restless or overly active); and peer problems (has trouble getting along with other children, is not liked by other children, does not get involved with others). The BPI is a widely used and tested measure developed by Nicholas Zill and James Peterson (1986) and draws from other popular child behavior scales including the Achenbach Behavior Problems Checklist (Achenbach & Edelbrock, 1981). Only children with a valid BPI score in a given person-year are included in the sample.

Parents’ Nonstandard Employment

The associations between the mother’s and the father’s shift schedules and child behavioral outcomes are examined because both cases may influence the relationships and interactions children have with their mothers and/or fathers and, by association, children’s behavioral outcomes. The focal independent variables—the father’s and mother’s shift schedules—are represented in four forms: (a) by a dummy variable capturing whether the parent reports working an evening shift (between 2 p.m. and 11 p.m.) during the observation period; (b) a dummy variable representing whether the parent reports working a night shift (between 9 p.m. and 7 a.m.); (c) a dummy variable indicating whether the parent reports working a rotating shift (a shift that changes from 1 day to the next), a split shift (two separate time periods are worked during a single day) or other shift (an ambiguous category reported by about 1% of parents), and (d) a dummy variable indicating whether the parent reports working an irregular shift, with the reference group represented by mothers and fathers who work day shifts. Rotating, split shifts and “other” shifts were combined because they consistently exhibited similar directions and magnitudes in regression analyses and to help aid with identification, given that relatively few parents work split or “other” shifts. Irregular shifts capture hours that vary unpredictably, perhaps extending from 9 a.m. to 10 p.m. 1 day, and from 8 a.m. to 4 p.m. the next day. Among NLSY79 respondents, irregular shifts are particularly prevalent among relatively well-off families and managers, health service workers, and cashiers. A dummy variable representing whether the parent is unemployed is also included for mothers and fathers, and a dummy variable representing whether the mother is out of the labor force is included (this variable was not included for fathers because too few fathers were out of the labor force to separately examine this outcome from unemployment). Excluding nonemployed parents reduces the representativeness of the sample and leads to the sole inclusion of dual earner families. Additionally, its exclusion could lead to spurious conclusions if behavioral outcomes are associated with spells of nonemployment, rather than with nonstandard schedules.

Control variables are included for the child’s age (in years), the child’s race, the mother’s years of educational attainment, the logged total net family income, whether the mother works part-time (less than 35 hours/week but greater than 0 hours/week),² and whether the father works part-time, with the reference group being parents who work full-time or overtime.³ Variables were also included for whether the mother works a professional job and whether the father works a professional job. Professional job status was defined on the basis of the Census’ 1970 occupation codes, which utilize the 1970 Code Duncan SEI Score to define “professional, technical, and kindred workers.” Almost all covariates exhibit very high response rates of upward of 90%. The most important exception to this is reported family income, for which about 16% of respondents gave no answer. Child-person-years with missing answers to the dependent and/or independent variables are excluded from the analysis. However, the results remained substantially similar when multiple imputation was used.⁴

Analytical Strategy

The longitudinal nature of the NLSY79-CS allows for the relationships between parental nonstandard schedules and child behavioral outcomes to be examined over time. Consequently, this analysis utilizes growth curve models estimated using Stata 14’s mixed command. Family and child random intercepts are included so that child-person-period observations are nested within individual children and individual children are nested within families, thereby helping account for the nonindependence of the multiple observations obtained for each child and for the multiple children within families (each family claims about 1.4 children in this sample). Moreover, the child random effects are allowed to vary for each child over time, thereby allowing each child to have their own estimated intercept and slope for these relationships. In doing this, the growth curve models yield results that are virtually numerically equivalent to those produced by structural equation models, with the only major difference, in this case, being that mixed effects models better handle higher orders of nesting, thereby allowing me to nest observations within children and families (Curran, Obeidat, & Losardo, 2010).

Multilevel growth curve models have important advantages for this analysis. In particular, they are able to estimate variation across individuals while accounting for changes over time within individuals (Curran et al., 2010). Multilevel models are also able to accommodate multiple levels of nesting, such as, in this case, observations within children and children within families. Multilevel growth curve models also better handle missing data and groups with small numbers of observations by using all participants, even those with one wave, relative to other types of longitudinal models such as fixed effects models. Moreover, multilevel growth curve models are able to estimate how the influence of a predictor variable changes over time. Multilevel growth curve models therefore help reduce bias and more accurately estimate the potentially changing influence of parental nonstandard schedules on children over time than cross-sectional or even other types of longitudinal models.

To assess whether the associations between parental nonstandard schedules and child behavior problems depend on the child’s gender, separate samples of girls and boys are considered. This decision was made on the basis of a Chow test, which suggested that the model effects on child behavior systematically differed by the child’s gender (F = 6.25; p < .01) and were strong enough to merit splitting the sample rather than including gender interaction effects and analyzing the sample jointly. Separately analyzing the sample by gender also helps reduce the burden of analysis that would result from analyzing the fully interacted effects of gender, age, shift status, and, for the supplementary analyses, work hours.

The sample is also examined for potential age effects and, in supplementary analyses, work hour effects. Toward this end, interactions are estimated between a continuous measure representing the child’s age in years and each parental shift schedule articulation. These results did not differ when different categorizations of age were used instead. Supplementary analysis models also include three-way, shift schedule-work hour-child age effect, interactions, which fully interact each parental shift schedule articulation, a dummy variable representing part-time status (parent works less than 35 hours/week and more than 0 hours/week), and the continuous child age measure.

Results

Descriptive statistics averaged across all years for children aged 5 to 15 years with fathers in the household are presented in Table 1 (the descriptive statistics therefore represent the average values for the full set of person-period observations). In this sample, approximately 19% to 20% of children have mothers who are employed in nonstandard schedules in a given year and approximately 21% of children have fathers who are employed in nonstandard schedules in a given year, proportions that are comparable with the national average (McMenamin, 2007).

Table 1.

Descriptive Statistics for Children Aged 5 to 15 Years With Fathers in the Household.

	Girls (n = 1,950; 4,836 person-years)		Boys (n = 2,044; 5,043 person-years)
	M	SD	M	SD
Outcome variable
Behavioral problem index	101.50	14.17	103.63	14.42
Mother’s shift schedule
Day shift	0.669	0.460	0.651	0.408
Evening shift	0.049	0.215	0.054	0.225
Night shift	0.022	0.146	0.026	0.160
Rotating/split shift	0.039	0.193	0.040	0.196
Irregular shift	0.075	0.264	0.084	0.278
Mother unemployed	0.009	0.093	0.008	0.086
Mother out of labor force	0.142	0.349	0.137	0.344
Father’s shift schedule
Day shift	0.738	0.419	0.745	0.422
Evening shift	0.036	0.187	0.041	0.199
Night shift	0.043	0.203	0.047	0.212
Rotating/split shift	0.052	0.222	0.049	0.215
Irregular shift	0.077	0.267	0.071	0.256
Father not employed	0.053	0.224	0.046	0.210
Characteristics
Race (percentage of sample)
Hispanic	0.196	0.397	0.191	0.393
Non-Hispanic Black	0.165	0.371	0.160	0.366
Non-Hispanic White	0.638	0.481	0.649	0.477
Child’s age	9.367	2.853	9.400	2.865
Mother works part-time	0.268	0.443	0.262	0.440
Father works part-time	0.038	0.190	0.031	0.174
Mother works professional job	0.361	0.480	0.350	0.477
Father works professional job	0.319	0.466	0.312	0.463
Mother’s years of education	13.52	2.414	13.52	2.492
Logged total net family income	10.80	1.203	10.81	1.094

Source. National Longitudinal Survey of Youth-1979 and National Longitudinal Survey of Youth-1979-Child Supplement.

Moreover, approximately 20% of children identify as Hispanic, about 16% identify as non-Hispanic Black, and 64% identify as non-Hispanic White, these proportions are within a few percentage points of the national averages. Finally, this sample exhibits high rates of employment, low rates of unemployment, and relatively high incomes, likely as a result of my sample’s restriction to two-parent families. I may therefore be missing some disadvantaged families who may have particular difficulties in adapting to nonstandard schedules. Nevertheless, numerous studies have specifically examined the association between nonstandard schedules and child outcomes among families in poverty and single-parent families (Dockery et al., 2009; Han, 2008; Han & Waldfogel, 2007; Li et al., 2014), their studies can therefore fill the gaps in this study’s findings.

This investigation begins with bivariate analyses that examine whether BPI scores are related to parents’ employment schedules and if these relationships differ by the gender of the child.

Models 1 in Table 2 indicate that different associations are, indeed, observed for girls and boys. Specifically, mothers’ evening shifts are associated with fewer behavior problems among boys, and mothers’ rotating/split shifts are associated with more behavior problems among boys. For girls, in contrast, mothers’ night shifts are associated with marginally significantly fewer behavior problems, as is having a mother who is out of the labor force. These bivariate results provide some initial indication that the relationships between nonstandard schedules and behavior problems may differ by the gender of the child and that mothers’ nonstandard schedules may be more strongly associated with children’s behavior problems than fathers’ nonstandard schedules. These bivariate relationships are only suggestive, however, because there are a multitude of factors that are related to both children’s behavior problems and parental nonstandard schedules—such as income, work hours, occupational status, and mothers’ educational attainment. Models 2 in Table 2 present results from multivariate analyses to account for these potentially confounding relationships.

Table 2.

Results From Mixed Effects Models.

	Girls			Boys
	Model 1	Model 2	Model 3	Model 1	Model 2	Model 3
Mother’s shift schedule
Day shift	Ref.	Ref.	Ref.	Ref.	Ref.	Ref.
Evening shift	0.059	0.110	-4.600⁺	-1.793*^†	-1.513*^†	−4.001^+ †
Night shift	−2.311^+ †	−2.464*^†	−4.014	−0.757	−0.871	−3.294
Rotating/split shift	0.920	0.935	-5.476*^†	2.422**	2.839***	2.417
Irregular Shift	−0.114	0.150	−2.823	−0.455	0.058	−2.695
Unemployed	0.777	−0.030	0.538	1.920	0.812	−11.138⁺
Out of Labor Force	−1.120*	−1.313*	−1.258	−0.752	−0.948⁺	0.163
Father’s shift schedule
Day shift	Ref.	Ref.	Ref.	Ref.	Ref.	Ref.
Evening shift	1.332	0.998	−0.415	−0.122	−0.652	−3.971
Night shift	0.456	0.138	4.215	−0.987	−1.815*	0.480
Rotating/split shift	0.396	0.404	4.272^+ †	−0.102	−0.232	1.790
Irregular shift	0.570	0.660	0.021	−0.159	−0.007	−2.838
Unemployed	0.899	0.000	1.853	2.185*	0.400	1.746
Characteristics
Child age		0.337***	0.295***		0.354***	0.296***
White		Ref.	Ref.		Ref.	Ref.
Black		1.191	1.129		0.098	0.019
Hispanic		−0.852	−0.914		−0.996	−1.050
Mother’s education. attainment		−0.668***	−0.666***		−0.745***	−0.736***
Mother’s professional status		0.433	0.437		0.269	0.243
Father’s professional status		−0.729	−0.743		−0.654	−0.703
Log total net income		−0.148	−0.146		−0.546**	−0.559**
Part-time		−0.195	−0.214		−0.664⁺	−0.643
Spouse part-time		−0.611	−0.554		−1.027	−1.061
Year 2000		−1.832***	−1.864***		−2.302***	−2.372***
Mother’s shift interactions
Evening shift × age			0.523*			0.295
Night shift × age			0.155			0.250
Rotate/split shift × age			0.691*^†			0.047
Irregular shift × age			0.332⁺			0.298
Unemployed × age			−0.042			1.314*^†
Out of labor force × age			−0.006			−0.124
Father’s shift interactions
Evening shift × age			0.144			0.376
Night shift × age			−0.436			−0.237
Rotate/split shift × age			−0.419⁺			−0.217
Irregular shift × age			0.070			0.313
Unemployed × age			−0.194			−0.139
Constant	101.972*** (273.40)	109.748*** (44.58)	110.124*** (43.97)	103.810*** (303.25)	117.343*** (47.77)	117.925*** (46.93)
Observations	4,836	4,836	4,836	5,043	5,043	5,043
Number of children	1,950	1,950	1,950	2,044	2,044	2,044

Source. National Longitudinal Survey of Youth-1979 and National Longitudinal Survey of Youth-1979-Child Supplement.

p < .1. *p < .05. **p < .01. ***p < .001. ^†Differences between genders significant at p < .05.

The findings in Models 2 indicate that, controlling for a host of relevant characteristics, mothers’ evening shifts continue to be associated with fewer behavior problems among boys, and mothers’ rotating/split shifts continue to be associated with more behavior problems among boys. After accounting for the full set of covariates, fathers’ night shifts also become associated with significantly fewer behavior problems among boys. For girls, the results remain unchanged in the multivariate models: mothers’ night shifts remain negatively and significantly associated with behavior problems and girls whose mothers are out of the labor force also exhibit fewer behavior problems. These results are somewhat surprising, demonstrating that mothers’ evening shifts and fathers’ night shifts appear to offer behavioral benefits for boys, and that mothers’ night shifts appear to offer behavioral benefits for girls. It is largely parents’ rotating and split shifts, which are characterized by instability, that exhibit harmful associations with boys’ behavioral problems, perhaps indicating that it is this instability and unpredictability that creates problems for parents and children, as hypothesized above.

As posited by the third question, the age of the child may also influence these relationships. On one hand, parents may make particularly great efforts to shield younger children from any negative effects associated with nonstandard schedules, or they may specifically choose to work nonstandard schedules when children are younger and in need of more frequent care and monitoring. If this is the case, nonstandard schedules may be associated with few harmful and even some beneficial relationships among younger children. Moreover, the effect of nonstandard schedules on adolescents could be particularly harmful, if these shift schedules take parents away from the home during periods of time when adolescents may be at the greatest risk of engaging in delinquent behavior, after school and in the evenings, for example (Flannery, Williams, & Vazsonyi, 1999; Gottfredson, Gottfredson, & Weisman, 2001; Richardson, Radziszewska, Dent, & Flay, 1993). On the other hand, nonstandard schedules may be associated with worse outcomes for young children if young children are particularly vulnerable to changes in parents’ work hours and, as a result, changes in the amount and/or quality of time parents spend with children. To assess these potential age effects, interactions are conducted between shift schedule articulations and a continuous measure of child’s age.

The results from these analyses indicate that the associations between parental nonstandard schedules and children’s behavioral outcomes differ by age, but this is largely only the case for girls. Additionally, these associations vary in unique ways that also depend on the type of shift schedule and the gender of the parent employed in that schedule. To illustrate the results from these models more clearly, the results from Model 3 for girls are illustrated in Figures 1 and 2, which depict predicted BPI scores for girls as they age. Separate lines are graphed for each nonstandard shift schedule articulation that is significantly associated with behavioral problems and that exhibits a significant interactive association with age. Figure 1 graphs these relationships for mothers’ shift schedules and Figure 2 graphs these relationships for fathers’ shift schedules. Lines are also included for the reference group—mothers’ and fathers’ day shifts—for comparison.

Figure 1.

Predicted behavior problems index scores by mother’s shift schedules calculated from Model 3 in Table 2 for girls.

Figure 2.

Predicted behavior problems index scores by fathers’ shift schedules calculated from Model 3 in Table 2 for girls.

In Figure 1, we see that after including interactions between parents’ shift schedules and girls’ ages, young girls exhibit fewer behavior problems when their mothers work evening and irregular shifts compared with when their mothers work day shifts. At older ages, these associations decline and exhibit signs of reversing, particularly once girls become teenagers. Likewise, the predicted BPI scores indicate that mothers’ rotating/split shifts are associated with more behavior problem, particularly among adolescent girls, a relationship that is consistent with previous literature. In contrast, fathers’ rotating/split shifts are associated with more behavior problems among younger girls than among older girls, perhaps indicating that rotating/split shifts are particularly disruptive to the time fathers would otherwise devote to younger daughters (Figure 2).

The relationships between parents’ nonstandard schedules and boys’ behavior problems do not, in contrast, appear to vary by age. The direction of the relationships suggests that, similar to girls, any benefits associated with nonstandard schedules are largely experienced at younger ages and these associations decrease as children age, but the nonsignificance of these relationships means that any hypotheses about these relationships are highly tenuous.

These results may be influenced by changes in the sample sizes of children exposed to nonstandard schedules at different ages and/or by particularly small samples of children exposed to nonstandard schedules at certain ages. To examine this possibility, separate bivariate regressions were conducted regressing BPI scores and parental shift schedule articulations for 5- to 6-year-old children, 7- to 8-year-old girls/boys, 9- to 10-year-old girls/boys, 11- to 12-year-old girls/boys, and 13- to 14-year-old girls/boys. Though these results are less stable and consistent than the main results, because of the smaller sample sizes, they are reasonably consistent with the findings described above. The main exception is that mothers’ irregular shifts for girls and fathers’ evening shifts for boys have relatively large fluctuations in sample sizes across ages and, as a result, fluctuations in the direction and significance of their associations with behavior problems. Consequently, irregular shifts are only significantly associated (p < .05) with girls’ behavioral outcomes when girls are younger than 9 years and are associated with increases in behavioral problems except when girls are 9 to 10 years. Fathers’ evening shifts are only significantly associated with boys’ behavioral problems when they are 13 to 15 years and are associated with increases in behavior problems at older, but not younger, ages. The results for these shift schedule articulations should therefore be viewed with some caution.

Supplementary Analyses

It is possible that the beneficial association between parents’ nonstandard schedules and children’s behavior problems arises largely among parents who are employed in part-time shifts. Indeed, this could also explain why it appears to be young children who benefit the most from parents’ nonstandard schedules, because parents are more likely to have the need and desire to flexibly schedule employment in order to spend more time supervising and caring for young children. Consequently, three-way interactions were conducted between parents’ nonstandard shift schedule articulations, children’s ages, and part-time work status. Unfortunately, the sample sizes of parents employed in part-time nonstandard shifts is rather small (less than 100 observations for most types of mothers’ shift schedule articulations and less than 50 observations for most types of fathers’ shift schedule articulations), hindering the ability to definitively identify these relationships. Nevertheless, both the bivariate and multivariate interactive analyses indicated that these interactive relationships were nonsignificant and unstable, with one important exception. Specifically, it appears to be mothers’ part-time rotating/split shifts that tend to be associated with fewer behavior problems among young girls, while full-time rotating/split shifts are largely associated with more behavior problems relative to part-time rotating/split shifts and day shifts. These results are illustrated in the appendix (Figure A1), which graphs girls’ predicted BPI scores as they age, with separate lines included for mothers’ part- and full-time day and rotating/split shifts (full table of results available on request).

Additionally, a dummy variable indicating if the mother and father both worked nonstandard schedules was included in the models, though this variable was not significantly related to behavioral problems for girls or boys, nor do it moderate any of the relationships. Moreover, the sample size of mothers and fathers that both work nonstandard schedules was small enough to make estimates utilizing this indicator relatively unstable.

Discussion

These results paint a nuanced picture of the associations between parental nonstandard schedules and children’s behavior problems, illustrating important contextual roles for parents’ and children’s genders and for children’s ages. Indeed, mothers’ evening and irregular shifts are associated with significantly fewer behavior problems for young girls, but these beneficial relationships disappear and even reverse as girls age. Mothers’ evening and night shifts also offer behavioral benefits for boys and girls, respectively, regardless of their age, and father’s night shifts are associated with fewer behavior problems among boys, regardless of their age.

In contrast, parental rotating/split shifts tend to exhibit harmful associations with girls’ and boys’ behavior problems. Specifically, mothers’ rotating/split shifts are associated with more behavior problems among boys (regardless of age) and they are associated with more behavior problems among older girls. There is suggestive evidence that these relationships are particularly strong for mothers who work full-time rotating/split shifts. Likewise, fathers’ rotating/split shifts are associated with more behavior problems among girls, particularly when they are young.

The answers to the five questions posed at the beginning of this article are therefore all affirmative: nonstandard schedules have associations with children’s behavior problems, with unstable shifts such as rotating and split shifts corresponding to consistently positive (or harmful) associations with children’s behavioral problems, while other types of nonstandard schedules tend to be associated with fewer behavior problems, particularly among young children. These results depend on the gender of the child and, in some cases, the age of the child. These results also differ depending on the type of nonstandard schedule the parent is employed in and whether it is the mother or father who is employed in a nonstandard schedule.

The Conceptual Resources Framework may provide an explanation for these results. As explained in the “Theoretical Background” section, it is possible that parental nonstandard schedules primarily influence children’s behavioral outcomes through the provision of psychological capital. Children are more likely to model the mental health and behavior of their same-gender parent. Therefore, the finding that mothers’ nonstandard schedules exhibit a larger and more consistent multitude of associations with girls’ behavioral outcomes than with boys’ behavioral outcomes, could reflect this psychological modeling. Mothers employed in nonstandard schedules that allow them greater flexibility in negotiating work and family may find themselves more satisfied with their work/life balance and exhibit more positive mental health outcomes as a result. In contrast, shift schedules that are unstable and less likely to be controlled by the worker correspond to greater stress (Fenwick & Tausig, 2001) and worse mental health outcomes (Bara & Arber, 2001), which may cause parents to provide less psychological capital. Parents who are stressed and/or exhausted may also find it more difficult to engage their children in stimulating activities, thereby harming human capital formation and also corresponding to fewer conceptual resources. This could explain why both mothers’ and fathers’ rotating/split shifts exhibit positive associations with children’s behavior problems.

Moreover, nonstandard schedules may influence parents’ provision of other types of conceptual resources, such as time and the provision of human capital. The finding that mothers’ shift schedules tend to be more strongly associated with children’s behavioral outcomes than fathers’ nonstandard schedules may reflect the possibility that mothers specifically choose some types of shift schedules, such as evening or night shifts, in order to spend more time with their children, corresponding to better behavioral outcomes. Indeed, as aforementioned, relatively large percentages of mothers choose to work nonstandard schedules in order to balance work and child care (Presser, 2003). Parents who spend more time with their children may also be able to spend more time providing human capital, by reading to children, for example, thereby corresponding to higher levels of conceptual resources in this realm as well. It is possible that fathers’ nonstandard schedules are less likely to correspond to changes in behavioral problems because fathers are less likely than mothers to choose shift schedules that specifically allow them to spend more time with children, consequently providing fewer opportunities for paternal nonstandard schedules to influence the provision of conceptual resources. Future research should continue to investigate the mechanisms behind these relationships.

The findings from this study provide evidence that some types of nonstandard schedules offer resources for flexibility among two-parent families, particularly those with young children. Given that much of the literature has focused on the negative influence of nonstandard schedules on child and family well-being, these findings are notable and important for parents attempting to balance work and family. While these findings would seem to depart from the literature on nonstandard shifts and child behavioral outcomes, they are not necessarily inconsistent with these findings. In a related study, Han et al. (2010) find that mothers’ evening and night shifts significantly increase delinquent behaviors such as drug use and sexual activity among adolescents (two-parent and single-parent families were both examined). Likewise, Strazdins et al. (2006) examine Canadian dual-earner families and find that mothers’ and fathers’ nonstandard schedules increase behavioral problems in 2- to 11-year-old children. While the findings in my study seemingly conflict with the findings in these and other previous studies, the studies aforementioned differ in their focuses from my own, indicating that my findings are not necessarily incompatible with those of previous studies. For example, Han et al. (2010) focus on adolescents and risky behaviors such as drug use and teen sex, behaviors that differ in magnitude and frequency from the range of problem behaviors exhibited by the majority of children and young adolescents that are the focus of this study. Indeed, it is possible that nonstandard schedules enable risky behaviors in teenagers predisposed to these behaviors by reducing parent monitoring, for example, while not increasing or even decreasing more common behavior problems among children, such as aggressiveness or hyperactivity. My findings that the beneficial associations between nonstandard schedules and children’s behavioral problems largely exist among children of young ages further illustrates that my findings and those of Han et al. (2010) are not incompatible.

Strazdins et al. (2006) also operationalize nonstandard schedules as a binary indicator (standard/nonstandard). Examining a wider range of nonstandard schedules allows for a more nuanced sense of why a particular schedule might be harmful and which schedules might have no or even beneficial effects. Additionally, neither of these studies examines whether the relationships differ by the gender or age of the child. It is therefore not entirely surprising that my study found that some nonstandard schedules have advantageous associations, while other studies do not, given this study’s use of longitudinal data and its attention to two-parent families, gender and age effects, and the varying influences of different shift schedule articulations.

Finally, my findings that parental rotating/split/other shifts are associated with more behavior problems among children is consistent with the findings of previous research and the findings of the aforementioned studies that nonstandard schedules are associated with worse child outcomes. It is possible that these types of nonstandard schedules are particularly harmful for family functioning because of their instability and because of the relative unpredictability of work hours, likely increasing parental stress and hindering parents’ abilities to reliably organize time with family members.

Contributions

These results have important implications for understanding how nonstandard schedules and parental employment more broadly influence family functioning. The United States offers few accommodations for working parents: businesses are not required to offer maternity leave and when they do so, the benefits are often much smaller than they are in other developed countries (Etehad & Lin, 2016); paternity leave is virtually nonexistent, and child care is extremely costly and often not convenient to access (Gould & Cooke, 2015). In the face of these obstacles, nonstandard schedules may offer important resources for flexibility for two-parent families, allowing parents to tag-team parent or to schedule their work hours in order to allow relatives to watch their children. In these cases, nonstandard schedules may offer benefits to families and signal the value to policymakers and businesses of increasing flexibility and the schedule options employees can choose from (distinct from schedules options imposed on employees). That being said, nonstandard schedules have well-documented associations with worse health, marital strife, and stress among the individuals working them. They should therefore be viewed as one tool for flexibility, with a multitude of others—such as parental leave and affordable child care—not yet being fully utilized.

This study also offers important contributions to the study of parental employment, nonstandard schedules, and children’s behavioral outcomes. Very few studies have analyzed whether the relationships between parental nonstandard schedules and child behavioral outcomes depend on the gender of the child or the parent (Li et al., 2014), and none that I know of analyze this gendered effect with such a large sample and wide range of ages. However, these findings demonstrate that nonstandard schedules do have different associations with behavior problems for boys and girls, younger and older children, and mothers and fathers. The former two findings also indicate that we should be cautious in our interpretations of studies that jointly examine girls and boys and younger and older children and that the influence of nonstandard schedules on children may be masked if gender and age effects are not examined.

Finally, this study found that it is important to separately examine different types of shift schedules. Examining nonstandard schedules as a binary category masks the unique benefits and constraints of different types of nonstandard schedules, such as the instability associated with rotating and split shifts, or the potential for night shift workers to spend more time with their children during the day, and it is only by identifying these benefits and constraints that the mechanisms behind these relationship can be illuminated.

For many families, nonstandard schedules may offer much-needed flexibility, even if these schedules come with strains in health, scheduling, and marital relationships. Nonstandard schedules may allow parents to off-set their schedules and tag-team parent, they may allow parents to save money on child care costs or encourage fathers to take on more parenting duties, enhancing feelings of parental effectiveness. Children may have closer relationships with extended family if these kin provide babysitting or child care in place of child care centers. At the same time, nonstandard schedules, particularly those that are relatively unstable and subject to continuous change, may disrupt family functioning and, as a result, increase children’s behavior problems. These findings therefore suggest the importance of stability for family functioning and child well-being. By documenting significant variation by child’s gender and age, mother’s and father’s employment, and type of nonstandard work schedule, the findings from this study set the stage for future research on the consequences of nonstandard work schedules for the well-being of children. Especially, important will be discerning the mechanisms that are responsible for the patterns revealed here.

Footnotes

Appendix

Table A1.

Number of Observations at Each Wave by Child Gender.

	Girls	Boys
1992	678	734
1994	709	792
1996	747	773
1998	749	783
2000	569	592
2002	662	667
2004	576	551
2006	146	151
Total	4,836	5,043

Acknowledgements

I am grateful to Stew Tolnay and Julie Brines for their continued mentorship and invaluable insights and feedback. I also thank the anonymous reviewers for their time and valuable feedback.

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Partial support for this research came from a Shanahan Endowment Fellowship and a Eunice Kennedy Shriver National Institute of Child Health and Human Development training grant, T32 HD007543, and a Eunice Kennedy Shriver National Institute of Child Health and Human Development research infrastructure grant, R24 HD042828, to the Center for Studies in Demography & Ecology at the University of Washington.

Notes

References

Achenbach

Edelbrock

(1981). Behavioral problems and competencies reported by parents of normal and disturbed children aged four through sixteen. Monographs of the Society for Research in Child Development, 46(1), 1-82.

Amato

Rivera

(1999). Paternal involvement and children’s behavior problems. Journal of Marriage and the Family, 61, 375-385.

Bara

A.-C.

Arber

(2009). Working shifts and mental health: Findings from the British Household Panel Survey (1995-2005). Scandinavian Journal of Work, Environment & Health, 35, 361-367.

Barnett

R. C.

Gareis

K. C.

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

Barton

Aldridge

Smith

(1998). The emotional impact of shift work on the children of shift workers. Scandinavian Journal of Work, Environment & Health, 24, 146-150.

Brayfield

(1995). Juggling jobs and kids: The impact of employment schedules on fathers’ caring for children. Journal of Marriage and the Family, 57, 321-332.

Brooks-Gunn

(1995). Children in families in communities: Risk and intervention in the Brofenbrenner tradition. In Moen

Elder

G. H.

Lusher

(Eds.), Examining lives in context: Perspective on the ecology of human development (pp. 467-519). Washington, DC: American Psychological Association.

Bureau of Labor Statistics. (2013). Employment projections 2012-2022. Washington, DC: Bureau of Labor Statistics.

Bussey

Bandura

(1999). Social cognitive theory of gender development and differentiation. Psychological Review, 106, 676-713.

10.

Cripps

Zyromski

(2009). Adolescents’ psychological well-being and perceived parental involvement: Implications for parental involvement in middle schools. Research in Middle Level Education, 33(4), 1-13.

11.

Curran

Obeidat

Losardo

(2010). Twelve frequently asked questions about growth curve modeling. Journal of Cognitive Development, 11, 121-136.

12.

Daniel

S.S.

Grzywacz

J.G.

Leerkes

Tucker

Han

W.J.

(2009). Nonstandard maternal work schedules during infancy: Implications for children’s early behavior problems. Infant Behavior and Development, 32, 195-207.

13.

Deater-Deckard

Dodge

(1997). Externalizing behavior problems and discipline revisited: Nonlinear effects and variation by culture, context, and gender. Psychological Inquiry, 8, 161-175.

14.

Dockery

Kendall

(2009). Parents’ work patterns and adolescent mental health. Social Science & Medicine, 68, 689-698.

15.

Englund

Luckner

Whaley

Egeland

(2004). Children’s achievement in early elementary school: Longitudinal effects of parental involvement, expectations, and quality of assistance. Journal of Educational Psychology, 96, 723-730.

16.

Etehad

Lin

(2016, August 13). The world is getting better at paid maternity leave. The U.S. is not. The Washington Post. Retrieved from https://www.washingtonpost.com/news/worldviews/wp/2016/08/13/the-world-is-getting-better-at-paid-maternity-leave-the-u-s-is-not/?utm_term=.5f143d98cf02

17.

Fenwick

Tausig

(2001). Scheduling stress: Family and health outcomes of shift work and schedule control. American Behavioral Scientist, 44, 1179-1198.

18.

Flannery

Williams

Vazsonyi

(1999). Who are they with and what are they doing? Delinquent behavior, substance use, and early adolescents’ after-school time. American Journal of Orthopsychiatry, 69, 247-253.

19.

Gilman

Huebner

(2003). A review of life satisfaction research with children and adolescents. School Psychology Quarterly, 18, 192-205.

20.

Gottfredson

Weisman

(2001). The timing of delinquent behavior and its implications for after-school programs. Criminology & Public Policy, 1, 61-86.

21.

Gould

Cooke

(2015, October). High quality child care is out of reach for working families (Issue Brief No. 404). Washington, DC: Economic Policy Institute.

22.

Han

W.J.

(2005). Maternal nonstandard work schedules and child cognitive outcomes. Child Development, 76, 137-154.

23.

Han

W. J.

(2008). Shift work and child behavioral outcomes. Work, Employment, and Society, 22, 67-87.

24.

Han

W. J.

Miller

Waldfogel

(2010). Parental work schedules and adolescent risky behaviors. Developmental Psychology, 46, 1245-1267.

25.

Han

W. J.

Waldfogel

(2007). Parental work schedules, family process, and early adolescents’ risky behavior. Children and Youth Services Review, 29, 1249-1266.

26.

Harris

K. M.

Furstenberg

F. F.

Marmer

(1998). Paternal involvement with adolescents in intact families: The influence of fathers over the life course. Demography, 35, 201-216.

27.

Hattery

(2001). Tag-team parenting: Costs and benefits of utilizing nonoverlapping shift work in families with young children. Families in Society: Journal of Contemporary Social Services, 82, 419-427.

28.

Hosking

Western

(2008). The effects of non-standard employment on work–family conflict. Journal of Sociology, 44, 5-26.

29.

Huebner

E. S.

Alderman

G. L.

(1993). Convergent and discriminant validation of a children’s life satisfaction scale: Its relationship to self- and teacher-reported psychological problems and school functioning. Social Indicators Research, 30, 71-82.

30.

Joshi

Bogen

(2007). Nonstandard schedules and young children’s behavioral outcomes among working low-income families. Journal of Marriage and Family, 69, 139-156.

31.

Kalil

Ziol-Gues

K. M.

Epstein

J. L.

(2010). Nonstandard work and marital instability: Evidence from the national longitudinal survey of youth. Journal of Marriage and Family, 72, 1289-1300.

32.

Kantermann

Juda

Vetter

Roenneberg

(2010). Shift-work research: Where do we stand, where should we go? Sleep and Biological Rhythms, 8, 95-105.

33.

Lesnard

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

34.

Johnson

Han

Andrews

Kendall

Strazdins

Dockery

(2014). Parents’ nonstandard work schedules and child well-being: A critical review of the literature. Journal of Primary Prevention, 35, 53-73.

35.

Lindsay

Maher

Bardoel

(2009). Modified maternalism: Nurses and their families managing work and care. Journal of Comparative Family Studies, 40, 661-675.

36.

Lundberg

McLanahan

Rose

(2007). Child gender and father involvement in fragile families. Demography, 44, 79-92.

37.

Mammen

(2011). Fathers’ time investments in children: Do sons get more? Journal of Population Economics, 24, 839-871.

38.

Mattejat

Remschimidt

(2008). The children of mentally ill parents. Deutsches Arzteblatt International, 105, 413-418. doi:10.3238/arztebl.2008.0413

39.

Maume

Sebastian

R. A.

Bardo

A. R.

(2010). Gender, work–family responsibilities, and sleep. Gender & Society, 24, 746-766.

40.

McMenamin

(2007, December). A time to work: Recent trends in shift work and flexible schedules. Monthly Labor Review. Retrieved from https://www.bls.gov/opub/mlr/2007/12/art1full.pdf

41.

Miedel

W. T.

Reynolds

A. J.

(1999). Parent involvement in early intervention for disadvantaged children: Does it matter? Journal of School Psychology, 37, 379-402.

42.

NICHD Early Child Care Research. (2005). Child care and development: Results from the NICHD study of early child care and youth development. New York, NY: Guilford Press.

43.

Presser

(2003). Working in a 24/7 economy: Challenges for American families. New York, NY: Russell Sage Foundation.

44.

Raley

Bianchi

(2006). Sons, daughters, and family processes: Does gender of children matter? Annual Review of Sociology, 32, 401-421.

45.

Richardson

Radziszewska

Dent

Flay

(1993). Relationship between after-school care of adolescents and substance use, risk taking, depressed mood, and academic achievement. Pediatrics, 92, 32-38.

46.

Starrels

(1994). Gender differences in parent-child relations. Journal of Family Issues, 15, 148-165.

47.

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, 394-410.

48.

Waldfogel

Craigie

T.-A.

Brooks-Gunn

(2010). Fragile families and child wellbeing. The Future of Children, 20, 87-112.

49.

Wight

Raley

Bianchi

(2008). Time for children, one’s spouse and oneself among parents who work nonstandard hours. Social Forces, 87, 243-269.

50.

Williams

Kelly

F. D.

(2005). Relationships among involvement, attachment, and behavioral problems in adolescence: Examining father’s influence. Journal of Early Adolescence, 25, 168-196.

51.

Zill

Peterson

(1986). Behavior problems index. Washington, DC: Child Trends.