The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.
About the Series
Assessing the New Federalism is a multiyear Urban Institute project designed to analyze the devolution of responsibility for social programs from the federal government to the states, focusing primarily on health care, income security, employment and training programs, and social services. Researchers monitor program changes and fiscal developments. In collaboration with Child Trends, the project studies changes in family well-being. The project aims to provide timely, nonpartisan information to inform public debate and to help state and local decisionmakers carry out their new responsibilities more effectively.
Key components of the project include a household survey, studies of policies in 13 states, and a database with information on all states and the District of Columbia, available at the Urban Institute's Web site (http://www.urban.org). This paper is one in a series of occasional papers analyzing information from these and other sources.
Contents
Executive Summary
Introduction
Data and Methods
Data and Sample Selection
Measures
Analytic Method
Results
Low-Income Children under Age 6
Low-Income Children Age 6 to 11
Low-Income Children Age 12 to 17
Summary and Conclusion
Appendix Tables
Notes
References
About the Author
Executive Summary
Over the past decade, increasing numbers of low-income parents have
entered the labor force. How parents combine their market work and their
child-rearing responsibilities has become an important concern for policymakers
and researchers. One reason for the interest is that working parents'
ability to meet the competing demands on their time may influence their
children's well-being. To date, the majority of the research on the relationships
between parent work and child well-being has examined the effects of
maternal employment on child development. Reviews of this literature typically
conclude that maternal employment is not related to negative achievement
and behavior outcomes for school-age and older children. Investigations
of the influence of maternal employment on contemporaneous and
longer-term outcomes for young children yield inconsistent results. Further,
recent evidence from evaluations of welfare-to-work programs suggests that
children of welfare leavers may neither directly benefit nor suffer from a
change in their mother's work status. Income, rather than hours of maternal
work, may be more important for this subgroup of children.
As the size of the welfare population continues to wane, it becomes
increasingly important to shift some of the research focus away from studies
of welfare leavers and onto studies of the larger population of low-income
families. Making this shift requires a consideration of not just maternal work
effort but paternal work effort as well. Using data from the National Survey
of America's Families (NSAF), this paper examines contemporaneous relationships
between patterns of parent work and positive child outcomes
among low-income families by asking three questions. How do parents
organize their work schedules? Are there relationships between patterns of
parent work and child well-being? Are the relationships the same for children
of single parents and children of married parents?
The paper finds that most low-income children have at least one parent
who works full-time. In general, levels of parent work are not associated with
positive child outcomes, even when family income is considered. A few
exceptions, however, do exist. High levels of parent work are negatively associated
with indicators of parent involvement among low-income preschoolers
of single parents. How parents schedule their work seems most important
for low-income, school-age children of married parents. For all children, positive
outcomes are more strongly associated with both family and child characteristics
than with patterns of parent work. Results in the paper are categorized
by child's age. For comparison, some of the key associations between
child well-being and parent work and home environment are summarized by
marital status below.
Key Findings by Parents' Marital Status
Low-Income Children of Single Parents
- Full-time work is common among low-income single parents. Nearly
two-fifths of low-income children under age 6 with single parents have a
parent who works full-time; 48 percent of low-income children ages 6 to
11 with single parents have a parent who works full-time; and 54 percent
of low-income adolescents with single parents have a parent who works
full-time.
- Full-time work is negatively associated with indicators of parent involvement
for low-income children under age 6 with single parents. But levels
of work are not related to positive outcomes for older children of single
parents. Working the night shift is unassociated with either parent
involvement or child well-being for low-income children of single
parents.
- Involvement in extracurricular activities is less likely among low-income
children age 6 to 17 when their single parent has less than a high school
education. Interestingly, this measure of parent education is not strongly
related to any other outcome for low-income children of single parents.
- The presence of siblings or other children in the family reduces the likelihood
that a young child of a low-income single parent is read to frequently
by family members. Other adults in the family, however, increase
the likelihood of frequent reading. While the presence of other children
is unrelated to outcomes for older children, having other adults in the
household increases the probability that a child age 6 to 11 has few
behavioral problems.
- Young children of low-income single parents in poor health are less likely
to be read to frequently by family members. Parent health, however, is
unrelated to outcomes for older children (between the ages of 6 and 17).
In contrast, poor parent mental health is associated with a reduced likelihood
of favorable outcomes among low-income, school-age children
and adolescents but unrelated to parent involvement for the younger
children of single parents.
Low-Income Children of Married Parents
- Most low-income children of married parents have at least one parent
who works full-time: 90 percent of children under age 6; 87 percent of
children age 6 to 11; and 85 percent of children age 12 to 17. Not surprisingly,
the proportion of low-income children of married parents who
have a full-time working parent married to another worker is higher for
school-age children and adolescents (36 percent) than for children under
6 years old (26 percent).
- Levels of parent work and parent work schedules have stronger relationships
with child outcomes among low-income, school-age children of
married parents than among either their younger or older counterparts.
Having a parent who works the night shift is associated with a reduction
in the likelihood that a low-income, school-age child participates in
extracurricular activities. But the likelihood of activity involvement
increases among this group of children when their married parents both
work, one full-time and the other part-time. Activity involvement also
increases among low-income, school-age children as the proportion of
total parent work hours worked by the mother increases.
- The share of parental work hours worked by the mother, however, is negatively
associated with school engagement among low-income, school-age
children. Low-income, school-age children of married parents who
try to arrange their work hours so one of them can care for their
child(ren) are more likely to be highly engaged in school than their counterparts
whose parents do not arrange their work hours.
- Low levels of parent education are associated with reductions in parent
involvement for low-income children under age 6 of married parents.
Older low-income children with married parents who have a parent who
did not finish high school are less likely than their counterparts with more
educated parents to participate in extracurricular activities.
- Other adults in the family increase both the likelihood that young children
of low-income married parents go on outings and the probability
that adolescents of low-income married parents have few behavioral
problems. Other children in the family, however, reduce the likelihood
that young children are either read to frequently or taken on outings, as
well as the likelihood of involvement in extracurricular activities for low-income
children age 6 to 11 of married parents.
- Poor maternal health is both positively and negatively related to outcomes
for low-income children under age 12 of married parents. Frequent
reading increases, while being included on outings decreases, for
children under age 6 when their married mother is in poor health. Low-income
children age 6 to 11 with a married mother in poor health are
more likely to be highly engaged in school but less likely to have few
behavioral problems than their counterparts with healthy mothers.
Introduction
Over the past decade, increasing numbers of low-income parents have
entered the labor force. How parents combine their market work and their
child-rearing responsibilities has become an important concern for policymakers
and researchers. Working parents' ability to meet the competing
demands on their time may influence their children's well-being. The majority
of the research that examines the relationships between parent work and
child well-being examines the effects of maternal employment on child development.
Reviews of this literature typically conclude that maternal employment
is not related to negative achievement and behavior outcomes for
school-age and older children (Belsky 1990; Hoffman 1989; Perry-Jenkins,
Repetti, and Crouter 2000). A number of studies, even those using the same
dataset, come to different conclusions about the contemporaneous and
longer-term impacts of maternal employment on young children (for examples,
see Belsky and Eggebeen 1991; Han, Waldfogel, and Brooks-Gunn
2001; Parcel and Menaghan 1994; Vandell and Ramanan 1992). In her
review of the literature and reanalysis of the data, Harvey (1999) concludes
that the influence of early parental employment on child development varies
by the age of the child and the marital and economic status of the parents. In
general, however, she finds few consistent effects, either negative or positive,
of early parental employment on child well-being.
Recent evidence from evaluations of welfare-to-work programs suggests
that children of welfare leavers may neither directly benefit nor suffer from a
change in their mother's work status. Income, rather than hours of maternal
work, may be more important for this subgroup of children. In general, welfare-
to-work programs that both increased work and increased income had
either no effects or positive effects on the children of the welfare leavers,
while programs that did not result in increased incomes had generally negative
effects on the children (Hamilton, Freedman, and McGroder 2000;
Morris et al. 2001; Sherman 2001). A survey of women transitioning off of
welfare finds relatively few relationships between changes in maternal work
and parenting behavior (Kalil, Dunifon, and Danziger 2001).
As the size of the welfare population continues to wane, it becomes
increasingly important to shift some of the research focus away from studies
of welfare leavers and onto studies of the larger population of low-income
families. Making this shift requires a consideration of not just maternal work
effort but paternal work effort as well. Married parents make conscious deci-
sions about the number of hours that each parent works and how they will
organize their work schedules. For example, a couple may decide that they
both would like to have full-time jobs, but they would also prefer to have one
parent at home with their child for a portion of the typical work day. This
couple may choose to have one parent work a night shift while the other
works during the day. Their decision about how to schedule their market
work hours may influence their child's well-being. The relationship could be
positive, by providing more parent time to the child, or negative, by increasing
stress in the household.
Both single and married parents face labor supply decisions and labor
demand constraints. How they resolve these trade-offs influences the amount
and type of resources they can provide for their children. While more time at
the office may provide additional income and other material resources, it also
reduces the amount of time parents can spend with their children. Working
parents can be positive role models for children; however, parent work may
also increase family stress. It is not clear from previous research or theory
whether high levels of parent work should be associated with either positive
or negative outcomes for children. In addition, other factors, such as the
quality of child care and the presence of other adults in the family, can mediate
the influence of parent work on child well-being.
Using data from the National Survey of America's Families (NSAF), this
paper examines contemporaneous relationships between patterns of parent
work and positive child outcomes among low-income families by asking three
questions. How do parents organize their work schedules? Are there
relationships between patterns of parent work and child well-being? Are the
relationships the same for children of single parents and children of married
parents?
The paper finds that most low-income children have at least one parent
who works full-time. In general, levels of parent work are not associated with
positive child outcomes, even when family income is considered. A few
exceptions, however, do exist. High levels of parent work are negatively associated
with indicators of parent involvement among low-income preschoolers
with single parents. How parents schedule their work seems most important
for school-age children with married parents. For all children, positive
outcomes are more strongly associated with both family and child characteristics
than with patterns of parent work.
The next section describes the data used in the analyses, defines the parent
work and child outcome measures, and outlines the analytic method used
in the paper. Next are the results, reporting the findings for preschoolers
(children under age 6), school-age children (children age 6 to 11), and adolescents
(children age 12 to 17) separately. The paper concludes with a summary
of its findings.
Data and Methods
Data and Sample Selection
Data for this paper come from the 1999 National Survey of America's
Families (NSAF). The NSAF is a survey of the economic, health, and social
characteristics of Americans, with an emphasis on low-income families with
children.1 Data from the survey are nationally representative of the civilian
noninstitutionalized population under age 65 and their families. The NSAF
contains an oversample of families with incomes below 200 percent of the
federal poverty level (FPL), allowing detailed analyses of subgroups of low-income
families.
In households with children under the age of 18, the NSAF interviewers
select up to two children for in-depth interviews: one under the age of 6 and
another between the ages of 6 and 17. The adult in the household who is
most knowledgeable (MKA) about each selected child answers questions
about family members' economic, health, and social characteristics. For most
children, the MKA is the child's biological mother.
Much of the prior research on the associations between parent work and
child well-being has either focused exclusively on families transitioning from
welfare to work or has analyzed outcomes for children in specific age ranges,
regardless of their welfare status. In contrast, this paper examines children of
all ages who live in low-income families. Here low-income includes all children
living in families with incomes less than 200 percent of FPL.2
Marital status influences the choices parents can make about how to pattern
their market work. As a result, this paper separates low-income children
into two groups according to the marital status of their parents (married versus
single). To be included in the married parents' sample, low-income children
must live with both parents at the time of the survey, and their parents
must have a legal marital relationship. Parents may be biological, step, or
adoptive. Children who live with cohabiting partners are excluded from the
analysis.3 Children in the single-parent sample live with only one of their parents, and their custodial parent may be neither married nor cohabiting. Parents
of these children may be widowed, divorced, separated, or never
married.4
Measures
Patterns of Parent Work
To describe parent work effort, I developed two separate categorizations of
parent work—one for children of single parents and one for children of married
parents. For children of single parents, the market work effort of their
parents is classified as either part-time (PT—usually works more than 1 but
less than 35 hours per week); full-time (FT—usually works at least 35 hours
per week); or no work (NW—usually works zero hours per week).5
For children of married parents, the categorical measures of parent work
are designed to capture the combination of both the mother's and the
father's labor supply under the assumption that couples plan their labor supply
jointly. The five categories of market work effort are based on the parents'
current "usual hours of work":
Full-time/full-time (FT/FT): Both parents work at least 35 hours per week.
Full-time/part-time (FT/PT): One parent works at least 35 hours per week
and the other parent works between 1 and 34 hours per week.
Full-time/no work (FT/NW): One parent works at least 35 hours per week
and the other parent is not employed.
Part-time/[part-time/no work] (PT/[PT/NW]): One parent works between
1 and 34 hours per week and the other parent either works between 1 and
34 hours per week or is not employed.
No work/no work (NW/NW): Neither parent is employed.
An additional set of variables captures various characteristics of the parents'
work schedules. These variables are described in the analytic method
section below.
Child Outcome Measures
The child outcome variables in the NSAF vary by the age of the child. Thus,
the analysis that follows is divided not only by the marital status of the parents
but also by the age of the child. For children under the age of 6, this
paper explores the frequency of interactions with family members. For
school-age children and adolescents, the analysis focuses on behavioral and
school outcomes. Below, I describe each outcome measure.6
Children under Age 6. The two well-being measures for children under
6 (preschoolers)—frequent reading and frequent outings—measure cognitive
stimulation provided to the child by family members. These are not child
outcome measures but rather indicators of mediating factors. For example,
children who are read to frequently when they are young have higher test
scores when they are school-aged (Wells 1985). In this paper, the outcome
measures for preschoolers are labeled "parent involvement."
Frequent reading: Identifies children who are read to or told stories frequently
(six or more days a week) by family members.
Frequent outings: Identifies children who are taken on frequent outings
(at least once a day) by family members.
Children Age 6 to 17. The outcome measures for children age 6 to 11
(school-aged) and 12 to 17 (adolescents) are conceptually the same, but they
are measured differently for the two groups. The three measures are as
follows:
Few behavioral problems: Identifies children with low levels of behavioral
and emotional problems, using a set of questions developed as an indicator
of children's mental health for the National Health Interview Survey.7
Engaged in school: Identifies children who are highly engaged in school,
using a scale measure of school engagement.8
Involved in activities: Identifies children who were involved in at least one
extracurricular activity over the past year. Activities include sports, clubs, and
lessons.
The analysis focuses on positive outcomes for children. For all the outcome
measures except activity involvement, the absence of a positive outcome
does not imply a negative outcome. For example, most children who
do not have "few behavioral problems" (a positive outcome) also do not have
"a lot of behavioral problems" (a negative outcome). Instead, the majority of
children have some problems but not enough to be considered troublesome.
9,10
Analytic Method
The analysis in this paper is organized by the age of the child. In each of the
results sections, descriptions of parents' work effort are followed by a summary
of parenting/child outcomes by marital status. Although the main
interest of the paper is to explore relationships between parent work and
child well-being, other factors act upon children and parents and may influence
not only child well-being but also how parent work affects child well-being.
To help disentangle some of the work/outcome relationships, a set of
multivariate analyses is undertaken to control for other factors that may influence
child well-being.11
Because I am interested in understanding the associations between child
outcomes and the amount of parent work, as well as their work schedules, the
multivariate analyses are restricted to families with some work effort. Thus
the results describe the relationships between child well-being and different
levels of parent work rather than the relationships between child well-being
and no parent work versus some parent work. For each child outcome, two
logistic regression equations are estimated—one for children of single parents
and one for children of married parents. The results from these equations isolate
the size and statistical strength of the relationship between, for example,
parent work and child well-being from the effect of the other variables
included in the equation. In other words, we estimate the relationship
between parent work and child well-being net of many of the other factors
we believe are associated with child well-being.
The variables included in the multivariate analyses fall into three main
categories: parent work, family environment, and child characteristics. The
parent work variables are measured as follows:
Categorical work variables: As defined above under Patterns of Parent
Work, the omitted category is either FT/NW for children with married parents
or PT for children with single parents.12
Percentage of total parent hours worked by mother: The percentage of total
parent hours of work (sum of both of the parents' usual hours of work) that
are usually worked by the child's mother. Used to assess the availability of the
mother relative to the father in the regressions for children of married
parents.
A parent works odd hours: If at least one parent's usual work schedule is
between the hours of 6 p.m. and 6 a.m., the child receives a 1 for this
variable.
Parents arranged work hours: For children under the age of 12 who have
two working parents, this variable indicates that over the past month the parents
arranged their work schedules so they could take turns caring for the
child. (Used only in the specifications for children of married parents.)
The family environment variables are family income, the number of other
children in the family, and indicators of the type of primary child care
arrangement (for preschoolers), poor parent mental health, parent aggravation,
parent and child argue a lot (for adolescents), another adult in the
family, and a parent has less than a high school education. Included child
characteristics are age and indicators of disability, race, ethnicity, and sex.
Appendix table 1 includes definitions for each of the family environment and
child characteristic variables, along with each variable's metric.
Many of the family environment variables may be related to work. For
example, high levels of work may be associated with high levels of parent
aggravation. Including highly correlated variables in a multivariate regression
can make it difficult to determine the separate influence of the variables on
the outcome. In the data, however, none of the family environment and child
characteristics are highly correlated with levels of parent work. Although a
number of the correlation coefficients are statistically significant, their levels
tend to be low—nearly every correlation coefficient has an absolute value less
than 0.20. For example, for school-age children with single parents, the correlation
coefficient for full-time work and parent aggravation is -0.06.
On their own, coefficients from logistic regressions can be difficult to
interpret. As a result, this paper uses predicted probabilities to demonstrate
the size of the relationships between child outcomes, parent work, and other
factors.13 To calculate predicted probabilities, the coefficients for each variable are multiplied by a value, typically the sample average, for each variable.
The sum of these products is transformed into a probability.14 To examine the
influence of one of the included variables, the value for that variable is set to
1 if the variable is a categorical value or some level of interest if the variable
is a continuous variable. For example, if we were interested in exploring the
association between frequent reading and having a parent who works odd
hours, we would use the regression results to calculate a predicted probability
that the child is read to frequently by including the average values for all
the variables in the regression equation but fixing the parent works odd hours
variable at 1.
Results
The following results are not meant to suggest a behavioral relationship, but
they do suggest associations between parent work patterns, other variables,
and child outcomes. In this section, I discuss results for preschoolers, school-age
children, and teenagers, respectively. Within each age group, a discussion
of the patterns of parent work for each marital status precedes a description
of the average outcomes for children by their parents' marital status. After
describing the univariate results, I present and discuss results from the multivariate
analyses. Each age group section concludes with a brief summary.
Low-Income Children under Age 6
Patterns of Parent Work
Single Parents. The majority (55.6 percent) of low-income children
under age 6 who live with single parents have a parent who works: 39.1 percent
have a parent who works full-time, while 16.5 percent have a parent who
works between 1 and 34 hours per week.15 The remaining 44.3 percent of
low-income preschoolers of single parents have parents who do not work.
Among children under age 6 of a single parent who works, the average
weekly hours of parent work is 36 hours (see appendix table A2a).
Married Parents. Sixty-four percent of low-income preschoolers of married
parents have one parent who works full-time and another parent who
does not work. More than one in four, however, have two parents who
work—13.2 percent have two parents who work full-time, and 12.6 percent
have a parent who works full-time and a parent who works part-time. Among
low-income preschoolers with married parents at least one of whom works,
their parents work, on average, a combined 55 hours per week (see appendix
table A2b).
How Are Low-Income Preschoolers Faring?
More than one-third (36.9 percent) of low-income children under age 6 who
live in single-parent families are read to frequently by family members, and
nearly 24 percent are taken on daily outings by family members. Among low-income
preschoolers with married parents, 42.8 percent are read to frequently
by family members, and 23.1 percent are taken on daily outings.
Surprisingly, these percentages are not much larger than the percentages for
young children of single parents. Given that there are typically more adults
and, by definition, one more parent in these children's families, it seems reasonable
to expect that children of married parents might benefit from more
interactions with family members.
Relationships between Parent Work and Child Well-Being
In addition to parent work, many other factors influence the quality and
quantity of parent-child interactions. For example, having additional children
to care for can take some time away from a parent's interaction with a child.
If we can statistically eliminate the influence of other factors on parent
involvement, a clearer picture of the association between parent work and
child outcomes may result.
To further explore the associations between parent work and the two
measures of parent involvement, I use a multivariate regression to predict
parent involvement with children, holding measures of parent work, the
child's environment, and the child's characteristics constant. The results from
the regressions are then used to predict the probability of a positive outcome,
assuming different levels of parent work as well as different levels of the other
intervening variables.16
Levels of Parent Work. For low-income preschoolers of single parents,
full-time work is associated with less parent involvement than part-time work.
The top portion of table 1 indicates that we would expect 36.0 percent of
low-income preschoolers with a single parent who works full-time to be read
to frequently by family members, compared with 49.0 percent of low-income
preschoolers with a single parent who works part-time. Likewise, the likelihood
that a child is taken on frequent outings is estimated to be much larger
for low-income preschoolers with part-time than full-time working single
parents (26.0 percent compared with 15.4 percent). These percentages are
calculated on the assumption that preschoolers of full-time and part-time
working parents have the same family and child characteristics. In other
words, I assume that they differ only in the amount of hours their parents
work. The predicted probabilities, then, allow a comparison of the influence
of different levels of work net of other intervening factors.
|
| Table 1. Predicted Percentage of Children under Age 6 with Positive Outcomes, by Parent Work Status (Low-Income Children with Working Parents, 1999) |
|
| |
Frequent reading |
Frequent outings |
| Single parents |
| Full-time |
36.0* |
15.4** |
| Part-time |
49.0 |
26.0 |
| Married parents |
| Full-time/full-time |
37.4 |
22.1 |
| Full-time/part-time |
36.3 |
19.7 |
| Full-time/no work |
42.5 |
20.4 |
| Part-time/other |
53.1 |
25.2 |
|
Source: Urban Institute calculations from the 1999 National Survey of America's Families.
Notes: Predicted probabilities are calculated from the results of multivariate regressions that control for parent work, family environment, and child characteristics. See text for details.
* significantly different from the part-time estimate at the .10 level.
** significantly different from the part-time estimate at the .05 level.
No values for children of single parents were significantly different from the part-time estimate at the .01 level.
No values for children of married parents were significantly different from the full-time/no work estimate at the .10, .05, or .01 level. |
In contrast, for low-income preschoolers of married parents, high levels
of parent work are not associated with less parent involvement than lower
levels of parent work are (bottom portion of table 1). Low-income preschoolers
who have two full-time working parents have a higher likelihood of
being taken on daily outings by family members and a lower likelihood of
being read to frequently by family members than preschoolers with a parent
who works full-time and a parent who does not work (22.1 percent compared
with 20.4 percent, and 37.4 percent compared with 42.5 percent).
However, none of the estimated outcomes for low-income preschoolers with
married parents with different levels of work are statistically different from
the estimated outcomes for low-income preschoolers with a parent who
works full-time and a parent who does not. In addition, none of the parent
work schedule variables, such as working odd hours or arranging hours, are
strongly associated with parent involvement.
Family Environment and Child Characteristics. Some of the family
environment and child characteristics included in the analyses are associated
with child outcomes. The size and strength of the associations vary by parent
marital status. Figures 1 and 2 illustrate the relative size of the relationships
between parent involvement and these other factors. The figures display
only the associations that are statistically significant. Each factor is listed on
the vertical axis and its relationship with frequent outings is shown by a light
gray bar, while its relationship with frequent reading is shown by a black bar.
If a factor does not have both bars next to it, its relationship with the missing
parent involvement measure is statistically insignificant.
Figure 1 presents results for low-income preschoolers with single parents
and figure 2 presents results for low-income preschoolers with married parents.
Each figure includes the change in the likelihood that preschoolers with
parents who work full-time experience one of the measures of parent
involvement. For example, to calculate the change in the likelihood that a
low-income preschooler with a single, full-time working parent is taken on
frequent outings based on whether the child is in center care, I estimate the
likelihood of the positive outcome assuming every child has a parent who
works full-time and uses center care as the child's primary care arrangement
(31.1 percent). From this predicted probability, I subtract the average predicted
probability in table 1 (15.4 percent). The net result (15.7 percentage
points) is shown in the first light gray bar on figure 1.17 Because the association between center care and frequent reading among preschoolers with single
parents is statistically insignificant, center care does not have a black bar
next to it.
Poor parent health, Food Stamp receipt, and additional children in the
family are all associated with a reduced probability that low-income
preschoolers of single parents are read to frequently by family members. The
percentage point reduction ranges from 9.2 points (Food Stamp receipt) to
13.1 points (poor parent health). Conversely, receipt of Temporary Assistance
for Needy Families (TANF), other adults in the family, and child disability
are all associated with an increased probability that low-income
preschoolers of single parents are read to frequently by family members. In
this case, the percentage point increase ranges from 9.4 points (other adults)
to 25.5 points (child is disabled). Outings are positively associated with both
center and relative care; children who are placed in these forms of care are,
on average, 16 percentage points more likely to go on daily outings with family
members.
Nanny care, a parent who did not finish high school, additional children
in the family, and being Hispanic are all associated with a reduction in the
likelihood that low-income preschoolers of married parents are read to frequently
by family members (figure 2). Nanny care and being Hispanic are
associated with the largest reductions, 10.8 and 13.1 points, respectively. A
mother in poor health and age of child are positively associated with frequent
reading by family members.
Conversely, a mother in poor health is associated with a reduction in the
likelihood that a low-income preschooler with married parents is taken on
frequent outings by family members. Perhaps the parent(s) substitute reading
for outside activities when health limitations preclude the mother from
leaving the house. In addition to maternal health, low parent education,
additional children, and being Hispanic are all associated with a reduced likelihood
that a low-income preschooler with married parents is taken on frequent
outings. The effect sizes, however, are relatively small. Having other
adults in the family increases by 12.7 percentage points the likelihood that
low-income preschoolers of married parents who both work full-time are
taken on frequent outings by family members.
Summary of Results for Low-Income Preschoolers
Most low-income preschoolers have a parent who works. Ninety percent of
low-income preschoolers with married parents have a parent who works
full-time. Nearly two-fifths of low-income preschoolers with single parents
have a parent who works full-time. Levels of parent work are not strongly
associated with parent involvement for children of married parents. However,
full-time work is negatively associated with both measures of parent involvement
for children of single parents.
Family environment and child characteristics are related to parent
involvement for low-income preschoolers. The relationships, however, differ
by parent marital status. For example, type of child care is strongly associated
with going on outings for preschoolers of single parents and not statistically
related to going on outings for preschoolers of married parents. In addition,
low levels of parent education are more strongly associated with parent
involvement for low-income preschoolers of married parents than for low-income
preschoolers of single parents. Mother's health status is positively
associated with reading among low-income preschoolers of married parents
but negatively associated with outings among this same group of children. In
contrast, among low-income preschoolers of single parents, parent health is
negatively associated with frequent reading.
Low-Income Children Age 6 to 11
Patterns of Parent Work
Single Parents. More than two-thirds of low-income children age 6 to
11 of single parents have a parent who works: 47.8 percent have a parent who
works full-time, and 19.8 percent have a parent who works part-time.18 The
parents of the remaining 32.5 percent of low-income, school-age children of
single parents do not work. Among school-age children with a single parent
who works, the average amount of parent work is 37 hours per week (see
appendix table A3a).
Married Parents. Fifty-one percent of low-income, school-age children
of married parents have one parent who works full-time and another parent
who does not work. More than one in three children, however, have two parents
who work—20.9 percent have two parents who work full-time and
another 15.0 percent have a parent who works full-time and a parent who
works part-time. Among low-income, school-age children of married parents
at least one of whom works, the parents work, on average, a combined 58
hours per week (see appendix table A3b).
How Are Low-Income, School-Age Children Faring?
One in four low-income, school-age children of single parents exhibit few
behavioral problems, 37.4 percent are highly engaged in school, and 66.1
percent are involved in activities. Thirty-five percent of low-income, school-age
children of married parents exhibit few behavioral problems; 41.5 percent
are engaged in school, and 71.4 percent are involved in some extracurricular
activity. Across all outcomes, the average likelihood that low-income,
school-age children of married parents have a positive outcome is higher than
that for low-income, school-age children of single parents.
Relationships between Parent Work and Child Well-Being
Levels of Parent Work. Even when we consider other factors that may
influence child well-being, levels of parent work are not statistically associated
with positive behavior, school engagement, or activity for low-income,
school-age children of single parents (top portion of table 2).19 In fact, the predicted probabilities for positive behavior and activity involvement are
nearly the same for school-age children of a single full-time worker and for
school-age children of a single part-time worker (20.3 percent compared
with 21.0 percent and 68.1 percent compared with 68.9 percent). Further,
having a parent who typically works a night shift is also not associated with
child well-being.
|
| Table 2. Predicted Percentage of Children Age 6 to 11 with Positive Outcomes, by Parent Work Status (Low-Income Children with Working Parents, 1999) |
|
| |
Few behavioral problems |
Engaged in school |
Involved in activities |
| |
|
|
|
| Single parents |
| Full-time |
20.3 |
37.4 |
68.1 |
| Part-time |
21.0 |
31.0 |
68.9 |
| Married parents |
| Full-time/full-time |
33.8 |
36.2 |
75.3 |
| Full-time/part-time |
30.3 |
40.7 |
84.4* |
| Full-time/no work |
28.5 |
45.1 |
72.4 |
| Part-time/other |
28.1 |
37.8 |
68.0 |
|
Source: Urban Institute calculations from the 1999 National Survey of America's Families.
Notes: Predicted probabilities are calculated from the results of multivariate regressions that control for parent work, family environment, and child characteristics. See text for details.
* significantly different from the full-time/no work estimate at the .10 level.
No values for children of single parents were significantly different from the part-time estimates at the .10, .05, or .01 levels.
No values for children of married parents were significantly different from the full-time/no work estimates at the .05 or .01 levels. |
As was the case with low-income preschoolers, the results for low-income,
school-age children of married parents differ from the results for
low-income, school-age children with single parents. For this group of children,
how parents organize their work schedules is somewhat associated with
contemporaneous child outcomes (bottom portion of table 2). Children who
have a parent who works full-time married to a parent who works part-time
are more likely to be involved in extracurricular activities than children who
have one parent who works full-time and one nonworking parent (84.4 percent
compared with 72.4 percent). In addition, how parents organize their
work schedules is associated with child well-being for low-income, school-age
children of married parents. As discussed below, the share of parent hours
of work that are worked by a child's mother, night shift work, and having
parents who arrange their work schedules to care for their children are all
associated with some of the child well-being measures for this group of
children.
Other Factors. Although levels of parent work are not statistically associated
with child well-being for low-income, school-age children of single
parents, some family environment and child characteristics measures have
strong relationships with positive behavior, school engagement, and activity
involvement (figure 3). Parent aggravation, poor parent mental health, child
disability, and child age are all associated with a reduced likelihood that a
low-income, school-age child of a full-time working single parent exhibits
few behavioral problems. The percentage point reduction ranges from 7.0
points (child is 10 years old) to 15.1 points (child is disabled). Conversely,
having other adults in the family and being black are associated with an
increased probability that low-income, school-age children of single parents
are classified as having few behavioral problems. The percentage point
increase is 8.9 points for black and 10.2 points for other adults in the family.
Parent aggravation and child disability are also associated with a reduced
probability that a low-income, school-age child of a single parent is highly
engaged in school. For low-income, school-age children of single parents
who work full-time, the percentage point reductions are 19.4 points for parent
aggravation and 23.4 points for child disability. Female school-age children
are more likely to be engaged in school than male school-age children.
Having a parent who did not finish high school reduces the probability
that a low-income, school-age child is involved in extracurricular activities by
more than a third (24.2 percentage points) for children of single parents who
work full-time. Age is the only other characteristic included in the regression
associated with involvement in extracurricular activities. As low-income,
school-age children of single parents age, they are more likely to participate
in extracurricular activities.
Levels of parent work are generally not associated with child well-being
among low-income, school-age children of married parents. However, how
parents schedule their hours of work is related to both school engagement
and activity involvement. Having a mother who works at least half of all of
the parent hours of work is associated with an increased likelihood that a
child is involved in activities and a decreased likelihood that the child is highly
engaged in school (figure 4). A low-income, school-age child whose parents
arrange their work hours so one of them can care for the child is more likely
to be highly engaged in school. Low-income, school-age children with a
married parent who works nonstandard hours, however, are less likely to
engage in extracurricular activities. The results do not suggest a strong relationship
between patterns of parent work and positive behavior among low-income,
school-age children of married parents.
Interestingly, low-income, school-age children with married mothers in
poor health are less likely to exhibit positive behaviors but more likely to be
positively engaged in school. Having a married parent who is in poor mental
health, however, is related to a reduced probability that low-income children
age 6 to 11 either have few behavioral problems or are highly engaged in
school. Parent aggravation is associated with a reduced probability that a
child exhibits positive behavior, but not statistically related to the other two
outcomes. Disabled children are much less likely to have few behavioral
problems or be highly engaged in school.20 Female students are more likely to be highly engaged in school than male students.
As well as being associated with patterns of parent work, activity involvement
among low-income, school-age children of married parents is related to
parents' education, the number of other children in the family, and the age
of the child. Having a parent who did not finish high school is associated
with a 19 percentage point reduction in the likelihood that a low-income,
school-age child of married parents is involved in activities. Having other
children in the family reduces the probability as well, but not by nearly as
much as parent education. As was the case with children of single parents,
older school-age children of married parents are more likely to participate in
extracurricular activities than younger ones are.
Summary of Results for Low-Income, School-Age Children
A majority of low-income, school-age children see a parent go to work fulltime
each week. Nearly half of low-income, school-age children with single
parents and 87 percent of low-income, school-age children with married parents
have a parent who works full-time. Although parents of low-income,
school-age children have higher levels of market work than parents of low-income
preschoolers, contemporaneous relationships between levels of parent
work and child well-being for children in this age group are not strong.
However, how parents schedule their work hours is related to child well-being
for school-age children with married parents. Particularly interesting
are the relationships between the share of parent work hours worked by the
mother and school engagement, on the one hand, and activity involvement
on the other. The relationship is negative for school engagement and positive
for activity involvement. The positive relationship between a mother's
share of hours and activity involvement could suggest that these parents are
using extracurricular activities as a form of child care when the mother is
working.
Low-Income Children Age 12 to 17
Patterns of Parent Work
Single Parents. Among low-income adolescents of single parents, 54.0
percent have a parent who works full-time, 13.4 percent have a parent who
works part-time, and 32.6 percent have a parent who does not work.21 The
average hours of parent work among low-income adolescents with a single
parent who works is 39 hours (see appendix table A4a). While the single parents
of low-income adolescents tend to work more than the single parents of
very young children, the differences are not that large—67.4 percent of low-income
adolescents of single parents see their parent go to work, compared
with 55.6 percent of low-income preschoolers and 67.6 percent of low-income,
school-age children. The proportion of low-income children with
parents who work full-time, however, seems to increase throughout the age
range—from 39.1 percent for preschoolers to 47.8 percent for school-age
children and 54.0 percent for adolescents.
Married Parents. Nearly half (48.1 percent) of all low-income adolescents
of married parents have one parent who works full-time and another
parent who does not work. And 36.4 percent have two parents who work—
21.1 percent have two parents who work full-time and another 15.3 percent
have a parent who works full-time and a parent who works part-time. Among
low-income adolescent children of married parents at least one of whom
works, the parents work, on average, a combined 59 hours per week (see
appendix table A4a). Interestingly, the pattern of parent work for low-income
adolescents with married parents is very similar to the pattern for low-income,
school-age children with married parents.
How Are Low-Income Adolescents Faring?
On average, 23.1 percent of low-income adolescents of single parents exhibit
few behavioral problems, 33.7 percent are highly engaged in school, and
73.7 percent are involved in activities. More than 31 percent of adolescents
of married parents are identified as having few behavioral problems, 35.6 percent
are engaged in school, and 74.4 percent are involved in some extracurricular
activity. As was the case among low-income, school-age children,
across all outcomes, the average likelihood that low-income adolescents of
married parents have a positive outcome is higher than that for low-income
adolescents of single parents.
Relationships between Parent Work and Adolescent Well-Being
Levels of Parent Work. The predicted probabilities in table 3 suggest
that no strong relationships exist between levels of parent work and child
well-being among low-income adolescents of working parents, even when
other intervening factors are taken into consideration.22 Low-income adolescents
with a single parent who works full-time are more likely to have few
behavioral problems and be highly engaged in school (18.0 percent and 31.8
percent compared with 16.6 percent and 29.1 percent). However, the differences
are not statistically significant. Although the results for low-income
adolescents of married parents seem to suggest a negative association with
levels of parent work and school engagement, none of the differences is
statistically significant. In fact, levels of parent work are not statistically associated
with any of the outcomes for low-income adolescents of either married
or single parents. Once again, family environment and child characteristics
have stronger connections with child outcomes than do levels of parent work.
|
| Table 3. Predicted Percentage of Children Age 12 to 17 with Positive Outcomes, by Parent Work Status (Low-Income Children with Working Parents, 1999) |
|
| |
Few behavioral problems |
Engaged in school |
Involved in activities |
| |
|
|
|
| Single parents |
| Full-time |
18.0 |
31.8 |
77.6 |
| Part-time |
16.6 |
29.1 |
82.2 |
| Married parents |
| Full-time/full-time |
23.4 |
24.7 |
78.2 |
| Full-time/part-time |
29.4 |
29.2 |
74.6 |
| Full-time/no work |
26.2 |
32.3 |
77.7 |
| Part-time/other |
19.1 |
38.2 |
79.8 |
|
Source: Urban Institute calculations from the 1999 National Survey of America's Families.
Notes: Predicted probabilities are calculated from the results of multivariate regressions that control for parent work, family environment, and child characteristics. See text for details.
No values for children of single parents were significantly different from the part-time estimates at the .10, .05, or .01 levels, and no values for children of married parents were significantly different from the full-time/no work estimates at the .10, .05, or .01 levels. |
Other Factors. Frequent parent-child arguments, parent aggravation,
poor parent mental health, and child disability are all independently associated
with a reduced likelihood that a low-income adolescent of a working single
parent exhibits few behavioral problems (figure 5). The percentage point
reduction for children of single parents who work full-time ranges from 7.1
points (parent in poor mental health) to 15.4 points (parent and child argue
a lot). Low-income black adolescents of working single parents are more
likely to have few behavioral problems than their white counterparts. Contrary
to expectations, income has a small but negative relationship with both
few behavioral problems and school engagement, implying that low-income
adolescents with relatively less family income have more positive outcomes
than low-income adolescents with relatively more family income. The sizes of
the associations are quite small, and they are more than offset by the relationships
between child well-being and other family and child characteristics.23
Having a highly aggravated parent is associated not only with a reduced
likelihood of having few behavioral problems but also with a reduced likelihood
that a low-income adolescent of a working single parent is engaged in
school or involved in activities. Of course, it is important to mention that all
the child outcomes are reported by the parent, and a parent's mental state or
feelings about parenting can influence how much the parent knows about his
or her teenager. Once again, being Hispanic is associated with reduced
school engagement and activity involvement compared with white adolescents.
Lower parent education is also linked to a lower probability of activity
involvement, as is age. Adolescent girls are more likely to be highly involved
in school than adolescent boys.
In contrast to the relationship among school-age children, having a
mother who works at least half of all of the parent hours of work is positively
associated with school engagement among low-income adolescents of married
parents (figure 6). No other work variables included in the multivariate
regressions are strongly associated with any of the outcomes and, once again,
environmental and personal characteristics have stronger associations with
outcomes than parent work schedules. In addition to the relative work effort
of the low-income adolescent's mother, being female is associated with high
levels of school engagement. Conversely, having a parent who did not finish
high school and being disabled are associated with reduced probabilities of
being highly engaged in school. Lower levels of parent education are also
associated with a reduced likelihood that a low-income adolescent with working
parents participates in extracurricular activities.
Indicators of the parent-child relationship are negatively associated with
the likelihood that a low-income adolescent has few behavioral problems, as
reported by the parent. Arguing a lot with a parent is associated with a 17.6
percentage point (75 percent) reduction in the probability that a low-income
adolescent of married parents has few behavioral problems. Similarly, high
parent aggravation is associated with a reduced likelihood of having few
behavioral problems. The relationships between having few behavioral problems
and both poor parent mental health and child disability are, once again,
large and negative. However, as mentioned previously, the negative relationship
between child disability and child behavior could indicate a problem
with the outcome measure. Having adults in the family other than the married
parents is associated with having few behavioral problems. Adolescent
girls of low-income married parents are more likely than adolescent boys to
have few behavioral problems.
Summary of Results for Low-Income Adolescents
Work levels among single parents of low-income adolescents are higher than
the levels among single parents of low-income, school-age children. More
than half of the low-income adolescents with single parents have a parent
who works full-time. Among low-income adolescents of married parents,
however, levels of parent work are similar to the levels among school-age
children. Neither levels nor schedules of parent work have strong associations
with positive outcomes among low-income adolescents. The one exception
is a positive link between the share of parent hours of work worked by the
mother and school engagement among low-income adolescents of married
parents. Interestingly, this is the inverse of the relationship found among
school-age children. The positive association could suggest that low-income
adolescents view their mothers' working positively and model that behavior
by engaging in school. Alternatively, it might suggest that mothers who have
adolescents who are engaged students feel more able to engage in market
work than mothers of unengaged students.
As was the case for the younger children, family and child characteristics
have stronger relationships with well-being among low-income adolescents
than do parent work measures. The relationships often differ by marital status.
For example, low parent education has more negative relationships with
well-being for low-income adolescents of married parents, while parent
aggravation is more strongly associated with negative outcomes for low-income
adolescents of single parents.
Summary and Conclusion
In general, among low-income children, family and child characteristics are
more strongly associated with positive child well-being than are patterns of
parent work. Table 4 summarizes the results. Minus signs (-) indicate a negative
association between the child outcome listed in the column heading and
the characteristics listed in the row heading. Plus signs (+) indicate a positive
relationship. The number of minus or plus signs indicates the statistical
strength of the relationship.
|
| Table 4. Summary of Findings |
|
| Low-Income Children of Married Parents |
|
|
|
|
|
|
|
|
| |
Children age 0-5 |
Children age 6-11 |
Children age 12-17 |
| |
|
|
|
| |
Reading |
Outings |
Behavior |
School |
Activities |
Behavior |
School |
Activities |
| Parent Work Variable |
|
|
|
|
|
|
|
|
| Full-time/full-time |
|
|
|
|
|
|
|
|
| Full-time/part-time |
|
|
|
|
+ |
|
|
|
| Part time/part-time/no work |
|
|
|
|
|
|
|
|
| A parent works odd hours |
|
|
|
- - - |
|
|
|
| Percent hours worked by mother |
|
|
|
- - |
++ |
|
++ |
|
| Arranged hours for child care |
|
|
|
+ |
|
NA |
NA |
NA |
| |
|
|
|
|
|
|
|
|
| Environment |
|
|
|
|
|
|
|
|
| Center care |
|
|
NA |
NA |
NA |
NA |
NA |
NA |
| Nanny care |
- |
|
NA |
NA |
NA |
NA |
NA |
NA |
| Relative care |
|
|
NA |
NA |
NA |
NA |
NA |
NA |
| Family day care |
|
|
NA |
NA |
NA |
NA |
NA |
NA |
| MKA and child argue a lot |
NA |
NA |
NA |
NA |
NA |
- - |
|
|
| Mother in poor health |
+ |
- - |
- - |
+ |
|
|
|
|
| Father in poor health |
|
|
|
|
|
|
|
|
| TANF receipt |
|
|
|
|
|
|
|
|
| Food stamp receipt |
|
|
|
- |
|
|
|
|
| A parent did not finish high school |
- - - |
- - |
|
|
- - - |
|
- - |
- - - |
| MKA is aggravated |
|
|
- - - |
|
|
- - - |
|
|
| MKA experiences poor mental health |
|
|
- - |
- - |
|
- |
|
|
| Ln (social family income) |
|
|
|
|
|
|
|
|
| Another adult lives in social family |
|
++ |
|
|
|
+ + |
|
|
| Number of other children |
- - - |
- |
|
|
- - - |
|
|
|
| |
|
|
|
|
|
|
|
|
| Child Characteristics |
|
|
|
|
|
|
|
|
| Disabled |
|
|
- - - |
- - - |
|
- - - |
- - - |
|
| Black |
|
|
|
|
|
|
|
|
| Hispanic |
- - - |
- - |
|
|
|
|
|
|
| Female |
|
|
|
+++ |
|
+ |
+++ |
|
| Age of child |
+ |
|
|
|
+++ |
|
|
|
| |
|
|
|
|
|
|
|
|
| Low-Income Children of Single Parents |
|
|
|
|
|
|
|
|
| Full-time |
- |
-- |
|
|
|
|
|
|
| Parent works odd hours |
|
|
|
|
|
|
|
|
| |
|
|
|
|
|
|
|
|
| Environment |
|
|
|
|
|
|
|
|
| Center care |
|
++ |
NA |
NA |
NA |
NA |
NA |
NA |
| Nanny care |
|
|
NA |
NA |
NA |
NA |
NA |
NA |
| Relative care |
|
+++ |
NA |
NA |
NA |
NA |
NA |
NA |
| Family day care |
|
|
NA |
NA |
NA |
NA |
NA |
NA |
| MKA and child argue a lot |
NA |
NA |
NA |
NA |
NA |
- - - |
|
|
| Parent in poor health |
- |
|
|
|
|
|
|
|
| TANF receipt |
+ |
|
|
|
|
|
|
|
| Food stamp receipt |
- - |
|
|
|
|
|
|
|
| A parent did not finish high school |
|
|
|
|
- - - |
|
|
- - - |
| MKA is aggravated |
|
|
- - - |
- - - |
|
- - |
- - - |
- - - |
| MKA experiences poor mental health |
|
|
- - - |
|
|
- - - |
|
|
| Ln (social family income) |
|
|
|
|
|
- |
- |
|
| Another adult lives in social family |
+ |
|
++ |
|
|
|
|
|
| Number of other children |
- - - |
|
|
|
|
|
|
|
| |
|
|
|
|
|
|
|
|
| Child Characteristics |
|
|
|
|
|
|
|
|
| Disabled |
- - |
|
- - |
- - - |
|
- - - |
|
|
| Black |
|
|
+++ |
|
|
+ |
|
|
| Hispanic |
- - |
|
|
|
|
|
- - |
- - |
| Female |
|
|
|
+++ |
|
|
++ |
|
| Age of child |
|
|
- - - |
|
+++ |
|
|
- - - |
|
Source: Urban Institute calculations from the 1999 National Survey of America's Families.
Notes: Minus(-)/plus(+) signs designate a statistically significant relationship in the multivariate analysis. The number of - / + signs indicates the statistical strength of the relationship as follows: - / + significant at the 10% level; - - / ++ significant at the 5% level; - - - / +++ significant at the 1% level.
Ln = natural log.
MKA = most knowledgeable adult.
NA = not applicable
TANF = Temporary Assistance for Needy Families. |
Table 4 highlights the differences in the patterns of association across age
groups and marital status. Many of these differences were discussed in the
age-group summary sections; this section highlights a few of the interesting
comparisons. For example, full-time work is negatively associated with parent
involvement for low-income preschoolers of single parents, but neither
parent work levels nor schedules are related to parent involvement among
low-income preschoolers of married parents. In contrast, both levels and
schedules are linked with some child outcomes for low-income, school-age
children of married parents, while parent work is not strongly associated with
outcomes for low-income, school-age children of single parents.
The association between low levels of parent education and child well-being
is typically stronger for low-income children of married parents than
for low-income children of single parents. On the other hand, parent aggravation
is more strongly associated with outcomes for low-income children of
single parents, particularly adolescents. Perhaps an additional parent can
serve as a buffer between the aggravated parent and the child.
This paper set out to examine the contemporaneous relationships
between parent work and child well-being. Although the paper finds a lot of
parent work in low-income families, it does not uncover many associations
between levels of parent work and child well-being. In all instances, family
environment and child characteristics have stronger links to child well-being
than does parent work. For policymakers looking to encourage work among
low-income families, the message from these results is fairly positive. However,
for policymakers concerned about improving child well-being among
low-income working families, the news is less good. Parent and child characteristics
are strongly related to child well-being, and many of these factors are
not easily influenced by policy. Both research and work/family programs
should begin to focus some of their resources on understanding and meeting
the special needs of less educated and more unstable working parents.
The finding—or non-finding—of no clear relationship between parent
work and child well-being is not too surprising.24 Parent characteristics, such as levels of education and mental health, influence the well-being of children. These environmental factors may have a more immediate impact on children
than levels of parent work, the effects of which may only be measured after
the passage of time. Furthermore, job characteristics, such as access to family
leave and flexible work schedules, may have more of an effect on parenting
and child well-being than levels of work. Future research should examine
job type and measures of job quality as possible intervening variables in the
relationships between parent work and child well-being. Alternatively, mediating
parent characteristics may be more stable than levels of parent work
and, as a result, the influence of these characteristics on child well-being may
have had a longer time to develop. Cross-sectional data do not allow us to
measure the historical timing of changes in either parent work or the family
environment. These life histories, however, may be crucial to understanding
how current patterns of market work relate to current levels of child well-being.
Future research should use longitudinal data to help uncover pathways
toward positive child outcomes and to examine how patterns of parent work
might influence these pathways.
Appendix Tables
(see pdf version)
Notes
1. Additional information about the NSAF is available on its web site (http://newfederalism.urban.org).
2. "Family" is defined more broadly in this analysis than in analyses that use Census Bureau data. In addition to persons who are related to the sampled child by blood, adoption, or marriage, the child's family includes (1) the most knowledgeable adult (MKA) about the child in the household (usually but not always a parent); (2) any relative of the MKA if the MKA is not related to the child; (3) any unmarried partners of any of the child's relatives and of the MKA; and (4) anyone related by blood, marriage, or adoption to these partners.
3. Fewer than 6 percent of all children live with cohabiting parents (Acs and Nelson 2001), and it is unclear whether these families function more like married-parent or single-parent families. In fact, teenagers who live with cohabiting parents often fare worse than teenagers who live with either a single parent or with both biological parents (Nelson, Clark, and Acs 2001). Including children who live with cohabiting parents in either sample could lead to significant selection bias.
4. It is very plausible that the outcomes for children of ever-married single parents will differ from those of never-married single parents. Including an indicator for ever-married in the multivariate analysis does not change the results, and the coefficient on the variable is never significant. (Sensitivity analysis results are available from the author.)
5. Using stricter definitions of part-time work (e.g., the parent works between 8 and 35 hours per week) does not change the results.
6. For detailed information about the child outcome measures, see Ehrle and Moore (1999). The NSAF uses abbreviated scales and indices from exiting surveys to measure child well-being during a telephone interview. Using telephone-based surveys to measure child well-being can help researchers monitor children on a large, national scale. The NSAF, however, is one of the first surveys to use this method for measuring child well-being, so the measures may evolve over time. The results in this paper should be viewed as a preliminary examination of the associations between parent work and child well-being as measured in a telephone survey.
7. The NSAF scale is tailored for two age groups (age 6 to 11 and age 12 to 17) to accommodate developmental differences. To be identified as having few behavioral problems, a child's scale value would have to equal the maximum score of 18, indicating that the respondent answered "never true" to all six scale items.
8. Four questions are asked of the MKA about the child to assess the child's interest in and willingness to do school work. A score equal to or greater than 15 out of a possible 16 points suggests high engagement in school. To receive a score of 15, the MKA had to respond that the child is in the most engaged category on at least three of the four items.
9. An example from the data helps clarify this point. Among school-age, low-income children with single parents, 25 percent have few behavioral problems and 11 percent have many behavioral problems; 37 percent are highly engaged in school and 23 percent are not engaged in school; and 66 percent are involved in extracurricular activities and 34 percent are not.
10. When the same models are run to predict negative outcomes, the results remain qualitatively the same. Results from these regressions are available from the author upon request.
11. Given the cross-sectional nature of the data and the complexity of the relationships between parent work and child outcomes, this analysis examines multivariate associations rather than developing a behavioral model.
12. In an alternative set of specifications, parent work effort was included as a continuous variable equal to the sum of the usual hours of work for the child's parent(s). The pattern of results from this set of specifications was essentially the same as the pattern arising from the specifications that contain the categorical parent work variables.
13. With the exception of the categorical parent work variables, I present predicted probabilities for different levels of included independent variables only if the coefficient for the variable is statistically significant. However, I include all the variables and their coefficients when calculating the predicted probabilities. The coefficients, standard errors, and sample means are in appendix tables A2, A3, and A4, respectively.
14. The formula for transforming the sum of the products into a probability is as follows: 
equals the sum of the products of the coefficients and the variable values.
15. The distributions of parent work levels and child outcomes come from Urban Institute calculations of the 1999 NSAF. Tables of the results are available upon request from the author.
16. See the discussion in the Methodology section for more details about the regression models. The regression results are run only on the sample of children who have at least one parent who works at least part-time—children of nonworking parents are excluded from the multivariate sample. The full set of regression results is in appendix table A2.
17. Researchers often display standardized changes in predicted probabilities. To standardize the change, the raw difference is divided by the standard deviation of the sample mean of the outcome, for example, frequent reading. I did not make this adjustment in this paper because the standard deviations for all the outcomes are quite similar. Thus, the standardized results are not very different from the undstandardized results, and they are more complicated to discuss.
18. The distributions of parent work levels and child outcomes come from Urban Institute calculations of the 1999 NSAF. Tables of the results are available upon request from the author.
19. The regression results are run only on the sample of children who have at least one parent who works at least part-time—children of nonworking parents are excluded from the multivariate sample. The full set of regression results is in appendix table A3.
20. The size of the effect of child disability on the likelihood of having few behavioral problems—a 75 percent reduction in the probability of a positive outcome—is unusually large. The behavioral problems scale may not be appropriate for children with disabilities.
21. The distributions of parent work levels and child outcomes come from Urban Institute calculations of the 1999 NSAF. Tables of the results are available upon request from the author.
22. The regression results are only run on the sample of children who have at least one parent who works at least part-time—children of nonworking parents are excluded from the multivariate sample. The full set of regression results is in appendix table A4.
23. While the income relationship is counterintuitive, it is quite robust. The restriction of the sample to only low-income children of working parents may be partially driving the result.
24. It is important to reiterate that the NSAF collects data through telephone interviews. It is one of the first surveys to measure child well-being over the telephone as part of a larger data collection effort. Furthermore, the NSAF uses modified versions of scales and indices that are used in other surveys. An early analysis of the measures concludes that they work fairly well (Ehrle and Moore 1999). However, researchers will continue to assess the validity and reliability of the measures.
References
Acs, Gregory, and Sandi Nelson. 2001. "'Honey, I'm Home.' Changes in Living Arrangements in the Late 1990s." Washington, D.C.: The Urban Institute. Assessing the New Federalism Policy Brief B-38.
Belsky, Jay. 1990. "Parental and Nonparental Care and Children's Socioemotional Development: A Decade in Review." Journal of Marriage and the Family 52: 885–903.
Belsky, Jay, and David Eggebeen. 1991. "Early and Extensive Maternal Employment and Young Children's Socioemotional Development: Children of the National Longitudinal Survey of Youth." Journal of Marriage and the Family 53: 1083–1110.
Ehrle, Jennifer, and Kristen Anderson Moore. 1999. 1997 NSAF Benchmarking Measures of Child and Family Well-Being. Washington, D.C.: The Urban Institute. Assessing the New Federalism NSAF Methodology Report No. 6.
Hamilton, Gayle, Stephen Freedman, and Sharon M. McGroder. 2000. Do Mandatory Welfare-to-Work Programs Affect the Well-Being of Children? A Synthesis of Child Research Conducted as Part of the National Evaluation of Welfare-to-Work Strategies. Washington, D.C.: U.S. Department of Health and Human Services, Administration for Children and Families and Office of the Assistant Secretary for Planning and Evaluation; and U.S. Department of Education, Office of the Under Secretary and Office of Vocational and Adult Education.
Han, Wen-Jui, Jane Waldfogel, and Jeanne Brooks-Gunn. 2001. "The Effects of Early Maternal Employment on Later Cognitive and Behavioral Outcomes." Journal of Marriage and the Family 63: 336–54.
Harvey, Elizabeth. 1999. "Short-Term and Long-Term Effects of Early Parental Employment on Children of the National Longitudinal Survey of Youth." Developmental Psychology 35(2): 445–59.
Hoffman, Lois W. 1989. "Effects of Maternal Employment in the Two-Parent Family." American Psychologist 44: 283–92.
Kalil, Ariel, Rachel E. Dunifon, and Sandra K. Danziger. 2001. "Does Maternal Employment Mandated by Welfare-Reform Affect Parenting Behavior?" Paper presented at the 23rd annual research conference of the Association for Public Policy Analysis and Management, Washington, D.C., November 1–3.
Morris, Pamela A., Aletha C. Huston, Greg J. Duncan, Danielle A. Crosby, and Johannes M. Bos. 2001. How Welfare and Work Policies Affect Children: A Synthesis of Research. New York: Manpower Demonstration Research Corporation.
Nelson, Sandi, Rebecca L. Clark, and Gregory Acs. 2001. "Beyond the Two-Parent Family: How Teenagers Fare in Cohabiting Couple and Blended Families." Washington, D.C.: The Urban Institute. Assessing the New Federalism Policy Brief B-31.
Parcel, Toby L., and Elizabeth G. Menaghan. 1994. "Early Parental Work, Family Social Capital, and Early Childhood Outcomes." American Journal of Sociology 99(4): 972–1009.
Perry-Jenkins, Maureen, Rena L. Repetti, and Ann C. Crouter. 2000. "Work and Family in the 1990s." Journal of Marriage and the Family 62(4): 981–98.
Sherman, Arloc. 2001. How Children Fare in Welfare Experiments Appears to Hinge on Income. Washington, D.C.: Children's Defense Fund.
Vandell, Deborah, and Janaki Ramanan. 1992. "Effects of Early and Recent Maternal Employment on Children from Low-Income Families." Child Development 63: 938–49.
Wells, Gordon. 1985. "Preschool Literacy-Related Activities and Success in School." In Literacy, Language and Learning, edited by David Olson, Nancy Torrance, and Angela Hildyard (229–55). Cambridge, England: Cambridge University Press.
About the Author
Katherin Ross Phillips is a research associate in the Urban Institute's Income and
Benefits Policy Center. Her research focuses on low-income workers with children.
She is currently studying the relationships between parental work and child well-being
and the effects of family policy on parental work effort.
Latest Research
Browse by Topic:
PDF
Printer-friendly summary
