Number B-23 in Series, "New Federalism: National Survey of America's Families"
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.
The preamble of the 1996 welfare reform law details numerous negative consequences associated with out-of-wedlock births, single-parent families, and welfare receipt that helped motivate the 1996 overhaul. With this law, Congress ended the existing Aid to Families with Dependent Children (AFDC) program and replaced it with Temporary Assistance for Needy Families (TANF). TANF is intended to (1) provide assistance for needy families to care for their children; (2) end welfare dependence by promoting job preparation, work, and marriage; (3) reduce the rate of nonmarital births; and (4) increase the proportion of two-parent families.
Many of the family characteristics that welfare reform is intended to change, such as poverty, parental employment, nonmarital birth, marriage, and welfare dependence, are also likely to affect the well-being of children living in low-income families. For that reason it is important to track changes in child well-being as welfare reform progresses. Here we examine state-level data on social, economic, and child well-being measures at the outset of welfare reform. The 13 states highlighted here are those being studied in depth as a part of the Assessing the New Federalism project.
States' Social and Economic Challenges at the Onset of Welfare Reform
The National Survey of America's Families (NSAF) reveals that social and economic conditions relevant to welfare reform are quite diverse among the 13 states being studied.
Income. Nationwide, about one out of every five children in 1996 lived in families with incomes below the poverty threshold (table 1), and about two out of every five children lived in low-income families—those that have incomes below twice the poverty threshold. The proportion of low-income children within the 13 states ranged from 29.4 to 57.9 percent.
TABLE 1. Percentage of Children under 18 Living in Families with Various Characteristics,
by State, 1997 |
| |
AL |
CA |
CO |
FL |
MA |
MI |
MN |
MS |
NJ |
NY |
TX |
WA |
WI |
US |
| All income levels |
| Below poverty, 1996 |
27.3 |
28.8 |
14.7 |
22.1 |
16.0 |
13.9 |
11.8 |
34.0 |
13.4 |
24.5 |
25.5 |
15.4 |
11.6 |
20.6 |
| Under 200% poverty, 1996 |
48.4 |
50.5 |
34.7 |
48.6 |
30.7 |
34.2 |
29.5 |
57.9 |
29.4 |
43.8 |
49.9 |
35.9 |
32.4 |
42.8 |
| Lived with one parent |
36.9 |
31.8 |
24.8 |
37.4 |
27.6 |
28.1 |
21.0 |
43.7 |
25.7 |
34.7 |
31.6 |
23.6 |
26.0 |
30.1 |
| Received AFDC in 1996 or 1997 |
7.3 |
17.2 |
6.2 |
12.8 |
10.5 |
11.0 |
8.4 |
14.5 |
7.2 |
9.7 |
11.4 |
9.0 |
6.9 |
11.0 |
| Unemployed parent(s) |
13.7 |
13.3 |
6.6 |
11.2 |
11.2 |
7.1 |
5.5 |
16.8 |
9.6 |
15.7 |
11.7 |
8.0 |
5.5 |
10.4 |
| Under 200% poverty |
| Lived with one parent |
60.6 |
44.1 |
43.1 |
56.5 |
58.3 |
55.9 |
45.1 |
62.8 |
53.5 |
55.8 |
46.9 |
40.3 |
51.1 |
48.7 |
| Received AFDC in 1996 or 1997 |
14.5 |
31.4 |
15.6 |
23.3 |
32.4 |
27.5 |
26.4 |
24.5 |
21.7 |
21.0 |
21.6 |
20.9 |
18.7 |
23.6 |
| Unemployed parent(s) |
27.1 |
24.6 |
16.3 |
20.3 |
33.3 |
18.9 |
18.1 |
27 |
28.0 |
33.2 |
21.8 |
19.2 |
15.2 |
22.4 |
| Note: Figures in bold represent statistically significant differences from the national average at the 0.05 confidence level. |
Family Structure. Almost a third of all children in the United States lived with only one parent. Among the 13 NSAF states, percentages of children living with only one parent ranged from 21.0 to 43.7 percent. Looking specifically at low- income children (second panel of table 1), almost half lived with a single parent in 1997. Among the 13 states, from 40.3 to 62.8 percent of low-income children lived with just one parent. As a group, low-income children were three to four times more likely to live in single-parent homes than children with family incomes at or above two times the poverty threshold (not shown in table 1).
Welfare Receipt. The NSAF also asked about whether families were currently receiving AFDC at the time of the survey in 1997 and whether they had received AFDC in 1996. Eleven percent of all children lived in families that received AFDC in 1996 or 1997. These percentages ranged from 6.2 to 17.2 percent in the 13 NSAF states. Almost one-quarter of low-income children received AFDC in 1996 or 1997.
Employment. The percentage of children who lived with unemployed parents ranged from 5.5 to 16.8 in the 13 states. Among low-income children, just over a fifth (22.4 percent) had unemployed parents. In contrast, 1.5 percent of higher-income children lived with unemployed parents (not shown in table 1). Among low-income children in the 13 NSAF states, from 15.2 to 33.3 percent had unemployed parents.
Looking collectively at the challenges facing the states in terms of welfare reform in 1997, some states appeared better off than others. For example, children in Colorado, Minnesota, Washington, and Wisconsin were significantly less likely than children in the nation as a whole to live in poor families, in families receiving AFDC in 1996 or 1997, or in families with single or unemployed parents. In contrast, children in Alabama, California, Florida, Mississippi, New York, and Texas were either more likely than or equally likely as other American children to live in these circumstances (except that Alabama children's families were less likely to receive AFDC).
Low-income children disproportionately faced social and economic conditions targeted by TANF. Almost half of all low-income children lived in single-parent families, almost a quarter received welfare in 1996 or 1997, and more than a fifth lived with unemployed parents. Five of the 13 states - Alabama, California, Florida, Mississippi, and Texas - also had a comparatively high proportion of low-income children. Thus, the stated TANF target populations of low-income children carried different levels of disadvantage at the outset of welfare reform.
Children's Well-Being—Variation by State
State differences in the well-being of children are not as large as those for poverty, welfare dependence, single parenthood, and employment. Nevertheless, data from the NSAF indicate that outcomes for children varied across states and also differed significantly from the national average in one or two of the indicators of child well-being in each state, as shown in table 2.
| TABLE 2. Percentage of Children with Various Outcomes, by State, 1997 |
| |
AL |
CA |
CO |
FL |
MA |
MI |
MN |
MS |
NJ |
NY |
TX |
WA |
WI |
US |
| Behavioral and emotional problems |
| Ages 6-11 |
7.6 |
5.6 |
5.9 |
7.9 |
8.8 |
7.1 |
6.9 |
9.0 |
6.3 |
8.5 |
9.0 |
5.6 |
7.2 |
6.6 |
| Ages 12-17 |
9.5 |
8.0 |
7.1 |
8.8 |
7.3 |
7.6 |
9.1 |
11.9 |
6.2 |
6.7 |
8.8 |
6.7 |
8.3 |
8.8 |
| Fair/poor health, ages 0-17 |
5.4 |
7.3 |
4.3 |
4.6 |
2.8 |
3.5 |
2.9 |
6.0 |
3.8 |
4.5 |
7.6 |
3.7 |
3.0 |
4.6 |
| Low school engagement, ages 6-17 |
24.2 |
19.8 |
20.2 |
23.7 |
18.1 |
20.3 |
21.4 |
28.1 |
17.0 |
19.6 |
19.0 |
17.5 |
18.8 |
20.4 |
| Skipped school 2+ times, ages 12-17 |
8.3 |
16.3 |
16.8 |
10.9 |
8.4 |
11.3 |
8.8 |
8.4 |
8.0 |
11.0 |
8.3 |
10.1 |
10.1 |
10.5 |
| Suspended or expelled, ages 12-17 |
19.8 |
11.8 |
14.1 |
13.9 |
10.8 |
14.7 |
11.2 |
21.4 |
12.8 |
10.1 |
12.3 |
11.7 |
11.4 |
13.9 |
| Note: Figures in bold represent statistically significant differences from the national average at the 0.05 confidence level. |
Behavioral and Emotional Problems. The NSAF asked parents a series of questions about their children's behavioral and emotional well-being.1 In 1997, 6.6 percent of children ages 6 to 11 and 8.8 percent of children ages 12 to 17 exhibited high levels of behavioral and emotional problems. None of the 13 states differed significantly from the national average.
School. The NSAF also asked caregivers about their children's interest in and willingness to do schoolwork.2 A fifth of American children ages 6 to 17 had low levels of school engagement. Among the 13 states, the proportion of poorly engaged children ranged from 17.0 to 28.1 percent. Two additional school- related survey items queried parents about whether children had skipped school or been suspended or expelled in the 12 months prior to the survey. Nationwide, 10.5 percent of children ages 12 to 17 had skipped school two or more times, and 13.9 percent had been suspended or expelled. The percentage of children in the 13 states who had skipped school ranged from 8.0 to 16.8. Throughout the 13 states, between 10.1 and 21.4 percent of children had been suspended or expelled.
Health. Among all U.S. children under age 18, 4.6 percent were reported by their parents to be in fair or poor health, ranging from 2.8 to 7.6 percent among the 13 NSAF states.
As we have seen, states have varying percentages of children living in poverty, receiving welfare, and living with single or unemployed parents. If these conditions exacerbate problems for children, then we would expect negative outcomes to be more common in states that face a disproportionately large share of welfare challenges. However, a comparison of tables 1 and 2 suggests that associations are not exceptionally strong at the state level. On the other hand, when families are the focus of the analysis, associations between socioeconomic conditions and child outcomes are stronger.
Relationship between Family Socioeconomic Conditions and Children's Well-Being
In this section, we compare the outcomes of children throughout the United States whose families faced the challenges of poverty, single parenthood, unemployment, or AFDC receipt with those of children whose families did not face these challenges to the same extent.
Poverty and Child Outcomes. While the causal effect of income is the focus of considerable debate in the research community, studies show consistent correlations between poverty and poor child outcomes (Brooks-Gunn and Duncan 1997; Mayer 1997; McLoyd 1998; National Research Council 1993). NSAF data indicate that poor children are more likely to experience behavioral and emotional problems, fair or poor health, and school problems than are nonpoor children in the United States (table 3).
TABLE 3. Percentage of Children Living in Various Family Circumstances
Who Have Negative Outcomes, U.S., 1997 |
| |
Behavioral and Emotional Problems |
Fair or Poor Health |
Low School Engagement |
Skipped School |
Suspended or Expelled |
| |
Ages 6-11 |
Ages 12-17 |
Ages 0-17 |
Ages 6-17 |
Ages 12-17 |
Ages 12-17 |
| Below poverty |
11.4 |
18.5 |
10.1 |
28.5 |
18.1 |
24.6 |
| Above poverty |
5.4 |
6.7 |
3.1 |
18.6 |
8.9 |
11.6 |
| AFDC in 1996 or 1997 |
14.2 |
22.8 |
8.6 |
33.7 |
27.0 |
34.0 |
| No AFDC in 1996 or 1997 |
5.7 |
7.5 |
4.1 |
19.1 |
9.0 |
12.1 |
| Unemployed parent(s) |
14.2 |
20.6 |
10.0 |
30.9 |
18.1 |
32.6 |
| Employed parent(s) |
5.8 |
7.4 |
4.0 |
19.3 |
9.6 |
11.8 |
| One Parent |
9.6 |
16.0 |
7.2 |
28.5 |
18.9 |
23.3 |
| Two parents |
5.2 |
5.1 |
3.4 |
16.7 |
6.2 |
9.1 |
| Note: The differences between population groupings are statistically significant for all of the child outcomes in the table at the 0.05 confidence level. |
Welfare Dependency and Child Outcomes. The U.S. Department of Health and Human Services has identified a wide range of risk factors that are associated with welfare dependence, although research has not proven whether these factors are causes, consequences, or simply correlates of welfare receipt (Department of Health and Human Services 1998). The evidence shown in table 3 indicates that children whose families received AFDC in 1996 or at the time of the survey in 1997 were disproportionately more likely to experience behavioral and emotional problems, fair or poor health, and school problems.
Parents' Employment and Child Outcomes. Employment is important not only as a means for supporting one's family economically but also as a framework for daily behavior (Wilson 1997). Data from the NSAF provide evidence that children with unemployed parents are more likely to experience all of the negative outcomes in table 3.
Parent Absence and Child Outcomes. Parent absence is related to negative outcomes in areas such as test scores, educational attainment, and behavioral and psychological problems among children even after accounting for socioeconomic differences (McLanahan 1997). NSAF data indicate that, as expected, children who live with a single parent are more likely to experience behavioral and emotional problems, fair or poor health, and school problems than are children who live with two parents (table 3).
In sum, analyses of NSAF data show that for each of the five child outcomes measured (table 3), American children living in families with the characteristics deemed harmful by the architects of welfare reform did indeed fare significantly worse than other children.
Do State Socioeconomic Conditions Explain State Variation in the Well-Being of Children?
How much less would child well-being vary across states if socioeconomic conditions were more similar? To answer this question, we have created adjusted estimates of child well-being. These are standardized estimates created by ascribing to each state levels of family poverty, welfare receipt, and single parenthood equal to those for the nation as a whole.3
Looking at these adjusted percentages (table 4), the proportions of children experiencing negative outcomes in the 13 states move closer to the national averages in most cases. After controlling for poverty, single parenthood, and welfare receipt, children in Florida, Michigan, Minnesota, New Jersey, and Washington no longer exhibit outcomes that differ significantly from the national averages.4
TABLE 4. Adjusted Percentages of Children with Various Outcomes, Controlling for
the Percentage of Children Living in Poverty, Living with Only One Parent,
and Receiving AFDC in 1996 or 1997, by State, 1997 |
| |
AL |
CA |
CO |
FL |
MA |
MI |
MN |
MS |
NJ |
NY |
TX |
WA |
WI |
US |
| Behavior and mental health problems |
| Ages 6-11 |
8.2 |
5.3 |
6.5 |
7.3 |
9.5 |
6.8 |
8.1 |
7.2a |
6.9 |
8.3 |
9.3 |
5.6 |
8.4 |
6.6 |
| Ages 12-17 |
9.9a |
7.8 |
7.9 |
8.2 |
7.9 |
9.1 |
10.9 |
11.1 |
6.5 |
7.7 |
9.4 |
8.2 |
9.9 |
8.8 |
| Fair or poor health, ages 0-17 |
5.2 |
6.8 |
5.3 |
4.4 |
3.4 |
4.2 |
3.6 |
5.7 |
4.5 |
4.5 |
7.3 |
4.2 |
3.6 |
4.6 |
| Low school engagement, ages 6-17 |
24.9 |
19.7 |
21.1 |
23.2 |
19.2 |
20.9 |
23.4 |
26.3 |
18.6 |
19.5 |
19.4 |
18.6 |
20.9 |
20.4 |
| Skipped school, ages 12-17 |
* |
16.0 |
17.5 |
10.2 |
9.8 |
11.8 |
10.7 |
7.9 |
8.6 |
11 |
8.2 |
11.8 |
11.5 |
10.5 |
| Suspended or expelled, ages 12-17 |
* |
11.5 |
15.7 |
13.1 |
12.2 |
16.8 |
13.2 |
19.3 |
12.8 |
9.6 |
12.3 |
14.1 |
14.5 |
13.9 |
Notes: Figures in bold represent statistically significant differences from the national average at the 0.05 confidence level.
a. Statistical significance of the difference from the national average could not be calculated due to small sample sizes.
*Predicted percentage could not be calculated due to small sample sizes. |
However, some notable exceptions remain. For example, even if Massachusetts had the same rates of poverty, welfare receipt, and single parenthood as the United States, behavioral and emotional problems among children ages 6 to 11 would be somewhat more common in Massachusetts than in the rest of the nation. California, Massachusetts, Texas, and Wisconsin would continue to have higher proportions of children with fair or poor health than the rest of the nation. Low school engagement would be more likely in Alabama and Mississippi than in the rest of the nation. California and Colorado children would be disproportionately more likely to have skipped school, but Mississippi and Texas children would be disproportionately less likely to have skipped school, relative to the rest of the nation. And Mississippi children would be more likely, but New York children less likely, than other American children to have been suspended or expelled. Clearly, other factors at the state, community, and family levels also affect children's well-being.
Conclusion
Analyses of the 1997 NSAF data show that states and families face a diversity of socioeconomic conditions. Children's well-being is strongly associated with the socioeconomic status of families. Their well-being varies across the states as well, though to a lesser extent. By examining these topics together, this brief provides a holistic sense of the status of children at the family and state levels.
As states approached welfare reform, they faced populations with different needs, and this may be reflected in the design of their welfare programs and the degree of success they enjoy from those programs. Those states suffering from fewer disadvantages may find it easier to succeed in helping those who are in need, for example.
Some of the differences in levels of child well-being across states can be attributed to differences in TANF-relevant family characteristics such as poverty, single parenthood, and welfare receipt. Differences in child well-being are reduced, but do not go away, when these factors are controlled. This evidence suggests that if welfare policy helps reduce experiences of poverty, single parenthood, unemployment, and receipt of cash assistance, child outcomes across states might improve. However, controlling for differences among states on these socioeconomic factors does not fully explain differences in child outcomes—variation in child well-being remains, and some states are likely to continue to have significantly better or worse child outcomes regardless of improvements in socioeconomic conditions until other factors also are addressed.
Conclusions should be drawn with caution, because causal relationships cannot be inferred from cross-sectional survey data. Also, welfare reform may affect factors that are important for child well-being that are not measured by the NSAF. It may also be that children's experiences of the social and economic factors differ across states so that the relationship of poverty, single parenthood, welfare receipt, and parental employment to child outcomes also may differ. Thus, as social policy evolves, policymakers should continue to track both socioeconomic conditions and child well-being at the state level in order to learn more about their interaction and adjust policies appropriately.
Endnotes
The behavioral and emotional problems scale was based on questions from a mental health indicator included in the National Health Interview Survey. For more information, see Ehrle and Moore 1999.
Lisa Bridges and Jim Connell of the Institute for Research and Reform in Education in California created the school engagement index. For more information, see Ehrle and Moore 1999.
To create these "standardized rates" (Fleiss, J.L. 1981. Statistical Methods for Rates and Proportions. New York: John Wiley and Sons, Inc.), the percentages of children experiencing various negative outcomes in each state "are adjusted using control totals so that the effects of population composition are eliminated when making comparisons between groups" (Westat, Inc. 1998. WesVar Handbook. Rockville, Md.: Westat).
Because the survey we are using for analysis has a complex sampling design, we use a statistical software package, WesVar, to calculate standard errors and t-tests for significant differences. WesVar takes advantage of the replicate weights included in the survey file to take account of the design features of the survey (Westat, Inc. 1998. WesVar Handbook. Rockville, Md.: Westat). However, WesVar is unable to calculate the statistical significance level of differences between adjusted percentages, although it does produce standard errors. Thus, we used the standard errors produced by WesVar to manually calculate t-tests of significant differences between the state adjusted percentages and the national (i.e., balance of U.S.) adjusted percentages. These manual calculations slightly underestimate the value of t; therefore, some significant differences in our analysis might be deemed insignificant if we could take full advantage of the design features of the survey. It is important to note, however, that this condition would apply only to estimates in table 4 that are marginally significant.
References
Brooks-Gunn, J., and G. Duncan. 1997. "The Effects of Poverty on Children." The Future of Children 7 (2): 55-71. Ehrle, Jennifer L., and Kristin A. Moore. 1999. 1997 NSAF Benchmarking Measures of Child and Family Well-Being.Washington, D.C.: The Urban Institute. National Survey of America's Families Methodology Report No. 6. (Accessed February 15, 2000.)
Mayer, S.E. 1997. What Money Can't Buy: Family Income and Children's Life Chances. Cambridge, Mass.: Harvard University Press. McLanahan, S.S. 1997. "Parent Absence or Poverty: Which Matters More?" In Consequences of Growing Up Poor, edited by G. Duncan and J. Brooks-Gunn (35-48). New York: Russell Sage Press.
McLoyd, V. C. 1998. "Socioeconomic Disadvantage and Child Development." American Psychologist 53 (2): 185-204.
National Research Council. 1993. Understanding Child Abuse and Neglect. Washington, D.C.: National Academy Press.
U.S. Department of Health and Human Services. 1998. Indicators of Welfare Dependence: Annual Report to Congress. http://aspe.hhs.gov/hsp/indicators98/indicators98.htm. (Accessed February 2,
2000.)
Wilson, W.J. 1997. When Work Disappears: The World of the New Urban Poor. New York: Alfred A. Knopf.
About the Authors
Sharon Vandivere is a research analyst at Child Trends. She has cowritten papers on child and family well-being and coedits the Child Indicator, a newsletter on social indicators. She recently earned a master's degree in public policy from Georgetown University.
Kristin Anderson Moore is president and a senior scholar at Child Trends. Dr. Moore is a social psychologist who studies trends in child and family well-being, positive development, adolescent parenthood, family processes and family structure, and the effects of welfare and poverty on children. She played a key role in developing a number of measures for the National Survey of America's Families.
Brett Brown is a senior research associate and area director for social indicators research at Child Trends. Dr. Brown manages projects related to the development and use of social indicators of child and family well-being at the international, national, and state levels. He is a member of the core working group on adolescent health for the CDC's Healthy People 2010 project.
About the Series
This series is a product of Assessing the New Federalism, a multiyear project to monitor and assess the devolution of social programs from the federal to the state and local levels. Alan Weil is the project director. The project analyzes changes in income support, social services, and health programs. In collaboration with Child Trends, the project studies child and family well-being.
The project has received funding from The Annie E. Casey Foundation, the W.K. Kellogg Foundation, The Robert Wood Johnson Foundation, The Henry J. Kaiser Family Foundation, The Ford Foundation, The David and Lucile Packard Foundation, The John D. and Catherine T. MacArthur Foundation, the Charles Stewart Mott Foundation, The McKnight Foundation, The Commonwealth Fund, the Stuart Foundation, the Weingart Foundation, The Fund for New Jersey, The Lynde and Harry Bradley Foundation, the Joyce Foundation, and The Rockefeller Foundation.
This series is dedicated to the memory of Steven D. Gold, who was codirector of Assessing the New Federalism until his death in August 1996.
Latest Research
Browse by Topic:
PDF
Printer-friendly summary
