Donate
Independent, objective, nonpartisan research

R 703FFR

Authors

R 703FFR

Tagged with:

Publication PDFs

Database

This is the content currently stored in the post and postmeta tables.

View live version

object(Timber\Post)#3711 (44) { ["ImageClass"]=> string(12) "Timber\Image" ["PostClass"]=> string(11) "Timber\Post" ["TermClass"]=> string(11) "Timber\Term" ["object_type"]=> string(4) "post" ["custom"]=> array(5) { ["_wp_attached_file"]=> string(12) "R_703FFR.pdf" ["wpmf_size"]=> string(6) "687927" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(188432) "The Well-Being of California’s Children ••• Frank F. Furstenberg Maureen R. Waller Hongyu Wang 2003 PUBLIC POLICY INSTITUTE OF CALIFORNIA Library of Congress Cataloging-in-Publication Data Furstenberg, Frank F., 1940- The well-being of California’s children / Frank F. Furstenberg, Maureen R. Waller, Hongyu Wang. p. cm. Uses the 1999 National Survey of America’s Families as a source of data collected on 1,917 children living in California households to examine variations in the well-being of children in the state and to compare their overall welfare to the welfare of children in other states. Includes bibliographical references. ISBN: 1-58213-066-3 1. Children—California—Social conditions. 2. Children—California— Economic conditions. 3. Children—Health and hygiene—California. 4. Children—California—Statistics. 5. Children—United States—States. 6. Family life surveys—California. 7. National Survey of America’s Families. I. Waller, Maureen Rosamond, 1967- II. Wang, Hongyu, 1971- III. Title. HQ792.C2F87 2003 305.23'09794—dc21 2003007577 Copyright © 2003 by Public Policy Institute of California All rights reserved San Francisco, CA Short sections of text, not to exceed three paragraphs, may be quoted without written permission provided that full attribution is given to the source and the above copyright notice is included. PPIC does not take or support positions on any ballot measure or state and federal legislation nor does it endorse or support any political parties or candidates for public office. Research publications reflect the views of the authors and do not necessarily reflect the views of the staff, officers, or Board of Directors of the Public Policy Institute of California. Foreword Descriptions of California’s enormous immigrant population are usually presented with a mixture of awe and anxiety. Awe because what California has accomplished in absorbing immigrants over the past two decades is truly remarkable and breathtaking in scale. Anxiety because no one can be really certain that this “experiment” in social change will not result in some insurmountable challenges. Nevertheless, California is moving boldly into the future, driven by a resilient economy and riding on a demographic transformation that is unprecedented in American history. The challenges generated by large-scale immigration are beginning to come into sharper focus. Families without health insurance, everincreasing demands on the public education system, and the need to increase the supply of affordable housing are all public policy issues that have come up for intense public debate. More recently, it has become apparent that an increasing number of children in California have parents born outside the United States. In 1999, almost half of all children under age six had a foreign-born parent. More important, about 20 percent of California’s children live in poverty, compared to about 17 percent in the nation as a whole. Thanks to the availability of a large national survey, Frank Furstenberg, Maureen Waller, and Hongyu Wang have been able to take a careful look at children in California and have provided a remarkable profile in The Well-Being of California’s Children. The authors present findings on the physical health of children, their emotional adjustment, their attachment to school, and the degree to which they are involved in pro-social activities. Overall, children in California appear to be faring slightly worse on a number of indicators of their well-being and are not outperforming children in the rest of the nation on any of the indicators examined. The overall health status of children in California is somewhat lower than that of children living elsewhere. iii Although many of California’s children are faring well, the authors conclude that a substantial minority might not be adequately served by existing services in the state. For the most part, these children come from the poorest families in California, have parents with low levels of educational attainment, or live in Hispanic immigrant households. The authors observe that “there is a conspicuous lack of mental health assistance for children with severe problems. The tremendous cost of treating antisocial behavior and substance abuse among adults suggests that early intervention is justified.” They also note that California would do well to strengthen in-school and after-school activities that help older and less-affluent adolescents stay connected to school. In other words, all taxpayers in California face the challenge of helping their children face the future in the best possible circumstances. And again, this is yet another public policy issue facing state and local governments that are already facing a cycle of substantial cuts in services. Where will the money come from? What about privatization? What is the public role in helping children? What other services will have to be cut to help the children? These are the questions raised by this important set of findings about a population that is central to the future of California’s civic, economic, and social health. We can only hope that by presenting the facts and raising the questions, the political process will render a set of solutions appropriate for both our own and future generations. David W. Lyon President and CEO Public Policy Institute of California iv Summary More than one in eight children in the United States lives in California. In recent years, California’s citizens and lawmakers have given a great deal of attention to addressing the needs of these children, and some programs have directed considerable resources toward enhancing their welfare and development. Given the growth in programs aimed at improving children’s well-being, it has become increasingly important to take stock of how children in the state are faring. However, our ability to assess child-centered policies has been seriously hindered because most national studies that include indicators of children’s well-being have not been large enough to yield reliable information on children at the state level. This report uses a new source of information—the National Survey of America’s Families (NSAF)—that was explicitly designed to measure children’s welfare and well-being and that provides a large enough sample of California families to assess the status of children in the state. In particular, the report draws on information collected for 1,917 children living in California households that participated in the 1999 NSAF to examine variations in the well-being of children in the state and how they are faring in comparison to children living elsewhere. An important objective of this report is to provide a baseline of information about children’s well-being in 1999 against which similar indicators can be reexamined in subsequent years. Moreover, this analysis provides information about the needs of special subgroups of children and suggests to policymakers ways to target the populations that are in particular need of assistance. The public cares whether children are healthy, happy, and secure for the same reasons that they care about the welfare of any other age group. However, because children are also one of the most vulnerable demographic groups in the population and because they are born into unequal circumstances, public concern also focuses on whether children v have the kind of resources that will enhance their welfare and development and, if disparities in these resources exist, how they can be alleviated. The public is further concerned with how children are faring because adult performance is strongly linked to the quality and quantity of material and emotional investment that children receive (Cairns, Elder, and Costello, 1996; Haveman and Wolfe, 1994). When children are adequately nurtured, they are also more likely to become productive and involved members of society as they grow older. In California, about one in five children is living in a poor household. Poverty has been found to have negative effects on children’s cognitive abilities and achievement that may persist in adulthood (Brooks-Gunn and Duncan, 1997; Duncan and Brooks-Gunn, 1997). Children in California are also from diverse racial and ethnic backgrounds. Because nonwhite families in California and the nation often have lower incomes than white families, there is also concern that children in these families disproportionately lack sufficient resources to meet their developmental needs. Immigrant families may face additional barriers to accessing these resources. State and local policies are increasingly being considered to address the gap between advantaged and disadvantaged populations, especially as the federal government has turned more responsibility for social welfare programs over to the states. How Are California’s Children Faring? In this report, we look at parents’ assessments of how their children are doing in four important areas of child development: physical health, emotional and behavioral adjustment, attachment to school, and positive social involvement. We examine whether these indicators of well-being vary by children’s gender, age, race/ethnicity, parents’ nativity and education, and family income. In addition to examining children’s outcomes as simple percentages, we look at how the percentages corresponding to children’s membership in racial/ethnic, Hispanic immigrant, and socioeconomic groups change after they are adjusted to control for children’s characteristics. The adjusted percentages allow us to see the unique contribution of children’s characteristics and to evaluate whether differences between children living in California and vi those living elsewhere are meaningful after their characteristics are taken into account. Children’s Physical Health Perhaps the most fundamental indicator of children’s welfare is their health status. If children are in poor health, this would likely also have a negative effect on the other measures of their well-being we observe. We examine an indicator of children’s general health status shown to be strongly correlated with specific health problems and to be a robust predictor of future health status. The analysis also considers whether children have a debilitating health condition and their use of health services. In our analysis, about 78 percent of children in California are reported to be in very good or excellent health and only 8 percent have a chronic health problem. At the same time, over one-third of children had not seen a physician for routine preventative care in the last year and over one-fifth had not seen a dentist. Although children’s overall health status is generally positive, we find strong differences in children’s health among demographic and socioeconomic subgroups in the state. Specifically, children’s health status declines with age; yet, their visits to physicians drop off during adolescence as their needs increase, particularly among males. Our results also show that children whose parents are Hispanic immigrants and who are less-educated are in worse health than other children but are receiving less routine medical care. Although poor children are also reported to be in worse health, they seem to have more access to medical, but not dental, care than other children. We find that about one-third of children at high risk (whose parents are Hispanic immigrants, have the lowest levels of education, or who live in poor or near-poor families) had not seen a dentist in the past year. Children’s Emotional Adjustment One standard indicator of children’s adjustment during childhood is a behavioral problems inventory. The NSAF inventory was derived from a longer checklist of symptoms and is a reliable predictor of future emotional and behavioral problems. In particular, children with more frequent symptoms (as reported by parents) are more likely to encounter vii mental health problems and to engage in deviant behavior in adolescence and adulthood (Achenbach and Edelbrock, 1979; Achenbach, McConaughy, and Howell, 1987). Our findings show that more than 10 percent of the children in California are experiencing serious behavioral problems that put them at risk for social and mental health problems later in life but that only 5 percent of these children have received mental health services. The gap between children’s needs and treatment is higher in California than in other parts of the United States. This gap and the disparity between California and other states remain after the composition of children within and outside the state is taken into account. Within subgroups of children, we find that the level of children’s need for mental health services does not correspond with their receipt of these services. Namely, males and children from lower socioeconomic status families show more problems than other children but do not receive more treatment. These disparities are even more apparent for children with severe problems and indicate that poorer children are particularly underserved. Children’s Educational Attachment The ability to function in school is one of the earliest and most powerful predictors of whether children will develop their human capital (i.e., the skills and knowledge that are associated with productive employment later in life). This process of commitment to and confidence in learning begins early in life—as early, in fact, as the preschool years—and continues throughout childhood and adolescence. Research has shown that children’s work habits in the classroom predict their educational achievement and their later attachment to schooling and eventually to the workforce (Brint, 1998; Danziger and Waldfogel, 2000). We find that about one-third of children in California are not highly engaged in school, about 13 percent were expelled or suspended in the last year, and about one-fifth skipped school in the last year. Within subgroups of children, we find that males and older children are less likely to be engaged in school and are more likely to have voluntary or involuntary absences than other students. Our results also show that family income is strongly related to children’s attachment to school, viii particularly in the area of school absence, suspension, and expulsion. Although we observe racial/ethnic variations in children’s school engagement, these differences are markedly reduced when their other characteristics are held constant. Children’s Involvement in Pro-Social Activities Developmental psychologists are paying increasing attention to children’s involvement in pro-social activities such as clubs, sports, and after-school lessons, because these activities help to build their capacity to establish strong relationships with peers. They also foster a variety of nonacademic skills such as communication and development of talents that build self-esteem, a sense of efficacy, and psychological resourcefulness. Other dividends of positive social involvement include increasing children’s ability to form lasting personal connections that may be helpful in times of trouble and shielding children from the influence of negative peers (Carnegie Council on Adolescent Development, 1992; Eccles and Barber, 1999; Mahoney 2000; Pierce, Hamm, and Vandell, 1999). Our results indicate that close to one-quarter of children were not involved in any pro-social activities in the last year. We do not find that children’s level of involvement varies in important ways by their gender or age, but we do find that involvement varies considerably by children’s socioeconomic status. Children whose parents do not have a high school diploma and children living in poor families are less likely than other children to participate in social activities beneficial to their development. Differences in participation by children’s race/ethnicity diminish considerably after these and other factors are controlled for. Access to these activities might be limited both by the cost of these programs and by their availability to low-income children. Children in California and the Rest of the Nation Overall, we find that children in California appear to be faring slightly worse on a number of indicators of their well-being and are not outperforming children in the rest of the nation on any of the indicators we examine. Compared to children living elsewhere, the overall health status of children in California is somewhat lower. Children with ix behavioral problems are less likely to receive mental health care. Furthermore, California children skip school more often and are less involved in activities that promote their social development. After controlling for the compositional differences between children within and outside the state, we find that the gap between children’s need for mental health services and their use of these services remains higher in California than in the rest of the nation. However, disparities in children’s health status, school absences, and pro-social involvement largely disappear. These results suggest that across-state differences in children’s well-being may be due to the demographic and socioeconomic makeup of children who reside in California, except in the area of mental health services. Implications for Policy Although many of California’s children are faring well, our results suggest that a substantial minority may not be adequately served by existing services in the state. For the most part, these children come from the poorest families in California, have parents with low levels of educational attainment, or live in Hispanic immigrant families. These families may lack the resources, knowledge, and social connections to obtain services for their children. Our findings about the well-being of children in California have broad implications for providing services to children within the state and for monitoring whether these services are effective in serving children’s needs. The first set of policy implications stems from our observations about the nature of children’s needs. Although children whose parents are Hispanic immigrants and who are less-educated are in worse health than other children, they are receiving less routine medical care. Children from poor families are also in worse health but appear to have access to health care. The importance of serving healthy adolescents in higher-income families who are not receiving routine health checkups may be less apparent. Yet, physicians can do an effective job of screening for risky behavior such as unprotected sex, alcohol and drug use, or depression. We discovered that adolescent males were particularly likely not to have received routine health care in the previous year. x Although children from poorer families often use medical services, they are receiving much less help in obtaining preventive dental care. The lack of dental care is especially evident for Hispanic children whose parents were born outside the United States, children whose parents have low educational levels, and those whose families have poverty or nearpoverty-level incomes. There is an evident need to extend preventive services to this population, which may provide cost savings in the long run if serious dental problems are avoided later in life. There is also a conspicuous lack of mental health assistance for children with severe problems. All income groups have some gap between the need for and provision of services among children who exhibit high levels of problem behavior; however, children in the poorest families are particularly underserved. Children with emotional and behavioral problems often experience problems in the classroom and the community. Many of them will experience more severe problems later in life if they remain untreated. The tremendous costs of treating antisocial behavior and substance abuse among adults suggest that early intervention is justified even if it helps to reduce only modestly the incidence of problem behaviors. Policymakers might want to consider the possibility of directing public information to underserved communities to increase parents’ awareness of mental health services. Service providers may be more effective in reaching underserved populations by locating service sites closer to the populations at risk. It is not news that school behavior problems are more common among males, especially as they reach adolescence. Yet, programs that keep older males attached to school are in short supply. California could strengthen the in-school and after-school activities that help males to stay connected to school. Tracking absenteeism and behavioral problems can identify youth at risk of dropping out. We need to craft more experiments that can help middle school students prepare to make a successful transition to high school, whether by providing additional tutoring, mentoring, or social activities. Finally, our results indicate that an extraordinarily high proportion of youth are not engaged in any extracurricular activities in their preadolescent and adolescent years. The absence of programs for youth in this formative period likely accounts for part of the wide income xi disparity in the proportion of youth who are not involved in clubs, associations, lessons, sports, and the like. If programs do not adequately serve poorer children, this puts them at a disadvantage for gaining the benefits offered by extracurricular activities, including greater school attachment, skill development, and social ties. Particularly in the wake of welfare reform, there is a manifest need to serve low-income adolescents in after-school activities because their parents may have entered the labor force, leaving them unattended in the interim between the close of the school day and the end of the workday. The second important set of policy implications that stems from this analysis relates to how we can improve information on children’s wellbeing. First, we should develop indicators appropriate for very young children and, second, use existing indicators to follow cohorts of children over the next decade as they reach school age to see the effects of early health and development programs and services. It is reasonable to expect that some of these programs may have had positive effects on the younger cohort of children now in their preschool years. Unfortunately, we do not have sufficient data to measure the well-being of younger children at the state level and are not tracking children’s well-being over time. The regular measurement of children’s well-being through surveys such as the NSAF, together with rigorous program evaluations and the use of longitudinal administrative data, can help monitor whether the services provided are reaching the target populations and whether they are effective in raising levels of well-being. By beginning to monitor the success of our children, we can establish a benchmark for assessing whether California is providing sufficient public investment in children and can gain a glimpse into the future health, happiness, and productivity of our state’s population. xii Contents Foreword ..................................... Summary..................................... Figures ...................................... Tables ....................................... iii v xv xvii 1. INTRODUCTION ........................... 1 2. DATA AND METHODS ....................... Sample Characteristics.......................... Measures of Children’s Well-Being.................. Physical Health .............................. Emotional Adjustment ......................... Educational Attachment ........................ Pro-Social Involvement ......................... 9 11 14 14 17 18 19 3. PHYSICAL HEALTH ......................... Gender and Age.............................. Family Characteristics .......................... Conclusion ................................. 21 21 24 32 4. EMOTIONAL ADJUSTMENT ................... Gender and Age.............................. Family Characteristics .......................... Conclusion ................................. 35 37 38 43 5. EDUCATIONAL ATTACHMENT ................ Gender and Age.............................. Family Characteristics .......................... Conclusion ................................. 6. PRO-SOCIAL INVOLVEMENT .................. Gender and Age.............................. Family Characteristics .......................... Conclusion ................................. 45 46 48 53 55 56 57 62 xiii 7. HOW CALIFORNIA’S CHILDREN COMPARE TO THOSE IN THE REST OF THE NATION........... Physical Health .............................. Emotional Adjustment ......................... Educational Attachment ........................ Pro-Social Involvement ......................... Conclusion ................................. 8. CONCLUSION AND POLICY IMPLICATIONS ...... Children’s Physical Health ....................... Children’s Emotional Adjustment .................. Children’s Educational Attachment ................. Children’s Involvement in Pro-Social Activities .......... Children in California and the Rest of the Nation ........ Appendix A: Detailed Comparisons of Scale Items for California and the Rest of the Nation ....................... Bibliography .................................. About the Authors ............................... Related PPIC Publications .......................... 63 64 65 69 70 71 75 75 76 76 77 77 83 87 95 97 xiv Figures 3.1. Children with Chronic Health Condition That Limits Their Activity, by Gender and Age Group (Weighted %) ................................... 4.1. Children with Severe Behavioral Problems Who Received Mental Health Services in Past 12 Months, by Family Income (Weighted %) .................. 5.1. Children Who Skipped School Two or More Times or Were Suspended or Expelled in the Past 12 Months, by Gender and Age Group (Weighted %)............. 7.1. Children’s Health Status in California and the Rest of the United States (Adjusted %) ................. 7.2a. Children Who Received Mental Health Services in Past 12 Months, by Severity of Behavioral Problems, California and the Rest of the United States (Weighted %) ................................... 7.2b. Children Who Received Mental Health Services in Past 12 Months, by Severity of Behavioral Problems, California and the Rest of the United States (Adjusted %) ................................... 7.3. Children Who Participate in Pro-Social Activities, by Number, California and the Rest of the United States (Adjusted %) ............................. 23 42 48 66 67 68 72 xv Tables 2.1. NSAF Sample Characteristics of Children in California and the Rest of the United States (Weighted %) ...... 2.2. NSAF Measures of Children’s Well-Being .......... 3.1. Health Status of Children in California (Weighted %) .. 3.2a. Health Status of Children in California, by Race/ Ethnicity, Level of Parent’s Education, and Family Income (Weighted %) ....................... 3.2b. Health Status of Children in California by Race/ Ethnicity, Level of Parent’s Education, and Family Income (Adjusted %) ....................... 4.1. Children’s Behavioral Problems, by Gender and Age (Weighted %) ............................ 4.2a. Children’s Behavioral Problems by Race/Ethnicity, Parent’s Education, and Family Income (Weighted %) .. 4.2b. Children’s Behavioral Problems by Race/Ethnicity, Parent’s Education, and Family Income (Adjusted %) ... 5.1. Children’s School Engagement, Frequency of Missing School, and Suspension/Expulsion Rates, by Gender and Age (Weighted %) ......................... 5.2a. Children’s School Engagement, Frequency of Missing School, and Suspension/Expulsion Rates, by Race/Ethnicity, Parent’s Education, and Family Income (Weighted %) ............................ 5.2b. Children’s School Engagement, Frequency of Missing School, and Suspension/Expulsion Rates, by Race/Ethnicity, Parent’s Education, and Family Income (Adjusted %) ............................. 6.1. Children’s Pro-Social Involvement, by Gender and Age (Weighted %) ............................ 6.2a. Children’s Pro-Social Involvement by Race/Ethnicity, Parent’s Education, and Family Income (Weighted %) .. 12 15 22 26 30 38 40 41 46 49 52 57 59 xvii 6.2b. Children’s Pro-Social Involvement by Race/Ethnicity, Parent’s Education, and Family Income (Adjusted %) ... 7.1. Health Status of Children in California and the Rest of the United States (Weighted %)................. 7.2. Children’s Behavioral Problems in California and the Rest of the United States (Weighted %) ............ 7.3. Children’s School Engagement in California and the Rest of the United States (Weighted %) ........... 7.4. Children’s Pro-Social Involvement in California and the Rest of the United States (Weighted %) ............ A.1. Emotional Adjustment of Children in California and the Rest of the United States (Weighted %) ........... A.2. School Commitment of Children in California and the Rest of the United States (Weighted %) ............ A.3. Pro-Social Involvement of Children in California and the Rest of the United States (Weighted %) ............ 60 64 66 69 71 84 85 85 xviii 1. Introduction California, the nation’s most populous state, has over nine million children. These children are among the most ethnically and racially diverse in the country, and almost half of the children under age six in California have a parent who was born outside the United States. By 2020, Hispanic children are expected to constitute the largest group of children in the state (Reed and Tafoya, 2001). California’s children are more likely to be poor than those in the nation as a whole, with about 20 percent living in poverty, compared to about 17 percent in the United States (U.S. Census Bureau, 2002). No population group in this country is the object of more public concern and scrutiny than America’s children and youth (Haveman and Wolfe, 1994; Mason, Skolnick, and Sugarman, 1998; Takanishi and Hamburg, 1997). The public cares whether children are healthy, happy, and secure for the same reasons that they care about the welfare of any other age group. But because children are also one of the most vulnerable demographic groups in the population, they receive more public attention (if not always financial aid and social services) than nonelderly adults, who are considerably less dependent on others for their survival and development. The public also cares about children’s well-being because how children fare in early and middle childhood—not to mention how they do in their adolescent years—is strongly related to how they will function as adults. A wealth of research from different social science disciplines has demonstrated that adult performance is strongly and causally linked to the quality and quantity of material and emotional investment received during childhood (Cairns, Elder, and Costello, 1996; Haveman and Wolfe, 1994). Children do better in later life when they are engaged in multiple contexts in which they are monitored, supported, and provided with opportunities to acquire cognitive, emotional, and physical skills. 1 At the same time, some national data indicate growing inequality in access to contexts in which children can develop these skills (Danziger and Gottschalk, 1993, 1995). Thus, public concern also focuses on whether children have the kind of resources that will enhance their welfare and development and, when disparities in these resources exist, how they can be alleviated. Poor families often experience greater stress than economically advantaged families and may be less able to provide their children the kind of resources related to healthy development. Research shows that poverty has a negative effect on children’s cognitive abilities and achievement, particularly if children live in extreme poverty or their families are poor for many years (Brooks-Gunn and Duncan, 1997; Duncan and Brooks-Gunn, 1997). Some of these effects also appear to persist in adulthood, when these youth later attempt to find stable and remunerative employment, to form families, and to participate in civic life. Because nonwhite families in California and the nation often have lower incomes than white families (Reyes, 2001), there is also concern that children in these families disproportionately lack sufficient resources to meet their developmental needs. Immigrant families may face additional barriers to accessing these resources. State and local policies are increasingly being considered to address the gap between advantaged and disadvantaged populations, especially as the federal government has turned more responsibility for social welfare programs over to the states (Millstein, Petersen, and Nightingale, 1993). Finally, the public cares about whether children are being adequately nurtured because we want our young people to become productive and involved members of society when they reach adulthood. When children do not receive this nourishment, they are far less likely as adults to be able to contribute to the quality of life and the economic security of the state. Indeed, the failure of children to develop the knowledge, skills, and values that make them successful adults leads to a weakening of the social fabric and places a burden on taxpayers and fellow citizens. In recent years, California’s citizens and lawmakers have given a great deal of attention to the needs of children, and some programs have directed considerable resources toward enhancing their welfare and development. The creation of three recent programs in the areas of child development, health, and after-school care illustrates how some of these 2 concerns have been addressed. First, with the passage of the Proposition 10 initiative in 1998, citizens voted to impose a 50 cent tax on tobacco, largely to fund early childhood development programs aimed at mothers during prenatal care and children through age five. The aim of this initiative is to provide comprehensive, integrated services that promote children’s development and ensure that they enter school ready to learn.1 In fiscal year 2000–01, the tobacco tax generated about $650 million in revenue for Proposition 10 activities (California Children and Families Commission, 2002).2 New efforts to expand children’s access to health coverage have also emerged in recent years. For example, the Healthy Families Program, created in 1997, is a state- and federally funded program to provide lowcost health, dental, and vision coverage to children living in low-income families who do not qualify for Medi-Cal but who have incomes below 250 percent of the poverty level. 3 Enrollment in Healthy Families is expected to reach 624,000 children by June 30, 2003, and the 2002–03 California budget allocates $672.2 million in combined state and federal funding for this program (California Department of Finance, 2002). Increased access to health care for children can help families obtain preventive health services that might discourage the onset of health problems in later life as well as offer medical assistance for children’s acute and chronic health problems. Both within and outside the K–12 educational system, California has been actively expanding a variety of resources and social services aimed at ____________ 1See http://www.ccfc.ca.gov/ for more information about the California Children and Families Commission created by Proposition 10. 2Of course, child care represents another important area of spending on children’s development. The 2002–03 state budget includes $3.1 billion for a variety of child care and related services (California Department of Finance, 2002). 3Healthy Families is the name of the State Children’s Health Insurance Program in California. As the state expanded health coverage under the Healthy Families Program, it also increased income eligibility in Medi-Cal for children up to age 18 and provided coverage for pregnant women and infants with incomes between 200 and 300 percent of the poverty level through the Access for Infants and Mothers Program (Brown et al., 2002, pp. 43–44). In January 2002, the U.S. Department of Health and Human Services also approved a federal waiver to cover 300,000 California parents with incomes up to 200 percent of the poverty level under the Healthy Families Program (U.S. Department of Health and Human Services, 2002). 3 both preschoolers and school-age children. Some of these services are designed to serve youngsters from low-income families and those whose parents are recent immigrants with limited language skills, low educational levels, and poor connections to mainstream institutions. For example, the state’s Before and After School Learning and Safe Neighborhoods Partnerships Program was created in 1998 to provide literacy and educational enrichment and to provide a safe atmosphere for children in kindergarten through junior high. Proposition 49, approved by voters in the November 2002 election, requires a specific spending level for this program, up to $550 million annually.4 These and other efforts to promote children’s welfare have generated interest in developing stable and reliable measures of children’s wellbeing to determine the success of the growing number of state and county services to improve parent education, child care, health care, and effective intervention for families at risk. However, our ability to assess child-centered policies has been seriously hindered by the absence of good data at the state level, despite the fact that many of the most important policies affecting children are administered by states. Although interest in children’s well-being has been keen for many years, only relatively recently have researchers and policymakers begun to systematically track information about this issue using nationally representative data (Ben-Arieh et al., 2001). At the state and local levels, progress in assembling information that would allow policymakers and the public to gauge the performance of children has been even slower, largely because of the absence of reliable information about children’s well-being. Most national studies that include indicators of children’s well-being have not been large enough to yield reliable information on children at the state level. Some important information about children’s demographic and socioeconomic characteristics as well as their early health status is available from vital statistics and Census data.5 Children ____________ 4Proposition 49 also changed the name of this program to the After School Education and Safety Program. 5See also Reed and Tafoya (2001), Reed and Bailey (2002), and Johnson (2003) for statistical portraits of children ages five and under in California, including information about population trends and projections, family structure, births to teen parents, parental 4 Now also assembles government and survey data for its annual report card and biannual county data book on California children’s education, health, family economic resources, and safety (Children Now, 2002a, 2002b). Because these datasets do not include measures of children’s psychological and social development, which would be useful to policymakers in gauging children’s well-being, one of the overarching recommendations of these reports is that more data on the well-being of young children in California be collected and analyzed.6 In this study, we use a new source of information—the National Survey of America’s Families (NSAF)—that was explicitly designed to measure children’s well-being and that provides a large enough sample of California families to assess the status of children in the state. We draw on information collected for 1,917 children living in California households that participated in the 1999 NSAF to examine variations in the well-being of children in the state and to compare their overall welfare to that of children in other states. In particular, we identify a series of indicators that measure different dimensions of their well-being, including their physical health status, emotional adjustment, educational attachment, and involvement in social activities that promote their development. The sample is large enough to allow us to examine how children are doing by looking at their demographic and socioeconomic characteristics—information typically lacking in data gained from national surveys and government records. Although data from the NSAF and other surveys have some obvious limitations associated with how accurately respondents report information, they offer some distinct advantages over information exclusively collected from administrative sources. For example, most administrative records offer information in a single domain of children’s welfare, such as their school performance, health, or involvement in the juvenile justice system. Although it might be possible to assemble an ________________________________________________________ education, family income, receipt of public assistance, and health insurance and vaccinations. 6Both publications identified areas where improvements have been made for children in the state, such as in declines in infant mortality, and where needs remain, such as in the area of high-quality child care. For more information, see http://www. childrennow.org. 5 archive with administrative reports from a variety of agencies, this task has proved difficult in practice. Through surveys of parents and children, it is possible to collect a variety of different indicators of children’s welfare with relative ease; indeed, it is possible to obtain data that simply are not available elsewhere. The strength of the NSAF is that it includes a broad array of indicators of children’s well-being not readily available from other data sources that can be tracked over time. Both the size of the California sample and NSAF’s unique measures of children’s well-being make the survey extremely well-suited for our study of children in the state. There are some potential problems in relying on reports exclusively from parents. For example, some parents may be reluctant to report negative outcomes for their child or to accurately state problems with their child’s behavior. The small amount of literature on this source of bias suggests that parent reports can sometimes overstate or understate, depending on social class or ethnicity. Nonetheless, it appears that parents are generally honest in their evaluations, albeit limited in their knowledge of how the child behaves outside the home. Furthermore, it is worth knowing how parents view their children’s well-being because parents are important gatekeepers to services offered by schools and social agencies. Policymakers and practitioners need information on how many children and families are in need of services, how many are being served, and, most of all, how well services are working to achieve the objectives of the health, educational, and social programs that are being mounted on behalf of improving the well-being of children.7 This report, representing a first step in providing valuable information on the well____________ 7For example, a package of legislation was put forward in 2002 that aimed to improve the health of California’s students. This legislation followed the release of a report from the Select Committee on California Children’s School Readiness and Health indicating that children are having trouble learning in school, in part because of health problems. The report recognized the lack of indicators of children’s health available and suggested that more empirical studies and measures of children’s health were needed to track the health status of school-age children (Bustillo, 2002). 6 being of California’s children,8 has several objectives: First, it provides a baseline of information about children’s well-being in 1999 against which similar indicators can be reexamined in subsequent years. This will allow an assessment of the overall success of increased investment in the development of children and in producing healthy, socially involved, and psychologically sound young people. The report also provides information about the needs of special subgroups of children and suggests to policymakers ways to target the populations that may be in particular need of health or social services. Finally, this report can inform policymakers and the public about how children in California are faring relative to those in the rest of the nation. In the next chapter, we describe the NSAF in more detail and provide an overall picture of how the survey data were collected and of the children included in the sample. We also discuss the measures used to assess children’s well-being. Chapters 3 through 6 present information on each of the domains of children’s well-being we analyze and give a more detailed picture of how children’s well-being varies by their gender, age, race/ethnicity, parents’ nativity and education, and family income. Chapter 7 compares California’s children to those in the rest of the United States. The last chapter sums up what we have learned and discusses some of the policy implications of this analysis. ____________ 8Complementary methods of tracking the well-being of children and the effectiveness of services in the state might include program evaluations and the linking of longitudinal administrative records. 7 2. Data and Methods Our analysis of the well-being of children in California draws on data from the 1999 National Survey of America’s Families. The NSAF is a national survey that contains information on the economic, health, and social characteristics of children and adults under age 65 in the United States. The survey collected information on over 100,000 people living in 42,000 U.S. households.1 This study uses information collected on 1,917 children living in 1,536 California households. It compares how these children are faring in comparison to 34,021 children living in 28,034 households in the rest of the United States. Because the NSAF was designed by the Urban Institute and Child Trends to examine the well-being of children and families following the devolution of welfare and other social policies to the states, 13 states, including California, were oversampled for the survey. This sampling design was intended to make the NSAF representative of the nonelderly population in these 13 states and in the nation as a whole.2 As such, the ____________ 1The first wave of the survey was conducted in 1997. In this report, we do not present results for the 1997 survey primarily because this earlier survey presents potential difficulties with examining how immigration status is related to children’s well-being—a key issue to interpreting differential outcomes among children in California. Specifically, the large discrepancy in the number of foreign-born and immigrant respondents in the 1997 NSAF compared to the number in other data sources prevents us from presenting results for immigration status with complete confidence. To correct for this problem, the 1999 NSAF changed the wording of the question used to determine the nativity of respondents by collecting information about the country of birth for each household member (Wang, Cantor, and Vaden-Kiernan, 2000). 2The survey is representative of the noninstitutionalized, civilian population of individuals under age 65 in the nation. It is also representative of this population in Alabama, California, Colorado, Florida, Massachusetts, Michigan, Minnesota, Mississippi, New Jersey, New York, Texas, Washington, and Wisconsin. For an overview of the survey, see http://www.urban.org/Content/Research/NewFederalism/NSAF/ Overview/NSAFOverview.htm. For more detailed discussions of the survey’s methodology, see Judkins et al. (2001), Safir, Scheuren, and Wang (2001), and Wang, Cantor, and Vaden-Kiernan (2000). 9 NSAF sample in California is large enough to produce reliable estimates of child and family well-being measures in the state. Because the NSAF oversamples low-income households, it also allows us to compare the outcomes of children living in higher- and lower-income families in the state. The NSAF used a random digit dialing technique to generate a sample of households. In areas with limited telephone coverage, the survey supplemented the telephone sample with an area probability sample of households without telephones. From the households in the sample, 50,355 extended interviews were conducted in the entire NSAF sample, including 48,679 telephone interviews and 1,676 in-person interviews (Vaden-Kiernan et al., 2000). The weighted household response rate for the entire NSAF sample was about 62 percent and the response rate for the California sample was about 56 percent. These response rates compare favorably to other surveys of this type (Brick et al., 2000).3 As part of the selection process, the survey screened households to identify the characteristics of its members. In households with children, up to two children could be randomly selected for analysis. However, only one child under age six and one child between the ages of six and 17 could be selected. In about 76 percent of the California households with children in the study, an interview was conducted for only one child; in the remaining 24 percent of households, interviews were conducted for two children. After selecting these children, the NSAF directed questions about their well-being to the person in the household who knew the most about the child—almost always the child’s parent.4 About 79 ____________ 3The telephone response rate for the entire sample was about 61 percent and for the California sample was about 55 percent. The area household response rate was about 86 percent for the entire sample and 91 percent for the California sample (Brick et al., 2000). 4In California, about 92 percent of the adult respondents were the biological parents of the child, 2 percent were stepparents, 2 percent were grandparents, 1 percent were adoptive parents, 1 percent were the partners of the parent, 1 percent were siblings, and less than 1 percent were foster parents, aunts, or uncles. Most children (58 percent) were living in families with two, married biological parents or with a single parent who was their biological mother (19 percent). 10 percent of these adult respondents were female. For purpose of this report, we shall refer to the person reporting on the child as the child’s parent although in a small minority of interviews, the most knowledgeable adult was a surrogate parent. Sample Characteristics Table 2.1 illustrates selected characteristics of children, their parents, and their households in the 1999 NSAF, comparing characteristics of the sample in California to those in the rest of the nation.5 As this table shows, a little over half of the sample of children in California and the rest of the United States is male, and the sample is fairly evenly divided among children in the three age groups (i.e., 0–5, 6–11, and 12–17) for which child well-being is measured in our study.6 The majority of children in the California sample have parents who are either white (43 percent), foreign-born Hispanic (27 percent), or U.S-born Hispanic (15 percent). Only a small percentage of children in the California NSAF have black or Asian parents (7 percent each). Because survey interviews were conducted in English and Spanish, households where neither English nor Spanish was spoken could not participate in the survey (Black and Safir, 2001), possibly leading to some underrepresentation of Asian children.7 Unfortunately, the small number of children with Native American parents in the California sample prevent us from ____________ 5These frequencies are weighted to account for the original possibility of the household being selected, for subsampling of respondents, and for nonresponse. The weight is also adjusted to correct for undercoverage using 1990 Census information and is adjusted for Census undercount. 6Data from the 2000 Census indicate that about 32 percent of children in California are under age 6, 36 percent are ages 6–11, and 32 percent are ages 12–17 (U.S. Census Bureau, 2002). 7We look at the race and immigrant status of the parent rather than the child because children living in Hispanic immigrant families are expected to have different outcomes than children living in nonimmigrant families. Data from the 2000 Census indicate that about 44 percent of children in California are Hispanic, 35 percent are white, 7 percent are black, 9 percent are Asian, 0.5 percent are Native American, and 4.5 percent are identified as other, including children with two or more racial/ethnic classifications (U.S. Census Bureau, 2002). 11 Table 2.1 NSAF Sample Characteristics of Children in California and the Rest of the United States (Weighted %) Gender Male Female Age 0–5 6–11 12–17 Race/ethnicity of parent Asian Black Hispanic, foreign-born Hispanic, U.S.-born White Other Rest of the California United States 52 51 48 49 35 33 35 34 30 33 72 7 13 27 5 15 6 43 72 11 Foreign-born parent % Hispanic % Non-Hispanic Parent’s education Less than high school High school or GED More than high school 30 9 68 48 32 52 20 11 29 36 51 53 Family income relative to poverty level (%) 300 21 18 22 23 18 20 38 40 Sample size 1,917 34,021 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. 12 comparing how these children are faring in relation to children from other racial and ethnic groups in the state.8 Compared to the sample in the rest of the United States, the California sample has more Hispanic and Asian children and fewer white and black children who are not of Hispanic origin. The California sample also has a much larger proportion of children from immigrant families than does the sample for the rest of the nation, with about 30 percent of children in the California sample having a foreign-born parent. Over two-thirds of foreign-born parents in California are Hispanic, compared to less than half in the rest of the nation. In both California and the rest of the nation, most other foreign-born parents are Asian (about one-fifth). Children in California are also slightly younger than children living elsewhere as a result of the higher birth rate of the immigrant population (Johnson, Hill, and Heim, 2001). It is also important to examine parents’ education and household income to better understand the socioeconomic circumstances in which the children are living.9 In California as in the rest of the nation, about half of children’s parents (that is, parents interviewed in the survey) had a high school diploma or less. However, almost twice the proportion of parents in California did not have a high school diploma or General Equivalency Degree (GED) as those in the rest of the nation. This discrepancy is likely related to the number of foreign-born Hispanic respondents in the California sample.10 Children in the California ____________ 8Specifically, 24 children in the California sample had parents who identified themselves as Native American, American Indian, Aleutian, or Eskimo, representing just over 1 percent of the sample. Although these children were kept in the analysis, we could not display results for Native American children separately in the tables. 9See Converse, Safir, and Scheuren (2001, pp. D1–2) for comparisons of estimations for adult nonelderly employment earnings and household size distributions in the Current Population Survey (CPS) and the NSAF. The authors report that when they analyze these measures (as well as measures of family composition, work experience, income, and poverty) by key demographic characteristics, the NSAF and CPS estimates are similar and generally within normal sampling variation. 10About 59 percent of foreign-born Hispanic parents in the California NSAF do not have a high school diploma or GED. 13 sample are also somewhat more likely than children in the rest of the nation to be living in households with incomes below the poverty level. Specifically, about 21 percent of California children live in households that have incomes below the official poverty line and an additional 22 percent live in households that are nearly poor, with incomes between 100 and 200 percent of the poverty line. Measures of Children’s Well-Being The NSAF collected information on a wide array of items that pertain to children’s well-being. For purposes of this report, we decided to restrict our analysis to those indicators that directly refer to child outcomes—that is, measures of the child’s current behavior or adjustment in the family, school, and larger community. Most of these measures are well-established indicators of children’s developmental status that have been used in previous national surveys.11 We selected indicators from four important cornerstones of children’s development: physical health, emotional adjustment, attachment to school, and pro-social involvement (see Table 2.2). Depending on the type of item in the survey, we used either separate questions or indices made up of a set of related questions. However, the discussion below describes each item individually, whether or not it was analyzed separately or as part of a summative index or scale. Because the items were intended to be age-appropriate, different questions were often asked for children in different age groups. The remainder of this chapter describes the components of each of the four domains that we examine in later chapters of this report. Physical Health The analysis includes four measures of children’s health. The first indicator is a summary question designed to provide a general assessment of the child’s physical health. The primary caregiver was asked to rate the child’s health on a scale from poor to excellent. This single measure ____________ 11For a more detailed discussion of the measures as well as the quality of the data, the internal reliability of the scales, and construct validity, see Ehrle and Moore (1999). 14 Table 2.2 NSAF Measures of Children’s Well-Being Children 0–5 Years of Age Children 6–11 Years of Age Children 12–17 Years of Age Physical Health Now, I’d like to talk about child’s health status. In general, would you say child’s health is 1. Excellent 2. Very good 3. Good 4. Fair 5. Poor. About how many of (his/her) visits to a doctor or other medical professionals that you just told me about were for well-child care, such as checkups? During the past 12 months, how many times did child see a dentist or dental hygienist? Does child have a physical, learning, or mental health condition that [limits (his/her) participation in the usual kinds of activities done by most children (his/her) age/limits (his/her) ability to do regular school work]? 1. Yes 2. No Emotional Adjustment I am going to read a list of items that sometimes describe children. For each item please tell me if it has been often true, sometimes true, or never true for child during the past month. a. (He/she) doesn’t get along with other kids. b. (He/she) can’t concentrate or pay attention for long. c. (He/she) has been unhappy, sad, or depressed. 15 I am going to read a list of items that sometimes describe children. For each item please tell me if it has been often true, sometimes true, or never true for child during the past month. a. (He/she) feels worthless or inferior. b. (He/she) has been nervous, high-strung, or tense. c. (He/she) acts too young for (his/her) age. I am going to read a list of items that sometimes describe children. For each item please tell me if it has been often true, sometimes true, or never true for child during the past month. a. (He/she) has trouble sleeping. b. (He/she) lies or cheats. c. (He/she) does poorly at schoolwork. During the past 12 months, how many times has child received mental health services, including mental health services received from a doctor, mental health counselor, or therapist? Table 2.2 (continued) Educational Attachment For each of the following statements, please tell me if you think it describes child all of the time, most of the time, some of the time, or none of the time. a. Child cares about doing well in school? b. Child only works on schoolwork when forced to? c. Child does just enough schoolwork to get by? d. Child always does homework? 16 During the past 12 months has child been suspended or expelled from school? This includes both in-school and out-of-school suspensions. 1. Yes 2. No During this past 12 months, how many times has child skipped school, cut classes without your permission, or refused to go to school? Was it . . . 0. Never 1. Once 2. Two or more times Pro-Social Involvement In the last year, has child been on a sports team either in or out of school? 1. Yes 2. No In the last year, has child taken lessons after school or on weekends in subjects like music, dance, language, or computers? 1.Yes 2. No In the last year, has child participated in any clubs or organizations after school, or on weekends, such as scouts, a religious group or Girls or Boys club? 1. Yes 2. No In the last year, has child participated in any clubs or organizations after school, or on weekends, such as a youth group or student government, drama, band or chorus, or a religious or community group? 1. Yes 2. No is commonly used in surveys and has been shown to be strongly correlated with specific health problems and to be a robust predictor of future health status.12 In addition to this overall evaluation of the child’s health, informants were also asked about the number of times the child had seen a doctor or medical health professional in the preceding 12 months for well-child care, such as checkups. This item provides both a measure of access to routine health examinations as well as an indication of the child’s health needs. Informants were also asked about the number of times children had seen a dentist in the past year. Finally, respondents were asked if the child had a physical, learning, or mental health condition that limits his or her activity. Questions about visits to a dentist and mental health professional were asked only for children ages three and older. Emotional Adjustment One standard indicator of children’s adjustment during childhood is a behavioral problems inventory derived from a checklist of symptoms originally developed by T. M. Achenbach (Achenbach and Edelbrock, 1979; Achenbach, McConaughy, and Howell, 1987). This inventory has shown that children with more frequent symptoms (as reported by parents) are more likely to encounter mental health problems and to engage in deviant behavior in adolescence and adulthood. These behaviors are typically examined together because each individual symptom is far less informative about the child’s emotional status than the index as a whole. Some studies have suggested that children may experience “internal symptoms” that relate to later mood disorders and depression, whereas other children manifest “external symptoms” by acting up or displaying aggressive behaviors (Compas and Hammen, 1994; Maccoby, 1998; Peterson and Zill, 1986). In largely descriptive ____________ 12The California Health Interview Survey (CHIS) shows results for children’s health status that are comparable to those of the NSAF for California, with most children indicated to be in very good or excellent health. However, fewer children in the CHIS are reported to be in excellent health and more are reported to be in fair, good, or very good health than those in the NSAF (based on authors’ calculations using the AskCHIS data query system available at http://www.chis.ucla.edu/). Unlike the NSAF, children ages 12–17 were asked to report their own health status in the CHIS. 17 studies such as this one, most researchers sum up the entire list of symptoms as an indicator of emotional and social adjustment. The NSAF uses a short-form adaptation of Achenbach’s longer list of symptoms. This measure has been used in previous studies conducted by telephone and seems to be both reliable and a good predictor of future emotional and behavioral problems.13 In addition, we also included a measure of emotional health—i.e., whether or not a child had received mental health services in the past year. Educational Attachment The ability to function in school is one of the earliest and most powerful predictors of whether children will develop their human capital, or skills and knowledge that improve the possibility for productive employment later in life. This process of commitment to and confidence in learning begins early in life—as early, in fact, as the preschool years— and continues throughout childhood and adolescence. Research has shown that children’s habits in the classroom predict their early educational achievement and their later attachment to schooling and eventually to the workforce (Brint, 1998; Danziger and Waldfogel, 2000). For example, Danziger and Waldfogel (2000) find that early school achievement forecasts long-term employment patterns and earnings ability. The NSAF includes four questions on children’s concern about doing well in school and their commitment to doing their schoolwork. These items have a high reliability (that is, they are strongly intercorrelated) and therefore can be used as a single scale of school engagement.14 The separate items listed in Table 2.2 are summed ____________ 13Ehrle and Moore (1999) used the 1997 NSAF data to benchmark this measure against a similar measure in the National Longitudinal Survey of Youth (NLSY). They found similar patterns across socioeconomic groups, with behavioral and emotional problems increasing with lower socioeconomic status. However, because children in the NLSY were more disadvantaged than those in the NSAF, children in the NLSY had a higher incidence of behavioral and emotional problems. This was consistent with the authors’ expectations. 14Ehrle and Moore’s (1999, pp. 1–2) report does not provide a benchmark comparison of the educational engagement scale because this measure had not yet been used in other national surveys at the time of their analysis. However, they show that the 18 together in a scale ranging from 0 to 4. The NSAF also includes information about voluntary and involuntary school absences for children ages 12–17 that provides indicators of negative school behaviors. These questions ask whether children have been suspended or expelled from school and the number of times they have skipped school in the past year. Pro-Social Involvement Developmental psychologists are paying increasing attention to children’s involvement in social activities because this involvement helps to build their capacity to establish strong relationships with peers; increases nonacademic skills such as communication and development of talents; and fosters self-esteem, a sense of efficacy, and psychological resourcefulness. Other dividends of positive social involvement include increasing children’s ability to form lasting connections that may be helpful in times of trouble and shielding children from the influence of negative peers (Carnegie Council on Adolescent Development, 1992; Eccles and Barber, 1999; Mahoney 2000; Pierce, Hamm, and Vandell, 1999). The NSAF includes a set of items that asks parents whether their children are involved in clubs or sports teams or take lessons outside of school to cultivate skills.15 Because many children were not involved in all three of these activities, we summed up the number of children who were involved in zero, one, or two or more types of social groups and activities. ________________________________________________________ measure varies as expected in the subgroup analysis, with lower socioeconomic groups showing lower school engagement. In addition, the authors note that the quality of the data in regard to missing data, distribution, and the alpha—a statistical measure of reliability—also indicates that the data can be used with confidence. 15The NSAF measures of pro-social involvement were benchmarked against those in the Survey of Income and Program Participation (SIPP) and the National Education Longitudinal Survey (NELS) (Ehrle and Moore, 1999). However, these comparisons were not exact in that the question wording differed somewhat across surveys and because the NELS sample differed from the NSAF sample. Ehrle and Moore found that more children in the NSAF were involved in activities than were those in the SIPP and NELS, likely because the NSAF question was more inclusive. Patterns by socioeconomic groups were similar across all three surveys and varied in the expected direction, with lower socioeconomic groups reporting less involvement. See Ehrle and Moore (1999, pp. 1-5; 7-4 to 7-6). 19 As we noted above, one obvious limitation of the NSAF data is that all of the measures we employ rely exclusively on one informant—nearly always the child’s parent. Ordinarily, and especially for younger children, this strategy poses no problems because parents are known to provide reasonably reliable and valid reports of how their offspring are doing. However, parents’ ability to detect problems in children’s behavior, particularly outside the home, may be less accurate when children reach early adolescence. Consequently, we suspect that parents’ reports for the older children in the NSAF sample may be more positive than those that could have been obtained if information had been gathered from teachers or children themselves. 20 3. Physical Health This chapter examines how the demographic and socioeconomic characteristics of California’s children and their families relate to indicators of children’s physical health. The investigation of children’s health indicators is important for two reasons. First, it offers useful information to those concerned with designing health policy and providing health care to children in California by directing attention to segments of the population where health needs are greatest and most underserved. Second, the findings in this chapter help explain the health status differential between California and other states that we identify and discuss in Chapters 7 and 8. We begin by examining how health indicators vary with the gender and age of children and then turn to health variations associated with family characteristics, including race and ethnicity, parental education, and family income. Gender and Age Research indicates that adult males experience higher rates of mortality and morbidity than females (Knudsen and McNown, 1993; Waldron 1995). However, national surveys do not show that the general health status of children differs greatly by gender on measures identical to those in the NSAF (e.g., National Center for Health Statistics, 2002). Accordingly, we would be surprised to discover large health differences between boys and girls in California as reported by parents. Consistent with other national surveys, the NSAF findings show that differences in boys’ and girls’ overall health status are not substantial in California (see Table 3.1). At the same time, parents report that boys (9 percent) are somewhat more likely than girls (6 percent) to have a health condition that limits their activity. This modest difference in chronic health problems does not seem to be reflected in parents’ overall evaluations of their son’s health, perhaps because some of these children had temporary conditions (such as a broken leg) that incapacitates them 21 Table 3.1 Health Status of Children in California (Weighted %) Child’s current health status Poor Fair Good Very good Excellent Gender Age Group Female Male ≤ 5 6–11 12–17 Total 1 11 01 1 6 55 57 5 17 15 16 15 17 16 25 25 21 29 25 25 52 54 58 50 50 53 Has condition that limits activity 6 93 8 11 8 Well-child doctor visit during past 12 months No One Two or more 35 37 16 39 56 36 37 36 33 43 34 37 28 27 51 18 11 27 Saw a dentist during past 12 months 76 77 66 81 77 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. 77 but does not seriously compromise their health. Despite these differences, boys and girls had comparable levels of well-child visits as indicated by their parents. Similarly, we find that boys and girls had almost identical levels of routine dental care. Thus, it appears that access—or lack of access—to health care does not vary by gender. Table 3.1 also presents information about children’s health status by their age group. Although national data indicate few changes in children’s health status between early childhood and adolescence, our evidence points to a very small deterioration of health status among older children. Specifically, more children under age 12 (79 percent) than over age 12 (75 percent) are in excellent or very good health, according to their parents’ reports. Slightly more adolescent children (8 percent) than younger children (5–6 percent) are reported to be in fair or poor health. We see that the incidence of chronic health conditions that limit children’s activity is also higher among older children, rising steadily from 3 percent in early 22 childhood, to 8 percent in middle childhood, to 11 percent in the adolescent years. As Figure 3.1 shows, the increase is much steeper for boys than for girls, especially as they move from middle childhood to adolescence. By their teen years, 16 percent of males have a health condition that limits their routine activities—a figure more than twice as high as that for adolescent females. When we turn to the health indicators relating to the provision of care, we see that older children in California are far less likely than younger children to have had any contact with a physician in the past year (Table 3.1). The strong decline in routine health care we observe among older children is especially disturbing because parents report that their adolescents are in poorer health. According to these reports, children age five and younger had the most health visits, with more than twice as many having some contact with a physician or other medical professional as children age 12 and older. In particular, approximately 16 percent of children under age six had not seen a physician or other medical professional in the past year, compared to about 39 percent of 18 16 Boys 14 Girls 16 Percentage 12 10 9 8 77 6 4 4 2 2 0 0–5 6–11 Age group 12–17 Figure 3.1—Children with Chronic Health Condition That Limits Their Activity, by Gender and Age Group (Weighted %) 23 children between the ages of six and 11 and 56 percent of children between the ages of 12 and 17. Even though the majority of older children do not appear to be experiencing health problems, few receive routine examinations that could prevent or detect such problems should they occur. The absence of contact with a physician also means that relatively few children receive the benefit of screening for mental health problems, sexual and reproductive health concerns, substance abuse, and a variety of other medical issues that arise in the adolescent years. Adolescent boys are especially likely not to have received routine medical care such as checkups, with about 62 percent of boys ages 12 to 17 reported to have had no medical visits, compared to 50 percent of girls in this age group (figures not shown in table). A somewhat different pattern emerges when we look at dental visits. As might be expected, dental checkups are lowest for children between the ages of three and five: approximately one-third of children in this age group had not seen a dentist in the previous year.1 This figure drops to about one-fifth in middle childhood and rises again to about one-quarter for adolescents. As with the pattern we observe for physician visits, we again find that parents of older children report that more males than females have not visited a dentist. About 27 percent of adolescent boys did not see a dentist in the past year, compared to about 18 percent of adolescent girls (figures not shown in table). Family Characteristics There is a large literature showing that characteristics related to children’s family background strongly affect their health outcomes (Millstein, Petersen, and Nightingale, 1993; Zuckerman and Kahm, 2000). For example, nonwhites have greater health needs and less access to health care (Elo and Preston, 1996; Hayward et al., 2000; Williams and Collins, 1995). This is due, at least in part, to the socioeconomic circumstances of these families, who are less connected to the health system, who may be less aware of the need for preventive care, and who may lack important resources such as the time and money to get routine health visits. ____________ 1The question about dental visits was asked only for children ages three and older. 24 We begin by examining whether indicators of children’s health vary by their racial/ethnic descent, their parent’s level of education, and their family income. When looking at Hispanic ethnicity, we divided the sample into Hispanic children whose parents (based on the status of the parent interviewed) were born in the United States and those whose parents were born outside the United States. In Table 3.2a, we see that parents’ evaluations of their children’s health are somewhat lower for children of black and native-born Hispanic parents and distinctly lower for children of foreign-born Hispanics. Whereas only 2 percent of white and Asian children are in fair or poor health according to their parents’ reports, the figure rises to 5 percent for black children, 8 percent for children of native-born Hispanics, and 14 percent for Hispanic children whose parents were born outside the United States. A similar pattern is evident if we look at the proportion of racial/ethnic groups who report that their children are in very good or excellent health. About 54 percent of Hispanic children in immigrant families appear in the top two categories compared to 80 percent of Hispanic children in nonimmigrant families, 83 percent of black children, and approximately 90 percent of white and Asian children. Sharp differences in children’s health status are also apparent when we examine children’s health by variations in their parents’ education and family income. About 55 percent of children whose parents have less than a high school education are reported to be in excellent or very good health as compared to 77 percent of children whose parents have a high school diploma and 88 percent of children whose parents have education beyond high school. We see the similarly steep gradients in health differences by family income, particularly between children in the lowest income category (whose families have incomes below the poverty line) and those in the highest category (whose families have incomes over 300 percent of the poverty line). Specifically, only 64 percent of children in the poorest families are reported by their parents to be in very good or excellent health compared to 90 percent of those in the highest income category. We observe much less variation among children by their race/ ethnicity, parents’ education, and family income when we look at a 25 Table 3.2a Health Status of Children in California, by Race/Ethnicity, Level of Parent’s Education, and Family Income (Weighted %) Child’s current health status Poor Fair Good Very good Excellent Asian Race/Ethnicity Hispanic Foreign- Hispanic Black Born U.S.-Born White Parent’s Education Less More Than Than High High High School School School Family Income < 100% 100– 200– of 200% of 300% of Poverty Poverty Poverty > 300% of Poverty Total 10 15 7 11 31 31 60 52 1 13 32 20 34 1 7 12 29 51 02 2 14 10 30 25 18 64 37 000 5 2 12 17 9 23 29 26 22 48 62 42 1 7 23 22 47 1 3 15 27 54 11 25 8 16 28 25 62 53 26 Has condition that limits activity 4 8 5 11 9 6 9 7 8 10 7 67 Well-child doctor visit during past 12 months No One Two or more 37 36 28 20 42 38 40 31 29 30 35 35 39 42 40 28 21 30 32 36 35 38 40 34 30 24 32 38 31 31 35 39 25 36 36 41 37 23 27 Saw a dentist during past 12 months 83 89 56 79 84 SOURCE: 1999 National Survey of America’s Families. NOTE: Perentages may not sum to 100 because of rounding. 59 73 86 63 69 82 86 78 second indicator of children’s health—whether or not they have health conditions that limit their physical activity. Differences in children’s health on this indicator are smaller and do not follow consistent or expected patterns, suggesting that these health problems are not strongly linked to socioeconomic status. However, this indicator suggests that about one out of ten Hispanic children living in nonimmigrant families and the same proportion of children living in “near poor” families (whose incomes fall between 100 and 200 percent of the poverty line) experience these debilitating conditions. The indicators measuring children’s health visits with physicians or other health care professionals show a somewhat different association with family background. Black and Hispanic children in nonimmigrant families are as likely as other children to have received routine medical care once in the past year but are also more likely to have visited a physician or medical professional two or more times. This finding suggests that actual health concerns are driving the high level of contact with physicians. About 60 percent of Hispanic children with foreignborn parents who had seen a doctor two or more times in the past year were reported to be in less than excellent health, compared to 46 percent of Hispanic children with native-born parents, 45 percent of black children, 33 percent of white children, and 28 percent of Asian children (figures not shown in table). Although this question asks parents to report about “well-child” care, parents may be reporting all medical visits, including those for more serious or chronic conditions. It is also possible that this question is understood somewhat differently by different racial/ethnic groups. In contrast to the pattern of well-child care, white and Asian children had typically received dental care in the past 12 months. However, Hispanic children, particularly those whose parents were born outside the United States, were less likely than other children to have received any dental care. Only 56 percent of Hispanic children living in immigrant families had seen a dentist in the past year, according to their parents’ reports, compared to 89 percent of black, 84 percent of white, and 83 percent of children of Asian descent. The relationship of parental education to well-child visits to doctors is also complex. About 42 percent of children whose parents do not have 27 a high school diploma did not receive routine medical care in the preceding year. At the same time, children whose parents have only a high school diploma received slightly more medical care than children whose parents have gone beyond high school in their education. We find less variation by family income in the percentage of children who received any care, although children living in higher-income families were less likely to have seen a physician or medical professional two or more times in the past year. Again, we assume that this is because the children in higher-income families are in better health. As expected, children from families with lower socioeconomic status, as defined by parental education and income, were much less likely to have seen a dentist in the past year. Since family income, parental education, and race/ethnicity are strongly interrelated, we next estimate the separate effect of each of the background factors, which allows us to adjust the percentages in a multivariate analysis. Although a bivariate crosstab analysis of children’s health outcomes by variables such as their race and ethnicity is informative, it fails to account for the influence of potentially confounding variables such as family income, parental education, and children’s age and gender. To account for such effects, we calculated adjusted percentages through a method that is similar to ordinary least squares (OLS) regression analysis in controlling for background factors but that yields substantively meaningful percentages that take account of confounding variables. When we show adjusted percentages, they are, in effect, regression estimates that control for children’s age, gender, parental race/ethnicity and immigration status, parental education, and family income.2 We used unweighted data when calculating the adjusted ____________ 2We used a procedure in Stata Version 7.0 that adjusts percentages through a command called adjmean. Using this command, we calculate the bivariate relationship between each outcome variable and each demographic or socioeconomic status variable of interest. This allows us to predict the likelihood of children falling into each category of the outcome variable while holding all other demographic or socioeconomic covariates at their mean. Dummy variables are created for each category of the outcome variable, and the outcomes are expressed in terms of an adjusted mean that falls between 0 and 1. These means are then multiplied by 100. 28 percentages because the weights incorporate some of the same variables that are being used in the procedure for adjusting the percentages. This adjusted-percentages procedure allows us to evaluate the unique contribution of each component of children’s demographic and family background characteristics. In other words, the adjusted percentages show the influence of race/ethnicity on children’s health outcomes, net of gender, age, parental education, and family income. Similarly, we observe the effect of parental education and family income net of other characteristics. If we continue to see differences between children after these other characteristics are controlled for, we can have more confidence that the differences are meaningfully related to children’s membership in particular subgroups. In Table 3.2b, we look at the same associations between children’s health status and health visits and their family characteristics reported in the last table. However, the numbers in this table show the net effect of each characteristic on children’s health outcomes. As these findings suggest, race and ethnicity continue to influence parents’ health evaluations of their children once we account for other characteristics of children and their families although differences between groups are reduced somewhat. White children are the most likely to receive excellent health ratings from their parents, closely followed by children of native-born Hispanic parents, blacks, and Asians. Hispanic children whose parents were born outside the United States are least likely to receive excellent ratings and most likely to receive poor or fair health ratings from their parents. We also find that both parental education and family income are independently related to the health status of children, net of other characteristics. After taking children’s demographic and family characteristics into account, children’s health status rises steadily as their parents’ education and family income go up. As these data show, children’s health status is sharply differentiated by both parental education and family income, suggesting that parents’ knowledge and access to health services both play an important part in their child’s health outcomes. 29 Table 3.2b Health Status of Children in California, by Race/Ethnicity, Level of Parent’s Education, and Family Income (Adjusted %) Child’s current health status Poor Fair Good Very good Excellent Asian Race/Ethnicity Hispanic Foreign- Hispanic Black Born U.S.-Born White Parent’s Education Less More Than Than High High High School School School Family Income < 100% 100– 200– of 200% of 300% of Poverty Poverty Poverty > 300% of Poverty Total 21 1 1 1 2 1 1 0 1 1 3 7 10 5 3 11 5 3 9 7 4 16 8 28 11 11 28 16 11 20 18 17 31 32 21 29 24 19 28 26 24 24 27 47 53 40 53 61 41 50 59 47 50 51 11 46 12 15 25 28 58 51 30 Has condition that limits activity 78 5 9 10 9 11 8 9 11 8 67 Well-child doctor visit during past 12 months No 45 21 35 31 40 40 35 35 33 39 38 One 30 43 39 36 37 31 37 40 40 34 37 Two or more 25 37 26 33 23 29 27 25 27 27 25 35 34 38 37 27 29 Saw a dentist during past 12 months 73 90 67 79 79 67 72 82 70 68 79 SOURCE: 1999 National Survey of America’s Families. NOTE: Perentages may not sum to 100 because of rounding. 82 78 It is interesting to note that parents’ reports of “objective” health conditions that limit children’s activities do not follow the same pattern as their health ratings. Again, we find little variation in these conditions by children’s race and ethnicity, except for children of Hispanic immigrants for whom it is lower. Similarly, parents’ education does not appear to be associated with children’s health limitations and family income appears to be only slightly related to these limitations. Although somewhat puzzling, this finding might suggest that parents with lower socioeconomic status or who are Hispanic immigrants may underreport their children’s chronic conditions. As described above, some of these limiting conditions may be temporary and may not compromise children’s overall health status in the eyes of their parents. After taking other factors into account, we find that children whose parents are in the lowest education category are more likely to have had no contact with a physician or health professional for routine care. However, they are also slightly more likely to have received care two or more times. Children whose parents are Asian, foreign-born Hispanic, and white are less likely to have seen a physician, and black and Hispanic children in nonimmigrant families are more likely to have seen a physician two or more times. It is of particular concern that Hispanic children in immigrant families are not receiving routine care because far fewer are described as being in very good or excellent health. Although the number of Asians in the study is too low to separate reliably children into subgroups by ethnicity and origin, we suspect that variation across these subgroups is substantial (see Reyes, 2001). We see a somewhat different pattern in the adjusted percentages for family income than we saw in the unadjusted figures. In particular, we find that children whose families have incomes between 100 and 300 percent of the poverty line are less likely to have received routine medical care than those at the higher and lower end of the income spectrum. Some of these parents do not qualify for state-subsidized health insurance (i.e., those between 250–300 percent of the poverty line) and others may not know that they qualify, thus limiting their contact with physicians. Again, another possible explanation is that higher-income parents may not feel that their child needs a routine health visit because they are in better health. 31 In the adjusted percentages, we continue to find that black children are the most likely to have seen a dentist in the last year, followed by white children and Hispanic children living in nonimmigrant families, and Asian children. Hispanic children in immigrant families are the least likely to have had a dental visit in the previous 12 months. The disparities between children of foreign-born Hispanic parents and those of other racial/ethnic groups are reduced when children’s other characteristics are held constant; however, they remain pronounced. As these findings indicate, Hispanic children in immigrant families and Asian children are less likely to have had a dental visit than other ethnic groups, especially blacks. Again, we suspect that children of foreign-born Asian parents have fewer dental checkups than other Asian children, but the numbers are too small to provide a reliable estimate of this difference. As might be expected, both family income and parental education are positively related to dental checkups, net of other characteristics. About two-thirds of children whose parents are high school dropouts and of poor or near-poor children have seen a dentist in the previous 12 months. This compares to over four-fifths of children whose parents have had post-high school education and whose families have incomes above 300 percent of the poverty line. When we adjust for other factors, these figures show a troubling pattern: Children in immigrant Hispanic, Asian, less-educated, and poor or near-poor families have limited access to dental care. Conclusion About 36 percent of children in California have not received routine medical care in the past year. Consistent with previous studies, this chapter reveals a strong disparity between children’s health needs and their use of health services that is strongly related to their age. As children’s health concerns and problems increase, especially in adolescence, their health visits drop off sharply. The deficit in both physician and dental care during the adolescent years is especially pronounced for males. Although adolescent males have greater health needs, they tend to use services less than adolescent females and younger children generally. 32 Children’s race and ethnicity and the socioeconomic characteristics of their families are also strongly associated with their health outcomes. Children living in Hispanic immigrant families, those living in poor families, and those whose parents are less educated are in worse overall health than other children. Although poor children seem to have comparatively good access to medical care, children with foreign-born Hispanic and less-educated parents often have not had contact with a physician or other medical health professional for well-child care in the last year. When we adjust for other factors, Asian children also have a lower likelihood of receiving routine medical care. Children are more likely to have received routine medical care when they live in families at higher and lower ends of the income spectrum. It is possible that the children of better-off parents and who are white are not getting enough routine health care; however, they are likely not to need as much care as other children. It is also possible that poor children are more likely to have health insurance coverage than children whose families have incomes between 100 and 300 percent of the poverty line. At the same time, children living in lower socioeconomic status families are far less likely than other children to have received routine dental checkups. More than 30 percent of children whose parents are Hispanic immigrants, whose parents have the lowest levels of education, or who live in poor and near-poor families have not seen a dentist in the past year. 33 4. Emotional Adjustment Most parents care deeply about their children’s emotional state and ability to get along with others—qualities that also matter greatly in how well children function inside the home, in the community, and in school. For example, behavioral problems can disrupt the learning of children and their classmates as well as increase the burden on teachers and staff. Children’s emotional and social adjustment problems are also strongly associated with subsequent mental health disorders, substance abuse, and problems with the criminal justice system. Thus, concern about children’s maladjustment by parents, teachers, and other caregivers is well justified. Moreover, the high social and economic costs to the larger society that are incurred when children function poorly make this area an important issue for public policy. As a result, considerable public and private resources are devoted to ensuring that the mental health needs and behavioral adjustment problems of children in California and the rest of the nation are adequately addressed (Carnegie Corporation of New York, 1996; Wolfe, 1995). As described in Chapter 2, we examine two general indicators of children’s socioemotional functioning. The first is a commonly used checklist of items to tap different domains of age-appropriate social and emotional adjustment. The first measure combines six NSAF items into a single scale to measure each child’s overall level of problems. We know from previous studies that a scale of behavioral problems that sums up particular symptoms of maladjustment is a more trustworthy predictor of children’s likelihood of encountering emotional or social problems in the future than any of the single items alone. When put together in a scale, the indicators provide a reliable measure of the child’s overall 35 behavioral adjustment.1 The NSAF includes items to measure six problem behaviors. Three of these items were asked about children ages six and older. An additional set of three items was asked about older children and another, three-item set was asked about younger children.2 The checklist was not administered for children below the age of six. To construct this scale, we separated the children into three categories depending on the frequency and severity of their problems. Well-adjusted children or those with few problems are reported not to experience any problems “often” or to show problems on more than two items “sometimes.” Poorly adjusted children or those with severe problems are reported to often have problems on at least two of the six items or to sometimes have problems on at least four of the six items. Children with moderate problems are reported to fall into these two extremes. It is important to keep in mind that parents are asked to assess their children’s behavior, so this measure is likely to reflect primarily children’s behavior within the household. At the same time, parents probably also base their evaluations on reports from other adults, such as relatives, neighbors, and teachers, who can assess their children’s behavior outside the home. The other indicator is a single question in the NSAF that asks ____________ 1A scale is the sum of the scores of the individual variables. In generating the scale, a score was created for every observation for which there was a response to at least one variable and divided by the variables over which the sum was calculated (StataCorp, 1997). Cronbach’s alpha measures the reliability of a scale. Specifically, it measures how well the variables measure a single, unidimensional latent construct. As the average interitem correlation increases, the alpha will also increase. The alpha is also affected by the number of variables in the scale (UCLA Academic Technology Services, 2001). 2We created one subscale for children ages 6–11 by taking the sum of six related items: doesn’t get along with other kids, can’t concentrate for long, has been sad or depressed, feels worthless or inferior, has been nervous or tense, and acts too young for his age. The alpha, or reliability score for this scale is .77. We create a second subscale for children ages 12–17 by taking the sum of another six related items: doesn’t get along with other kids, can’t concentrate for long, has been sad or depressed, has trouble sleeping, lies or cheats, and does poorly at school work. We created the scale for children’s behavioral problems by taking the sum of these two subscales and splitting it into three categories. The alpha for this scale is .73. 36 whether children have received services from a mental health counselor in the past 12 months. Gender and Age Most research shows that boys and girls manifest different symptoms of mental health problems (Maccoby, 1998). Boys are inclined to “act out” their problems whereas girls are more likely to respond to problems by withdrawing. Overall, boys tend to show more symptoms of poor mental health and are more likely than girls to engage in problem behaviors in childhood and adolescence. Generally, problems for both boys and girls increase with age. Although we can examine gender differences, the particular items employed in NSAF are designed to be age-specific, possibly making it less likely that we would detect increases in problem behavior in higher age groups. Our findings reveal that over two-thirds of the girls in California do not have notable behavioral problems and that only 8 percent reveal severe problems, according to their parents’ reports (see Table 4.1). By contrast, a somewhat lower percentage of boys are reported to have no behavioral problems, and 13 percent have severe problems. These results are consistent with previous findings on children’s emotional health. Contrary to our expectations, however, the results in Table 4.1 show no increase in problem behavior as children move from early childhood to adolescence.3 Looking at the proportion of children who have received mental health services in the previous 12 months, we observe some discrepancies between children’s need for services (as indicated by the responses to the behavioral-problem checklist) and their receipt of services (Table 4.1). Although parents report that boys have more severe symptoms than girls, boys are no more likely to have received mental health services in the previous 12 months. Overall, only 5 percent of boys and girls have received any counseling. Among children with the most serious symptoms, approximately 25 percent of girls and 23 percent of boys have ____________ 3Questions about children’s behavioral problems were not asked for children ages five and under. 37 Table 4.1 Children’s Behavioral Problems, by Gender and Age (Weighted %) Child behavioral problems scale Low Moderate Severe Gender Age Group Female Male ≤ 5 6–11 12–17 Total 69 64 22 23 8 13 66 67 23 23 11 11 66 23 11 Received mental health services in past 12 months 5 52 5 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. 6 5 received mental health services (figures not shown). Receipt of services is almost identical among children in middle childhood and adolescence, with about one in twenty receiving any treatment in the previous year. Not surprisingly, the percentage of children ages three to five who have received service is even lower, at 2 percent.4 Family Characteristics Abundant evidence demonstrates that children living in families with low socioeconomic status are much more likely to experience behavioral problems than other children (National Research Council, 1993; Wolfe, 1995). There are many reasons for this. Poverty itself is linked to higher levels of family stress. However, low-income parents also have fewer resources to address problems when they arise. Parents with less income and education may also have a greater incidence of social and mental health problems that in turn affect their parenting skills or their ability to respond to children’s needs. Children in low-income families are also more likely to be living in single-parent households and thus to experience greater family flux and strain resulting from such living arrangements. ____________ 4The question about mental health services was not asked for children under age three. 38 Research that has examined whether there is a relationship between race/ethnicity and behavioral problems in children is much less conclusive. Some racial/ethnic groups may be more prone to certain types of problems than others, but there is no consensus that the level of problem behavior among children, especially younger children, varies by racial/ethnic group, particularly when socioeconomic status is taken into account (Furstenberg et al., 1999). Even before we adjust for children’s other characteristics, we do not observe large racial/ethnic differences in our index of children’s behavioral adjustment (see Table 4.2a). The major difference across these groups is that Asian children have fewer severe problems than other children. There are clearer differences in children’s problem behavior when we examine this indicator by their parent’s educational level and family income, confirming the results of previous studies. According to their parents’ accounts, children living in poor households and who have less-educated parents are more likely to experience problems. It should be noted that the vast majority of children—including those living in families with the lowest incomes and levels of education—are not encountering any serious problems. Still, one in seven of those children whose parents did not complete high school and almost one in five of those children living in the poorest families have severe behavioral problems that might put them at risk of experiencing social and mental health problems in later life. After we adjust for the association between children’s demographic and family characteristics (in Table 4.2b), the relationship between parental education and children’s behavioral adjustment diminishes. We still see that the children of parents who had education beyond high school do slightly better than those whose parents had less education. The adjusted percentages show that the relationship between children’s behavioral problems and family income persists, but this is pronounced only for very poor children. Thus, it appears that problem behaviors among children, at least as reported by their parents, are relatively evenly distributed across racial/ethnic and socioeconomic groups, except for children living in very poor families where about 15 percent of children are reported to have severe problems. 39 Table 4.2a Children’s Behavioral Problems, by Race/Ethnicity, Parent’s Education, and Family Income (Weighted %) 40 Child behavioral problems Low Moderate Severe Asian Race/Ethnicity Hispanic Hispanic Foreign- U.S.Black Born Born White Parent’s Education Less More Than Than High High High School School School Family Income 100– 200– < 100% 200% 300% of of of Poverty Poverty Poverty > 300% of Poverty Total 69 67 61 64 69 60 60 72 52 64 66 75 66 24 21 28 24 20 25 28 19 30 26 24 17 23 6 12 11 12 11 14 12 9 19 10 9 8 11 Received mental health services in the past 12 months 34 2 6 6 46 4 5 54 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. 55 Table 4.2b Children’s Behavioral Problems, by Race/Ethnicity, Parent’s Education, and Family Income (Adjusted %) 41 Child behavioral problems Low Moderate Severe Race/Ethnicity Hispanic Hispanic Foreign- U.S.Asian Black Born Born White Parent’s Education Less More Than Than High High High School School School Family Income 100– 200– < 100% 200% 300% of of of Poverty Poverty Poverty > 300% of Poverty Total 61 68 28 21 11 12 65 26 9 67 22 11 66 65 62 68 56 22 24 26 22 30 12 12 12 10 15 68 65 24 24 9 11 69 65 20 24 11 11 Received mental health services in the past 12 months 54 3 6 5 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. 5 5 44 44 65 When we examine children’s use of mental health services, we find that not all racial/ethnic groups are equally likely to access services (see Table 4.2a). For example, 6 percent of white children and those whose parents are native-born Hispanics, 4 percent of black children, 3 percent of Asian children, and only 2 percent of children with foreign-born Hispanic parents have used mental health services. However, these findings are more illustrative of the fact that few children overall are receiving services, particularly those in need. Although problems occur disproportionately to children of poorer and less-educated parents, we can see that their use of mental health services is no greater than that of children whose parents are more affluent. Indeed, after adjusting for other characteristics, Table 4.2b indicates that children living in the highest-income families actually use services more often than those in lower-income families. When we examine family income simultaneously with children’s need for and use of services, we find that children who live in families with incomes below 200 percent of the poverty line and experience severe problems have a much lower probability of receiving mental health services than children in families with higher incomes (see Figure 4.1). Only half as many children in lower-income families with serious 35 30 < 200% of poverty ≥ 200% of poverty 25 29 Percentage 20 15 15 10 5 0 Receiving mental health services Figure 4.1—Children with Severe Behavioral Problems Who Received Mental Health Services in Past 12 Months, by Family Income (Weighted %) 42 problems received mental health services in the past 12 months as children in higher-income families (15 percent compared to 29 percent). Similarly, children with high levels of problem behavior are much less likely to be treated if their parents have lower levels of education or are nonwhite (figures not shown). These findings indicate that children most at risk for experiencing problems in later life are being underserved because their parents may lack knowledge about mental health services, have fewer social connections, or have much more limited economic resources to obtain treatment. Conclusion Although only a small minority of children in California are experiencing severe behavioral problems, as reported by their parents, the rates of problem behavior are significant enough to be of public concern. Within subgroups of children, we find that the level of children’s need for mental health services does not correspond with their receipt of these services. Namely, males and children from lower socioeconomic status families show more problems than other children but do not receive more treatment. Particularly troubling is the low level of treatment provided to children of lower-income families whose parents say the children are displaying serious symptoms of problem behavior. Only about 16 percent of children living in or near poverty who have serious behavioral problems are receiving services. Children with severe problems are almost twice as likely to receive treatment if their families are at least 200 percent above the poverty line. 43 5. Educational Attachment Previous studies have shown that children’s attachment to education begins to form early in life—often before they enter elementary school (Entwisle, Alexander, and Olson, 1997). This research indicates that children’s experience in the home sets the stage for their success in school. California policy recognizes this fact through its support of preschool programs that aim to reduce disparities in children’s preparation for school based on their family backgrounds. It is also well known that the quality of children’s early schooling establishes the skills and motivation they need to perform well throughout their educational careers. Indeed, children’s attachment to school is an important precondition for success in middle school, high school, and beyond. Alternatively, children who lack confidence in and commitment to schooling are more likely to drop out of high school and less likely to go to college (Neisser, 1986; Lamborn et al., 1992). As discussed in Chapter 2, the NSAF included a series of items to measure children’s attachment to school, as reported by parents. The first of these is a four-item scale that asks parents to rate their children’s engagement in school. To construct the scale for children’s school engagement, we took the sum of four related items: cares about doing well in school, works on schoolwork only when forced to, does just enough schoolwork to get by, always does homework. We then split the scale into four categories to represent low, medium, high, and very high engagement.1 A second indicator measures the frequency of skipping school among children age 12 and older. On a third indicator, parents of these older children in the sample were also asked whether their child had been suspended or expelled from school in the past year. This chapter examines how the indicators of school attachment vary by characteristics of California’s children and their families. ____________ 1The alpha for the scale of school commitment is .71. 45 Gender and Age Previous research finds that children’s attachment to school varies greatly within demographic subgroups of the population. In particular, a large literature demonstrates that boys have more difficulty than girls engaging in school and show higher levels of problem behavior in school. Moreover, these gender differences widen with age, particularly as children move from primary to middle school. As young people gain more autonomy from their parents, and parents tend to have less contact with teachers, children’s disengagement from school typically rises. These differences reach their highest levels during the adolescent years when children enter high school (Eckstrom et al., 1987; Natriello, 1987). The distribution of the three indicators of school engagement by gender and age is shown in Table 5.1. The NSAF results closely mirror findings of previous studies indicating that males are noticeably more likely than females to have lower levels of school attachment. In California, many more males than females (38 percent compared to 22 Table 5.1 Children’s School Engagement, Frequency of Missing School, and Suspension/Expulsion Rates, by Gender and Age (Weighted %) School engagement scale Low Moderate High Very high Gender Age Group Female Male 6–11 12–14 15–17 12–17 Total 4 94 18 29 23 30 32 36 48 30 37 10 7 24 24 26 31 40 38 Times skipped school past 12 monthsa 0 81 79 1 87 2 or more 11 13 88 71 6 10 6 19 80 8 12 Suspended/expelled in past 12 monthsa 10 17 8 20 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. aThese questions were not asked of the 6–11 age group. 13 46 percent) have only low or moderate levels of engagement in school. Males are slightly more likely than females to have skipped school two or more times in the past year (13 percent compared to 11 percent). Males are also much more likely to have been suspended or expelled in the past 12 months (17 percent compared to 10 percent). Research has found that dropping out of high school is somewhat more common for males than females and that all of these indicators of school disengagement predict high school dropout (U.S. Department of Education, 2002). As noted above, school attachment generally declines as children move from primary school to middle school and from middle school to high school. We find that low engagement is more than twice as high among children ages 12 and older as among those ages 11 and younger (10 percent compared to 4 percent) in California. The question about skipping school was asked only of children in the oldest age group (ages 12 and above), in part because it does not occur frequently among children in elementary school. About 8 percent of students between the ages of 12 and 17 have skipped school at least once, and an additional 12 percent have skipped school two or more times in the past year. Frequent school skipping rises sharply among students moving from middle school to high school, increasing from 6 percent among children who are ages 12 to 14 to 19 percent of children between ages 15 and 17. Suspension and expulsion rates also tend to rise with age. Again, questions were asked for children ages 12 and older in the NSAF; however, we can compare the students who have likely reached high school (ages 15 to 17) with those still likely to be in middle school (ages 12 to 14). The proportion of children who have been suspended or expelled in the past 12 months climbs from 8 percent among 12 to 14 year olds to 20 percent among 15 to 17 year olds (see Table 5.1). Overall, these indicators show a troubling pattern of school detachment that is especially marked for males and older students. Figure 5.1 plots the patterns of skipping school (two or more times) and suspension or expulsion in the past year for males and females separately by age. As this figure shows, the increase by age is relatively modest for females but is quite pronounced for males. About 23 percent of high school age boys have skipped school more than twice in the past year and 25 percent have 47 30 Skipped two or more times in past 12 months (male) Suspended/expelled in last 25 12 months (male) Skipped two or more times in past 12 months (female) 20 Suspended/expelled in last 12 months (female) 15 25 23 15 14 Percentage 10 9 8 7 54 0 12–14 15–17 Age group Figure 5.1—Children Who Skipped School Two or More Times or Were Suspended or Expelled in the Past 12 Months, by Gender and Age Group (Weighted %) been suspended or expelled in the previous 12 months, compared to 15 percent of girls who have skipped school and 14 percent who have been suspended or expelled. Family Characteristics Children’s school attachment is also strongly associated with their family backgrounds as shown in Table 5.2a. Children’s scores on the school engagement index are higher for white and Asian children than for Hispanic or black children, indicating that children in the latter groups are more likely to be withdrawing from school. Almost threequarters of white and Asian children are reported be highly or very highly engaged in school, compared to about two-thirds of black and all Hispanic children. A higher percentage of foreign-born Hispanic parents report that their children are highly engaged in school (69 percent) than native-born Hispanics (63 percent). 48 Table 5.2a Children’s School Engagement, Frequency of Missing School, and Suspension/Expulsion Rates, by Race/Ethnicity, Parent’s Education, and Family Income (Weighted %) School engagement scale Low Moderate High Very high Asian 3 24 30 43 Race/Ethnicity Hispanic Hispanic Foreign- U.S.Black Born Born White Parent’s Education Family Income Less More 100– 200– Than Than < 100% 200% 300% > 300% High High High of of of of School School School Poverty Poverty Poverty Poverty Total 76 8 8 5 10 5 4 7 7 86 31 26 29 19 31 20 22 32 27 15 21 26 29 35 35 29 33 32 30 31 27 37 32 32 33 34 28 44 31 38 43 33 38 41 40 36 49 Times skipped school in past 12 months 0 93 79 74 73 83 77 70 86 69 76 82 87 80 1 0 10 8 10 8 8 12 5 12 4 9 7 8 2 or more 7 12 18 18 9 15 18 8 19 20 9 6 12 Suspended/expelled in past 12 months 3 18 19 18 10 19 23 6 29 20 8 4 13 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. Consistent with previous research, our analysis of the association between children’s socioeconomic status and their school attachment (presented in Table 5.2a) reveals that children of less-educated and lower-income families experience greater problems in school. Children whose parents attained some postsecondary education were notably more engaged in school than those whose parents were only high school graduates. Children of high school graduates, in turn, were more engaged than children whose parents never completed high school. Similarly, children’s engagement in school generally rises with family income, although this pattern is not entirely consistent. We see that approximately 33 percent of children living in poor families are reported to have a very high level of engagement compared to about 40 percent of children in families with incomes at least 200 percent above the poverty line. At the same time, about 29 percent of children in the highestincome families are reported to have a low or moderate level engagement, compared to 36 percent of the families living in poverty. The same pattern is apparent when we look at voluntary school absences (see Table 5.2a). Skipping school one or more times is somewhat less prevalent among white children and much less prevalent among Asian children than it is among black and Hispanic children. On this same measure, we find that black children had only somewhat higher rates of skipping school two or more times (12 percent) and Hispanic children had sharply higher rates of doing so (18 percent), compared to white (9 percent) and Asian (7 percent) children. Skipping school occurs less frequently among children whose parents have attained education beyond high school than among children whose parents have a high school education or less. The strongest correlate of frequent school skipping is family income. Children in the two lowest-income groups have more than three times the rate of skipping school frequently as those in the highest-income group. Finally, we see similar patterns for school suspension or expulsion— our other indicator of problems in school—with Hispanic and black children and the most disadvantaged children showing more problems than other children. Almost twice as many black and Hispanic children than white children were suspended or expelled from school. The differential was even greater for Asian children, for whom suspension or 50 expulsion was a rare event. Children of the most educated parents are less than one-third as likely to have been suspended or expelled in the past year as children whose parents did not have education beyond high school. Family income is also strongly associated with children’s probability of being suspended or expelled from school. In the lowestincome group, 29 percent of children were suspended or expelled from school compared to just 4 percent of children in the highest-income group. After controlling for children’s other characteristics, family income is the strongest correlate of problem behavior in school for children in California, although it is only moderately associated with the measure of low engagement (see Table 5.2b). The differences we observed in children’s school engagement between white and nonwhite children are substantially reduced, after adjusting for other characteristics. It is interesting to note that Hispanic children whose parents were born outside the United States appear to be just as engaged in school as white children in the adjusted figures and are actually more engaged in school than Hispanic children living in nonimmigrant families.2 Differences by parents’ educational level also diminish, particularly among children whose parents have at least a high school diploma. Similarly, differences by family income also decline when parental education and race and ethnicity are held constant. The same pattern is evident for children’s voluntary school absence and suspension/expulsion. When we take account of children’s demographic and family characteristics, we observe smaller differences by race/ethnicity, although white children still appear less likely to skip school frequently. Similarly, in the adjusted percentages, differences by parental education are no longer as apparent. Somewhat unexpectedly, however, children whose parents have a high school education have a higher level of suspensions/expulsions than those whose parents have ____________ 2Previous research on immigrant children in California indicates that as time and generation in the United States increase, the time children spend on homework and their grade point average decrease (e.g., Rumbaut, 1995, p. 44). There is also some evidence to suggest that second- or third-generation Hispanic children become disaffected with the educational system and perform more poorly than foreign-born Hispanic children (Suárez-Orozco and Suárez-Orozco, 1995, p. 187). 51 Table 5.2b Children’s School Engagement, Frequency of Missing School, and Suspension/Expulsion Rates, by Race/Ethnicity, Parent’s Education, and Family Income (Adjusted %) School engagement scale Low Moderate High Very high Asian 6 28 35 31 Race/Ethnicity Hispanic Hispanic Foreign- U.S.Black Born Born White Parent’s Education Family Income Less More 100– 200– Than Than < 100% 200% 300% > 300% High High High of of of of School School Schoo Poverty Poverty Poverty Poverty Total 648 24 23 25 38 34 31 32 40 36 97 20 26 31 33 40 34 10 6 5 19 22 29 34 31 31 37 41 35 6 7 97 24 16 21 24 29 35 32 34 41 41 38 36 52 Times skipped school in past 12 months 0 81 83 77 77 1 13 8 10 2 or more 18 14 16 13 83 81 7 10 99 79 82 72 9 5 10 13 12 18 79 77 87 80 5 10 7 6 17 14 6 14 Suspended/expelled in past 12 months 8 13 10 13 12 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. 8 17 10 27 18 11 4 11 both more and less education. Family income continues to have a strong association with children’s school skipping and suspension or expulsion, with more than three times the number of children in the lowest-income categories skipping frequently and six times the number being suspended or expelled compared to those in the highest-income families. Conclusion Our results show that most children in California are highly engaged in school and are not experiencing problems in their classrooms, according to their parents’ reports. However, we find that children’s demographic characteristics are associated with low engagement in education and disaffection from school. Consistent with previous research, boys and older children are experiencing much higher rates of detachment. Family income remains very strongly related to children having problems in school, particularly in regard to voluntary and involuntary school absences. Although we discovered that black and Hispanic children have more problems in school than white and Asian children, this association appears to have much to do with their socioeconomic situations. It is important to stress that we are not indicating that low family income alone is the source of children’s school problems. Other conditions related to poverty may account for this association, such as the greater likelihood of children living in a poor school district, residing in a singleparent household, or having a parent with fewer psychosocial resources. 53 6. Pro-Social Involvement Both practitioners who provide services to children and researchers who study the effect of these services on children’s development have become increasingly interested in children’s involvement in pro-social groups such as clubs and associations, after-school activities, and recreational opportunities. There is growing evidence that children’s involvement in these kinds of activities outside formal schooling offers them many advantages (Carnegie Council on Adolescent Development, 1992; Eccles and Barber, 1999; Mahoney, 2000; Pierce, Hamm, and Vandell, 1999). These activities cultivate personal talents and interpersonal skills that play an important part in promoting good mental health and social functioning among children. Involvement in groups, lessons, and activities builds what sociologists refer to as cultural and social capital—knowledge about the way the world works, as well as social connections and sponsorship. For example, it provides children with the opportunity to interact with adults who may serve as mentors and counselors to children outside family and school settings. Finally, pro-social activities often shield children from risky behaviors and negative peer influences by providing them with extensive adult supervision and opportunities to engage in positive peer relationships. The measure of social involvement that we use is based on several questions in the NSAF that asked parents about whether their school-age children participated in clubs, associations, or sports teams, or took lessons. These different activities were summed up in an index of social involvement ranging from no involvement to participation in all three types of activities. We examine whether children participate in multiple activities not only because this variation may be beneficial to the children’s development, but also because children who participate in 55 more activities may be spending more supervised time with adults.1 However, we cannot determine how much time children spend in activities from the survey measure. Gender and Age Previous studies have shown that the nature of pro-social involvement sometimes differs for males and females. Girls are somewhat more likely to take lessons outside school, and boys are somewhat more inclined to participate in athletic activities. However, research gives us little reason to expect large gender differences in the level of activities in which children are engaged. Although the types of activities that children participate in change as they get older, the volume of activity does not decline from middle childhood to adolescence (Furstenberg et al., 1999). Our findings from the NSAF indicate that close to one-quarter of all children ages 6–17 in California are not involved in any activities such as clubs, teams, or lessons.2 The findings show little variation by gender or age in the number of activities in which children participate (Table 6.1). Although differences are not large between males and females, a higher percentage of girls than boys are reported not to participate in pro-social involvement (27 percent compared to 21 percent). This may be because more boys are involved in sports activities than girls (61 compared to 42 percent). Some girls may not participate in activities because parents exercise more control over their daughters than their sons in communities that are perceived to be dangerous. It is also possible that activities are more available to boys than to girls. The distribution in children’s level of activities is very similar by age group, indicating little decline in activities as children move from middle childhood into their adolescent years. In fact, children in the older age group are somewhat more likely than those in the younger group to engage in two to three activities. ____________ 1It is also possible that involvement in activities could have the negative effect of preventing children from spending time on their homework or from engaging in unstructured play. 2These questions were not asked for children ages five and under. 56 Table 6.1 Children’s Pro-Social Involvement, by Gender and Age (Weighted %) Number of prosocial activities 0 1 2 3 Gender Age Group Female Male 6–11 12–17 Total 27 21 23 24 32 36 36 31 28 30 28 31 14 12 13 14 24 34 29 13 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. Family Characteristics There is more reason to expect variations in children’s level of prosocial involvement by race/ethnicity and socioeconomic status. In particular, previous studies have shown that children in some locations have limited access to community-based programs (National Research Council and the Institute of Medicine, 2002; Wynn et al., 1988). In addition, lower-income families may be less likely to place their children in pro-social activities because they have less information, connections, and resources (e.g., time, money, and transportation) to avail themselves of clubs, recreational activities, and lessons even when they are potentially available. Conversely, more-educated and affluent parents often possess greater knowledge of where to find programs, have the means to get their children placed in programs, and can afford to pay for activities or lessons when they are not free of charge. They may also be more aware of the benefits of such programs for children and less fearful of letting children be supervised by other adults outside the home. Less information is available on the effect of race and ethnicity on social involvement. It is possible that language barriers might restrict access to such activities for the children of foreign-born parents. Hispanic and black children who live in disadvantaged communities may also be constrained by the availability of programs in their 57 neighborhoods. It would not be surprising, then, to discover racial/ethnic variations in the involvement of children in pro-social activities. We compare children’s social involvement by race/ethnicity, parental educational, and family income in Tables 6.2a and 6.2b. Looking first at basic distributions in Table 6.2a it is evident that large differences in the degree of pro-social involvement exist among children of different racial and ethnic groups, with children who have foreign-born Hispanic parents reported to have the lowest level of involvement. Fully 40 percent of Hispanic children in immigrant families, 29 percent of Hispanic children in nonimmigrant families, and 24 percent of black children are reported to participate in no activities compared to only 16 percent of white children and 12 percent of Asian children. At the other end of the index, about 50 percent of white children, about 45 percent of Asian children, and about 41 percent of black children participate in two or more types of activities, compared to 39 percent of Hispanic children in nonimmigrant families and 28 percent of Hispanic children in immigrant families. The lower participation of Hispanic children in immigrant families may be related to the educational levels of Hispanic parents who were born outside the United States (see Chapter 2). In fact, children of parents with less than a high school education have extremely low levels of participation in pro-social activities. Approximately 45 percent of these children are not involved in activities whereas about 22 percent participate in two or more activities. By contrast, only 14 percent of the children whose parents have more than a high school education are not engaged in activities and 52 percent are involved in two or more activities. The level of activity among children whose parents have only a high school diploma falls squarely between these two groups. Similarly, family income reveals a steep gradient of participation among children in lower- and higher-income families. For example, 43 percent of children in very poor families are reported to have no involvement whereas 25 percent participate in two or more activities; only 13 percent of children in the highest-income families have no involvement and 52 percent participate in two or more activities. These represent extraordinarily large differences in children’s participation rates 58 Table 6.2a Children’s Pro-Social Involvement, by Race/Ethnicity, Parent’s Education, and Family Income (Weighted %) 59 Race/Ethnicity Number of pro-social activities 0 1 2 3 Asian 12 42 32 13 Black 24 35 29 12 Hispanic Hispanic Foreign- U.S.- Born Born White 40 29 16 33 32 34 21 26 34 7 13 16 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. Parent’s Education Less More Than Than High High High School School School Family Income 100– 200– < 100% 200% 300% > 300% of of of of Poverty Poverty Poverty Poverty Total 45 27 14 43 28 21 13 24 33 34 34 32 35 34 34 34 14 30 34 16 26 31 36 29 8 9 18 9 11 14 16 13 Table 6.2b Children’s Pro-Social Involvement, by Race/Ethnicity, Parent’s Education, and Family Income (Adjusted %) 60 Race/Ethnicity Asian Black Number of pro-social activities 0 20 25 1 42 34 2 22 32 3 16 10 Hispanic Hispanic Foreign- U.S.- Born Born White 28 21 30 32 30 32 11 14 21 34 33 12 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. Parent’s Education Less More Than Than High High High School School School Family Income 100– 200– < 100% 200% 300% > 300% of of of of Poverty Poverty Poverty Poverty Total 37 26 16 33 25 20 19 20 33 35 32 31 36 36 32 37 21 31 35 24 29 30 36 30 8 8 16 12 11 13 13 13 and presumably in their opportunities for receiving the benefits of social involvement. The results presented in Table 6.2b correct for the mutual association of demographic and family characteristics. Again, the adjusted percentages show us the unique contribution of race/ethnicity, parental education, and family income on the level of children’s prosocial activities. As in our analysis of other indicators of children’s wellbeing, we observe much less variation in social involvement across racial/ethnic groups after children’s other demographic and socioeconomic characteristics are taken into account. On the other hand, both parental education and family income remain strongly related to children’s level of activities. The adjusted percentages show that the children of the least-educated parents are more than twice as likely not to engage in any pro-social opportunities and only half as likely to be in the highest category of social involvement as children whose parents have education beyond high school. Even taking account of parental education and other differences, family income is also strongly linked to children’s social involvement. About one-third of poor children have no involvement in activities compared to about one-fifth of children in the highest-income families. Similarly, slightly over one-third of poor children engage in two or more activities compared to almost half of children from the highest-income families. Thus, we find that socioeconomic status is strongly associated with the likelihood of children’s engagement in pro-social activities. As suggested above, several factors may be at work in producing these large variations in children’s social involvement, such as the socioeconomic status of their families. Access may be limited by the availability of services and the cost of services. We know from other research that the distribution of both extracurricular programs in schools and recreational programs in neighborhoods are much more available in affluent than in disadvantaged neighborhoods. When services are not freely available, many poor parents cannot afford to pay for them. And even when they are available, less-educated and lower-income parents may not know about them, may have problems arranging to transport their children to the programs, may have fears about letting their children go outside the household or immediate neighborhood, or may 61 need their children to provide in-home services. Any and all of these factors could help to account for the large differences we observe. Conclusion Close to one-quarter of all children ages 6–17 in California are not involved in any activities such as clubs, teams, or lessons. Consistent with previous research, we do not find that the level of children’s involvement varies much by the their gender or age; however, girls are somewhat more likely than boys to not be involved in any activities. We find much larger variations in involvement by children’s socioeconomic status. Both income and education are strongly related to the likelihood of children participating in pro-social activities. Differences in participation by children’s race and ethnicity are greatly reduced after these and other factors are controlled for. Children from poor families and those with less-educated parents are strikingly less likely to engage in any activities whereas those from more affluent families and living with a well-educated parent are highly likely to engage in at least some activities. About half of the most advantaged children participate in two or more types of activities, compared to about one-third of the least advantaged children. We suspect that the sources of these differences stem from both the limited resources of parents and the lower availability of programs for children in poorer communities. 62 7. How California’s Children Compare to Those in the Rest of the Nation In previous chapters, we examined variations in health, emotional adjustment, school attachment, and pro-social involvement among various subgroups of children within the state of California. This chapter investigates how children in California are faring relative to those living in the rest of the nation. Reports by the Urban Institute have examined this topic but have not taken into account how the composition of the population in California compares to that in other parts of the United States, therefore making it difficult to assess whether any variations in children’s well-being they observe are due to demographic differences or to other conditions distinctive to California such as public policies or the availability of services. More than one in eight children in the nation live in California. Because such a large proportion of all U.S. children live in the state, we might expect to find only small differences between the welfare of the children in California and that of children in the rest of the nation. On the other hand, the extremely diverse nature of the population and the precarious state of many recent immigrant families lead us to suspect that some differences related to children’s demographic, social, and economic characteristics might exist. In particular, to the extent that poverty, low parental education, and Hispanic ethnicity and nativity are associated with children’s outcomes, we expect that children in California are not doing as well as children in the rest of the nation. Our analysis in this chapter shows some support for each of these expectations. In some cases, California’s children resemble and at other times differ from those in the rest of the country, depending on the outcome we examine. 63 Physical Health Our findings indicate that the overall health status of children in California is somewhat inferior to that of children in the rest of the nation. About 22 percent of California parents report that their children’s health is less than very good compared to 17 percent of parents living in the rest of the nation (see Table 7.1). Most of this difference results because a higher proportion of parents in California than in the rest of the nation say that their children’s health is “good” instead of “very good.” The incidence of extreme health problems among children is not greater in this state than elsewhere. Consistent with parents’ overall evaluations, the proportion of those reporting that their children have a health problem that limits their activity is similar in California and the rest of the nation (8 percent compared to 9 percent). Table 7.1 Health Status of Children in California and the Rest of the United States (Weighted %) Child current health status Poor Fair Good Very good Excellent Child has condition that limits activity Rest of the California United States 11 54 16 12 25 27 53 56 89 Well-child doctor visit during past 12 months No 36 34 One 37 41 Two or more 27 25 Saw dentist in past 12 months 77 79 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. 64 It is important to note that about 36 percent of children in California and 34 percent of children in the rest of the nation had not seen a doctor or other medical health professional for a well-child visit in the previous 12 months. Similarly, approximately 23 percent of California children had also not seen a dentist in the past 12 months compared to 21 percent in the rest of the nation. These findings show that a sizable minority of children in California do not receive routine, preventive health and dental care as is also true of children elsewhere in the United States. Given the general similarities in the health status of children in California and the rest of the nation, it is perhaps not surprising that California parents indicate that their children are seeing physicians only slightly more frequently for routine health care than parents elsewhere. Approximately 27 percent of children in the state had seen a doctor or other medical health professional two or more times in the past 12 months compared to one-quarter in the rest of the nation. We next examine whether these modest differences in children’s health status are maintained after accounting for differences in the characteristics of children and their families. When we make adjustments to control for demographic differences between California and the rest of the nation, no sizable disparities remain. Thus, the statelevel difference in health is apparently attributable to the different demographic composition of California’s families compared to that of the rest of the nation (see Figure 7.1.) Emotional Adjustment When we turn to indicators of the mental health and emotional wellbeing of children, we again compare the needs of children in California to those of children in the rest of the country. As shown in Table 7.2, children’s scores on the behavioral problem scale are almost identical in California and in the rest of the country.1 Slightly more than one-tenth of children have high levels of emotional or behavioral problems whereas about two-thirds have low levels of these problems, according to their parents’ reports. At the same time, we see that the proportion of children in California who received some form of mental health counseling in the ____________ 1See Appendix Table A.1 for the distribution of individual scale items. 65 Percentage 60 California 50 Rest of the U.S. 40 56 55 30 28 25 20 10 11 0 Poor 44 Fair 14 13 Good Health status Very good Excellent Figure 7.1—Children’s Health Status in California and the Rest of the United States (Adjusted %) Table 7.2 Children’s Behavioral Problems in California and the Rest of the United States (Weighted %) Behavioral problems scale Low Moderate Severe Rest of the California United States 66 67 23 22 11 11 Received mental health services in past 12 months 5 7 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. 66 past year is a little lower than elsewhere—5 percent in this state compared to 7 percent in the rest of the nation. In California as elsewhere, most children who are displaying serious behavioral adjustment problems are not being treated. However, as shown in Figure 7.2a, the situation is worse for children in California where only about 21 percent of the children with severe symptoms have received mental health services compared to 30 percent of children living in other states. Figure 7.2a also shows that children with moderate symptoms in California are about two-thirds as likely to be served as the children in the rest of the nation. In particular, 8 percent of children in California have received mental health services compared to 12 percent of children elsewhere. These differences between treatment are disturbing, particularly for children with severe problems, because it means that children with the same level of behavioral problems are less likely to be receiving treatment in California. 35 California 30 Rest of the U.S. 30 25 21 20 Percentage 15 12 10 8 5 23 0 Low Moderate Behavioral problem Severe Figure 7.2a—Children Who Received Mental Health Services in Past 12 Months, by Severity of Behavioral Problems, California and the Rest of the United States (Weighted %) 67 After taking into account the demographic and family characteristics of children within and outside the state, the gap between children’s need for mental health services and use of those services remains high in California, with less than one-quarter of children with severe problems receiving services (see Figure 7.2b). The disparity between California and other states closes only slightly when we look at use of mental health services among children with severe problems, which drops from a 9 to a 7 percent difference. It would seem that fewer children in California are getting needed mental health services than in other states, even taking into account racial/ethnic, educational, and family income differences. This indicates the possibility that California is performing less well than many other states in providing mental health services to children at risk of later problem behaviors. 35 California 30 Rest of the U.S. 31 25 24 Percentage 20 15 10 54 2 0 Low 12 6 Moderate Behavioral problem Severe Figure 7.2b—Children Who Received Mental Health Services in Past 12 Months, by Severity of Behavioral Problems, California and the Rest of the United States (Adjusted %) 68 Educational Attachment The NSAF included a variety of measures designed to assess the level of attachment to school among children who were attending primary school or above, that is, children ages six and older. As discussed in Chapter 5, four items tapping the child’s interest in and commitment to school are included in the survey that was combined into an index measuring the children’s educational attachment (see Table 7.3).2 We find that children’s engagement in school in California is similar to that in the rest of the nation. Over two-thirds of parents in California and the rest of the nation indicate that their children have a high or very high level of engagement in school. However, a somewhat higher proportion of children are not as highly involved in school in California as children in other states. About 31 percent of children in California are Table 7.3 Children’s School Engagement in California and the Rest of the United States (Weighted %) School engagement scale Low Medium High Very high California 7 24 32 38 Rest of the United States 8 20 32 39 Times skipped school in past 12 months (ages 12–17) 0 1 2 or more 80 8 12 85 6 9 Expelled/suspended from school 13 14 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. ____________ 2See Appendix Table A.2 for the distribution of individual scale items. 69 reported to have low or moderate school engagement in comparison to 28 percent of children elsewhere. As can be seen in Table 7.3, California’s children are about as likely as those elsewhere to exhibit problems that require suspension or expulsion (13 and 14 percent, respectively). However, when we look at the number of times the child skipped school in the previous year, we observe larger differences between California and the rest of the nation. Twenty percent of California children skipped school at least once compared to 15 percent of children elsewhere. The percentage of children who skip school frequently (two or more times) is also more common in California, 12 percent compared to 9 percent. The modest differences in skipping school disappear, however, when we adjust for demographic variations between California and the rest of the nation (not shown). The adjusted percentages are almost identical on all the indicators of educational attachment when demographic differences are taken into account. Thus, it appears that California’s children are faring about as well (or as poorly) in regard to their school behaviors as children in other parts of the country. Pro-Social Involvement The NSAF also measures children’s involvement in activities outside school. To construct the index for social involvement, we took the sum of the three related items: participating in clubs, taking lessons after school, and playing on sport teams (Table 7.4). When we look at the individual measures of social involvement—we find sizable differences in the proportion of children participating in clubs, with children in California much less likely to be participants than children elsewhere (see Appendix Table A.3). More modest but still substantially lower numbers of children were on sports teams in California than elsewhere. On the other hand, California’s children were somewhat more likely to be taking lessons to develop a talent in such areas as music, dance, language, or computers. When combined into a single index, the results suggest a modest deficit in activities aimed at developing children’s social skills and personal development. Slightly less than one-quarter of California’s children are involved in none of the three activities compared to one-fifth 70 Table 7.4 Children’s Pro-Social Involvement in California and the Rest of the United States (Weighted %) No. of prosocial activities 0 1 2 3 Rest of the California United States 24 20 34 31 29 33 13 16 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. of children in the rest of the nation. Whereas close to half of the children in the rest of the nation were in two or more activities, just over two-fifths of California’s children reached that level of involvement. After we take children’s characteristics into account in Figure 7.3, we find these differences largely disappear. California’s children are slightly less likely to be involved in three or more activities but equally likely to be involved in none. Conclusion Our comparison of children in California with children in the rest of the nation provides the general impression that children in the state are slightly worse off on a number of indicators of well-being. The health ratings of children in California are a little lower than those of children elsewhere. Children in California are also less likely to have seen a physician or other medical professional for a well-child visit or to have received dental or mental health care. The disparity between mental health needs and mental health care in California is particularly troubling for children with severe mental health problems. Equally disturbing is the fact that at least one-third of children in California and in the rest of the country had not received any routine health care and at least onefifth (23 and 21 percent) had not received dental care in the past year. 71 100 14 90 80 70 33 60 3 17 2 1 0 33 Percentage 50 40 33 30 31 20 10 20 0 California 19 Rest of the U.S. Figure 7.3—Children Who Participate in Pro-Social Activities, by Number, California and the Rest of the United States (Adjusted %) Children in California also have a slightly lower level of engagement in school, skip school more often, and are less involved in pro-social activities that cultivate children’s social and personal skills. Although we did not observe any differences in other realms of behavior, none of the indicators showed that California’s children were outperforming children in other states. The question of whether the differences are due to demographic variations between the population of California and the rest of the nation does not yield an entirely straightforward answer. For most indicators, the discrepancies disappeared when we took into account the demographic and socioeconomic differences of children living in and outside the state. This is true for the difference in reported health, school engagement, and pro-social activities. It appears that the higher incidence of poor children, many of whom are recent immigrants to the country, contribute to the state-level deficits in children’s well-being. However, the availability of mental health services for children in need of treatment is an exception to this pattern. Children in California are 72 receiving fewer services and the gap in provision of services to children with severe problems is conspicuous even taking account of the demographic characteristics of the state. 73 8. Conclusion and Policy Implications This report provides a baseline of information about the well-being of California’s children in 1999 against which similar indicators can be reexamined in subsequent years. In particular, it has shown us how California’s children are faring on measures of their physical health, behavioral problems, school attachment, and social involvement in comparison to children in other states. The report has devoted even more attention to documenting how the well-being of children within the state varies by their demographic and socioeconomic characteristics. This examination of differences among children within the state illustrates where the greatest needs for services exist and gives policymakers ways to target the populations that are in need of this assistance. From these comparative analyses, we observe the following outcomes. Children’s Physical Health About 78 percent of children in California are reported to be in very good or excellent health and only 8 percent have a chronic health problem. At the same time, over one-third of children had not seen a physician for routine preventive care in the last year and over one-fifth had not seen a dentist. Although children’s overall health status is generally positive, we find strong differences in children’s health among demographic and socioeconomic subgroups in the state. Specifically, children’s health status declines with age, however, their visits to physicians decline during adolescence as their needs increase, particularly among males. Our results also show that children whose parents are Hispanic immigrants and who are less-educated are in worse health than other children but are receiving less routine medical care. Although poor 75 children are also reported to be in worse health, they seem to have more access to medical, but not dental, care than these other children. More than 30 percent of children whose parents are Hispanic immigrants, who have the lowest levels of education, or who live in poor and near poor families had not seen a dentist in the past year. Children’s Emotional Adjustment More than one out of ten children in California are experiencing serious behavioral problems that could put them at risk for social and mental health problems later in life, but only 5 percent have received mental health services. The gap between children’s needs and treatment is higher in California than in other parts of the United States. This gap and the disparity between California and other states remain after the composition of children within and outside the state is taken into account. Within subgroups of children, we find that the level of children’s need for mental health services does not correspond with their receipt of these services. Namely, males and children from lower socioeconomic status families have more problems than other children but do not receive more treatment. These disparities are even more troubling for children with severe mental health problems and indicate that poor children are particularly underserved. Children’s Educational Attachment About one-third of children in California are not highly engaged in school, about 13 percent were expelled or suspended in the last year, and about one-fifth skipped school during this time. Within subgroups of children, we find that males and older children are less likely to be engaged in school and are more likely than other students to have voluntary or involuntary absences. Our results also show that family income is strongly related to children’s attachment to school, particularly in the area of school absence, suspension, and expulsion. Although we observe variations in children’s school engagement by their race and ethnicity, these differences are markedly reduced when other characteristics are held constant. 76 Children’s Involvement in Pro-Social Activities Close to one-quarter of children were not involved in any pro-social activities in the last year. We do not find that children’s level of involvement varies importantly by their gender or age, but we do find that involvement varies considerably by children’s socioeconomic status. Children whose parents do not have a high school diploma and children living in poor families are less likely to participate in social activities beneficial to their development. Differences in participation by children’s race and ethnicity diminish considerably after these and other factors are controlled for. Access to these activities might be limited both by the cost of these programs and by their lack of availability to low-income children. Children in California and the Rest of the Nation Children in California appear to be faring slightly worse on a number of indicators of their well-being and are not outperforming children in the rest of the nation on any of the indicators we examine. Compared to children living elsewhere, the overall health status of children in California is somewhat lower. Children with behavioral problems are less likely to receive mental health care. Furthermore, they skip school more often, and they are less involved in activities that promote their social development. After controlling for the compositional differences between children within and outside the state, we find that the gap between children’s need for mental health services and their use of these services remains higher in California than in other states. However, disparities in children’s health status and school absences largely disappear. Some disparities remain in children’s participation in pro-social activities indicating a lower level of in-school and after-school activities available in the state. Many of California’s children are faring well; however, our results suggest that a substantial minority of children may not be adequately served by existing services in the state. For the most part, these children come from the poorest families in California, have parents with low levels of educational attainment, or have parents who are Hispanic immigrants. 77 These families may lack the resources, knowledge, and social connections to obtain services for their children. Our findings about the well-being of children in California have broad implications for those seeking to provide services to children within the state and for monitoring whether these services are effective in serving children’s needs. The first set of policy implications stems from our observations about the nature of children’s needs. Although children whose parents are Hispanic immigrants and who are less-educated are in worse health than other children, they are receiving less routine medical care. Children from poor families are also in worse health but appear to have access to health care. It is critical to reach out to low-income immigrant communities where children are in need of health services as well as to native-born children from low- to moderate-income families who may not be receiving adequate health care. It is at least worth investigating whether such visits might identify and treat incipient problems that might later prove to be costly to individuals, families, and society. The importance of serving healthy adolescents in higher-income families who are not receiving routine health checkups may be less apparent. Yet, physicians can do an effective job of screening for risky behavior such as unprotected sex, alcohol and drug use, or depression. We discovered that adolescent males were particularly likely not to have received routine health care in the previous year. Although children from poorer families often use medical services, they are receiving much less help in obtaining preventive dental care. The lack of dental care is especially evident for Hispanic children whose parents were born outside the United States, whose parents have low educational levels, and whose families have incomes at the poverty or near-poverty level. There is an evident need to extend preventive services to this population, which may provide cost savings in the long run if serious dental problems are avoided. There is also a conspicuous lack of mental health assistance for children with multiple behavioral problems. All income groups experience some gap between the need for services among children who exhibit high levels of problem behavior; however, children in the poorest families are particularly underserved. Children with emotional and behavioral problems often experience problems in the classroom and the 78 community. Many of them will experience more severe problems later in life if they remain untreated. The tremendous costs of treating antisocial behavior and substance abuse among adults suggest that early intervention is justified even if it helps to reduce only modestly the incidence of problem behaviors. Policymakers might want to consider the possibility of directing public information to underserved communities to increase parents’ awareness of mental health services. Service providers may be more effective in reaching underserved populations by locating service sites closer to the populations at risk. It is not news that school behavior problems are more common among males, especially as they reach adolescence. Yet programs that keep older youth attached to school are in short supply. California could strengthen the in-school and after-school activities that help young males to stay connected to school. Tracking absenteeism and behavioral problems can identify youth at risk of dropping out. We need to craft more experiments that can help middle school students prepare to make a successful transition to high school by providing additional tutoring, mentoring, or social activities. Finally, our results indicate that an extraordinarily high proportion of youth are not engaged in any extracurricular activities in their preadolescent and adolescent years. The absence of programs for youth in this formative period likely accounts for part of the wide income disparity in the proportion of youth who are not involved in clubs, associations, lessons, sports, and other such activities. If programs do not adequately serve poorer children, this puts them at a disadvantage for gaining the benefits offered by extracurricular activities including greater school attachment, skill development, and social ties. Particularly in the wake of welfare reform, there is a manifest need to engage low-income adolescents in after-school activities because their parents may have entered the labor force, leaving them unattended in the interim between the close of the school day and the end of the workday. California is a state with high levels of income inequality (Reed, 1999). Among the poor and near-poor, families cannot afford to provide some of the assistance that is routinely offered to children in more affluent areas. Public investment in our young people through the preschool years may be alleviating some of the burden placed on low- 79 income families with young children, but we cannot expect that that assistance in the preschool years will inoculate children from problems later on. Families with older children require the same sort of public aid to help their children maintain good health, develop skills and competencies, and retain high motivation and a belief that hard work will pay off. At present, California may not be providing this sort of assistance at a level that reaches low-income families and that effectively services communities of nonnative residents. The second important set of policy implications that stems from this analysis relates to how we can improve information on children’s wellbeing. First, we should develop indicators appropriate for very young children and, second, use existing indicators to follow cohorts of children over the next decade as they reach school age to see the effects of early health and development programs and services. It is reasonable to expect that some of these programs may have had positive effects on the younger cohort of children now in their preschool years. Unfortunately, we do not have sufficient data to measure the well-being of younger children at the state level and are not tracking children’s well-being over time. The regular measurement of children’s well-being through surveys such as the NSAF, together with rigorous program evaluations, can help monitor whether the services provided are reaching the target populations and whether they are effective in raising levels of well-being. By beginning to monitor the success of our children, we can establish a benchmark for assessing whether California is providing sufficient public investment in children and can gain a glimpse into the future health, happiness, and productivity of the state’s population. More attention to marshalling data relevant to the well-being of children involves periodic assessments of the sorts of indicators described in this report. Fortunately, a third wave of NSAF was conducted in 2002–03 and will provide a way to update the set of indicators that we have described. However, we believe that other avenues of data analysis should also be pursued such as the use of vital statistics information. In other localities, administrative records such as these have been linked across agencies to provide a longitudinal record of children and families’ experiences from birth to early adulthood. We foresee the possibility of agencies collaborating at both the state and local levels to produce 80 longitudinal information that can help policymakers better identify the need for services, the success of targeted interventions, and the changing well-being of successive cohorts of children throughout their early lives. 81 Appendix A Detailed Comparisons of Scale Items for California and the Rest of the Nation 83 Table A.1 Emotional Adjustment of Children in California and the Rest of the United States (Weighted %) Doesn’t get along with other kidsa Never true Sometimes true Often true Can't concentrate for longa Never true Sometimes true Often true Has been sad or depresseda Never true Sometimes true Often true Feels worthless or inferiorb Never true Sometimes true Often true Has been nervous or tenseb Never true Sometimes true Often true Acts too young for his ageb Never true Sometimes true Often true Has trouble sleepingc Never true Sometimes true Often true Lies or cheatsc Never true Sometimes true Often true Does poorly at school workc Never true Sometimes true Often true California 71 25 4 59 34 7 64 34 2 85 14 1 74 25 1 79 17 5 87 11 1 75 22 3 67 28 5 Rest of the United States 70 28 3 62 31 8 62 35 3 86 13 1 71 25 4 80 17 3 85 13 2 76 21 2 67 28 6 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. aAsked of respondents about children ages 6–17. bAsked of respondents about children ages 6–11. cAsked of respondents about children ages 12–17. 84 Table A.2 School Commitment of Children in California and the Rest of the United States (Weighted %) Cares to do well in school None of the time Some of the time Most of the time All of the time Always does homework None of the time Some of the time Most of the time All of the time Does schoolwork just to get by None of the time Some of the time Most of the time All of the time Only does schoolwork when forced None of the time Some of the time Most of the time All of the time California 1 16 28 54 2 14 17 67 14 16 24 45 8 12 28 52 Rest of the United States 2 18 30 51 3 15 20 62 10 13 25 52 8 13 31 48 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. Table A.3 Pro-Social Involvement of Children in California and the Rest of the United States (Weighted %) Participated in club in past 12 months Played on sports team in past 12 months Took lessons after school in past 12 months California 46 52 34 Rest of the United States 58 54 33 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. 85 Bibliography Achenbach, T. M., and C. S. Edelbrock, “The Child Behavior Profile: II, Boys Aged 12–16 and Girls Aged 6–11 and 12–16,” Journal of Clinical and Consulting Psychology, Vol. 47, 1979, pp. 223–233. Achenbach, T. M., S. H. McConaughy, and C. T. Howell, “Child/ Adolescent Behavioral and Emotional Problems: Implications of Cross-Informant for Situational Stability,” Psychological Bulletin, Vol. 101, 1987, pp. 213–232. Ben-Arieh, Asher, Natalie Hevener Kaufman, Arlene Bowers Andrews, Robert M. Goerge, Bong Joo Lee, and J. Lawrence Aber, Measuring and Monitoring Children’s Well-Being, Kluwer Academic Publishers, Boston, Massachusetts, 2001. Black, Tamara, and Adam Safir, “Assessing Nonresponse Bias in the National Survey of America’s Families,” in F. Scheuren et al., Report No. 7: 1999 Collection of Papers, Urban Institute, Washington, D.C., 2001, available at http://www.urban.org/content/Research/ NewFederalism/NSAF/Methodology/1999MethodologySeries/1999. htm. Brick, J. M., P. Broene, D. Cantor, D. Ferraro, T. Hankins, C. Rauch, and T. Strickler, No. 8: 1999 NSAF Response Rates and Methods Evaluation, Urban Institute, Washington, D.C., 2000, available at http://www.urban.org/content/Research/NewFederalism/NSAF/ Methodology/1999MethodologySeries/1999.htm. Brint, Steven, Schools and Societies, Pine Forge Press, Thousand Oaks, California, 1998. Brooks-Gunn, Jeanne, and Greg J. Duncan, “The Effects of Poverty on Children,” The Future of Children, Vol. 7, No. 2, Summer/Fall 1997. Brown, E. Richard, Ninez Ponze, Thomas Rice, and Shana Alex Lavarreda, The State of Health Insurance in California: Findings from 87 the 2001 California Health Interview Survey, UCLA Center for Health Policy Research, Los Angeles, California, June 2002. Bustillo, Miguel, “Bills Aim to Make Students Healthier,” Los Angeles Times, March 17, 2002. Cairns, Robert B., Glen H. Elder, Jr., and E. Jane Costello (eds.), Developmental Science, Cambridge University Press, Cambridge, England, New York, 1996. California Children and Families Commission, California Children and Families Commission Annual Report—Fiscal Year 2000/01, 2002, available at http://www.ccfc.ca.gov/fiscal.htm. California Department of Finance, California State Budget Highlights, 2002, available at www.dof.ca.gov. Carnegie Corporation of New York, From Birth to Fifteen: A Comprehensive Approach to Children’s Learning and Development, Report of the Carnegie Corporation of New York 1989–1996, Carnegie Corporation, New York, 1996. Carnegie Council on Adolescent Development, A Matter of Time: Risk and Opportunity in the Non-School Hours, Carnegie Council, New York, 1992. Children Now, The State of Our Children: Children’s Critical Early Years, 2002a, available at http://www.childrennow.org/. Children Now, California Report Card: Children’s Critical Early Years, 2002b, available at http://www.childrennow.org/. Compas, Bruce E., and Constance L. Hammen, “Child and Adolescent Depression: Covariation and Comorbidity in Development,” in Robert J. Haggerty, Lonnie R. Sherrod, Norman Garmezy, and Michael Rutter (eds.), Stress, Risk, and Resilience in Children and Adolescents: Processes, Mechanisms, and Interventions, Cambridge University Press, New York, 1994, pp. 225–267. Converse, N., A. Safir, and F. Scheuren, No. 11: 1999 Public Use File Data Documentation, Urban Institute, Washington, D.C., 2001, 88 available at http://www.urban.org/content/Research/NewFederalism/ NSAF/Methodology/1999MethodologySeries/1999.htm. Danziger, Sheldon, and Peter Gottschalk (eds.), Uneven Tides: Rising Inequality in America, Russell Sage Foundation, New York, 1993. Danziger, Sheldon, and Peter Gottschalk, America Unequal, Russell Sage Foundation, Harvard University Press, New York, Cambridge, Massachusetts, 1995. Danziger, Sheldon, and Jane Waldfogel (eds.), Securing the Future: Investing in Children from Birth to College, Russell Sage Foundation, New York, 2000. Duncan, Greg J., Jeanne Brooks-Gunn (eds.), Consequences of Growing Up Poor, Russell Sage Foundation, New York, 1997. Eccles, Jacquelynne S., and B. L. Barber, “Student Council, Volunteering, Basketball or Marching Band: What Kind of Extracurricular Involvement Matters?” Journal of Adolescent Research, Vol. 14, No. 1, 1999, pp. 10–43. Eckstrom, R. B., M. S. Goertz, J. M. Pollack, and D. A. Rock, “Who Drops Out of High School and Why? Findings from a National Study,” in Gary Natriello (ed.), School Dropouts: Patterns and Policies, Teachers College Press, New York, 1987, pp. 52–69. Ehrle, Jennifer, and Kristin Anderson Moore, 1997 NSAF Benchmarking Measures of Child and Family Well-Being: Report No. 6, Urban Institute, Washington, D.C., 1999, available at http://www.urban. org/UploadedPDF/Methodology_6.pdf. Elo, Irma T., and Samuel H. Preston, “Educational Differentials in Mortality in the United States: 1979–1985,” Social Science and Medicine, Vol. 42, No. 1, 1996, pp. 47–57. Entwisle, Doris R. , K. L. Alexander, and L. S. Olson, Children, Schools, and Inequality, Westview Press, Boulder, Colorado, 1997. Furstenberg, Frank F., Thomas D. Cook, Jacquelynne Eccles, Glen H. Elder, Jr., and Arnold Sameroff, Managing to Make It: Urban Families 89 and Adolescent Success, University of Chicago Press, Chicago, Illinois, 1999. Haveman, Robert, and Barbara Wolfe (eds.), Succeeding Generations: On the Effects of Investments in Children, Russell Sage Foundation, New York, 1994. Hayward, Mark D., Eileen M. Crimmins, Toni P. Miles, and Yu Yang, “The Significance of Socioeconomic Status in Explaining the Racial Gap in Chronic Health Conditions,” American Sociological Review, Vol. 65, No. 6, 2000, pp. 910–930. Johnson, Hans, Laura Hill, and Mary Heim, “New Trends in Newborns: Fertility Rates and Patterns in California,” California Counts, Vol. 3, No. 1, Public Policy Institute of California, San Francisco, California, 2001. Johnson, Hans P., “Teen Birth Rates: Trends and Patterns in California,” California Counts, Vol. 4, No. 3, Public Policy Institute of California, San Francisco, California, 2003. Judkins, D., J. Michael Brick, Pam Broene, David Ferraro, and Teresa Strickler, Report No. 2, 1999 NSAF Sample Design, The Urban Institute and Westat, 2001, available at http://test.urban.org/ UploadedPDF/1999_Methodology_2.pdf. Knudsen, C., and R. McNown, “Changing Causes of Death and Sex Differences in the USA,” Population Research and Policy Review, Vol. 12, 1993, pp. 27–41. Lamborn, S. D., B. B. Brown, N. S. Mounts, and L. Steinberg, “Putting School in Perspective: The Influence of Family, Peers, Extracurricular Participation, and Part-Time Work on Academic Engagement,” in Student Engagement and Achievement in American Secondary Schools, Teachers College Press, New York, 1992, pp. 153–181. Maccoby, Eleanor, The Two Sexes: Growing Up Apart, Coming Together, Harvard University Press, Cambridge, Massachusetts, 1998. Mahoney, J. L., “School Extracurricular Activity Participation as a Moderator in the Development of Antisocial Behavior,” Child Development, Vol. 71, No. 2, 2000, pp. 502–516. 90 Mason, Mary Ann, Arlene Skolnick, and Stephen D. Sugarman (eds.), All Our Families: New Policies for a New Century: A Report of the Berkeley Family Forum, Oxford University Press, New York, 1998. Millstein, Susan G., Anne C. Petersen, and Elena O. Nightingale (eds.), Promoting the Health of Adolescents: New Directions for the TwentyFirst Century, Oxford University Press, New York, 1993. Moore, Kristin Anderson, Juliet L. Hatcher, Sharon Vandivere, and Brett V. Brown, “Children’s Behavior and Well-Being,” 1999 Snapshots of America’s Families II, Urban Institute, Washington, D.C., 2000, available at http://www.urban.org/content/Research/NewFederalism/ NSAF/Snapshots/1999Results/ChildrensBehaviorandWellBeing/ Behavior.htm. National Center for Health Statistics, “Summary Health Statistics for U.S. Children: National Health Interview Survey, 1997,” 2002, available at http://www.cdc.gov/nchs/products/pubs/pubd/series/ sr10/pre-200/sr10_203.htm. National Research Council, Losing Generations: Adolescents in High-Risk Settings, National Academy Press, Washington, D.C., 1993. National Research Council and the Institute of Medicine, Community Programs to Promote Youth Development, National Academy Press, Washington, D.C., 2002. Natriello, Gary (ed.), School Dropouts: Patterns and Policies, Teachers College Press, New York, 1987. Neisser, Ulric, The School Achievement of Minority Children: New Perspectives, Lawrence Erlbaum Associates, Hillsdale, New Jersey, 1986. Peterson, James L., and Nicholas Zill, “Marital Disruption, Parent-Child Relationships, and Behavior Problems in Children,” Journal of Marriage and the Family, Vol. 48, No. 2, 1986, pp. 295–307. Pierce, K. M., J. V. Hamm, and D. L. Vandell, “Experiences in AfterSchool Programs and Children’s Adjustment in First-Grade Classrooms, Child Development, Vol. 65, 1999, pp. 440–456. 91 Reed, Deborah, “California’s Rising Income Inequality: Causes and Concerns,” Occasional Papers, Public Policy Institute of California, San Francisco, California, 1999. Reed, Deborah, and Amanda Bailey, “California’s Young Children: Demographic, Social, and Economic Conditions,” California Counts, Vol. 4, No. 2, Public Policy Institute of California, San Francisco, California, 2002. Reed, Deborah, and Sonya M. Tafoya, “Demographic, Social, and Economic Trends for Young Children in California,” Occasional Papers, Public Policy Institute of California, San Francisco, California, 2001. Reyes, Belinda I., “An Assessment of Social and Economic Well-Being,” in Belinda Reyes (ed.), A Portrait of Race and Ethnicity in California, Public Policy Institute of California, San Francisco, California, 2001. Rumbaut, Ruben G., “The New Californians: Comparative Research Findings on the Educational Progress of Immigrant Children,” in Ruben G. Rumbaut and Wayne A. Cornelius (eds.), California’s Immigrant Children: Theory, Research, and Implications for Educational Policy, Center for U.S.-Mexican Studies, University of California, San Diego, La Jolla, California, 1995. Safir, Adam, Fritz Scheuren, and Kevin Wang, “Survey Methods and Data Reliability, 1997 and 1999,” Urban Institute, Washington, D.C., 2001, available at http://www.urban.org/content/Research/ NewFederalism/NSAF/survey-methods.htm. StataCorp, Strata Statistical Software: Release 5.0, Stata Corporation, College Station, Texas, 1997. Suárez-Orozco, Marcelo M., and Carola E. Suárez-Orozco, “The Cultural Patterning of Achievement Motivation: A Comparison of Mexican, Mexican Immigrant, Mexican American, and Non-Latino White American Students,” in Ruben G. Rumbaut and Wayne A. Cornelius (eds.), California’s Immigrant Children: Theory, Research, and Implications for Educational Policy , Center for U.S.-Mexican Studies, University of California, San Diego, La Jolla, California, 1995. 92 Takanishi, Ruby, and David A. Hamburg (eds.), Preparing Adolescents for the Twenty-First Century: Challenges Facing Europe and the United States, Cambridge University Press, New York, 1997. UCLA Academic Technology Services, SPSS Frequently Asked Questions: What Does Cronbach’s Alpha Mean? 2001, available at www.ats.ucla. edu/stat/spss/faq/alpha.html. U.S. Census Bureau, Poverty Status in 1999 of Individuals: 2000, 2002, available at http://factfinder.census.gov/. U.S. Department of Education, National Center for Education Statistics, Dropout Rates in the United States, 2000, 2002, available at http:// nces.ed.gov/pubs2002/droppub_2001/. U.S. Department of Health and Human Services, “HHS Approves Plan to Expand Coverage to 300,000 Uninsured Californians Under New Waiver Initiative,” 2002, HHS press release, available at http://www. hhs.gov/news. Vaden-Kiernan, N., D. Cantor, P. Cunningham, S. Dipko, K. Malloy, and P. Warren, No. 9: 1999 NSAF Telephone Survey Methods, Urban Institute, Washington, D.C., 2000, available at http://www.urban. org/content/Research/NewFederalism/NSAF/Methodology/ 1999MethodologySeries/1999.htm. Waldron, Ingrid, “Contributions of Changing Gender Differences in Behavior and Social Roles to Changing Gender Differences in Mortality,” in D. Sabo and D. F. Gordon (eds.), Men’s Health and Illness, Sage Publications, Thousand Oaks, California, 1995, pp. 22– 45. Wang, Kevin, David Cantor, and Nancy Vaden-Kiernan, Report No. 1, 1999 NSAF Questionnaire, Urban Institute, Washington, D.C., 2000, available at http://www.urban.org/content/Research/NewFederalism/ NSAF/Methodology/1999MethodologySeries/1999.htm. Williams, David R., and Chiquita Collins, “US Socioeconomic and Racial Differences in Health: Patterns and Explanations,” Annual Review of Sociology, Vol. 21, 1995. 93 Wolfe, Barbara L., “Economic Issues of Health Care,” in P. Lindsay Chase-Lansdale and Jeanne Brooks-Gunn (eds.), Escape from Poverty: What Makes a Difference for Children? Cambridge University Press, New York, 1995. Wynn, J., H. Richman, R. A. Rubenstein, and J. Littell, Communities and Adolescents: An Exploration of Reciprocal Supports, Youth and America’s Future, William T. Grant Foundation, Commission on Work, Family and Citizenship, Washington, D.C., 1988. Zuckerman, Barry, and Robert Kahm, “Pathways to Early Child Health and Development,” in Sheldon Danziger and Jane Waldfogel (eds.), Securing the Future: Investing in Children from Birth to College, Russell Sage Foundation, New York, 2000, pp. 87–121. 94 About the Authors FRANK F. FURSTENBERG Frank F. Furstenberg, an adjunct fellow at the Public Policy Institute of California, is the Zellerbach Family Professor of Sociology and a research associate in the Population Studies Center at the University of Pennsylvania. He has published numerous articles on teenage sexuality, pregnancy, and childbearing and also on divorce, remarriage, and stepparenting. His current research is focusing on the family in the context of disadvantaged urban neighborhoods, adolescent sexual behavior, children’s well-being, urban education, and the transition to adulthood. His book publications include Managing to Make It: Urban Families in High-Risk Neighborhoods (with Thomas Cook, Jacquelynne Eccles, Glen H. Elder, Jr., and Arnold Sameroff, 1999), Divided Families (with Andrew J. Cherlin, 1991), and Adolescent Mothers in Later Life (with J. BrooksGunn and S. Philip Morgan, 1987). He holds a B.A. in sociology from Haverford College and a Ph.D. in sociology from Columbia University. MAUREEN WALLER Maureen Waller is an assistant professor in the Department of Policy Analysis and Management at Cornell University. Her research interests include poverty, the family, culture, and social policy. Her work has used qualitative and mixed method techniques to examine the meaning of marriage, fatherhood, and social policies in low-income communities. She holds a B.A. in political science from the University of Dayton and an M.A and Ph.D. in sociology from Princeton University. HONGYU WANG Hongyu Wang is a Ph.D candidate in the Department of Sociology at the University of Pennsylvania. She is interested in the study of the family, adolescents’ well-being, interracial dating, and methodology. She has published an article with Paul Amato on divorce adjustment in the Journal of Marriage and the Family. She holds a B.A. in library science from Peking University and an M.A. in sociology from the University of Nebraska–Lincoln. 95 Related PPIC Publications “At Home and in School: Racial and Ethnic Gaps in Educational Preparedness” California Counts: Population Trends and Profiles Volume 3, Number 2, November 2001 Jennifer Y. Cheng Student and School Indicators for Youth in California’s Central Valley Anne Danenberg, Christopher Jepsen, and Pedro Cerdán Expensive Children in Poor Families: The Intersection of Childhood Disabilities and Welfare Marcia K. Meyers, Henry E. Brady, and Eva Y. Seto “California’s Young Children: Demographic, Social, and Economic Conditions” California Counts: Population Trends and Profiles Volume 4, Number 2, November 2002 Deborah Reed and Amanda Bailey A Portrait of Race and Ethnicity in California: An Assessment of Social and Economic Well-Being Belinda I. Reyes (editor) et al. PPIC publications may be ordered by phone or from our website (800) 232-5343 [mainland U.S.] (415) 291-4400 [Canada, Hawaii, overseas] www.ppic.org 97" } ["___content":protected]=> string(102) "

R 703FFR

" ["_permalink":protected]=> string(81) "https://www.ppic.org/publication/the-well-being-of-californias-children/r_703ffr/" ["_next":protected]=> array(0) { } ["_prev":protected]=> array(0) { } ["_css_class":protected]=> NULL ["id"]=> int(8276) ["ID"]=> int(8276) ["post_author"]=> string(1) "1" ["post_content"]=> string(0) "" ["post_date"]=> string(19) "2017-05-20 02:36:10" ["post_excerpt"]=> string(0) "" ["post_parent"]=> int(3425) ["post_status"]=> string(7) "inherit" ["post_title"]=> string(8) "R 703FFR" ["post_type"]=> string(10) "attachment" ["slug"]=> string(8) "r_703ffr" ["__type":protected]=> NULL ["_wp_attached_file"]=> string(12) "R_703FFR.pdf" ["wpmf_size"]=> string(6) "687927" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(188432) "The Well-Being of California’s Children ••• Frank F. Furstenberg Maureen R. Waller Hongyu Wang 2003 PUBLIC POLICY INSTITUTE OF CALIFORNIA Library of Congress Cataloging-in-Publication Data Furstenberg, Frank F., 1940- The well-being of California’s children / Frank F. Furstenberg, Maureen R. Waller, Hongyu Wang. p. cm. Uses the 1999 National Survey of America’s Families as a source of data collected on 1,917 children living in California households to examine variations in the well-being of children in the state and to compare their overall welfare to the welfare of children in other states. Includes bibliographical references. ISBN: 1-58213-066-3 1. Children—California—Social conditions. 2. Children—California— Economic conditions. 3. Children—Health and hygiene—California. 4. Children—California—Statistics. 5. Children—United States—States. 6. Family life surveys—California. 7. National Survey of America’s Families. I. Waller, Maureen Rosamond, 1967- II. Wang, Hongyu, 1971- III. Title. HQ792.C2F87 2003 305.23'09794—dc21 2003007577 Copyright © 2003 by Public Policy Institute of California All rights reserved San Francisco, CA Short sections of text, not to exceed three paragraphs, may be quoted without written permission provided that full attribution is given to the source and the above copyright notice is included. PPIC does not take or support positions on any ballot measure or state and federal legislation nor does it endorse or support any political parties or candidates for public office. Research publications reflect the views of the authors and do not necessarily reflect the views of the staff, officers, or Board of Directors of the Public Policy Institute of California. Foreword Descriptions of California’s enormous immigrant population are usually presented with a mixture of awe and anxiety. Awe because what California has accomplished in absorbing immigrants over the past two decades is truly remarkable and breathtaking in scale. Anxiety because no one can be really certain that this “experiment” in social change will not result in some insurmountable challenges. Nevertheless, California is moving boldly into the future, driven by a resilient economy and riding on a demographic transformation that is unprecedented in American history. The challenges generated by large-scale immigration are beginning to come into sharper focus. Families without health insurance, everincreasing demands on the public education system, and the need to increase the supply of affordable housing are all public policy issues that have come up for intense public debate. More recently, it has become apparent that an increasing number of children in California have parents born outside the United States. In 1999, almost half of all children under age six had a foreign-born parent. More important, about 20 percent of California’s children live in poverty, compared to about 17 percent in the nation as a whole. Thanks to the availability of a large national survey, Frank Furstenberg, Maureen Waller, and Hongyu Wang have been able to take a careful look at children in California and have provided a remarkable profile in The Well-Being of California’s Children. The authors present findings on the physical health of children, their emotional adjustment, their attachment to school, and the degree to which they are involved in pro-social activities. Overall, children in California appear to be faring slightly worse on a number of indicators of their well-being and are not outperforming children in the rest of the nation on any of the indicators examined. The overall health status of children in California is somewhat lower than that of children living elsewhere. iii Although many of California’s children are faring well, the authors conclude that a substantial minority might not be adequately served by existing services in the state. For the most part, these children come from the poorest families in California, have parents with low levels of educational attainment, or live in Hispanic immigrant households. The authors observe that “there is a conspicuous lack of mental health assistance for children with severe problems. The tremendous cost of treating antisocial behavior and substance abuse among adults suggests that early intervention is justified.” They also note that California would do well to strengthen in-school and after-school activities that help older and less-affluent adolescents stay connected to school. In other words, all taxpayers in California face the challenge of helping their children face the future in the best possible circumstances. And again, this is yet another public policy issue facing state and local governments that are already facing a cycle of substantial cuts in services. Where will the money come from? What about privatization? What is the public role in helping children? What other services will have to be cut to help the children? These are the questions raised by this important set of findings about a population that is central to the future of California’s civic, economic, and social health. We can only hope that by presenting the facts and raising the questions, the political process will render a set of solutions appropriate for both our own and future generations. David W. Lyon President and CEO Public Policy Institute of California iv Summary More than one in eight children in the United States lives in California. In recent years, California’s citizens and lawmakers have given a great deal of attention to addressing the needs of these children, and some programs have directed considerable resources toward enhancing their welfare and development. Given the growth in programs aimed at improving children’s well-being, it has become increasingly important to take stock of how children in the state are faring. However, our ability to assess child-centered policies has been seriously hindered because most national studies that include indicators of children’s well-being have not been large enough to yield reliable information on children at the state level. This report uses a new source of information—the National Survey of America’s Families (NSAF)—that was explicitly designed to measure children’s welfare and well-being and that provides a large enough sample of California families to assess the status of children in the state. In particular, the report draws on information collected for 1,917 children living in California households that participated in the 1999 NSAF to examine variations in the well-being of children in the state and how they are faring in comparison to children living elsewhere. An important objective of this report is to provide a baseline of information about children’s well-being in 1999 against which similar indicators can be reexamined in subsequent years. Moreover, this analysis provides information about the needs of special subgroups of children and suggests to policymakers ways to target the populations that are in particular need of assistance. The public cares whether children are healthy, happy, and secure for the same reasons that they care about the welfare of any other age group. However, because children are also one of the most vulnerable demographic groups in the population and because they are born into unequal circumstances, public concern also focuses on whether children v have the kind of resources that will enhance their welfare and development and, if disparities in these resources exist, how they can be alleviated. The public is further concerned with how children are faring because adult performance is strongly linked to the quality and quantity of material and emotional investment that children receive (Cairns, Elder, and Costello, 1996; Haveman and Wolfe, 1994). When children are adequately nurtured, they are also more likely to become productive and involved members of society as they grow older. In California, about one in five children is living in a poor household. Poverty has been found to have negative effects on children’s cognitive abilities and achievement that may persist in adulthood (Brooks-Gunn and Duncan, 1997; Duncan and Brooks-Gunn, 1997). Children in California are also from diverse racial and ethnic backgrounds. Because nonwhite families in California and the nation often have lower incomes than white families, there is also concern that children in these families disproportionately lack sufficient resources to meet their developmental needs. Immigrant families may face additional barriers to accessing these resources. State and local policies are increasingly being considered to address the gap between advantaged and disadvantaged populations, especially as the federal government has turned more responsibility for social welfare programs over to the states. How Are California’s Children Faring? In this report, we look at parents’ assessments of how their children are doing in four important areas of child development: physical health, emotional and behavioral adjustment, attachment to school, and positive social involvement. We examine whether these indicators of well-being vary by children’s gender, age, race/ethnicity, parents’ nativity and education, and family income. In addition to examining children’s outcomes as simple percentages, we look at how the percentages corresponding to children’s membership in racial/ethnic, Hispanic immigrant, and socioeconomic groups change after they are adjusted to control for children’s characteristics. The adjusted percentages allow us to see the unique contribution of children’s characteristics and to evaluate whether differences between children living in California and vi those living elsewhere are meaningful after their characteristics are taken into account. Children’s Physical Health Perhaps the most fundamental indicator of children’s welfare is their health status. If children are in poor health, this would likely also have a negative effect on the other measures of their well-being we observe. We examine an indicator of children’s general health status shown to be strongly correlated with specific health problems and to be a robust predictor of future health status. The analysis also considers whether children have a debilitating health condition and their use of health services. In our analysis, about 78 percent of children in California are reported to be in very good or excellent health and only 8 percent have a chronic health problem. At the same time, over one-third of children had not seen a physician for routine preventative care in the last year and over one-fifth had not seen a dentist. Although children’s overall health status is generally positive, we find strong differences in children’s health among demographic and socioeconomic subgroups in the state. Specifically, children’s health status declines with age; yet, their visits to physicians drop off during adolescence as their needs increase, particularly among males. Our results also show that children whose parents are Hispanic immigrants and who are less-educated are in worse health than other children but are receiving less routine medical care. Although poor children are also reported to be in worse health, they seem to have more access to medical, but not dental, care than other children. We find that about one-third of children at high risk (whose parents are Hispanic immigrants, have the lowest levels of education, or who live in poor or near-poor families) had not seen a dentist in the past year. Children’s Emotional Adjustment One standard indicator of children’s adjustment during childhood is a behavioral problems inventory. The NSAF inventory was derived from a longer checklist of symptoms and is a reliable predictor of future emotional and behavioral problems. In particular, children with more frequent symptoms (as reported by parents) are more likely to encounter vii mental health problems and to engage in deviant behavior in adolescence and adulthood (Achenbach and Edelbrock, 1979; Achenbach, McConaughy, and Howell, 1987). Our findings show that more than 10 percent of the children in California are experiencing serious behavioral problems that put them at risk for social and mental health problems later in life but that only 5 percent of these children have received mental health services. The gap between children’s needs and treatment is higher in California than in other parts of the United States. This gap and the disparity between California and other states remain after the composition of children within and outside the state is taken into account. Within subgroups of children, we find that the level of children’s need for mental health services does not correspond with their receipt of these services. Namely, males and children from lower socioeconomic status families show more problems than other children but do not receive more treatment. These disparities are even more apparent for children with severe problems and indicate that poorer children are particularly underserved. Children’s Educational Attachment The ability to function in school is one of the earliest and most powerful predictors of whether children will develop their human capital (i.e., the skills and knowledge that are associated with productive employment later in life). This process of commitment to and confidence in learning begins early in life—as early, in fact, as the preschool years—and continues throughout childhood and adolescence. Research has shown that children’s work habits in the classroom predict their educational achievement and their later attachment to schooling and eventually to the workforce (Brint, 1998; Danziger and Waldfogel, 2000). We find that about one-third of children in California are not highly engaged in school, about 13 percent were expelled or suspended in the last year, and about one-fifth skipped school in the last year. Within subgroups of children, we find that males and older children are less likely to be engaged in school and are more likely to have voluntary or involuntary absences than other students. Our results also show that family income is strongly related to children’s attachment to school, viii particularly in the area of school absence, suspension, and expulsion. Although we observe racial/ethnic variations in children’s school engagement, these differences are markedly reduced when their other characteristics are held constant. Children’s Involvement in Pro-Social Activities Developmental psychologists are paying increasing attention to children’s involvement in pro-social activities such as clubs, sports, and after-school lessons, because these activities help to build their capacity to establish strong relationships with peers. They also foster a variety of nonacademic skills such as communication and development of talents that build self-esteem, a sense of efficacy, and psychological resourcefulness. Other dividends of positive social involvement include increasing children’s ability to form lasting personal connections that may be helpful in times of trouble and shielding children from the influence of negative peers (Carnegie Council on Adolescent Development, 1992; Eccles and Barber, 1999; Mahoney 2000; Pierce, Hamm, and Vandell, 1999). Our results indicate that close to one-quarter of children were not involved in any pro-social activities in the last year. We do not find that children’s level of involvement varies in important ways by their gender or age, but we do find that involvement varies considerably by children’s socioeconomic status. Children whose parents do not have a high school diploma and children living in poor families are less likely than other children to participate in social activities beneficial to their development. Differences in participation by children’s race/ethnicity diminish considerably after these and other factors are controlled for. Access to these activities might be limited both by the cost of these programs and by their availability to low-income children. Children in California and the Rest of the Nation Overall, we find that children in California appear to be faring slightly worse on a number of indicators of their well-being and are not outperforming children in the rest of the nation on any of the indicators we examine. Compared to children living elsewhere, the overall health status of children in California is somewhat lower. Children with ix behavioral problems are less likely to receive mental health care. Furthermore, California children skip school more often and are less involved in activities that promote their social development. After controlling for the compositional differences between children within and outside the state, we find that the gap between children’s need for mental health services and their use of these services remains higher in California than in the rest of the nation. However, disparities in children’s health status, school absences, and pro-social involvement largely disappear. These results suggest that across-state differences in children’s well-being may be due to the demographic and socioeconomic makeup of children who reside in California, except in the area of mental health services. Implications for Policy Although many of California’s children are faring well, our results suggest that a substantial minority may not be adequately served by existing services in the state. For the most part, these children come from the poorest families in California, have parents with low levels of educational attainment, or live in Hispanic immigrant families. These families may lack the resources, knowledge, and social connections to obtain services for their children. Our findings about the well-being of children in California have broad implications for providing services to children within the state and for monitoring whether these services are effective in serving children’s needs. The first set of policy implications stems from our observations about the nature of children’s needs. Although children whose parents are Hispanic immigrants and who are less-educated are in worse health than other children, they are receiving less routine medical care. Children from poor families are also in worse health but appear to have access to health care. The importance of serving healthy adolescents in higher-income families who are not receiving routine health checkups may be less apparent. Yet, physicians can do an effective job of screening for risky behavior such as unprotected sex, alcohol and drug use, or depression. We discovered that adolescent males were particularly likely not to have received routine health care in the previous year. x Although children from poorer families often use medical services, they are receiving much less help in obtaining preventive dental care. The lack of dental care is especially evident for Hispanic children whose parents were born outside the United States, children whose parents have low educational levels, and those whose families have poverty or nearpoverty-level incomes. There is an evident need to extend preventive services to this population, which may provide cost savings in the long run if serious dental problems are avoided later in life. There is also a conspicuous lack of mental health assistance for children with severe problems. All income groups have some gap between the need for and provision of services among children who exhibit high levels of problem behavior; however, children in the poorest families are particularly underserved. Children with emotional and behavioral problems often experience problems in the classroom and the community. Many of them will experience more severe problems later in life if they remain untreated. The tremendous costs of treating antisocial behavior and substance abuse among adults suggest that early intervention is justified even if it helps to reduce only modestly the incidence of problem behaviors. Policymakers might want to consider the possibility of directing public information to underserved communities to increase parents’ awareness of mental health services. Service providers may be more effective in reaching underserved populations by locating service sites closer to the populations at risk. It is not news that school behavior problems are more common among males, especially as they reach adolescence. Yet, programs that keep older males attached to school are in short supply. California could strengthen the in-school and after-school activities that help males to stay connected to school. Tracking absenteeism and behavioral problems can identify youth at risk of dropping out. We need to craft more experiments that can help middle school students prepare to make a successful transition to high school, whether by providing additional tutoring, mentoring, or social activities. Finally, our results indicate that an extraordinarily high proportion of youth are not engaged in any extracurricular activities in their preadolescent and adolescent years. The absence of programs for youth in this formative period likely accounts for part of the wide income xi disparity in the proportion of youth who are not involved in clubs, associations, lessons, sports, and the like. If programs do not adequately serve poorer children, this puts them at a disadvantage for gaining the benefits offered by extracurricular activities, including greater school attachment, skill development, and social ties. Particularly in the wake of welfare reform, there is a manifest need to serve low-income adolescents in after-school activities because their parents may have entered the labor force, leaving them unattended in the interim between the close of the school day and the end of the workday. The second important set of policy implications that stems from this analysis relates to how we can improve information on children’s wellbeing. First, we should develop indicators appropriate for very young children and, second, use existing indicators to follow cohorts of children over the next decade as they reach school age to see the effects of early health and development programs and services. It is reasonable to expect that some of these programs may have had positive effects on the younger cohort of children now in their preschool years. Unfortunately, we do not have sufficient data to measure the well-being of younger children at the state level and are not tracking children’s well-being over time. The regular measurement of children’s well-being through surveys such as the NSAF, together with rigorous program evaluations and the use of longitudinal administrative data, can help monitor whether the services provided are reaching the target populations and whether they are effective in raising levels of well-being. By beginning to monitor the success of our children, we can establish a benchmark for assessing whether California is providing sufficient public investment in children and can gain a glimpse into the future health, happiness, and productivity of our state’s population. xii Contents Foreword ..................................... Summary..................................... Figures ...................................... Tables ....................................... iii v xv xvii 1. INTRODUCTION ........................... 1 2. DATA AND METHODS ....................... Sample Characteristics.......................... Measures of Children’s Well-Being.................. Physical Health .............................. Emotional Adjustment ......................... Educational Attachment ........................ Pro-Social Involvement ......................... 9 11 14 14 17 18 19 3. PHYSICAL HEALTH ......................... Gender and Age.............................. Family Characteristics .......................... Conclusion ................................. 21 21 24 32 4. EMOTIONAL ADJUSTMENT ................... Gender and Age.............................. Family Characteristics .......................... Conclusion ................................. 35 37 38 43 5. EDUCATIONAL ATTACHMENT ................ Gender and Age.............................. Family Characteristics .......................... Conclusion ................................. 6. PRO-SOCIAL INVOLVEMENT .................. Gender and Age.............................. Family Characteristics .......................... Conclusion ................................. 45 46 48 53 55 56 57 62 xiii 7. HOW CALIFORNIA’S CHILDREN COMPARE TO THOSE IN THE REST OF THE NATION........... Physical Health .............................. Emotional Adjustment ......................... Educational Attachment ........................ Pro-Social Involvement ......................... Conclusion ................................. 8. CONCLUSION AND POLICY IMPLICATIONS ...... Children’s Physical Health ....................... Children’s Emotional Adjustment .................. Children’s Educational Attachment ................. Children’s Involvement in Pro-Social Activities .......... Children in California and the Rest of the Nation ........ Appendix A: Detailed Comparisons of Scale Items for California and the Rest of the Nation ....................... Bibliography .................................. About the Authors ............................... Related PPIC Publications .......................... 63 64 65 69 70 71 75 75 76 76 77 77 83 87 95 97 xiv Figures 3.1. Children with Chronic Health Condition That Limits Their Activity, by Gender and Age Group (Weighted %) ................................... 4.1. Children with Severe Behavioral Problems Who Received Mental Health Services in Past 12 Months, by Family Income (Weighted %) .................. 5.1. Children Who Skipped School Two or More Times or Were Suspended or Expelled in the Past 12 Months, by Gender and Age Group (Weighted %)............. 7.1. Children’s Health Status in California and the Rest of the United States (Adjusted %) ................. 7.2a. Children Who Received Mental Health Services in Past 12 Months, by Severity of Behavioral Problems, California and the Rest of the United States (Weighted %) ................................... 7.2b. Children Who Received Mental Health Services in Past 12 Months, by Severity of Behavioral Problems, California and the Rest of the United States (Adjusted %) ................................... 7.3. Children Who Participate in Pro-Social Activities, by Number, California and the Rest of the United States (Adjusted %) ............................. 23 42 48 66 67 68 72 xv Tables 2.1. NSAF Sample Characteristics of Children in California and the Rest of the United States (Weighted %) ...... 2.2. NSAF Measures of Children’s Well-Being .......... 3.1. Health Status of Children in California (Weighted %) .. 3.2a. Health Status of Children in California, by Race/ Ethnicity, Level of Parent’s Education, and Family Income (Weighted %) ....................... 3.2b. Health Status of Children in California by Race/ Ethnicity, Level of Parent’s Education, and Family Income (Adjusted %) ....................... 4.1. Children’s Behavioral Problems, by Gender and Age (Weighted %) ............................ 4.2a. Children’s Behavioral Problems by Race/Ethnicity, Parent’s Education, and Family Income (Weighted %) .. 4.2b. Children’s Behavioral Problems by Race/Ethnicity, Parent’s Education, and Family Income (Adjusted %) ... 5.1. Children’s School Engagement, Frequency of Missing School, and Suspension/Expulsion Rates, by Gender and Age (Weighted %) ......................... 5.2a. Children’s School Engagement, Frequency of Missing School, and Suspension/Expulsion Rates, by Race/Ethnicity, Parent’s Education, and Family Income (Weighted %) ............................ 5.2b. Children’s School Engagement, Frequency of Missing School, and Suspension/Expulsion Rates, by Race/Ethnicity, Parent’s Education, and Family Income (Adjusted %) ............................. 6.1. Children’s Pro-Social Involvement, by Gender and Age (Weighted %) ............................ 6.2a. Children’s Pro-Social Involvement by Race/Ethnicity, Parent’s Education, and Family Income (Weighted %) .. 12 15 22 26 30 38 40 41 46 49 52 57 59 xvii 6.2b. Children’s Pro-Social Involvement by Race/Ethnicity, Parent’s Education, and Family Income (Adjusted %) ... 7.1. Health Status of Children in California and the Rest of the United States (Weighted %)................. 7.2. Children’s Behavioral Problems in California and the Rest of the United States (Weighted %) ............ 7.3. Children’s School Engagement in California and the Rest of the United States (Weighted %) ........... 7.4. Children’s Pro-Social Involvement in California and the Rest of the United States (Weighted %) ............ A.1. Emotional Adjustment of Children in California and the Rest of the United States (Weighted %) ........... A.2. School Commitment of Children in California and the Rest of the United States (Weighted %) ............ A.3. Pro-Social Involvement of Children in California and the Rest of the United States (Weighted %) ............ 60 64 66 69 71 84 85 85 xviii 1. Introduction California, the nation’s most populous state, has over nine million children. These children are among the most ethnically and racially diverse in the country, and almost half of the children under age six in California have a parent who was born outside the United States. By 2020, Hispanic children are expected to constitute the largest group of children in the state (Reed and Tafoya, 2001). California’s children are more likely to be poor than those in the nation as a whole, with about 20 percent living in poverty, compared to about 17 percent in the United States (U.S. Census Bureau, 2002). No population group in this country is the object of more public concern and scrutiny than America’s children and youth (Haveman and Wolfe, 1994; Mason, Skolnick, and Sugarman, 1998; Takanishi and Hamburg, 1997). The public cares whether children are healthy, happy, and secure for the same reasons that they care about the welfare of any other age group. But because children are also one of the most vulnerable demographic groups in the population, they receive more public attention (if not always financial aid and social services) than nonelderly adults, who are considerably less dependent on others for their survival and development. The public also cares about children’s well-being because how children fare in early and middle childhood—not to mention how they do in their adolescent years—is strongly related to how they will function as adults. A wealth of research from different social science disciplines has demonstrated that adult performance is strongly and causally linked to the quality and quantity of material and emotional investment received during childhood (Cairns, Elder, and Costello, 1996; Haveman and Wolfe, 1994). Children do better in later life when they are engaged in multiple contexts in which they are monitored, supported, and provided with opportunities to acquire cognitive, emotional, and physical skills. 1 At the same time, some national data indicate growing inequality in access to contexts in which children can develop these skills (Danziger and Gottschalk, 1993, 1995). Thus, public concern also focuses on whether children have the kind of resources that will enhance their welfare and development and, when disparities in these resources exist, how they can be alleviated. Poor families often experience greater stress than economically advantaged families and may be less able to provide their children the kind of resources related to healthy development. Research shows that poverty has a negative effect on children’s cognitive abilities and achievement, particularly if children live in extreme poverty or their families are poor for many years (Brooks-Gunn and Duncan, 1997; Duncan and Brooks-Gunn, 1997). Some of these effects also appear to persist in adulthood, when these youth later attempt to find stable and remunerative employment, to form families, and to participate in civic life. Because nonwhite families in California and the nation often have lower incomes than white families (Reyes, 2001), there is also concern that children in these families disproportionately lack sufficient resources to meet their developmental needs. Immigrant families may face additional barriers to accessing these resources. State and local policies are increasingly being considered to address the gap between advantaged and disadvantaged populations, especially as the federal government has turned more responsibility for social welfare programs over to the states (Millstein, Petersen, and Nightingale, 1993). Finally, the public cares about whether children are being adequately nurtured because we want our young people to become productive and involved members of society when they reach adulthood. When children do not receive this nourishment, they are far less likely as adults to be able to contribute to the quality of life and the economic security of the state. Indeed, the failure of children to develop the knowledge, skills, and values that make them successful adults leads to a weakening of the social fabric and places a burden on taxpayers and fellow citizens. In recent years, California’s citizens and lawmakers have given a great deal of attention to the needs of children, and some programs have directed considerable resources toward enhancing their welfare and development. The creation of three recent programs in the areas of child development, health, and after-school care illustrates how some of these 2 concerns have been addressed. First, with the passage of the Proposition 10 initiative in 1998, citizens voted to impose a 50 cent tax on tobacco, largely to fund early childhood development programs aimed at mothers during prenatal care and children through age five. The aim of this initiative is to provide comprehensive, integrated services that promote children’s development and ensure that they enter school ready to learn.1 In fiscal year 2000–01, the tobacco tax generated about $650 million in revenue for Proposition 10 activities (California Children and Families Commission, 2002).2 New efforts to expand children’s access to health coverage have also emerged in recent years. For example, the Healthy Families Program, created in 1997, is a state- and federally funded program to provide lowcost health, dental, and vision coverage to children living in low-income families who do not qualify for Medi-Cal but who have incomes below 250 percent of the poverty level. 3 Enrollment in Healthy Families is expected to reach 624,000 children by June 30, 2003, and the 2002–03 California budget allocates $672.2 million in combined state and federal funding for this program (California Department of Finance, 2002). Increased access to health care for children can help families obtain preventive health services that might discourage the onset of health problems in later life as well as offer medical assistance for children’s acute and chronic health problems. Both within and outside the K–12 educational system, California has been actively expanding a variety of resources and social services aimed at ____________ 1See http://www.ccfc.ca.gov/ for more information about the California Children and Families Commission created by Proposition 10. 2Of course, child care represents another important area of spending on children’s development. The 2002–03 state budget includes $3.1 billion for a variety of child care and related services (California Department of Finance, 2002). 3Healthy Families is the name of the State Children’s Health Insurance Program in California. As the state expanded health coverage under the Healthy Families Program, it also increased income eligibility in Medi-Cal for children up to age 18 and provided coverage for pregnant women and infants with incomes between 200 and 300 percent of the poverty level through the Access for Infants and Mothers Program (Brown et al., 2002, pp. 43–44). In January 2002, the U.S. Department of Health and Human Services also approved a federal waiver to cover 300,000 California parents with incomes up to 200 percent of the poverty level under the Healthy Families Program (U.S. Department of Health and Human Services, 2002). 3 both preschoolers and school-age children. Some of these services are designed to serve youngsters from low-income families and those whose parents are recent immigrants with limited language skills, low educational levels, and poor connections to mainstream institutions. For example, the state’s Before and After School Learning and Safe Neighborhoods Partnerships Program was created in 1998 to provide literacy and educational enrichment and to provide a safe atmosphere for children in kindergarten through junior high. Proposition 49, approved by voters in the November 2002 election, requires a specific spending level for this program, up to $550 million annually.4 These and other efforts to promote children’s welfare have generated interest in developing stable and reliable measures of children’s wellbeing to determine the success of the growing number of state and county services to improve parent education, child care, health care, and effective intervention for families at risk. However, our ability to assess child-centered policies has been seriously hindered by the absence of good data at the state level, despite the fact that many of the most important policies affecting children are administered by states. Although interest in children’s well-being has been keen for many years, only relatively recently have researchers and policymakers begun to systematically track information about this issue using nationally representative data (Ben-Arieh et al., 2001). At the state and local levels, progress in assembling information that would allow policymakers and the public to gauge the performance of children has been even slower, largely because of the absence of reliable information about children’s well-being. Most national studies that include indicators of children’s well-being have not been large enough to yield reliable information on children at the state level. Some important information about children’s demographic and socioeconomic characteristics as well as their early health status is available from vital statistics and Census data.5 Children ____________ 4Proposition 49 also changed the name of this program to the After School Education and Safety Program. 5See also Reed and Tafoya (2001), Reed and Bailey (2002), and Johnson (2003) for statistical portraits of children ages five and under in California, including information about population trends and projections, family structure, births to teen parents, parental 4 Now also assembles government and survey data for its annual report card and biannual county data book on California children’s education, health, family economic resources, and safety (Children Now, 2002a, 2002b). Because these datasets do not include measures of children’s psychological and social development, which would be useful to policymakers in gauging children’s well-being, one of the overarching recommendations of these reports is that more data on the well-being of young children in California be collected and analyzed.6 In this study, we use a new source of information—the National Survey of America’s Families (NSAF)—that was explicitly designed to measure children’s well-being and that provides a large enough sample of California families to assess the status of children in the state. We draw on information collected for 1,917 children living in California households that participated in the 1999 NSAF to examine variations in the well-being of children in the state and to compare their overall welfare to that of children in other states. In particular, we identify a series of indicators that measure different dimensions of their well-being, including their physical health status, emotional adjustment, educational attachment, and involvement in social activities that promote their development. The sample is large enough to allow us to examine how children are doing by looking at their demographic and socioeconomic characteristics—information typically lacking in data gained from national surveys and government records. Although data from the NSAF and other surveys have some obvious limitations associated with how accurately respondents report information, they offer some distinct advantages over information exclusively collected from administrative sources. For example, most administrative records offer information in a single domain of children’s welfare, such as their school performance, health, or involvement in the juvenile justice system. Although it might be possible to assemble an ________________________________________________________ education, family income, receipt of public assistance, and health insurance and vaccinations. 6Both publications identified areas where improvements have been made for children in the state, such as in declines in infant mortality, and where needs remain, such as in the area of high-quality child care. For more information, see http://www. childrennow.org. 5 archive with administrative reports from a variety of agencies, this task has proved difficult in practice. Through surveys of parents and children, it is possible to collect a variety of different indicators of children’s welfare with relative ease; indeed, it is possible to obtain data that simply are not available elsewhere. The strength of the NSAF is that it includes a broad array of indicators of children’s well-being not readily available from other data sources that can be tracked over time. Both the size of the California sample and NSAF’s unique measures of children’s well-being make the survey extremely well-suited for our study of children in the state. There are some potential problems in relying on reports exclusively from parents. For example, some parents may be reluctant to report negative outcomes for their child or to accurately state problems with their child’s behavior. The small amount of literature on this source of bias suggests that parent reports can sometimes overstate or understate, depending on social class or ethnicity. Nonetheless, it appears that parents are generally honest in their evaluations, albeit limited in their knowledge of how the child behaves outside the home. Furthermore, it is worth knowing how parents view their children’s well-being because parents are important gatekeepers to services offered by schools and social agencies. Policymakers and practitioners need information on how many children and families are in need of services, how many are being served, and, most of all, how well services are working to achieve the objectives of the health, educational, and social programs that are being mounted on behalf of improving the well-being of children.7 This report, representing a first step in providing valuable information on the well____________ 7For example, a package of legislation was put forward in 2002 that aimed to improve the health of California’s students. This legislation followed the release of a report from the Select Committee on California Children’s School Readiness and Health indicating that children are having trouble learning in school, in part because of health problems. The report recognized the lack of indicators of children’s health available and suggested that more empirical studies and measures of children’s health were needed to track the health status of school-age children (Bustillo, 2002). 6 being of California’s children,8 has several objectives: First, it provides a baseline of information about children’s well-being in 1999 against which similar indicators can be reexamined in subsequent years. This will allow an assessment of the overall success of increased investment in the development of children and in producing healthy, socially involved, and psychologically sound young people. The report also provides information about the needs of special subgroups of children and suggests to policymakers ways to target the populations that may be in particular need of health or social services. Finally, this report can inform policymakers and the public about how children in California are faring relative to those in the rest of the nation. In the next chapter, we describe the NSAF in more detail and provide an overall picture of how the survey data were collected and of the children included in the sample. We also discuss the measures used to assess children’s well-being. Chapters 3 through 6 present information on each of the domains of children’s well-being we analyze and give a more detailed picture of how children’s well-being varies by their gender, age, race/ethnicity, parents’ nativity and education, and family income. Chapter 7 compares California’s children to those in the rest of the United States. The last chapter sums up what we have learned and discusses some of the policy implications of this analysis. ____________ 8Complementary methods of tracking the well-being of children and the effectiveness of services in the state might include program evaluations and the linking of longitudinal administrative records. 7 2. Data and Methods Our analysis of the well-being of children in California draws on data from the 1999 National Survey of America’s Families. The NSAF is a national survey that contains information on the economic, health, and social characteristics of children and adults under age 65 in the United States. The survey collected information on over 100,000 people living in 42,000 U.S. households.1 This study uses information collected on 1,917 children living in 1,536 California households. It compares how these children are faring in comparison to 34,021 children living in 28,034 households in the rest of the United States. Because the NSAF was designed by the Urban Institute and Child Trends to examine the well-being of children and families following the devolution of welfare and other social policies to the states, 13 states, including California, were oversampled for the survey. This sampling design was intended to make the NSAF representative of the nonelderly population in these 13 states and in the nation as a whole.2 As such, the ____________ 1The first wave of the survey was conducted in 1997. In this report, we do not present results for the 1997 survey primarily because this earlier survey presents potential difficulties with examining how immigration status is related to children’s well-being—a key issue to interpreting differential outcomes among children in California. Specifically, the large discrepancy in the number of foreign-born and immigrant respondents in the 1997 NSAF compared to the number in other data sources prevents us from presenting results for immigration status with complete confidence. To correct for this problem, the 1999 NSAF changed the wording of the question used to determine the nativity of respondents by collecting information about the country of birth for each household member (Wang, Cantor, and Vaden-Kiernan, 2000). 2The survey is representative of the noninstitutionalized, civilian population of individuals under age 65 in the nation. It is also representative of this population in Alabama, California, Colorado, Florida, Massachusetts, Michigan, Minnesota, Mississippi, New Jersey, New York, Texas, Washington, and Wisconsin. For an overview of the survey, see http://www.urban.org/Content/Research/NewFederalism/NSAF/ Overview/NSAFOverview.htm. For more detailed discussions of the survey’s methodology, see Judkins et al. (2001), Safir, Scheuren, and Wang (2001), and Wang, Cantor, and Vaden-Kiernan (2000). 9 NSAF sample in California is large enough to produce reliable estimates of child and family well-being measures in the state. Because the NSAF oversamples low-income households, it also allows us to compare the outcomes of children living in higher- and lower-income families in the state. The NSAF used a random digit dialing technique to generate a sample of households. In areas with limited telephone coverage, the survey supplemented the telephone sample with an area probability sample of households without telephones. From the households in the sample, 50,355 extended interviews were conducted in the entire NSAF sample, including 48,679 telephone interviews and 1,676 in-person interviews (Vaden-Kiernan et al., 2000). The weighted household response rate for the entire NSAF sample was about 62 percent and the response rate for the California sample was about 56 percent. These response rates compare favorably to other surveys of this type (Brick et al., 2000).3 As part of the selection process, the survey screened households to identify the characteristics of its members. In households with children, up to two children could be randomly selected for analysis. However, only one child under age six and one child between the ages of six and 17 could be selected. In about 76 percent of the California households with children in the study, an interview was conducted for only one child; in the remaining 24 percent of households, interviews were conducted for two children. After selecting these children, the NSAF directed questions about their well-being to the person in the household who knew the most about the child—almost always the child’s parent.4 About 79 ____________ 3The telephone response rate for the entire sample was about 61 percent and for the California sample was about 55 percent. The area household response rate was about 86 percent for the entire sample and 91 percent for the California sample (Brick et al., 2000). 4In California, about 92 percent of the adult respondents were the biological parents of the child, 2 percent were stepparents, 2 percent were grandparents, 1 percent were adoptive parents, 1 percent were the partners of the parent, 1 percent were siblings, and less than 1 percent were foster parents, aunts, or uncles. Most children (58 percent) were living in families with two, married biological parents or with a single parent who was their biological mother (19 percent). 10 percent of these adult respondents were female. For purpose of this report, we shall refer to the person reporting on the child as the child’s parent although in a small minority of interviews, the most knowledgeable adult was a surrogate parent. Sample Characteristics Table 2.1 illustrates selected characteristics of children, their parents, and their households in the 1999 NSAF, comparing characteristics of the sample in California to those in the rest of the nation.5 As this table shows, a little over half of the sample of children in California and the rest of the United States is male, and the sample is fairly evenly divided among children in the three age groups (i.e., 0–5, 6–11, and 12–17) for which child well-being is measured in our study.6 The majority of children in the California sample have parents who are either white (43 percent), foreign-born Hispanic (27 percent), or U.S-born Hispanic (15 percent). Only a small percentage of children in the California NSAF have black or Asian parents (7 percent each). Because survey interviews were conducted in English and Spanish, households where neither English nor Spanish was spoken could not participate in the survey (Black and Safir, 2001), possibly leading to some underrepresentation of Asian children.7 Unfortunately, the small number of children with Native American parents in the California sample prevent us from ____________ 5These frequencies are weighted to account for the original possibility of the household being selected, for subsampling of respondents, and for nonresponse. The weight is also adjusted to correct for undercoverage using 1990 Census information and is adjusted for Census undercount. 6Data from the 2000 Census indicate that about 32 percent of children in California are under age 6, 36 percent are ages 6–11, and 32 percent are ages 12–17 (U.S. Census Bureau, 2002). 7We look at the race and immigrant status of the parent rather than the child because children living in Hispanic immigrant families are expected to have different outcomes than children living in nonimmigrant families. Data from the 2000 Census indicate that about 44 percent of children in California are Hispanic, 35 percent are white, 7 percent are black, 9 percent are Asian, 0.5 percent are Native American, and 4.5 percent are identified as other, including children with two or more racial/ethnic classifications (U.S. Census Bureau, 2002). 11 Table 2.1 NSAF Sample Characteristics of Children in California and the Rest of the United States (Weighted %) Gender Male Female Age 0–5 6–11 12–17 Race/ethnicity of parent Asian Black Hispanic, foreign-born Hispanic, U.S.-born White Other Rest of the California United States 52 51 48 49 35 33 35 34 30 33 72 7 13 27 5 15 6 43 72 11 Foreign-born parent % Hispanic % Non-Hispanic Parent’s education Less than high school High school or GED More than high school 30 9 68 48 32 52 20 11 29 36 51 53 Family income relative to poverty level (%) 300 21 18 22 23 18 20 38 40 Sample size 1,917 34,021 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. 12 comparing how these children are faring in relation to children from other racial and ethnic groups in the state.8 Compared to the sample in the rest of the United States, the California sample has more Hispanic and Asian children and fewer white and black children who are not of Hispanic origin. The California sample also has a much larger proportion of children from immigrant families than does the sample for the rest of the nation, with about 30 percent of children in the California sample having a foreign-born parent. Over two-thirds of foreign-born parents in California are Hispanic, compared to less than half in the rest of the nation. In both California and the rest of the nation, most other foreign-born parents are Asian (about one-fifth). Children in California are also slightly younger than children living elsewhere as a result of the higher birth rate of the immigrant population (Johnson, Hill, and Heim, 2001). It is also important to examine parents’ education and household income to better understand the socioeconomic circumstances in which the children are living.9 In California as in the rest of the nation, about half of children’s parents (that is, parents interviewed in the survey) had a high school diploma or less. However, almost twice the proportion of parents in California did not have a high school diploma or General Equivalency Degree (GED) as those in the rest of the nation. This discrepancy is likely related to the number of foreign-born Hispanic respondents in the California sample.10 Children in the California ____________ 8Specifically, 24 children in the California sample had parents who identified themselves as Native American, American Indian, Aleutian, or Eskimo, representing just over 1 percent of the sample. Although these children were kept in the analysis, we could not display results for Native American children separately in the tables. 9See Converse, Safir, and Scheuren (2001, pp. D1–2) for comparisons of estimations for adult nonelderly employment earnings and household size distributions in the Current Population Survey (CPS) and the NSAF. The authors report that when they analyze these measures (as well as measures of family composition, work experience, income, and poverty) by key demographic characteristics, the NSAF and CPS estimates are similar and generally within normal sampling variation. 10About 59 percent of foreign-born Hispanic parents in the California NSAF do not have a high school diploma or GED. 13 sample are also somewhat more likely than children in the rest of the nation to be living in households with incomes below the poverty level. Specifically, about 21 percent of California children live in households that have incomes below the official poverty line and an additional 22 percent live in households that are nearly poor, with incomes between 100 and 200 percent of the poverty line. Measures of Children’s Well-Being The NSAF collected information on a wide array of items that pertain to children’s well-being. For purposes of this report, we decided to restrict our analysis to those indicators that directly refer to child outcomes—that is, measures of the child’s current behavior or adjustment in the family, school, and larger community. Most of these measures are well-established indicators of children’s developmental status that have been used in previous national surveys.11 We selected indicators from four important cornerstones of children’s development: physical health, emotional adjustment, attachment to school, and pro-social involvement (see Table 2.2). Depending on the type of item in the survey, we used either separate questions or indices made up of a set of related questions. However, the discussion below describes each item individually, whether or not it was analyzed separately or as part of a summative index or scale. Because the items were intended to be age-appropriate, different questions were often asked for children in different age groups. The remainder of this chapter describes the components of each of the four domains that we examine in later chapters of this report. Physical Health The analysis includes four measures of children’s health. The first indicator is a summary question designed to provide a general assessment of the child’s physical health. The primary caregiver was asked to rate the child’s health on a scale from poor to excellent. This single measure ____________ 11For a more detailed discussion of the measures as well as the quality of the data, the internal reliability of the scales, and construct validity, see Ehrle and Moore (1999). 14 Table 2.2 NSAF Measures of Children’s Well-Being Children 0–5 Years of Age Children 6–11 Years of Age Children 12–17 Years of Age Physical Health Now, I’d like to talk about child’s health status. In general, would you say child’s health is 1. Excellent 2. Very good 3. Good 4. Fair 5. Poor. About how many of (his/her) visits to a doctor or other medical professionals that you just told me about were for well-child care, such as checkups? During the past 12 months, how many times did child see a dentist or dental hygienist? Does child have a physical, learning, or mental health condition that [limits (his/her) participation in the usual kinds of activities done by most children (his/her) age/limits (his/her) ability to do regular school work]? 1. Yes 2. No Emotional Adjustment I am going to read a list of items that sometimes describe children. For each item please tell me if it has been often true, sometimes true, or never true for child during the past month. a. (He/she) doesn’t get along with other kids. b. (He/she) can’t concentrate or pay attention for long. c. (He/she) has been unhappy, sad, or depressed. 15 I am going to read a list of items that sometimes describe children. For each item please tell me if it has been often true, sometimes true, or never true for child during the past month. a. (He/she) feels worthless or inferior. b. (He/she) has been nervous, high-strung, or tense. c. (He/she) acts too young for (his/her) age. I am going to read a list of items that sometimes describe children. For each item please tell me if it has been often true, sometimes true, or never true for child during the past month. a. (He/she) has trouble sleeping. b. (He/she) lies or cheats. c. (He/she) does poorly at schoolwork. During the past 12 months, how many times has child received mental health services, including mental health services received from a doctor, mental health counselor, or therapist? Table 2.2 (continued) Educational Attachment For each of the following statements, please tell me if you think it describes child all of the time, most of the time, some of the time, or none of the time. a. Child cares about doing well in school? b. Child only works on schoolwork when forced to? c. Child does just enough schoolwork to get by? d. Child always does homework? 16 During the past 12 months has child been suspended or expelled from school? This includes both in-school and out-of-school suspensions. 1. Yes 2. No During this past 12 months, how many times has child skipped school, cut classes without your permission, or refused to go to school? Was it . . . 0. Never 1. Once 2. Two or more times Pro-Social Involvement In the last year, has child been on a sports team either in or out of school? 1. Yes 2. No In the last year, has child taken lessons after school or on weekends in subjects like music, dance, language, or computers? 1.Yes 2. No In the last year, has child participated in any clubs or organizations after school, or on weekends, such as scouts, a religious group or Girls or Boys club? 1. Yes 2. No In the last year, has child participated in any clubs or organizations after school, or on weekends, such as a youth group or student government, drama, band or chorus, or a religious or community group? 1. Yes 2. No is commonly used in surveys and has been shown to be strongly correlated with specific health problems and to be a robust predictor of future health status.12 In addition to this overall evaluation of the child’s health, informants were also asked about the number of times the child had seen a doctor or medical health professional in the preceding 12 months for well-child care, such as checkups. This item provides both a measure of access to routine health examinations as well as an indication of the child’s health needs. Informants were also asked about the number of times children had seen a dentist in the past year. Finally, respondents were asked if the child had a physical, learning, or mental health condition that limits his or her activity. Questions about visits to a dentist and mental health professional were asked only for children ages three and older. Emotional Adjustment One standard indicator of children’s adjustment during childhood is a behavioral problems inventory derived from a checklist of symptoms originally developed by T. M. Achenbach (Achenbach and Edelbrock, 1979; Achenbach, McConaughy, and Howell, 1987). This inventory has shown that children with more frequent symptoms (as reported by parents) are more likely to encounter mental health problems and to engage in deviant behavior in adolescence and adulthood. These behaviors are typically examined together because each individual symptom is far less informative about the child’s emotional status than the index as a whole. Some studies have suggested that children may experience “internal symptoms” that relate to later mood disorders and depression, whereas other children manifest “external symptoms” by acting up or displaying aggressive behaviors (Compas and Hammen, 1994; Maccoby, 1998; Peterson and Zill, 1986). In largely descriptive ____________ 12The California Health Interview Survey (CHIS) shows results for children’s health status that are comparable to those of the NSAF for California, with most children indicated to be in very good or excellent health. However, fewer children in the CHIS are reported to be in excellent health and more are reported to be in fair, good, or very good health than those in the NSAF (based on authors’ calculations using the AskCHIS data query system available at http://www.chis.ucla.edu/). Unlike the NSAF, children ages 12–17 were asked to report their own health status in the CHIS. 17 studies such as this one, most researchers sum up the entire list of symptoms as an indicator of emotional and social adjustment. The NSAF uses a short-form adaptation of Achenbach’s longer list of symptoms. This measure has been used in previous studies conducted by telephone and seems to be both reliable and a good predictor of future emotional and behavioral problems.13 In addition, we also included a measure of emotional health—i.e., whether or not a child had received mental health services in the past year. Educational Attachment The ability to function in school is one of the earliest and most powerful predictors of whether children will develop their human capital, or skills and knowledge that improve the possibility for productive employment later in life. This process of commitment to and confidence in learning begins early in life—as early, in fact, as the preschool years— and continues throughout childhood and adolescence. Research has shown that children’s habits in the classroom predict their early educational achievement and their later attachment to schooling and eventually to the workforce (Brint, 1998; Danziger and Waldfogel, 2000). For example, Danziger and Waldfogel (2000) find that early school achievement forecasts long-term employment patterns and earnings ability. The NSAF includes four questions on children’s concern about doing well in school and their commitment to doing their schoolwork. These items have a high reliability (that is, they are strongly intercorrelated) and therefore can be used as a single scale of school engagement.14 The separate items listed in Table 2.2 are summed ____________ 13Ehrle and Moore (1999) used the 1997 NSAF data to benchmark this measure against a similar measure in the National Longitudinal Survey of Youth (NLSY). They found similar patterns across socioeconomic groups, with behavioral and emotional problems increasing with lower socioeconomic status. However, because children in the NLSY were more disadvantaged than those in the NSAF, children in the NLSY had a higher incidence of behavioral and emotional problems. This was consistent with the authors’ expectations. 14Ehrle and Moore’s (1999, pp. 1–2) report does not provide a benchmark comparison of the educational engagement scale because this measure had not yet been used in other national surveys at the time of their analysis. However, they show that the 18 together in a scale ranging from 0 to 4. The NSAF also includes information about voluntary and involuntary school absences for children ages 12–17 that provides indicators of negative school behaviors. These questions ask whether children have been suspended or expelled from school and the number of times they have skipped school in the past year. Pro-Social Involvement Developmental psychologists are paying increasing attention to children’s involvement in social activities because this involvement helps to build their capacity to establish strong relationships with peers; increases nonacademic skills such as communication and development of talents; and fosters self-esteem, a sense of efficacy, and psychological resourcefulness. Other dividends of positive social involvement include increasing children’s ability to form lasting connections that may be helpful in times of trouble and shielding children from the influence of negative peers (Carnegie Council on Adolescent Development, 1992; Eccles and Barber, 1999; Mahoney 2000; Pierce, Hamm, and Vandell, 1999). The NSAF includes a set of items that asks parents whether their children are involved in clubs or sports teams or take lessons outside of school to cultivate skills.15 Because many children were not involved in all three of these activities, we summed up the number of children who were involved in zero, one, or two or more types of social groups and activities. ________________________________________________________ measure varies as expected in the subgroup analysis, with lower socioeconomic groups showing lower school engagement. In addition, the authors note that the quality of the data in regard to missing data, distribution, and the alpha—a statistical measure of reliability—also indicates that the data can be used with confidence. 15The NSAF measures of pro-social involvement were benchmarked against those in the Survey of Income and Program Participation (SIPP) and the National Education Longitudinal Survey (NELS) (Ehrle and Moore, 1999). However, these comparisons were not exact in that the question wording differed somewhat across surveys and because the NELS sample differed from the NSAF sample. Ehrle and Moore found that more children in the NSAF were involved in activities than were those in the SIPP and NELS, likely because the NSAF question was more inclusive. Patterns by socioeconomic groups were similar across all three surveys and varied in the expected direction, with lower socioeconomic groups reporting less involvement. See Ehrle and Moore (1999, pp. 1-5; 7-4 to 7-6). 19 As we noted above, one obvious limitation of the NSAF data is that all of the measures we employ rely exclusively on one informant—nearly always the child’s parent. Ordinarily, and especially for younger children, this strategy poses no problems because parents are known to provide reasonably reliable and valid reports of how their offspring are doing. However, parents’ ability to detect problems in children’s behavior, particularly outside the home, may be less accurate when children reach early adolescence. Consequently, we suspect that parents’ reports for the older children in the NSAF sample may be more positive than those that could have been obtained if information had been gathered from teachers or children themselves. 20 3. Physical Health This chapter examines how the demographic and socioeconomic characteristics of California’s children and their families relate to indicators of children’s physical health. The investigation of children’s health indicators is important for two reasons. First, it offers useful information to those concerned with designing health policy and providing health care to children in California by directing attention to segments of the population where health needs are greatest and most underserved. Second, the findings in this chapter help explain the health status differential between California and other states that we identify and discuss in Chapters 7 and 8. We begin by examining how health indicators vary with the gender and age of children and then turn to health variations associated with family characteristics, including race and ethnicity, parental education, and family income. Gender and Age Research indicates that adult males experience higher rates of mortality and morbidity than females (Knudsen and McNown, 1993; Waldron 1995). However, national surveys do not show that the general health status of children differs greatly by gender on measures identical to those in the NSAF (e.g., National Center for Health Statistics, 2002). Accordingly, we would be surprised to discover large health differences between boys and girls in California as reported by parents. Consistent with other national surveys, the NSAF findings show that differences in boys’ and girls’ overall health status are not substantial in California (see Table 3.1). At the same time, parents report that boys (9 percent) are somewhat more likely than girls (6 percent) to have a health condition that limits their activity. This modest difference in chronic health problems does not seem to be reflected in parents’ overall evaluations of their son’s health, perhaps because some of these children had temporary conditions (such as a broken leg) that incapacitates them 21 Table 3.1 Health Status of Children in California (Weighted %) Child’s current health status Poor Fair Good Very good Excellent Gender Age Group Female Male ≤ 5 6–11 12–17 Total 1 11 01 1 6 55 57 5 17 15 16 15 17 16 25 25 21 29 25 25 52 54 58 50 50 53 Has condition that limits activity 6 93 8 11 8 Well-child doctor visit during past 12 months No One Two or more 35 37 16 39 56 36 37 36 33 43 34 37 28 27 51 18 11 27 Saw a dentist during past 12 months 76 77 66 81 77 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. 77 but does not seriously compromise their health. Despite these differences, boys and girls had comparable levels of well-child visits as indicated by their parents. Similarly, we find that boys and girls had almost identical levels of routine dental care. Thus, it appears that access—or lack of access—to health care does not vary by gender. Table 3.1 also presents information about children’s health status by their age group. Although national data indicate few changes in children’s health status between early childhood and adolescence, our evidence points to a very small deterioration of health status among older children. Specifically, more children under age 12 (79 percent) than over age 12 (75 percent) are in excellent or very good health, according to their parents’ reports. Slightly more adolescent children (8 percent) than younger children (5–6 percent) are reported to be in fair or poor health. We see that the incidence of chronic health conditions that limit children’s activity is also higher among older children, rising steadily from 3 percent in early 22 childhood, to 8 percent in middle childhood, to 11 percent in the adolescent years. As Figure 3.1 shows, the increase is much steeper for boys than for girls, especially as they move from middle childhood to adolescence. By their teen years, 16 percent of males have a health condition that limits their routine activities—a figure more than twice as high as that for adolescent females. When we turn to the health indicators relating to the provision of care, we see that older children in California are far less likely than younger children to have had any contact with a physician in the past year (Table 3.1). The strong decline in routine health care we observe among older children is especially disturbing because parents report that their adolescents are in poorer health. According to these reports, children age five and younger had the most health visits, with more than twice as many having some contact with a physician or other medical professional as children age 12 and older. In particular, approximately 16 percent of children under age six had not seen a physician or other medical professional in the past year, compared to about 39 percent of 18 16 Boys 14 Girls 16 Percentage 12 10 9 8 77 6 4 4 2 2 0 0–5 6–11 Age group 12–17 Figure 3.1—Children with Chronic Health Condition That Limits Their Activity, by Gender and Age Group (Weighted %) 23 children between the ages of six and 11 and 56 percent of children between the ages of 12 and 17. Even though the majority of older children do not appear to be experiencing health problems, few receive routine examinations that could prevent or detect such problems should they occur. The absence of contact with a physician also means that relatively few children receive the benefit of screening for mental health problems, sexual and reproductive health concerns, substance abuse, and a variety of other medical issues that arise in the adolescent years. Adolescent boys are especially likely not to have received routine medical care such as checkups, with about 62 percent of boys ages 12 to 17 reported to have had no medical visits, compared to 50 percent of girls in this age group (figures not shown in table). A somewhat different pattern emerges when we look at dental visits. As might be expected, dental checkups are lowest for children between the ages of three and five: approximately one-third of children in this age group had not seen a dentist in the previous year.1 This figure drops to about one-fifth in middle childhood and rises again to about one-quarter for adolescents. As with the pattern we observe for physician visits, we again find that parents of older children report that more males than females have not visited a dentist. About 27 percent of adolescent boys did not see a dentist in the past year, compared to about 18 percent of adolescent girls (figures not shown in table). Family Characteristics There is a large literature showing that characteristics related to children’s family background strongly affect their health outcomes (Millstein, Petersen, and Nightingale, 1993; Zuckerman and Kahm, 2000). For example, nonwhites have greater health needs and less access to health care (Elo and Preston, 1996; Hayward et al., 2000; Williams and Collins, 1995). This is due, at least in part, to the socioeconomic circumstances of these families, who are less connected to the health system, who may be less aware of the need for preventive care, and who may lack important resources such as the time and money to get routine health visits. ____________ 1The question about dental visits was asked only for children ages three and older. 24 We begin by examining whether indicators of children’s health vary by their racial/ethnic descent, their parent’s level of education, and their family income. When looking at Hispanic ethnicity, we divided the sample into Hispanic children whose parents (based on the status of the parent interviewed) were born in the United States and those whose parents were born outside the United States. In Table 3.2a, we see that parents’ evaluations of their children’s health are somewhat lower for children of black and native-born Hispanic parents and distinctly lower for children of foreign-born Hispanics. Whereas only 2 percent of white and Asian children are in fair or poor health according to their parents’ reports, the figure rises to 5 percent for black children, 8 percent for children of native-born Hispanics, and 14 percent for Hispanic children whose parents were born outside the United States. A similar pattern is evident if we look at the proportion of racial/ethnic groups who report that their children are in very good or excellent health. About 54 percent of Hispanic children in immigrant families appear in the top two categories compared to 80 percent of Hispanic children in nonimmigrant families, 83 percent of black children, and approximately 90 percent of white and Asian children. Sharp differences in children’s health status are also apparent when we examine children’s health by variations in their parents’ education and family income. About 55 percent of children whose parents have less than a high school education are reported to be in excellent or very good health as compared to 77 percent of children whose parents have a high school diploma and 88 percent of children whose parents have education beyond high school. We see the similarly steep gradients in health differences by family income, particularly between children in the lowest income category (whose families have incomes below the poverty line) and those in the highest category (whose families have incomes over 300 percent of the poverty line). Specifically, only 64 percent of children in the poorest families are reported by their parents to be in very good or excellent health compared to 90 percent of those in the highest income category. We observe much less variation among children by their race/ ethnicity, parents’ education, and family income when we look at a 25 Table 3.2a Health Status of Children in California, by Race/Ethnicity, Level of Parent’s Education, and Family Income (Weighted %) Child’s current health status Poor Fair Good Very good Excellent Asian Race/Ethnicity Hispanic Foreign- Hispanic Black Born U.S.-Born White Parent’s Education Less More Than Than High High High School School School Family Income < 100% 100– 200– of 200% of 300% of Poverty Poverty Poverty > 300% of Poverty Total 10 15 7 11 31 31 60 52 1 13 32 20 34 1 7 12 29 51 02 2 14 10 30 25 18 64 37 000 5 2 12 17 9 23 29 26 22 48 62 42 1 7 23 22 47 1 3 15 27 54 11 25 8 16 28 25 62 53 26 Has condition that limits activity 4 8 5 11 9 6 9 7 8 10 7 67 Well-child doctor visit during past 12 months No One Two or more 37 36 28 20 42 38 40 31 29 30 35 35 39 42 40 28 21 30 32 36 35 38 40 34 30 24 32 38 31 31 35 39 25 36 36 41 37 23 27 Saw a dentist during past 12 months 83 89 56 79 84 SOURCE: 1999 National Survey of America’s Families. NOTE: Perentages may not sum to 100 because of rounding. 59 73 86 63 69 82 86 78 second indicator of children’s health—whether or not they have health conditions that limit their physical activity. Differences in children’s health on this indicator are smaller and do not follow consistent or expected patterns, suggesting that these health problems are not strongly linked to socioeconomic status. However, this indicator suggests that about one out of ten Hispanic children living in nonimmigrant families and the same proportion of children living in “near poor” families (whose incomes fall between 100 and 200 percent of the poverty line) experience these debilitating conditions. The indicators measuring children’s health visits with physicians or other health care professionals show a somewhat different association with family background. Black and Hispanic children in nonimmigrant families are as likely as other children to have received routine medical care once in the past year but are also more likely to have visited a physician or medical professional two or more times. This finding suggests that actual health concerns are driving the high level of contact with physicians. About 60 percent of Hispanic children with foreignborn parents who had seen a doctor two or more times in the past year were reported to be in less than excellent health, compared to 46 percent of Hispanic children with native-born parents, 45 percent of black children, 33 percent of white children, and 28 percent of Asian children (figures not shown in table). Although this question asks parents to report about “well-child” care, parents may be reporting all medical visits, including those for more serious or chronic conditions. It is also possible that this question is understood somewhat differently by different racial/ethnic groups. In contrast to the pattern of well-child care, white and Asian children had typically received dental care in the past 12 months. However, Hispanic children, particularly those whose parents were born outside the United States, were less likely than other children to have received any dental care. Only 56 percent of Hispanic children living in immigrant families had seen a dentist in the past year, according to their parents’ reports, compared to 89 percent of black, 84 percent of white, and 83 percent of children of Asian descent. The relationship of parental education to well-child visits to doctors is also complex. About 42 percent of children whose parents do not have 27 a high school diploma did not receive routine medical care in the preceding year. At the same time, children whose parents have only a high school diploma received slightly more medical care than children whose parents have gone beyond high school in their education. We find less variation by family income in the percentage of children who received any care, although children living in higher-income families were less likely to have seen a physician or medical professional two or more times in the past year. Again, we assume that this is because the children in higher-income families are in better health. As expected, children from families with lower socioeconomic status, as defined by parental education and income, were much less likely to have seen a dentist in the past year. Since family income, parental education, and race/ethnicity are strongly interrelated, we next estimate the separate effect of each of the background factors, which allows us to adjust the percentages in a multivariate analysis. Although a bivariate crosstab analysis of children’s health outcomes by variables such as their race and ethnicity is informative, it fails to account for the influence of potentially confounding variables such as family income, parental education, and children’s age and gender. To account for such effects, we calculated adjusted percentages through a method that is similar to ordinary least squares (OLS) regression analysis in controlling for background factors but that yields substantively meaningful percentages that take account of confounding variables. When we show adjusted percentages, they are, in effect, regression estimates that control for children’s age, gender, parental race/ethnicity and immigration status, parental education, and family income.2 We used unweighted data when calculating the adjusted ____________ 2We used a procedure in Stata Version 7.0 that adjusts percentages through a command called adjmean. Using this command, we calculate the bivariate relationship between each outcome variable and each demographic or socioeconomic status variable of interest. This allows us to predict the likelihood of children falling into each category of the outcome variable while holding all other demographic or socioeconomic covariates at their mean. Dummy variables are created for each category of the outcome variable, and the outcomes are expressed in terms of an adjusted mean that falls between 0 and 1. These means are then multiplied by 100. 28 percentages because the weights incorporate some of the same variables that are being used in the procedure for adjusting the percentages. This adjusted-percentages procedure allows us to evaluate the unique contribution of each component of children’s demographic and family background characteristics. In other words, the adjusted percentages show the influence of race/ethnicity on children’s health outcomes, net of gender, age, parental education, and family income. Similarly, we observe the effect of parental education and family income net of other characteristics. If we continue to see differences between children after these other characteristics are controlled for, we can have more confidence that the differences are meaningfully related to children’s membership in particular subgroups. In Table 3.2b, we look at the same associations between children’s health status and health visits and their family characteristics reported in the last table. However, the numbers in this table show the net effect of each characteristic on children’s health outcomes. As these findings suggest, race and ethnicity continue to influence parents’ health evaluations of their children once we account for other characteristics of children and their families although differences between groups are reduced somewhat. White children are the most likely to receive excellent health ratings from their parents, closely followed by children of native-born Hispanic parents, blacks, and Asians. Hispanic children whose parents were born outside the United States are least likely to receive excellent ratings and most likely to receive poor or fair health ratings from their parents. We also find that both parental education and family income are independently related to the health status of children, net of other characteristics. After taking children’s demographic and family characteristics into account, children’s health status rises steadily as their parents’ education and family income go up. As these data show, children’s health status is sharply differentiated by both parental education and family income, suggesting that parents’ knowledge and access to health services both play an important part in their child’s health outcomes. 29 Table 3.2b Health Status of Children in California, by Race/Ethnicity, Level of Parent’s Education, and Family Income (Adjusted %) Child’s current health status Poor Fair Good Very good Excellent Asian Race/Ethnicity Hispanic Foreign- Hispanic Black Born U.S.-Born White Parent’s Education Less More Than Than High High High School School School Family Income < 100% 100– 200– of 200% of 300% of Poverty Poverty Poverty > 300% of Poverty Total 21 1 1 1 2 1 1 0 1 1 3 7 10 5 3 11 5 3 9 7 4 16 8 28 11 11 28 16 11 20 18 17 31 32 21 29 24 19 28 26 24 24 27 47 53 40 53 61 41 50 59 47 50 51 11 46 12 15 25 28 58 51 30 Has condition that limits activity 78 5 9 10 9 11 8 9 11 8 67 Well-child doctor visit during past 12 months No 45 21 35 31 40 40 35 35 33 39 38 One 30 43 39 36 37 31 37 40 40 34 37 Two or more 25 37 26 33 23 29 27 25 27 27 25 35 34 38 37 27 29 Saw a dentist during past 12 months 73 90 67 79 79 67 72 82 70 68 79 SOURCE: 1999 National Survey of America’s Families. NOTE: Perentages may not sum to 100 because of rounding. 82 78 It is interesting to note that parents’ reports of “objective” health conditions that limit children’s activities do not follow the same pattern as their health ratings. Again, we find little variation in these conditions by children’s race and ethnicity, except for children of Hispanic immigrants for whom it is lower. Similarly, parents’ education does not appear to be associated with children’s health limitations and family income appears to be only slightly related to these limitations. Although somewhat puzzling, this finding might suggest that parents with lower socioeconomic status or who are Hispanic immigrants may underreport their children’s chronic conditions. As described above, some of these limiting conditions may be temporary and may not compromise children’s overall health status in the eyes of their parents. After taking other factors into account, we find that children whose parents are in the lowest education category are more likely to have had no contact with a physician or health professional for routine care. However, they are also slightly more likely to have received care two or more times. Children whose parents are Asian, foreign-born Hispanic, and white are less likely to have seen a physician, and black and Hispanic children in nonimmigrant families are more likely to have seen a physician two or more times. It is of particular concern that Hispanic children in immigrant families are not receiving routine care because far fewer are described as being in very good or excellent health. Although the number of Asians in the study is too low to separate reliably children into subgroups by ethnicity and origin, we suspect that variation across these subgroups is substantial (see Reyes, 2001). We see a somewhat different pattern in the adjusted percentages for family income than we saw in the unadjusted figures. In particular, we find that children whose families have incomes between 100 and 300 percent of the poverty line are less likely to have received routine medical care than those at the higher and lower end of the income spectrum. Some of these parents do not qualify for state-subsidized health insurance (i.e., those between 250–300 percent of the poverty line) and others may not know that they qualify, thus limiting their contact with physicians. Again, another possible explanation is that higher-income parents may not feel that their child needs a routine health visit because they are in better health. 31 In the adjusted percentages, we continue to find that black children are the most likely to have seen a dentist in the last year, followed by white children and Hispanic children living in nonimmigrant families, and Asian children. Hispanic children in immigrant families are the least likely to have had a dental visit in the previous 12 months. The disparities between children of foreign-born Hispanic parents and those of other racial/ethnic groups are reduced when children’s other characteristics are held constant; however, they remain pronounced. As these findings indicate, Hispanic children in immigrant families and Asian children are less likely to have had a dental visit than other ethnic groups, especially blacks. Again, we suspect that children of foreign-born Asian parents have fewer dental checkups than other Asian children, but the numbers are too small to provide a reliable estimate of this difference. As might be expected, both family income and parental education are positively related to dental checkups, net of other characteristics. About two-thirds of children whose parents are high school dropouts and of poor or near-poor children have seen a dentist in the previous 12 months. This compares to over four-fifths of children whose parents have had post-high school education and whose families have incomes above 300 percent of the poverty line. When we adjust for other factors, these figures show a troubling pattern: Children in immigrant Hispanic, Asian, less-educated, and poor or near-poor families have limited access to dental care. Conclusion About 36 percent of children in California have not received routine medical care in the past year. Consistent with previous studies, this chapter reveals a strong disparity between children’s health needs and their use of health services that is strongly related to their age. As children’s health concerns and problems increase, especially in adolescence, their health visits drop off sharply. The deficit in both physician and dental care during the adolescent years is especially pronounced for males. Although adolescent males have greater health needs, they tend to use services less than adolescent females and younger children generally. 32 Children’s race and ethnicity and the socioeconomic characteristics of their families are also strongly associated with their health outcomes. Children living in Hispanic immigrant families, those living in poor families, and those whose parents are less educated are in worse overall health than other children. Although poor children seem to have comparatively good access to medical care, children with foreign-born Hispanic and less-educated parents often have not had contact with a physician or other medical health professional for well-child care in the last year. When we adjust for other factors, Asian children also have a lower likelihood of receiving routine medical care. Children are more likely to have received routine medical care when they live in families at higher and lower ends of the income spectrum. It is possible that the children of better-off parents and who are white are not getting enough routine health care; however, they are likely not to need as much care as other children. It is also possible that poor children are more likely to have health insurance coverage than children whose families have incomes between 100 and 300 percent of the poverty line. At the same time, children living in lower socioeconomic status families are far less likely than other children to have received routine dental checkups. More than 30 percent of children whose parents are Hispanic immigrants, whose parents have the lowest levels of education, or who live in poor and near-poor families have not seen a dentist in the past year. 33 4. Emotional Adjustment Most parents care deeply about their children’s emotional state and ability to get along with others—qualities that also matter greatly in how well children function inside the home, in the community, and in school. For example, behavioral problems can disrupt the learning of children and their classmates as well as increase the burden on teachers and staff. Children’s emotional and social adjustment problems are also strongly associated with subsequent mental health disorders, substance abuse, and problems with the criminal justice system. Thus, concern about children’s maladjustment by parents, teachers, and other caregivers is well justified. Moreover, the high social and economic costs to the larger society that are incurred when children function poorly make this area an important issue for public policy. As a result, considerable public and private resources are devoted to ensuring that the mental health needs and behavioral adjustment problems of children in California and the rest of the nation are adequately addressed (Carnegie Corporation of New York, 1996; Wolfe, 1995). As described in Chapter 2, we examine two general indicators of children’s socioemotional functioning. The first is a commonly used checklist of items to tap different domains of age-appropriate social and emotional adjustment. The first measure combines six NSAF items into a single scale to measure each child’s overall level of problems. We know from previous studies that a scale of behavioral problems that sums up particular symptoms of maladjustment is a more trustworthy predictor of children’s likelihood of encountering emotional or social problems in the future than any of the single items alone. When put together in a scale, the indicators provide a reliable measure of the child’s overall 35 behavioral adjustment.1 The NSAF includes items to measure six problem behaviors. Three of these items were asked about children ages six and older. An additional set of three items was asked about older children and another, three-item set was asked about younger children.2 The checklist was not administered for children below the age of six. To construct this scale, we separated the children into three categories depending on the frequency and severity of their problems. Well-adjusted children or those with few problems are reported not to experience any problems “often” or to show problems on more than two items “sometimes.” Poorly adjusted children or those with severe problems are reported to often have problems on at least two of the six items or to sometimes have problems on at least four of the six items. Children with moderate problems are reported to fall into these two extremes. It is important to keep in mind that parents are asked to assess their children’s behavior, so this measure is likely to reflect primarily children’s behavior within the household. At the same time, parents probably also base their evaluations on reports from other adults, such as relatives, neighbors, and teachers, who can assess their children’s behavior outside the home. The other indicator is a single question in the NSAF that asks ____________ 1A scale is the sum of the scores of the individual variables. In generating the scale, a score was created for every observation for which there was a response to at least one variable and divided by the variables over which the sum was calculated (StataCorp, 1997). Cronbach’s alpha measures the reliability of a scale. Specifically, it measures how well the variables measure a single, unidimensional latent construct. As the average interitem correlation increases, the alpha will also increase. The alpha is also affected by the number of variables in the scale (UCLA Academic Technology Services, 2001). 2We created one subscale for children ages 6–11 by taking the sum of six related items: doesn’t get along with other kids, can’t concentrate for long, has been sad or depressed, feels worthless or inferior, has been nervous or tense, and acts too young for his age. The alpha, or reliability score for this scale is .77. We create a second subscale for children ages 12–17 by taking the sum of another six related items: doesn’t get along with other kids, can’t concentrate for long, has been sad or depressed, has trouble sleeping, lies or cheats, and does poorly at school work. We created the scale for children’s behavioral problems by taking the sum of these two subscales and splitting it into three categories. The alpha for this scale is .73. 36 whether children have received services from a mental health counselor in the past 12 months. Gender and Age Most research shows that boys and girls manifest different symptoms of mental health problems (Maccoby, 1998). Boys are inclined to “act out” their problems whereas girls are more likely to respond to problems by withdrawing. Overall, boys tend to show more symptoms of poor mental health and are more likely than girls to engage in problem behaviors in childhood and adolescence. Generally, problems for both boys and girls increase with age. Although we can examine gender differences, the particular items employed in NSAF are designed to be age-specific, possibly making it less likely that we would detect increases in problem behavior in higher age groups. Our findings reveal that over two-thirds of the girls in California do not have notable behavioral problems and that only 8 percent reveal severe problems, according to their parents’ reports (see Table 4.1). By contrast, a somewhat lower percentage of boys are reported to have no behavioral problems, and 13 percent have severe problems. These results are consistent with previous findings on children’s emotional health. Contrary to our expectations, however, the results in Table 4.1 show no increase in problem behavior as children move from early childhood to adolescence.3 Looking at the proportion of children who have received mental health services in the previous 12 months, we observe some discrepancies between children’s need for services (as indicated by the responses to the behavioral-problem checklist) and their receipt of services (Table 4.1). Although parents report that boys have more severe symptoms than girls, boys are no more likely to have received mental health services in the previous 12 months. Overall, only 5 percent of boys and girls have received any counseling. Among children with the most serious symptoms, approximately 25 percent of girls and 23 percent of boys have ____________ 3Questions about children’s behavioral problems were not asked for children ages five and under. 37 Table 4.1 Children’s Behavioral Problems, by Gender and Age (Weighted %) Child behavioral problems scale Low Moderate Severe Gender Age Group Female Male ≤ 5 6–11 12–17 Total 69 64 22 23 8 13 66 67 23 23 11 11 66 23 11 Received mental health services in past 12 months 5 52 5 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. 6 5 received mental health services (figures not shown). Receipt of services is almost identical among children in middle childhood and adolescence, with about one in twenty receiving any treatment in the previous year. Not surprisingly, the percentage of children ages three to five who have received service is even lower, at 2 percent.4 Family Characteristics Abundant evidence demonstrates that children living in families with low socioeconomic status are much more likely to experience behavioral problems than other children (National Research Council, 1993; Wolfe, 1995). There are many reasons for this. Poverty itself is linked to higher levels of family stress. However, low-income parents also have fewer resources to address problems when they arise. Parents with less income and education may also have a greater incidence of social and mental health problems that in turn affect their parenting skills or their ability to respond to children’s needs. Children in low-income families are also more likely to be living in single-parent households and thus to experience greater family flux and strain resulting from such living arrangements. ____________ 4The question about mental health services was not asked for children under age three. 38 Research that has examined whether there is a relationship between race/ethnicity and behavioral problems in children is much less conclusive. Some racial/ethnic groups may be more prone to certain types of problems than others, but there is no consensus that the level of problem behavior among children, especially younger children, varies by racial/ethnic group, particularly when socioeconomic status is taken into account (Furstenberg et al., 1999). Even before we adjust for children’s other characteristics, we do not observe large racial/ethnic differences in our index of children’s behavioral adjustment (see Table 4.2a). The major difference across these groups is that Asian children have fewer severe problems than other children. There are clearer differences in children’s problem behavior when we examine this indicator by their parent’s educational level and family income, confirming the results of previous studies. According to their parents’ accounts, children living in poor households and who have less-educated parents are more likely to experience problems. It should be noted that the vast majority of children—including those living in families with the lowest incomes and levels of education—are not encountering any serious problems. Still, one in seven of those children whose parents did not complete high school and almost one in five of those children living in the poorest families have severe behavioral problems that might put them at risk of experiencing social and mental health problems in later life. After we adjust for the association between children’s demographic and family characteristics (in Table 4.2b), the relationship between parental education and children’s behavioral adjustment diminishes. We still see that the children of parents who had education beyond high school do slightly better than those whose parents had less education. The adjusted percentages show that the relationship between children’s behavioral problems and family income persists, but this is pronounced only for very poor children. Thus, it appears that problem behaviors among children, at least as reported by their parents, are relatively evenly distributed across racial/ethnic and socioeconomic groups, except for children living in very poor families where about 15 percent of children are reported to have severe problems. 39 Table 4.2a Children’s Behavioral Problems, by Race/Ethnicity, Parent’s Education, and Family Income (Weighted %) 40 Child behavioral problems Low Moderate Severe Asian Race/Ethnicity Hispanic Hispanic Foreign- U.S.Black Born Born White Parent’s Education Less More Than Than High High High School School School Family Income 100– 200– < 100% 200% 300% of of of Poverty Poverty Poverty > 300% of Poverty Total 69 67 61 64 69 60 60 72 52 64 66 75 66 24 21 28 24 20 25 28 19 30 26 24 17 23 6 12 11 12 11 14 12 9 19 10 9 8 11 Received mental health services in the past 12 months 34 2 6 6 46 4 5 54 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. 55 Table 4.2b Children’s Behavioral Problems, by Race/Ethnicity, Parent’s Education, and Family Income (Adjusted %) 41 Child behavioral problems Low Moderate Severe Race/Ethnicity Hispanic Hispanic Foreign- U.S.Asian Black Born Born White Parent’s Education Less More Than Than High High High School School School Family Income 100– 200– < 100% 200% 300% of of of Poverty Poverty Poverty > 300% of Poverty Total 61 68 28 21 11 12 65 26 9 67 22 11 66 65 62 68 56 22 24 26 22 30 12 12 12 10 15 68 65 24 24 9 11 69 65 20 24 11 11 Received mental health services in the past 12 months 54 3 6 5 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. 5 5 44 44 65 When we examine children’s use of mental health services, we find that not all racial/ethnic groups are equally likely to access services (see Table 4.2a). For example, 6 percent of white children and those whose parents are native-born Hispanics, 4 percent of black children, 3 percent of Asian children, and only 2 percent of children with foreign-born Hispanic parents have used mental health services. However, these findings are more illustrative of the fact that few children overall are receiving services, particularly those in need. Although problems occur disproportionately to children of poorer and less-educated parents, we can see that their use of mental health services is no greater than that of children whose parents are more affluent. Indeed, after adjusting for other characteristics, Table 4.2b indicates that children living in the highest-income families actually use services more often than those in lower-income families. When we examine family income simultaneously with children’s need for and use of services, we find that children who live in families with incomes below 200 percent of the poverty line and experience severe problems have a much lower probability of receiving mental health services than children in families with higher incomes (see Figure 4.1). Only half as many children in lower-income families with serious 35 30 < 200% of poverty ≥ 200% of poverty 25 29 Percentage 20 15 15 10 5 0 Receiving mental health services Figure 4.1—Children with Severe Behavioral Problems Who Received Mental Health Services in Past 12 Months, by Family Income (Weighted %) 42 problems received mental health services in the past 12 months as children in higher-income families (15 percent compared to 29 percent). Similarly, children with high levels of problem behavior are much less likely to be treated if their parents have lower levels of education or are nonwhite (figures not shown). These findings indicate that children most at risk for experiencing problems in later life are being underserved because their parents may lack knowledge about mental health services, have fewer social connections, or have much more limited economic resources to obtain treatment. Conclusion Although only a small minority of children in California are experiencing severe behavioral problems, as reported by their parents, the rates of problem behavior are significant enough to be of public concern. Within subgroups of children, we find that the level of children’s need for mental health services does not correspond with their receipt of these services. Namely, males and children from lower socioeconomic status families show more problems than other children but do not receive more treatment. Particularly troubling is the low level of treatment provided to children of lower-income families whose parents say the children are displaying serious symptoms of problem behavior. Only about 16 percent of children living in or near poverty who have serious behavioral problems are receiving services. Children with severe problems are almost twice as likely to receive treatment if their families are at least 200 percent above the poverty line. 43 5. Educational Attachment Previous studies have shown that children’s attachment to education begins to form early in life—often before they enter elementary school (Entwisle, Alexander, and Olson, 1997). This research indicates that children’s experience in the home sets the stage for their success in school. California policy recognizes this fact through its support of preschool programs that aim to reduce disparities in children’s preparation for school based on their family backgrounds. It is also well known that the quality of children’s early schooling establishes the skills and motivation they need to perform well throughout their educational careers. Indeed, children’s attachment to school is an important precondition for success in middle school, high school, and beyond. Alternatively, children who lack confidence in and commitment to schooling are more likely to drop out of high school and less likely to go to college (Neisser, 1986; Lamborn et al., 1992). As discussed in Chapter 2, the NSAF included a series of items to measure children’s attachment to school, as reported by parents. The first of these is a four-item scale that asks parents to rate their children’s engagement in school. To construct the scale for children’s school engagement, we took the sum of four related items: cares about doing well in school, works on schoolwork only when forced to, does just enough schoolwork to get by, always does homework. We then split the scale into four categories to represent low, medium, high, and very high engagement.1 A second indicator measures the frequency of skipping school among children age 12 and older. On a third indicator, parents of these older children in the sample were also asked whether their child had been suspended or expelled from school in the past year. This chapter examines how the indicators of school attachment vary by characteristics of California’s children and their families. ____________ 1The alpha for the scale of school commitment is .71. 45 Gender and Age Previous research finds that children’s attachment to school varies greatly within demographic subgroups of the population. In particular, a large literature demonstrates that boys have more difficulty than girls engaging in school and show higher levels of problem behavior in school. Moreover, these gender differences widen with age, particularly as children move from primary to middle school. As young people gain more autonomy from their parents, and parents tend to have less contact with teachers, children’s disengagement from school typically rises. These differences reach their highest levels during the adolescent years when children enter high school (Eckstrom et al., 1987; Natriello, 1987). The distribution of the three indicators of school engagement by gender and age is shown in Table 5.1. The NSAF results closely mirror findings of previous studies indicating that males are noticeably more likely than females to have lower levels of school attachment. In California, many more males than females (38 percent compared to 22 Table 5.1 Children’s School Engagement, Frequency of Missing School, and Suspension/Expulsion Rates, by Gender and Age (Weighted %) School engagement scale Low Moderate High Very high Gender Age Group Female Male 6–11 12–14 15–17 12–17 Total 4 94 18 29 23 30 32 36 48 30 37 10 7 24 24 26 31 40 38 Times skipped school past 12 monthsa 0 81 79 1 87 2 or more 11 13 88 71 6 10 6 19 80 8 12 Suspended/expelled in past 12 monthsa 10 17 8 20 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. aThese questions were not asked of the 6–11 age group. 13 46 percent) have only low or moderate levels of engagement in school. Males are slightly more likely than females to have skipped school two or more times in the past year (13 percent compared to 11 percent). Males are also much more likely to have been suspended or expelled in the past 12 months (17 percent compared to 10 percent). Research has found that dropping out of high school is somewhat more common for males than females and that all of these indicators of school disengagement predict high school dropout (U.S. Department of Education, 2002). As noted above, school attachment generally declines as children move from primary school to middle school and from middle school to high school. We find that low engagement is more than twice as high among children ages 12 and older as among those ages 11 and younger (10 percent compared to 4 percent) in California. The question about skipping school was asked only of children in the oldest age group (ages 12 and above), in part because it does not occur frequently among children in elementary school. About 8 percent of students between the ages of 12 and 17 have skipped school at least once, and an additional 12 percent have skipped school two or more times in the past year. Frequent school skipping rises sharply among students moving from middle school to high school, increasing from 6 percent among children who are ages 12 to 14 to 19 percent of children between ages 15 and 17. Suspension and expulsion rates also tend to rise with age. Again, questions were asked for children ages 12 and older in the NSAF; however, we can compare the students who have likely reached high school (ages 15 to 17) with those still likely to be in middle school (ages 12 to 14). The proportion of children who have been suspended or expelled in the past 12 months climbs from 8 percent among 12 to 14 year olds to 20 percent among 15 to 17 year olds (see Table 5.1). Overall, these indicators show a troubling pattern of school detachment that is especially marked for males and older students. Figure 5.1 plots the patterns of skipping school (two or more times) and suspension or expulsion in the past year for males and females separately by age. As this figure shows, the increase by age is relatively modest for females but is quite pronounced for males. About 23 percent of high school age boys have skipped school more than twice in the past year and 25 percent have 47 30 Skipped two or more times in past 12 months (male) Suspended/expelled in last 25 12 months (male) Skipped two or more times in past 12 months (female) 20 Suspended/expelled in last 12 months (female) 15 25 23 15 14 Percentage 10 9 8 7 54 0 12–14 15–17 Age group Figure 5.1—Children Who Skipped School Two or More Times or Were Suspended or Expelled in the Past 12 Months, by Gender and Age Group (Weighted %) been suspended or expelled in the previous 12 months, compared to 15 percent of girls who have skipped school and 14 percent who have been suspended or expelled. Family Characteristics Children’s school attachment is also strongly associated with their family backgrounds as shown in Table 5.2a. Children’s scores on the school engagement index are higher for white and Asian children than for Hispanic or black children, indicating that children in the latter groups are more likely to be withdrawing from school. Almost threequarters of white and Asian children are reported be highly or very highly engaged in school, compared to about two-thirds of black and all Hispanic children. A higher percentage of foreign-born Hispanic parents report that their children are highly engaged in school (69 percent) than native-born Hispanics (63 percent). 48 Table 5.2a Children’s School Engagement, Frequency of Missing School, and Suspension/Expulsion Rates, by Race/Ethnicity, Parent’s Education, and Family Income (Weighted %) School engagement scale Low Moderate High Very high Asian 3 24 30 43 Race/Ethnicity Hispanic Hispanic Foreign- U.S.Black Born Born White Parent’s Education Family Income Less More 100– 200– Than Than < 100% 200% 300% > 300% High High High of of of of School School School Poverty Poverty Poverty Poverty Total 76 8 8 5 10 5 4 7 7 86 31 26 29 19 31 20 22 32 27 15 21 26 29 35 35 29 33 32 30 31 27 37 32 32 33 34 28 44 31 38 43 33 38 41 40 36 49 Times skipped school in past 12 months 0 93 79 74 73 83 77 70 86 69 76 82 87 80 1 0 10 8 10 8 8 12 5 12 4 9 7 8 2 or more 7 12 18 18 9 15 18 8 19 20 9 6 12 Suspended/expelled in past 12 months 3 18 19 18 10 19 23 6 29 20 8 4 13 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. Consistent with previous research, our analysis of the association between children’s socioeconomic status and their school attachment (presented in Table 5.2a) reveals that children of less-educated and lower-income families experience greater problems in school. Children whose parents attained some postsecondary education were notably more engaged in school than those whose parents were only high school graduates. Children of high school graduates, in turn, were more engaged than children whose parents never completed high school. Similarly, children’s engagement in school generally rises with family income, although this pattern is not entirely consistent. We see that approximately 33 percent of children living in poor families are reported to have a very high level of engagement compared to about 40 percent of children in families with incomes at least 200 percent above the poverty line. At the same time, about 29 percent of children in the highestincome families are reported to have a low or moderate level engagement, compared to 36 percent of the families living in poverty. The same pattern is apparent when we look at voluntary school absences (see Table 5.2a). Skipping school one or more times is somewhat less prevalent among white children and much less prevalent among Asian children than it is among black and Hispanic children. On this same measure, we find that black children had only somewhat higher rates of skipping school two or more times (12 percent) and Hispanic children had sharply higher rates of doing so (18 percent), compared to white (9 percent) and Asian (7 percent) children. Skipping school occurs less frequently among children whose parents have attained education beyond high school than among children whose parents have a high school education or less. The strongest correlate of frequent school skipping is family income. Children in the two lowest-income groups have more than three times the rate of skipping school frequently as those in the highest-income group. Finally, we see similar patterns for school suspension or expulsion— our other indicator of problems in school—with Hispanic and black children and the most disadvantaged children showing more problems than other children. Almost twice as many black and Hispanic children than white children were suspended or expelled from school. The differential was even greater for Asian children, for whom suspension or 50 expulsion was a rare event. Children of the most educated parents are less than one-third as likely to have been suspended or expelled in the past year as children whose parents did not have education beyond high school. Family income is also strongly associated with children’s probability of being suspended or expelled from school. In the lowestincome group, 29 percent of children were suspended or expelled from school compared to just 4 percent of children in the highest-income group. After controlling for children’s other characteristics, family income is the strongest correlate of problem behavior in school for children in California, although it is only moderately associated with the measure of low engagement (see Table 5.2b). The differences we observed in children’s school engagement between white and nonwhite children are substantially reduced, after adjusting for other characteristics. It is interesting to note that Hispanic children whose parents were born outside the United States appear to be just as engaged in school as white children in the adjusted figures and are actually more engaged in school than Hispanic children living in nonimmigrant families.2 Differences by parents’ educational level also diminish, particularly among children whose parents have at least a high school diploma. Similarly, differences by family income also decline when parental education and race and ethnicity are held constant. The same pattern is evident for children’s voluntary school absence and suspension/expulsion. When we take account of children’s demographic and family characteristics, we observe smaller differences by race/ethnicity, although white children still appear less likely to skip school frequently. Similarly, in the adjusted percentages, differences by parental education are no longer as apparent. Somewhat unexpectedly, however, children whose parents have a high school education have a higher level of suspensions/expulsions than those whose parents have ____________ 2Previous research on immigrant children in California indicates that as time and generation in the United States increase, the time children spend on homework and their grade point average decrease (e.g., Rumbaut, 1995, p. 44). There is also some evidence to suggest that second- or third-generation Hispanic children become disaffected with the educational system and perform more poorly than foreign-born Hispanic children (Suárez-Orozco and Suárez-Orozco, 1995, p. 187). 51 Table 5.2b Children’s School Engagement, Frequency of Missing School, and Suspension/Expulsion Rates, by Race/Ethnicity, Parent’s Education, and Family Income (Adjusted %) School engagement scale Low Moderate High Very high Asian 6 28 35 31 Race/Ethnicity Hispanic Hispanic Foreign- U.S.Black Born Born White Parent’s Education Family Income Less More 100– 200– Than Than < 100% 200% 300% > 300% High High High of of of of School School Schoo Poverty Poverty Poverty Poverty Total 648 24 23 25 38 34 31 32 40 36 97 20 26 31 33 40 34 10 6 5 19 22 29 34 31 31 37 41 35 6 7 97 24 16 21 24 29 35 32 34 41 41 38 36 52 Times skipped school in past 12 months 0 81 83 77 77 1 13 8 10 2 or more 18 14 16 13 83 81 7 10 99 79 82 72 9 5 10 13 12 18 79 77 87 80 5 10 7 6 17 14 6 14 Suspended/expelled in past 12 months 8 13 10 13 12 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. 8 17 10 27 18 11 4 11 both more and less education. Family income continues to have a strong association with children’s school skipping and suspension or expulsion, with more than three times the number of children in the lowest-income categories skipping frequently and six times the number being suspended or expelled compared to those in the highest-income families. Conclusion Our results show that most children in California are highly engaged in school and are not experiencing problems in their classrooms, according to their parents’ reports. However, we find that children’s demographic characteristics are associated with low engagement in education and disaffection from school. Consistent with previous research, boys and older children are experiencing much higher rates of detachment. Family income remains very strongly related to children having problems in school, particularly in regard to voluntary and involuntary school absences. Although we discovered that black and Hispanic children have more problems in school than white and Asian children, this association appears to have much to do with their socioeconomic situations. It is important to stress that we are not indicating that low family income alone is the source of children’s school problems. Other conditions related to poverty may account for this association, such as the greater likelihood of children living in a poor school district, residing in a singleparent household, or having a parent with fewer psychosocial resources. 53 6. Pro-Social Involvement Both practitioners who provide services to children and researchers who study the effect of these services on children’s development have become increasingly interested in children’s involvement in pro-social groups such as clubs and associations, after-school activities, and recreational opportunities. There is growing evidence that children’s involvement in these kinds of activities outside formal schooling offers them many advantages (Carnegie Council on Adolescent Development, 1992; Eccles and Barber, 1999; Mahoney, 2000; Pierce, Hamm, and Vandell, 1999). These activities cultivate personal talents and interpersonal skills that play an important part in promoting good mental health and social functioning among children. Involvement in groups, lessons, and activities builds what sociologists refer to as cultural and social capital—knowledge about the way the world works, as well as social connections and sponsorship. For example, it provides children with the opportunity to interact with adults who may serve as mentors and counselors to children outside family and school settings. Finally, pro-social activities often shield children from risky behaviors and negative peer influences by providing them with extensive adult supervision and opportunities to engage in positive peer relationships. The measure of social involvement that we use is based on several questions in the NSAF that asked parents about whether their school-age children participated in clubs, associations, or sports teams, or took lessons. These different activities were summed up in an index of social involvement ranging from no involvement to participation in all three types of activities. We examine whether children participate in multiple activities not only because this variation may be beneficial to the children’s development, but also because children who participate in 55 more activities may be spending more supervised time with adults.1 However, we cannot determine how much time children spend in activities from the survey measure. Gender and Age Previous studies have shown that the nature of pro-social involvement sometimes differs for males and females. Girls are somewhat more likely to take lessons outside school, and boys are somewhat more inclined to participate in athletic activities. However, research gives us little reason to expect large gender differences in the level of activities in which children are engaged. Although the types of activities that children participate in change as they get older, the volume of activity does not decline from middle childhood to adolescence (Furstenberg et al., 1999). Our findings from the NSAF indicate that close to one-quarter of all children ages 6–17 in California are not involved in any activities such as clubs, teams, or lessons.2 The findings show little variation by gender or age in the number of activities in which children participate (Table 6.1). Although differences are not large between males and females, a higher percentage of girls than boys are reported not to participate in pro-social involvement (27 percent compared to 21 percent). This may be because more boys are involved in sports activities than girls (61 compared to 42 percent). Some girls may not participate in activities because parents exercise more control over their daughters than their sons in communities that are perceived to be dangerous. It is also possible that activities are more available to boys than to girls. The distribution in children’s level of activities is very similar by age group, indicating little decline in activities as children move from middle childhood into their adolescent years. In fact, children in the older age group are somewhat more likely than those in the younger group to engage in two to three activities. ____________ 1It is also possible that involvement in activities could have the negative effect of preventing children from spending time on their homework or from engaging in unstructured play. 2These questions were not asked for children ages five and under. 56 Table 6.1 Children’s Pro-Social Involvement, by Gender and Age (Weighted %) Number of prosocial activities 0 1 2 3 Gender Age Group Female Male 6–11 12–17 Total 27 21 23 24 32 36 36 31 28 30 28 31 14 12 13 14 24 34 29 13 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. Family Characteristics There is more reason to expect variations in children’s level of prosocial involvement by race/ethnicity and socioeconomic status. In particular, previous studies have shown that children in some locations have limited access to community-based programs (National Research Council and the Institute of Medicine, 2002; Wynn et al., 1988). In addition, lower-income families may be less likely to place their children in pro-social activities because they have less information, connections, and resources (e.g., time, money, and transportation) to avail themselves of clubs, recreational activities, and lessons even when they are potentially available. Conversely, more-educated and affluent parents often possess greater knowledge of where to find programs, have the means to get their children placed in programs, and can afford to pay for activities or lessons when they are not free of charge. They may also be more aware of the benefits of such programs for children and less fearful of letting children be supervised by other adults outside the home. Less information is available on the effect of race and ethnicity on social involvement. It is possible that language barriers might restrict access to such activities for the children of foreign-born parents. Hispanic and black children who live in disadvantaged communities may also be constrained by the availability of programs in their 57 neighborhoods. It would not be surprising, then, to discover racial/ethnic variations in the involvement of children in pro-social activities. We compare children’s social involvement by race/ethnicity, parental educational, and family income in Tables 6.2a and 6.2b. Looking first at basic distributions in Table 6.2a it is evident that large differences in the degree of pro-social involvement exist among children of different racial and ethnic groups, with children who have foreign-born Hispanic parents reported to have the lowest level of involvement. Fully 40 percent of Hispanic children in immigrant families, 29 percent of Hispanic children in nonimmigrant families, and 24 percent of black children are reported to participate in no activities compared to only 16 percent of white children and 12 percent of Asian children. At the other end of the index, about 50 percent of white children, about 45 percent of Asian children, and about 41 percent of black children participate in two or more types of activities, compared to 39 percent of Hispanic children in nonimmigrant families and 28 percent of Hispanic children in immigrant families. The lower participation of Hispanic children in immigrant families may be related to the educational levels of Hispanic parents who were born outside the United States (see Chapter 2). In fact, children of parents with less than a high school education have extremely low levels of participation in pro-social activities. Approximately 45 percent of these children are not involved in activities whereas about 22 percent participate in two or more activities. By contrast, only 14 percent of the children whose parents have more than a high school education are not engaged in activities and 52 percent are involved in two or more activities. The level of activity among children whose parents have only a high school diploma falls squarely between these two groups. Similarly, family income reveals a steep gradient of participation among children in lower- and higher-income families. For example, 43 percent of children in very poor families are reported to have no involvement whereas 25 percent participate in two or more activities; only 13 percent of children in the highest-income families have no involvement and 52 percent participate in two or more activities. These represent extraordinarily large differences in children’s participation rates 58 Table 6.2a Children’s Pro-Social Involvement, by Race/Ethnicity, Parent’s Education, and Family Income (Weighted %) 59 Race/Ethnicity Number of pro-social activities 0 1 2 3 Asian 12 42 32 13 Black 24 35 29 12 Hispanic Hispanic Foreign- U.S.- Born Born White 40 29 16 33 32 34 21 26 34 7 13 16 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. Parent’s Education Less More Than Than High High High School School School Family Income 100– 200– < 100% 200% 300% > 300% of of of of Poverty Poverty Poverty Poverty Total 45 27 14 43 28 21 13 24 33 34 34 32 35 34 34 34 14 30 34 16 26 31 36 29 8 9 18 9 11 14 16 13 Table 6.2b Children’s Pro-Social Involvement, by Race/Ethnicity, Parent’s Education, and Family Income (Adjusted %) 60 Race/Ethnicity Asian Black Number of pro-social activities 0 20 25 1 42 34 2 22 32 3 16 10 Hispanic Hispanic Foreign- U.S.- Born Born White 28 21 30 32 30 32 11 14 21 34 33 12 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. Parent’s Education Less More Than Than High High High School School School Family Income 100– 200– < 100% 200% 300% > 300% of of of of Poverty Poverty Poverty Poverty Total 37 26 16 33 25 20 19 20 33 35 32 31 36 36 32 37 21 31 35 24 29 30 36 30 8 8 16 12 11 13 13 13 and presumably in their opportunities for receiving the benefits of social involvement. The results presented in Table 6.2b correct for the mutual association of demographic and family characteristics. Again, the adjusted percentages show us the unique contribution of race/ethnicity, parental education, and family income on the level of children’s prosocial activities. As in our analysis of other indicators of children’s wellbeing, we observe much less variation in social involvement across racial/ethnic groups after children’s other demographic and socioeconomic characteristics are taken into account. On the other hand, both parental education and family income remain strongly related to children’s level of activities. The adjusted percentages show that the children of the least-educated parents are more than twice as likely not to engage in any pro-social opportunities and only half as likely to be in the highest category of social involvement as children whose parents have education beyond high school. Even taking account of parental education and other differences, family income is also strongly linked to children’s social involvement. About one-third of poor children have no involvement in activities compared to about one-fifth of children in the highest-income families. Similarly, slightly over one-third of poor children engage in two or more activities compared to almost half of children from the highest-income families. Thus, we find that socioeconomic status is strongly associated with the likelihood of children’s engagement in pro-social activities. As suggested above, several factors may be at work in producing these large variations in children’s social involvement, such as the socioeconomic status of their families. Access may be limited by the availability of services and the cost of services. We know from other research that the distribution of both extracurricular programs in schools and recreational programs in neighborhoods are much more available in affluent than in disadvantaged neighborhoods. When services are not freely available, many poor parents cannot afford to pay for them. And even when they are available, less-educated and lower-income parents may not know about them, may have problems arranging to transport their children to the programs, may have fears about letting their children go outside the household or immediate neighborhood, or may 61 need their children to provide in-home services. Any and all of these factors could help to account for the large differences we observe. Conclusion Close to one-quarter of all children ages 6–17 in California are not involved in any activities such as clubs, teams, or lessons. Consistent with previous research, we do not find that the level of children’s involvement varies much by the their gender or age; however, girls are somewhat more likely than boys to not be involved in any activities. We find much larger variations in involvement by children’s socioeconomic status. Both income and education are strongly related to the likelihood of children participating in pro-social activities. Differences in participation by children’s race and ethnicity are greatly reduced after these and other factors are controlled for. Children from poor families and those with less-educated parents are strikingly less likely to engage in any activities whereas those from more affluent families and living with a well-educated parent are highly likely to engage in at least some activities. About half of the most advantaged children participate in two or more types of activities, compared to about one-third of the least advantaged children. We suspect that the sources of these differences stem from both the limited resources of parents and the lower availability of programs for children in poorer communities. 62 7. How California’s Children Compare to Those in the Rest of the Nation In previous chapters, we examined variations in health, emotional adjustment, school attachment, and pro-social involvement among various subgroups of children within the state of California. This chapter investigates how children in California are faring relative to those living in the rest of the nation. Reports by the Urban Institute have examined this topic but have not taken into account how the composition of the population in California compares to that in other parts of the United States, therefore making it difficult to assess whether any variations in children’s well-being they observe are due to demographic differences or to other conditions distinctive to California such as public policies or the availability of services. More than one in eight children in the nation live in California. Because such a large proportion of all U.S. children live in the state, we might expect to find only small differences between the welfare of the children in California and that of children in the rest of the nation. On the other hand, the extremely diverse nature of the population and the precarious state of many recent immigrant families lead us to suspect that some differences related to children’s demographic, social, and economic characteristics might exist. In particular, to the extent that poverty, low parental education, and Hispanic ethnicity and nativity are associated with children’s outcomes, we expect that children in California are not doing as well as children in the rest of the nation. Our analysis in this chapter shows some support for each of these expectations. In some cases, California’s children resemble and at other times differ from those in the rest of the country, depending on the outcome we examine. 63 Physical Health Our findings indicate that the overall health status of children in California is somewhat inferior to that of children in the rest of the nation. About 22 percent of California parents report that their children’s health is less than very good compared to 17 percent of parents living in the rest of the nation (see Table 7.1). Most of this difference results because a higher proportion of parents in California than in the rest of the nation say that their children’s health is “good” instead of “very good.” The incidence of extreme health problems among children is not greater in this state than elsewhere. Consistent with parents’ overall evaluations, the proportion of those reporting that their children have a health problem that limits their activity is similar in California and the rest of the nation (8 percent compared to 9 percent). Table 7.1 Health Status of Children in California and the Rest of the United States (Weighted %) Child current health status Poor Fair Good Very good Excellent Child has condition that limits activity Rest of the California United States 11 54 16 12 25 27 53 56 89 Well-child doctor visit during past 12 months No 36 34 One 37 41 Two or more 27 25 Saw dentist in past 12 months 77 79 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. 64 It is important to note that about 36 percent of children in California and 34 percent of children in the rest of the nation had not seen a doctor or other medical health professional for a well-child visit in the previous 12 months. Similarly, approximately 23 percent of California children had also not seen a dentist in the past 12 months compared to 21 percent in the rest of the nation. These findings show that a sizable minority of children in California do not receive routine, preventive health and dental care as is also true of children elsewhere in the United States. Given the general similarities in the health status of children in California and the rest of the nation, it is perhaps not surprising that California parents indicate that their children are seeing physicians only slightly more frequently for routine health care than parents elsewhere. Approximately 27 percent of children in the state had seen a doctor or other medical health professional two or more times in the past 12 months compared to one-quarter in the rest of the nation. We next examine whether these modest differences in children’s health status are maintained after accounting for differences in the characteristics of children and their families. When we make adjustments to control for demographic differences between California and the rest of the nation, no sizable disparities remain. Thus, the statelevel difference in health is apparently attributable to the different demographic composition of California’s families compared to that of the rest of the nation (see Figure 7.1.) Emotional Adjustment When we turn to indicators of the mental health and emotional wellbeing of children, we again compare the needs of children in California to those of children in the rest of the country. As shown in Table 7.2, children’s scores on the behavioral problem scale are almost identical in California and in the rest of the country.1 Slightly more than one-tenth of children have high levels of emotional or behavioral problems whereas about two-thirds have low levels of these problems, according to their parents’ reports. At the same time, we see that the proportion of children in California who received some form of mental health counseling in the ____________ 1See Appendix Table A.1 for the distribution of individual scale items. 65 Percentage 60 California 50 Rest of the U.S. 40 56 55 30 28 25 20 10 11 0 Poor 44 Fair 14 13 Good Health status Very good Excellent Figure 7.1—Children’s Health Status in California and the Rest of the United States (Adjusted %) Table 7.2 Children’s Behavioral Problems in California and the Rest of the United States (Weighted %) Behavioral problems scale Low Moderate Severe Rest of the California United States 66 67 23 22 11 11 Received mental health services in past 12 months 5 7 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. 66 past year is a little lower than elsewhere—5 percent in this state compared to 7 percent in the rest of the nation. In California as elsewhere, most children who are displaying serious behavioral adjustment problems are not being treated. However, as shown in Figure 7.2a, the situation is worse for children in California where only about 21 percent of the children with severe symptoms have received mental health services compared to 30 percent of children living in other states. Figure 7.2a also shows that children with moderate symptoms in California are about two-thirds as likely to be served as the children in the rest of the nation. In particular, 8 percent of children in California have received mental health services compared to 12 percent of children elsewhere. These differences between treatment are disturbing, particularly for children with severe problems, because it means that children with the same level of behavioral problems are less likely to be receiving treatment in California. 35 California 30 Rest of the U.S. 30 25 21 20 Percentage 15 12 10 8 5 23 0 Low Moderate Behavioral problem Severe Figure 7.2a—Children Who Received Mental Health Services in Past 12 Months, by Severity of Behavioral Problems, California and the Rest of the United States (Weighted %) 67 After taking into account the demographic and family characteristics of children within and outside the state, the gap between children’s need for mental health services and use of those services remains high in California, with less than one-quarter of children with severe problems receiving services (see Figure 7.2b). The disparity between California and other states closes only slightly when we look at use of mental health services among children with severe problems, which drops from a 9 to a 7 percent difference. It would seem that fewer children in California are getting needed mental health services than in other states, even taking into account racial/ethnic, educational, and family income differences. This indicates the possibility that California is performing less well than many other states in providing mental health services to children at risk of later problem behaviors. 35 California 30 Rest of the U.S. 31 25 24 Percentage 20 15 10 54 2 0 Low 12 6 Moderate Behavioral problem Severe Figure 7.2b—Children Who Received Mental Health Services in Past 12 Months, by Severity of Behavioral Problems, California and the Rest of the United States (Adjusted %) 68 Educational Attachment The NSAF included a variety of measures designed to assess the level of attachment to school among children who were attending primary school or above, that is, children ages six and older. As discussed in Chapter 5, four items tapping the child’s interest in and commitment to school are included in the survey that was combined into an index measuring the children’s educational attachment (see Table 7.3).2 We find that children’s engagement in school in California is similar to that in the rest of the nation. Over two-thirds of parents in California and the rest of the nation indicate that their children have a high or very high level of engagement in school. However, a somewhat higher proportion of children are not as highly involved in school in California as children in other states. About 31 percent of children in California are Table 7.3 Children’s School Engagement in California and the Rest of the United States (Weighted %) School engagement scale Low Medium High Very high California 7 24 32 38 Rest of the United States 8 20 32 39 Times skipped school in past 12 months (ages 12–17) 0 1 2 or more 80 8 12 85 6 9 Expelled/suspended from school 13 14 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. ____________ 2See Appendix Table A.2 for the distribution of individual scale items. 69 reported to have low or moderate school engagement in comparison to 28 percent of children elsewhere. As can be seen in Table 7.3, California’s children are about as likely as those elsewhere to exhibit problems that require suspension or expulsion (13 and 14 percent, respectively). However, when we look at the number of times the child skipped school in the previous year, we observe larger differences between California and the rest of the nation. Twenty percent of California children skipped school at least once compared to 15 percent of children elsewhere. The percentage of children who skip school frequently (two or more times) is also more common in California, 12 percent compared to 9 percent. The modest differences in skipping school disappear, however, when we adjust for demographic variations between California and the rest of the nation (not shown). The adjusted percentages are almost identical on all the indicators of educational attachment when demographic differences are taken into account. Thus, it appears that California’s children are faring about as well (or as poorly) in regard to their school behaviors as children in other parts of the country. Pro-Social Involvement The NSAF also measures children’s involvement in activities outside school. To construct the index for social involvement, we took the sum of the three related items: participating in clubs, taking lessons after school, and playing on sport teams (Table 7.4). When we look at the individual measures of social involvement—we find sizable differences in the proportion of children participating in clubs, with children in California much less likely to be participants than children elsewhere (see Appendix Table A.3). More modest but still substantially lower numbers of children were on sports teams in California than elsewhere. On the other hand, California’s children were somewhat more likely to be taking lessons to develop a talent in such areas as music, dance, language, or computers. When combined into a single index, the results suggest a modest deficit in activities aimed at developing children’s social skills and personal development. Slightly less than one-quarter of California’s children are involved in none of the three activities compared to one-fifth 70 Table 7.4 Children’s Pro-Social Involvement in California and the Rest of the United States (Weighted %) No. of prosocial activities 0 1 2 3 Rest of the California United States 24 20 34 31 29 33 13 16 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. of children in the rest of the nation. Whereas close to half of the children in the rest of the nation were in two or more activities, just over two-fifths of California’s children reached that level of involvement. After we take children’s characteristics into account in Figure 7.3, we find these differences largely disappear. California’s children are slightly less likely to be involved in three or more activities but equally likely to be involved in none. Conclusion Our comparison of children in California with children in the rest of the nation provides the general impression that children in the state are slightly worse off on a number of indicators of well-being. The health ratings of children in California are a little lower than those of children elsewhere. Children in California are also less likely to have seen a physician or other medical professional for a well-child visit or to have received dental or mental health care. The disparity between mental health needs and mental health care in California is particularly troubling for children with severe mental health problems. Equally disturbing is the fact that at least one-third of children in California and in the rest of the country had not received any routine health care and at least onefifth (23 and 21 percent) had not received dental care in the past year. 71 100 14 90 80 70 33 60 3 17 2 1 0 33 Percentage 50 40 33 30 31 20 10 20 0 California 19 Rest of the U.S. Figure 7.3—Children Who Participate in Pro-Social Activities, by Number, California and the Rest of the United States (Adjusted %) Children in California also have a slightly lower level of engagement in school, skip school more often, and are less involved in pro-social activities that cultivate children’s social and personal skills. Although we did not observe any differences in other realms of behavior, none of the indicators showed that California’s children were outperforming children in other states. The question of whether the differences are due to demographic variations between the population of California and the rest of the nation does not yield an entirely straightforward answer. For most indicators, the discrepancies disappeared when we took into account the demographic and socioeconomic differences of children living in and outside the state. This is true for the difference in reported health, school engagement, and pro-social activities. It appears that the higher incidence of poor children, many of whom are recent immigrants to the country, contribute to the state-level deficits in children’s well-being. However, the availability of mental health services for children in need of treatment is an exception to this pattern. Children in California are 72 receiving fewer services and the gap in provision of services to children with severe problems is conspicuous even taking account of the demographic characteristics of the state. 73 8. Conclusion and Policy Implications This report provides a baseline of information about the well-being of California’s children in 1999 against which similar indicators can be reexamined in subsequent years. In particular, it has shown us how California’s children are faring on measures of their physical health, behavioral problems, school attachment, and social involvement in comparison to children in other states. The report has devoted even more attention to documenting how the well-being of children within the state varies by their demographic and socioeconomic characteristics. This examination of differences among children within the state illustrates where the greatest needs for services exist and gives policymakers ways to target the populations that are in need of this assistance. From these comparative analyses, we observe the following outcomes. Children’s Physical Health About 78 percent of children in California are reported to be in very good or excellent health and only 8 percent have a chronic health problem. At the same time, over one-third of children had not seen a physician for routine preventive care in the last year and over one-fifth had not seen a dentist. Although children’s overall health status is generally positive, we find strong differences in children’s health among demographic and socioeconomic subgroups in the state. Specifically, children’s health status declines with age, however, their visits to physicians decline during adolescence as their needs increase, particularly among males. Our results also show that children whose parents are Hispanic immigrants and who are less-educated are in worse health than other children but are receiving less routine medical care. Although poor 75 children are also reported to be in worse health, they seem to have more access to medical, but not dental, care than these other children. More than 30 percent of children whose parents are Hispanic immigrants, who have the lowest levels of education, or who live in poor and near poor families had not seen a dentist in the past year. Children’s Emotional Adjustment More than one out of ten children in California are experiencing serious behavioral problems that could put them at risk for social and mental health problems later in life, but only 5 percent have received mental health services. The gap between children’s needs and treatment is higher in California than in other parts of the United States. This gap and the disparity between California and other states remain after the composition of children within and outside the state is taken into account. Within subgroups of children, we find that the level of children’s need for mental health services does not correspond with their receipt of these services. Namely, males and children from lower socioeconomic status families have more problems than other children but do not receive more treatment. These disparities are even more troubling for children with severe mental health problems and indicate that poor children are particularly underserved. Children’s Educational Attachment About one-third of children in California are not highly engaged in school, about 13 percent were expelled or suspended in the last year, and about one-fifth skipped school during this time. Within subgroups of children, we find that males and older children are less likely to be engaged in school and are more likely than other students to have voluntary or involuntary absences. Our results also show that family income is strongly related to children’s attachment to school, particularly in the area of school absence, suspension, and expulsion. Although we observe variations in children’s school engagement by their race and ethnicity, these differences are markedly reduced when other characteristics are held constant. 76 Children’s Involvement in Pro-Social Activities Close to one-quarter of children were not involved in any pro-social activities in the last year. We do not find that children’s level of involvement varies importantly by their gender or age, but we do find that involvement varies considerably by children’s socioeconomic status. Children whose parents do not have a high school diploma and children living in poor families are less likely to participate in social activities beneficial to their development. Differences in participation by children’s race and ethnicity diminish considerably after these and other factors are controlled for. Access to these activities might be limited both by the cost of these programs and by their lack of availability to low-income children. Children in California and the Rest of the Nation Children in California appear to be faring slightly worse on a number of indicators of their well-being and are not outperforming children in the rest of the nation on any of the indicators we examine. Compared to children living elsewhere, the overall health status of children in California is somewhat lower. Children with behavioral problems are less likely to receive mental health care. Furthermore, they skip school more often, and they are less involved in activities that promote their social development. After controlling for the compositional differences between children within and outside the state, we find that the gap between children’s need for mental health services and their use of these services remains higher in California than in other states. However, disparities in children’s health status and school absences largely disappear. Some disparities remain in children’s participation in pro-social activities indicating a lower level of in-school and after-school activities available in the state. Many of California’s children are faring well; however, our results suggest that a substantial minority of children may not be adequately served by existing services in the state. For the most part, these children come from the poorest families in California, have parents with low levels of educational attainment, or have parents who are Hispanic immigrants. 77 These families may lack the resources, knowledge, and social connections to obtain services for their children. Our findings about the well-being of children in California have broad implications for those seeking to provide services to children within the state and for monitoring whether these services are effective in serving children’s needs. The first set of policy implications stems from our observations about the nature of children’s needs. Although children whose parents are Hispanic immigrants and who are less-educated are in worse health than other children, they are receiving less routine medical care. Children from poor families are also in worse health but appear to have access to health care. It is critical to reach out to low-income immigrant communities where children are in need of health services as well as to native-born children from low- to moderate-income families who may not be receiving adequate health care. It is at least worth investigating whether such visits might identify and treat incipient problems that might later prove to be costly to individuals, families, and society. The importance of serving healthy adolescents in higher-income families who are not receiving routine health checkups may be less apparent. Yet, physicians can do an effective job of screening for risky behavior such as unprotected sex, alcohol and drug use, or depression. We discovered that adolescent males were particularly likely not to have received routine health care in the previous year. Although children from poorer families often use medical services, they are receiving much less help in obtaining preventive dental care. The lack of dental care is especially evident for Hispanic children whose parents were born outside the United States, whose parents have low educational levels, and whose families have incomes at the poverty or near-poverty level. There is an evident need to extend preventive services to this population, which may provide cost savings in the long run if serious dental problems are avoided. There is also a conspicuous lack of mental health assistance for children with multiple behavioral problems. All income groups experience some gap between the need for services among children who exhibit high levels of problem behavior; however, children in the poorest families are particularly underserved. Children with emotional and behavioral problems often experience problems in the classroom and the 78 community. Many of them will experience more severe problems later in life if they remain untreated. The tremendous costs of treating antisocial behavior and substance abuse among adults suggest that early intervention is justified even if it helps to reduce only modestly the incidence of problem behaviors. Policymakers might want to consider the possibility of directing public information to underserved communities to increase parents’ awareness of mental health services. Service providers may be more effective in reaching underserved populations by locating service sites closer to the populations at risk. It is not news that school behavior problems are more common among males, especially as they reach adolescence. Yet programs that keep older youth attached to school are in short supply. California could strengthen the in-school and after-school activities that help young males to stay connected to school. Tracking absenteeism and behavioral problems can identify youth at risk of dropping out. We need to craft more experiments that can help middle school students prepare to make a successful transition to high school by providing additional tutoring, mentoring, or social activities. Finally, our results indicate that an extraordinarily high proportion of youth are not engaged in any extracurricular activities in their preadolescent and adolescent years. The absence of programs for youth in this formative period likely accounts for part of the wide income disparity in the proportion of youth who are not involved in clubs, associations, lessons, sports, and other such activities. If programs do not adequately serve poorer children, this puts them at a disadvantage for gaining the benefits offered by extracurricular activities including greater school attachment, skill development, and social ties. Particularly in the wake of welfare reform, there is a manifest need to engage low-income adolescents in after-school activities because their parents may have entered the labor force, leaving them unattended in the interim between the close of the school day and the end of the workday. California is a state with high levels of income inequality (Reed, 1999). Among the poor and near-poor, families cannot afford to provide some of the assistance that is routinely offered to children in more affluent areas. Public investment in our young people through the preschool years may be alleviating some of the burden placed on low- 79 income families with young children, but we cannot expect that that assistance in the preschool years will inoculate children from problems later on. Families with older children require the same sort of public aid to help their children maintain good health, develop skills and competencies, and retain high motivation and a belief that hard work will pay off. At present, California may not be providing this sort of assistance at a level that reaches low-income families and that effectively services communities of nonnative residents. The second important set of policy implications that stems from this analysis relates to how we can improve information on children’s wellbeing. First, we should develop indicators appropriate for very young children and, second, use existing indicators to follow cohorts of children over the next decade as they reach school age to see the effects of early health and development programs and services. It is reasonable to expect that some of these programs may have had positive effects on the younger cohort of children now in their preschool years. Unfortunately, we do not have sufficient data to measure the well-being of younger children at the state level and are not tracking children’s well-being over time. The regular measurement of children’s well-being through surveys such as the NSAF, together with rigorous program evaluations, can help monitor whether the services provided are reaching the target populations and whether they are effective in raising levels of well-being. By beginning to monitor the success of our children, we can establish a benchmark for assessing whether California is providing sufficient public investment in children and can gain a glimpse into the future health, happiness, and productivity of the state’s population. More attention to marshalling data relevant to the well-being of children involves periodic assessments of the sorts of indicators described in this report. Fortunately, a third wave of NSAF was conducted in 2002–03 and will provide a way to update the set of indicators that we have described. However, we believe that other avenues of data analysis should also be pursued such as the use of vital statistics information. In other localities, administrative records such as these have been linked across agencies to provide a longitudinal record of children and families’ experiences from birth to early adulthood. We foresee the possibility of agencies collaborating at both the state and local levels to produce 80 longitudinal information that can help policymakers better identify the need for services, the success of targeted interventions, and the changing well-being of successive cohorts of children throughout their early lives. 81 Appendix A Detailed Comparisons of Scale Items for California and the Rest of the Nation 83 Table A.1 Emotional Adjustment of Children in California and the Rest of the United States (Weighted %) Doesn’t get along with other kidsa Never true Sometimes true Often true Can't concentrate for longa Never true Sometimes true Often true Has been sad or depresseda Never true Sometimes true Often true Feels worthless or inferiorb Never true Sometimes true Often true Has been nervous or tenseb Never true Sometimes true Often true Acts too young for his ageb Never true Sometimes true Often true Has trouble sleepingc Never true Sometimes true Often true Lies or cheatsc Never true Sometimes true Often true Does poorly at school workc Never true Sometimes true Often true California 71 25 4 59 34 7 64 34 2 85 14 1 74 25 1 79 17 5 87 11 1 75 22 3 67 28 5 Rest of the United States 70 28 3 62 31 8 62 35 3 86 13 1 71 25 4 80 17 3 85 13 2 76 21 2 67 28 6 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. aAsked of respondents about children ages 6–17. bAsked of respondents about children ages 6–11. cAsked of respondents about children ages 12–17. 84 Table A.2 School Commitment of Children in California and the Rest of the United States (Weighted %) Cares to do well in school None of the time Some of the time Most of the time All of the time Always does homework None of the time Some of the time Most of the time All of the time Does schoolwork just to get by None of the time Some of the time Most of the time All of the time Only does schoolwork when forced None of the time Some of the time Most of the time All of the time California 1 16 28 54 2 14 17 67 14 16 24 45 8 12 28 52 Rest of the United States 2 18 30 51 3 15 20 62 10 13 25 52 8 13 31 48 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. Table A.3 Pro-Social Involvement of Children in California and the Rest of the United States (Weighted %) Participated in club in past 12 months Played on sports team in past 12 months Took lessons after school in past 12 months California 46 52 34 Rest of the United States 58 54 33 SOURCE: 1999 National Survey of America’s Families. NOTE: Percentages may not sum to 100 because of rounding. 85 Bibliography Achenbach, T. M., and C. S. Edelbrock, “The Child Behavior Profile: II, Boys Aged 12–16 and Girls Aged 6–11 and 12–16,” Journal of Clinical and Consulting Psychology, Vol. 47, 1979, pp. 223–233. Achenbach, T. M., S. H. McConaughy, and C. T. Howell, “Child/ Adolescent Behavioral and Emotional Problems: Implications of Cross-Informant for Situational Stability,” Psychological Bulletin, Vol. 101, 1987, pp. 213–232. Ben-Arieh, Asher, Natalie Hevener Kaufman, Arlene Bowers Andrews, Robert M. Goerge, Bong Joo Lee, and J. Lawrence Aber, Measuring and Monitoring Children’s Well-Being, Kluwer Academic Publishers, Boston, Massachusetts, 2001. Black, Tamara, and Adam Safir, “Assessing Nonresponse Bias in the National Survey of America’s Families,” in F. Scheuren et al., Report No. 7: 1999 Collection of Papers, Urban Institute, Washington, D.C., 2001, available at http://www.urban.org/content/Research/ NewFederalism/NSAF/Methodology/1999MethodologySeries/1999. htm. Brick, J. M., P. Broene, D. Cantor, D. Ferraro, T. Hankins, C. Rauch, and T. Strickler, No. 8: 1999 NSAF Response Rates and Methods Evaluation, Urban Institute, Washington, D.C., 2000, available at http://www.urban.org/content/Research/NewFederalism/NSAF/ Methodology/1999MethodologySeries/1999.htm. Brint, Steven, Schools and Societies, Pine Forge Press, Thousand Oaks, California, 1998. Brooks-Gunn, Jeanne, and Greg J. Duncan, “The Effects of Poverty on Children,” The Future of Children, Vol. 7, No. 2, Summer/Fall 1997. Brown, E. Richard, Ninez Ponze, Thomas Rice, and Shana Alex Lavarreda, The State of Health Insurance in California: Findings from 87 the 2001 California Health Interview Survey, UCLA Center for Health Policy Research, Los Angeles, California, June 2002. Bustillo, Miguel, “Bills Aim to Make Students Healthier,” Los Angeles Times, March 17, 2002. Cairns, Robert B., Glen H. Elder, Jr., and E. Jane Costello (eds.), Developmental Science, Cambridge University Press, Cambridge, England, New York, 1996. California Children and Families Commission, California Children and Families Commission Annual Report—Fiscal Year 2000/01, 2002, available at http://www.ccfc.ca.gov/fiscal.htm. California Department of Finance, California State Budget Highlights, 2002, available at www.dof.ca.gov. Carnegie Corporation of New York, From Birth to Fifteen: A Comprehensive Approach to Children’s Learning and Development, Report of the Carnegie Corporation of New York 1989–1996, Carnegie Corporation, New York, 1996. Carnegie Council on Adolescent Development, A Matter of Time: Risk and Opportunity in the Non-School Hours, Carnegie Council, New York, 1992. Children Now, The State of Our Children: Children’s Critical Early Years, 2002a, available at http://www.childrennow.org/. Children Now, California Report Card: Children’s Critical Early Years, 2002b, available at http://www.childrennow.org/. Compas, Bruce E., and Constance L. Hammen, “Child and Adolescent Depression: Covariation and Comorbidity in Development,” in Robert J. Haggerty, Lonnie R. Sherrod, Norman Garmezy, and Michael Rutter (eds.), Stress, Risk, and Resilience in Children and Adolescents: Processes, Mechanisms, and Interventions, Cambridge University Press, New York, 1994, pp. 225–267. Converse, N., A. Safir, and F. Scheuren, No. 11: 1999 Public Use File Data Documentation, Urban Institute, Washington, D.C., 2001, 88 available at http://www.urban.org/content/Research/NewFederalism/ NSAF/Methodology/1999MethodologySeries/1999.htm. Danziger, Sheldon, and Peter Gottschalk (eds.), Uneven Tides: Rising Inequality in America, Russell Sage Foundation, New York, 1993. Danziger, Sheldon, and Peter Gottschalk, America Unequal, Russell Sage Foundation, Harvard University Press, New York, Cambridge, Massachusetts, 1995. Danziger, Sheldon, and Jane Waldfogel (eds.), Securing the Future: Investing in Children from Birth to College, Russell Sage Foundation, New York, 2000. Duncan, Greg J., Jeanne Brooks-Gunn (eds.), Consequences of Growing Up Poor, Russell Sage Foundation, New York, 1997. Eccles, Jacquelynne S., and B. L. Barber, “Student Council, Volunteering, Basketball or Marching Band: What Kind of Extracurricular Involvement Matters?” Journal of Adolescent Research, Vol. 14, No. 1, 1999, pp. 10–43. Eckstrom, R. B., M. S. Goertz, J. M. Pollack, and D. A. Rock, “Who Drops Out of High School and Why? Findings from a National Study,” in Gary Natriello (ed.), School Dropouts: Patterns and Policies, Teachers College Press, New York, 1987, pp. 52–69. Ehrle, Jennifer, and Kristin Anderson Moore, 1997 NSAF Benchmarking Measures of Child and Family Well-Being: Report No. 6, Urban Institute, Washington, D.C., 1999, available at http://www.urban. org/UploadedPDF/Methodology_6.pdf. Elo, Irma T., and Samuel H. Preston, “Educational Differentials in Mortality in the United States: 1979–1985,” Social Science and Medicine, Vol. 42, No. 1, 1996, pp. 47–57. Entwisle, Doris R. , K. L. Alexander, and L. S. Olson, Children, Schools, and Inequality, Westview Press, Boulder, Colorado, 1997. Furstenberg, Frank F., Thomas D. Cook, Jacquelynne Eccles, Glen H. Elder, Jr., and Arnold Sameroff, Managing to Make It: Urban Families 89 and Adolescent Success, University of Chicago Press, Chicago, Illinois, 1999. Haveman, Robert, and Barbara Wolfe (eds.), Succeeding Generations: On the Effects of Investments in Children, Russell Sage Foundation, New York, 1994. Hayward, Mark D., Eileen M. Crimmins, Toni P. Miles, and Yu Yang, “The Significance of Socioeconomic Status in Explaining the Racial Gap in Chronic Health Conditions,” American Sociological Review, Vol. 65, No. 6, 2000, pp. 910–930. Johnson, Hans, Laura Hill, and Mary Heim, “New Trends in Newborns: Fertility Rates and Patterns in California,” California Counts, Vol. 3, No. 1, Public Policy Institute of California, San Francisco, California, 2001. Johnson, Hans P., “Teen Birth Rates: Trends and Patterns in California,” California Counts, Vol. 4, No. 3, Public Policy Institute of California, San Francisco, California, 2003. Judkins, D., J. Michael Brick, Pam Broene, David Ferraro, and Teresa Strickler, Report No. 2, 1999 NSAF Sample Design, The Urban Institute and Westat, 2001, available at http://test.urban.org/ UploadedPDF/1999_Methodology_2.pdf. Knudsen, C., and R. McNown, “Changing Causes of Death and Sex Differences in the USA,” Population Research and Policy Review, Vol. 12, 1993, pp. 27–41. Lamborn, S. D., B. B. Brown, N. S. Mounts, and L. Steinberg, “Putting School in Perspective: The Influence of Family, Peers, Extracurricular Participation, and Part-Time Work on Academic Engagement,” in Student Engagement and Achievement in American Secondary Schools, Teachers College Press, New York, 1992, pp. 153–181. Maccoby, Eleanor, The Two Sexes: Growing Up Apart, Coming Together, Harvard University Press, Cambridge, Massachusetts, 1998. Mahoney, J. L., “School Extracurricular Activity Participation as a Moderator in the Development of Antisocial Behavior,” Child Development, Vol. 71, No. 2, 2000, pp. 502–516. 90 Mason, Mary Ann, Arlene Skolnick, and Stephen D. Sugarman (eds.), All Our Families: New Policies for a New Century: A Report of the Berkeley Family Forum, Oxford University Press, New York, 1998. Millstein, Susan G., Anne C. Petersen, and Elena O. Nightingale (eds.), Promoting the Health of Adolescents: New Directions for the TwentyFirst Century, Oxford University Press, New York, 1993. Moore, Kristin Anderson, Juliet L. Hatcher, Sharon Vandivere, and Brett V. Brown, “Children’s Behavior and Well-Being,” 1999 Snapshots of America’s Families II, Urban Institute, Washington, D.C., 2000, available at http://www.urban.org/content/Research/NewFederalism/ NSAF/Snapshots/1999Results/ChildrensBehaviorandWellBeing/ Behavior.htm. National Center for Health Statistics, “Summary Health Statistics for U.S. Children: National Health Interview Survey, 1997,” 2002, available at http://www.cdc.gov/nchs/products/pubs/pubd/series/ sr10/pre-200/sr10_203.htm. National Research Council, Losing Generations: Adolescents in High-Risk Settings, National Academy Press, Washington, D.C., 1993. National Research Council and the Institute of Medicine, Community Programs to Promote Youth Development, National Academy Press, Washington, D.C., 2002. Natriello, Gary (ed.), School Dropouts: Patterns and Policies, Teachers College Press, New York, 1987. Neisser, Ulric, The School Achievement of Minority Children: New Perspectives, Lawrence Erlbaum Associates, Hillsdale, New Jersey, 1986. Peterson, James L., and Nicholas Zill, “Marital Disruption, Parent-Child Relationships, and Behavior Problems in Children,” Journal of Marriage and the Family, Vol. 48, No. 2, 1986, pp. 295–307. Pierce, K. M., J. V. Hamm, and D. L. Vandell, “Experiences in AfterSchool Programs and Children’s Adjustment in First-Grade Classrooms, Child Development, Vol. 65, 1999, pp. 440–456. 91 Reed, Deborah, “California’s Rising Income Inequality: Causes and Concerns,” Occasional Papers, Public Policy Institute of California, San Francisco, California, 1999. Reed, Deborah, and Amanda Bailey, “California’s Young Children: Demographic, Social, and Economic Conditions,” California Counts, Vol. 4, No. 2, Public Policy Institute of California, San Francisco, California, 2002. Reed, Deborah, and Sonya M. Tafoya, “Demographic, Social, and Economic Trends for Young Children in California,” Occasional Papers, Public Policy Institute of California, San Francisco, California, 2001. Reyes, Belinda I., “An Assessment of Social and Economic Well-Being,” in Belinda Reyes (ed.), A Portrait of Race and Ethnicity in California, Public Policy Institute of California, San Francisco, California, 2001. Rumbaut, Ruben G., “The New Californians: Comparative Research Findings on the Educational Progress of Immigrant Children,” in Ruben G. Rumbaut and Wayne A. Cornelius (eds.), California’s Immigrant Children: Theory, Research, and Implications for Educational Policy, Center for U.S.-Mexican Studies, University of California, San Diego, La Jolla, California, 1995. Safir, Adam, Fritz Scheuren, and Kevin Wang, “Survey Methods and Data Reliability, 1997 and 1999,” Urban Institute, Washington, D.C., 2001, available at http://www.urban.org/content/Research/ NewFederalism/NSAF/survey-methods.htm. StataCorp, Strata Statistical Software: Release 5.0, Stata Corporation, College Station, Texas, 1997. Suárez-Orozco, Marcelo M., and Carola E. Suárez-Orozco, “The Cultural Patterning of Achievement Motivation: A Comparison of Mexican, Mexican Immigrant, Mexican American, and Non-Latino White American Students,” in Ruben G. Rumbaut and Wayne A. Cornelius (eds.), California’s Immigrant Children: Theory, Research, and Implications for Educational Policy , Center for U.S.-Mexican Studies, University of California, San Diego, La Jolla, California, 1995. 92 Takanishi, Ruby, and David A. Hamburg (eds.), Preparing Adolescents for the Twenty-First Century: Challenges Facing Europe and the United States, Cambridge University Press, New York, 1997. UCLA Academic Technology Services, SPSS Frequently Asked Questions: What Does Cronbach’s Alpha Mean? 2001, available at www.ats.ucla. edu/stat/spss/faq/alpha.html. U.S. Census Bureau, Poverty Status in 1999 of Individuals: 2000, 2002, available at http://factfinder.census.gov/. U.S. Department of Education, National Center for Education Statistics, Dropout Rates in the United States, 2000, 2002, available at http:// nces.ed.gov/pubs2002/droppub_2001/. U.S. Department of Health and Human Services, “HHS Approves Plan to Expand Coverage to 300,000 Uninsured Californians Under New Waiver Initiative,” 2002, HHS press release, available at http://www. hhs.gov/news. Vaden-Kiernan, N., D. Cantor, P. Cunningham, S. Dipko, K. Malloy, and P. Warren, No. 9: 1999 NSAF Telephone Survey Methods, Urban Institute, Washington, D.C., 2000, available at http://www.urban. org/content/Research/NewFederalism/NSAF/Methodology/ 1999MethodologySeries/1999.htm. Waldron, Ingrid, “Contributions of Changing Gender Differences in Behavior and Social Roles to Changing Gender Differences in Mortality,” in D. Sabo and D. F. Gordon (eds.), Men’s Health and Illness, Sage Publications, Thousand Oaks, California, 1995, pp. 22– 45. Wang, Kevin, David Cantor, and Nancy Vaden-Kiernan, Report No. 1, 1999 NSAF Questionnaire, Urban Institute, Washington, D.C., 2000, available at http://www.urban.org/content/Research/NewFederalism/ NSAF/Methodology/1999MethodologySeries/1999.htm. Williams, David R., and Chiquita Collins, “US Socioeconomic and Racial Differences in Health: Patterns and Explanations,” Annual Review of Sociology, Vol. 21, 1995. 93 Wolfe, Barbara L., “Economic Issues of Health Care,” in P. Lindsay Chase-Lansdale and Jeanne Brooks-Gunn (eds.), Escape from Poverty: What Makes a Difference for Children? Cambridge University Press, New York, 1995. Wynn, J., H. Richman, R. A. Rubenstein, and J. Littell, Communities and Adolescents: An Exploration of Reciprocal Supports, Youth and America’s Future, William T. Grant Foundation, Commission on Work, Family and Citizenship, Washington, D.C., 1988. Zuckerman, Barry, and Robert Kahm, “Pathways to Early Child Health and Development,” in Sheldon Danziger and Jane Waldfogel (eds.), Securing the Future: Investing in Children from Birth to College, Russell Sage Foundation, New York, 2000, pp. 87–121. 94 About the Authors FRANK F. FURSTENBERG Frank F. Furstenberg, an adjunct fellow at the Public Policy Institute of California, is the Zellerbach Family Professor of Sociology and a research associate in the Population Studies Center at the University of Pennsylvania. He has published numerous articles on teenage sexuality, pregnancy, and childbearing and also on divorce, remarriage, and stepparenting. His current research is focusing on the family in the context of disadvantaged urban neighborhoods, adolescent sexual behavior, children’s well-being, urban education, and the transition to adulthood. His book publications include Managing to Make It: Urban Families in High-Risk Neighborhoods (with Thomas Cook, Jacquelynne Eccles, Glen H. Elder, Jr., and Arnold Sameroff, 1999), Divided Families (with Andrew J. Cherlin, 1991), and Adolescent Mothers in Later Life (with J. BrooksGunn and S. Philip Morgan, 1987). He holds a B.A. in sociology from Haverford College and a Ph.D. in sociology from Columbia University. MAUREEN WALLER Maureen Waller is an assistant professor in the Department of Policy Analysis and Management at Cornell University. Her research interests include poverty, the family, culture, and social policy. Her work has used qualitative and mixed method techniques to examine the meaning of marriage, fatherhood, and social policies in low-income communities. She holds a B.A. in political science from the University of Dayton and an M.A and Ph.D. in sociology from Princeton University. HONGYU WANG Hongyu Wang is a Ph.D candidate in the Department of Sociology at the University of Pennsylvania. She is interested in the study of the family, adolescents’ well-being, interracial dating, and methodology. She has published an article with Paul Amato on divorce adjustment in the Journal of Marriage and the Family. She holds a B.A. in library science from Peking University and an M.A. in sociology from the University of Nebraska–Lincoln. 95 Related PPIC Publications “At Home and in School: Racial and Ethnic Gaps in Educational Preparedness” California Counts: Population Trends and Profiles Volume 3, Number 2, November 2001 Jennifer Y. Cheng Student and School Indicators for Youth in California’s Central Valley Anne Danenberg, Christopher Jepsen, and Pedro Cerdán Expensive Children in Poor Families: The Intersection of Childhood Disabilities and Welfare Marcia K. Meyers, Henry E. Brady, and Eva Y. Seto “California’s Young Children: Demographic, Social, and Economic Conditions” California Counts: Population Trends and Profiles Volume 4, Number 2, November 2002 Deborah Reed and Amanda Bailey A Portrait of Race and Ethnicity in California: An Assessment of Social and Economic Well-Being Belinda I. Reyes (editor) et al. PPIC publications may be ordered by phone or from our website (800) 232-5343 [mainland U.S.] (415) 291-4400 [Canada, Hawaii, overseas] www.ppic.org 97" ["post_date_gmt"]=> string(19) "2017-05-20 09:36:10" ["comment_status"]=> string(4) "open" ["ping_status"]=> string(6) "closed" ["post_password"]=> string(0) "" ["post_name"]=> string(8) "r_703ffr" ["to_ping"]=> string(0) "" ["pinged"]=> string(0) "" ["post_modified"]=> string(19) "2017-05-20 02:36:10" ["post_modified_gmt"]=> string(19) "2017-05-20 09:36:10" ["post_content_filtered"]=> string(0) "" ["guid"]=> string(50) "http://148.62.4.17/wp-content/uploads/R_703FFR.pdf" ["menu_order"]=> int(0) ["post_mime_type"]=> string(15) "application/pdf" ["comment_count"]=> string(1) "0" ["filter"]=> string(3) "raw" ["status"]=> string(7) "inherit" ["attachment_authors"]=> bool(false) }