White House Task Force Report

Table of Contents

Executive Summary

Chapter 1: Introduction: Our Aspirations for Disadvantaged Youth

Chapter 2: Better Management: Streamlining the Federal Response to Disadvantaged Youth

Chapter 3: Better Accountability: Producing Results, Not Just Promises

Chapter 4: Better Connections: Engaging Youth and Families

Chapter 5: Give Priority to the Neediest Youth: Caring for Special Target Populations

Appendices

Endnotes

Executive Summary

The complexity of the problems faced by disadvantaged youth is matched only by the complexity of the traditional Federal response to those problems. Both are confusing, complicated, and costly.

Ideally, we want the families and communities of young people to be able to supply all that they need-love, a secure childhood, adequate housing, access to health care, a good education, discipline of character, a sense of personal responsibility, and a commitment to their communities and their country. Most young Americans are raised in this kind of environment, and they grow up to be healthy, responsible, and productive citizens.

But there are children who have the same dreams for their own futures, whose daily realities make those dreams seem forever out of reach. The Federal government plays a significant role in helping to make up for those daily deficits for millions of disadvantaged youth.

Because of his commitment to the Nation's youth and to improving the effectiveness of Federal programs in general, the President created the White House Task Force for Disadvantaged Youth on December 23, 2002. [1] He directed the Task Force to develop for his consideration a comprehensive Federal response to the problems of youth failure, under existing authorities and programs, with a focus on enhanced agency accountability and effectiveness.

Our Aspirations for Disadvantaged Youth

Our comprehensive Federal response begins with our Vision for Youth in the form of a national youth policy framework. This is an outcome-focused approach designed to express what we as a country want for disadvantaged youth and for all children. Namely, we want them to grow up:

Healthy and Safe
Ready for Work, College, and Military Service
Ready for Marriage, Family, and Parenting
Ready for Civic Engagement and Service

Our national youth policy framework is designed to ensure that programs we invest in meet one or more of these four goals.

The Task Force was organized into committees around these four guiding principles. Each committee was charged with developing recommendations to improve Federal disadvantaged youth programs under existing authorities. While they were each assigned to review a different subset from among the 339 Federal youth programs that we identified, each of the committees ultimately came to similar conclusions: The best way to get the greatest outcomes for disadvantaged youth from the significant Federal funds invested was to focus on these four goals:

Better Management
Better Accountability
Better Connections
Give Priority to the Neediest Youth

Better Management:
Streamlining the Federal Response to Disadvantaged Youth

The Task Force developed a series of recommendations to address problematic management and coordination issues regarding the hundreds of Federal youth programs we identified. The recommendations in this section address some important issues that we discussed in our April, 2003 report, including problems with overlap and duplication as well as mission fragmentation. To begin to address these problems, we present a proposal for a Disadvantaged Youth Initiative, followed by recommendations on mission alignment, interagency coordination, and improving the Federal grants system.

Create a Disadvantaged Youth Initiative

Through the work discussed in our April, 2003 preliminary report, the Task Force identified the following issues that need to be addressed properly and as comprehensively as possible in order to increase the quality of Federal disadvantaged youth programs:

Reduce/eliminate overlap and duplication of services;
Focus policies and resources on Administration priorities;
Maximize the use of expertise that agencies already have;
Increase collaborative efforts;
Keep public health messages to youth consistent across agencies and programs;
Bring programs into the agency with whose mission they are most closely aligned;
Improve the quantity and quality of program evaluations;
Have a unified definition of "best practices;"
Develop a unified research agenda to identify best practices;
Encourage the development and use of similar performance measures for similar programs.

To properly address these issues, to help ensure that disadvantaged youth grow up to be healthy, productive adults, and to maximize the return on our Federal investment, the Task Force proposes the creation of a Disadvantaged Youth Policy Initiative, to be coordinated through the Executive Office of the President, to do the following:

Develop and coordinate policy, within existing policy processes and structures, to address the needs of disadvantaged youth;

Maximize interagency collaboration to use the significant expertise within specific Federal agencies;

Coordinate Federal research so we can fund programs that produce results that help disadvantaged youth;

Find and elevate models of "what works," through collaboration and coordination with existing agency structures, and help replicate them nationwide.

MISSION ALIGNMENT

Through the process of identifying the 339 Federal youth programs, we encountered several programs that were located in departments whose mission did not provide a clear and compelling reason for locating them within that agency. Our recommendations below reflect the belief that, clearly, the youth programs belong in an agency whose mission more closely matches theirs.

Move YouthBuild to the Department of Labor and Better Align Youth ChalleNGe with the Department of Labor

The Task Force recommends that the Department of Labor assume administrative responsibility for the YouthBuild program, currently administered by the Department of Housing and Urban Development (HUD), and work with the Department of Defense on better aligning the Youth ChalleNGe program with other youth programs funded by DOL. Each program is, at its core, an employment and training program for disadvantaged youth, and will benefit from administrative oversight in DOL within the Employment and Training Administration (ETA), since the strategic goals and mission of that department and agency align directly with the goals and mission of each program.

Move the Gang Resistance, Education and Training Program to OJJDP

The Task Force recommends that the Gang Resistance Education and Training Program, or "G.R.E.A.T.," currently housed in the Bureau of Alcohol, Tobacco, Firearms, and Explosives, be transferred within the Department of Justice to the Office of Juvenile Justice and Delinquency Prevention (OJJDP) in the Office of Justice Programs. This transfer would include technical assistance and other support resources, as well as the grants budget, and the basic content of the strategic plan. It will link G.R.E.A.T. into the many OJJDP programs designed to prevent gangs and violence and promote constructive behavior among young people.

INTERAGENCY COORDINATION

Interagency coordination should be accomplished around topic areas or special target populations. Where issue areas warrant the attention of multiple agencies, we recommend that an interagency group be created to ensure communication, coordination, and collaboration. The Federal government should also help facilitate interagency collaboration at the state and local levels as well, particularly since these levels of government receive the bulk of the Federal funds for youth-serving programs. The following are two examples of recommendations that support this approach. There is an additional recommendation regarding interagency coordination in Chapter 5.

Improve Coordination of Mentoring Programs

The Task Force recommends the creation of a Federal Interagency Workgroup on Mentoring to engage in a variety of activities identified in the report that would aid the coordination and collaboration of all mentoring programs and activities supported by Federal agencies.

Support State and Local Community Planning Process

The Task Force recommends that the relevant agencies contribute existing funds to provide interagency support for state and local government efforts to assess youth-related policies, programs, funding streams, indicators, and data in order to create and implement strategic plans for coordinated investment of Federal, state, and local dollars to improve outcomes for youth.

IMPROVE THE FEDERAL GRANTS SYSTEM

The current Federal grants process is in need of improvements to increase its value to these specific audiences: potential grantees, Federal program officers, and policymakers within the executive and legislative branches. This issue is critically important to those who care about disadvantaged youth, for the more they can take advantage of the resources of the Federal government and maximize their effectiveness, the more likely it is that there are going to be better outcomes for the youth who need help the most.

The Task Force has developed a series of recommendations to improve this current system. Ideally, a searchable database of all past, present, and future grant activity would be created, which could then become the comprehensive database on discretionary grant spending in the United States. From our point of view, this will require a series of steps, which should include the following:

Modernize the Catalog of Federal Domestic Assistance

The Task Force recommends that the Catalog of Federal Domestic Assistance (CFDA) format be revised and updated to fit the way grants are currently administered. This would enable potential grantees as well as policymakers to better track grant opportunities and related activities within youth-serving agencies.

Create a Centrally Available and Improved Grants Database

The Task Force recommends that the CFDA should be linked to various other databases which are currently available, and also made accessible through Grants.gov, the government-wide e-grants portal website.

Improve the FAADS Database

The Task Force recommends the following ways to improve the Federal Assistant Awards Data System (FAADS). First, include the EIN (employer identification number) as well as the DUNS [2] (Dun & Bradstreet, Inc., number) of each grant recipient. Both are unique identifiers and are required by statute and OMB policy to be submitted in applications for funding. Second, provide the key to the Federal Award Identifier Numbers provided by each agency so that the code can be understood by all. These changes will allow all users to identify specific grantees and determine which agencies and programs provided them with funds, and for what purposes.

Create a Resource Mapping Function for the Database

The Task Force recommends that grantees of all Federal youth-serving programs be required to provide the zip codes or GIS codes for all areas where they are providing services.

Research Eligibility of Faith-Based Grant Applicants

The Task Force recommends that the Department Centers for Faith-Based and Community Initiatives review the data from our Federal Youth Programs Survey relating to the applicant eligibility of faith-based groups. The goal of these reviews is to determine the reasons for the reported apparent ineligibility of faith-based groups compared to nonprofit organizations, as reported by approximately half the Federal youth program managers. The Department Centers should then take any steps that they may determine are necessary to follow up with program officers within their agencies to ensure that it is clear that faith-based applicants are equally eligible to apply.

Better Accountability:
Producing Results, Not Just Promises

Government likes to begin things - to declare grand new programs and causes. But good beginnings are not the measure of success. What matters in the end is completion. Performance. Results. Not just making promises, but making good on promises. [3]

President George W. Bush

The public policy world of youth programs suffers from a credibility gap. While there is the will among the public to help young people address the many difficult problems they face, there is a lack of consensus as to how to do it successfully. Unfortunately, the Federal government has been ineffective in helping to close that gap in the public's mind. [4]

Through the last four decades, there has been growing Federal involvement and a rapidly increasing infusion of funds designed to address numerous problems of youth, including substance abuse, violence, teen pregnancy, hunger and nutrition, school failure, and workforce preparation. In Fiscal Year 2003 alone, the Federal government is spending $223.5 billion to help needy children and their families, focusing on these and related issues (see Appendix D). State and local governments and private groups will contribute billions more.

As the President has noted, the focus needs to be on achieving results. Part of the responsibility for this lack of focus on results lies in the fact that the Federal government has often made funding decisions without clear evidence that what it is attempting to do will actually work. Thus, public faith in the efficacy of social programs to successfully address youth failure has eroded. They wonder, what really works? How can we know? This is important because, as one researcher has noted, "even the most perfect solution, if there were such a creature, needs to be recognized and believed in, in order to be adopted as durable policy." [5]

This section of our report includes a number of recommendations providing a look at how the Federal government can strengthen its role in the area of research and evaluation of youth programs. With these recommendations, we call for the Federal government to develop and implement a coherent and comprehensive plan designed to identify with confidence and adopt those practices that will successfully help youth.

The focus of the Task Force here was in two areas, the importance of which should not be underestimated. First, we had broad consensus of the need to improve the Federal role in helping to understand what works. Second, we also recognized that with a Federal investment in youth-serving programs of hundreds of billions of dollars annually, we needed to firmly hold programs accountable for results showing that they actually achieve what they were designed to accomplish. This means that we need well-designed evaluations of current programs so that those not achieving their goals can be quickly discontinued and their resources diverted to other priority needs.

UNDERSTANDING WHAT WORKS

The Task Force has developed several recommendations aimed at improving the quality of the information we have about what works to improve youth outcomes. The first recommendation (in several parts) addresses our concern that we create a more consistent set of guidelines for assessing the quality of program evaluations across agencies. The second puts forth a road map for guiding the direction of future Federally supported research on youth programs, and the third offers a suggestion on improving national survey data collection.

Develop a Unified Protocol for Federal "What Works" Clearinghouses

The Task Force recommends that a committee of the relevant Federal agencies develop a consistent approach to the assessment of youth program and policy evaluations, including the development of protocols. Random assignment experiments are considered the "gold standard" of evaluation because they can most clearly attribute outcomes to interventions. The Task Force strongly recognizes this gold standard and believes those evaluations should be given the greatest weight in shaping what we know about what works. Because individual agencies have different needs, the protocols to be developed need not be identical, just sufficiently consistent so that materials and findings can be shared among agencies with relative ease.

Build a Rigorous and Unified Disadvantaged Youth Research Agenda

The Task Force recommends that a cross-agency research agenda based on large, randomized field trials be created and implemented to assess the effectiveness of interventions to improve outcomes for disadvantaged youth. The design of these field trials must be based on comprehensive, systematic reviews of previous trials, and supported within existing program resources.

Improve Data Collected on the Well-Being of Families

The Task Force recommends that the Federal government seek opportunities to improve the quality of data collected on families in the national data collection systems in order to better monitor the well-being of families, track problems, identify how populations are changing, and provide direction with agenda-setting.

HOLDING PROGRAMS ACCOUNTABLE FOR RESULTS

Providing funds to grantees in order to support proven interventions does not in itself guarantee results. Youth programs must implement these programs correctly and must monitor their service delivery and program outcomes. Currently, similar youth programs rarely have similar performance measures in their Government Performance and Results Act (GPRA) plans, and nearly half do not measure performance at all. [6] In this section we present recommendations for developing and implementing common youth program performance standards and measures. We suggest that these will serve as a starting point for discussion and consensus-building among various stakeholders. This significant process, once completed, will allow policymakers to compare the outcomes of similar programs, no matter which agency they are in. It would also facilitate considerations of program consolidation, redirection of resources, and elimination of ineffective programs, where appropriate. In the interest of improving our ability to document the results of Federal investments, we also offer recommendations on addressing earmarks, and implementing the principles of the No Child Left Behind Act in Department of Defense schools.

Develop Standards for Measuring Grantee Performance

The Task Force recommends the development of uniform standards for measuring grantee performance for all Federal agencies that manage youth-serving programs. While it outlines a process for developing and implementing standard measures, the Task Force understands this is an ambitious goal and will likely require a sustained effort over time, including pilot testing and incremental implementation.

Implement Grantee-Level Performance Measurement Guidelines

The Task Force recommends launching a major effort to work with applicants and programs over the next several years to strengthen the accountability and performance of organizations receiving Federal funds to operate disadvantaged youth programs. The Task Force believes an increased emphasis on performance measurement as both a program management tool and a means by which to communicate program impact will improve the effectiveness of youth-serving programs, while providing Federal agencies the necessary information to hold grantees accountable for results.

Conduct Rigorous Oversight of Earmarked Grantees

The Task Force opposes earmarks for youth programs because they significantly reduce accountability, and they exclude potentially higher quality projects that could otherwise successfully compete for funds. This weakens what should be a strong focus on proven, positive short-term and long-term results for children and youth. The Task Force recommends that each Department with earmarked youth programs use a vigorous, comprehensive oversight and accountability system to oversee these programs.

Implement No Child Left Behind in Department of Defense Schools

The President's landmark legislation, the No Child Left Behind Act (NCLB), provided a new focus on accountability for all of the Nation's children. School districts can no longer focus on average performance, they must now ensure that every subgroup, including disadvantaged students, is making significant progress toward proficiency. Thus, for the first time in our Nation's history, disadvantaged students will be of prime concern to school districts across America. This backdrop gives new leverage to Federal efforts to coordinate services for disadvantaged youth.

The Task Force recommends that the Department of Defense consider implementing select, relevant provisions of the No Child Left Behind Act in Department of Defense Education Activity (DoDEA) schools, in conjunction with the Department of Education. Specific recommendations for consideration include: bringing the DOD system into line with NCLB provisions regarding the pillars of Standards and Accountability and a Focus on What Works; ensuring that the DoDEA strategic plan focuses on improving student achievement, particularly in the core subjects of reading and language arts and math; working to become a model of international excellence by implementing instructional practices that are based on rigorous scientific research; and assessing current levels of parental input and design mechanisms to further increase parental participation within the context of the "Parent Empowerment" pillar of NCLB.

Better Connections:
Engaging Youth and Families

Research has shown that in order to ensure their healthy development, adolescents need caring adults in their lives; opportunities to learn marketable skills and maintain good health; and opportunities to contribute meaningfully to their communities and society. [7] Generally, American families and communities are doing a good job of addressing these youth needs and opportunities. We make several recommendations aimed at validating and building on the strengths that exist in most families and communities. The first recommendation is born from the knowledge that parents play a pivotal role in guiding their children's development and should be supported in that role. The next two recommendations are aimed at providing opportunities for young people to contribute through service, recognizing the value of the assets they bring to their communities and to the Nation.

Increase Parent Involvement in Federal Youth Programs

The Task Force recommends that any Federal program that serves disadvantaged youth should endeavor, when appropriate, to involve parents as much as possible in the program. This means including parents in planning stages and in any advisory groups, as well as in the program itself.

Design a Youth Service Initiative

The Task Force recommends that a youth service initiative be designed that would allow older youth (college age) to display leadership by providing opportunities for them to serve children living in high poverty areas of the United States.

Recruit Youth for Federal Grant Review Panels

The Task Force recommends that college youth be recruited and included as participants on Federal panels that review youth program grants, where feasible.

Give Priority to the Neediest Youth:
Caring for Special Target Populations

While the Federal government is spending billions of dollars to address the problems of youth, the problem is often that too many of these dollars are spread out among too many youth. Although these actions may appear to be preventive, in fact it typically leads to under-serving or never even engaging the youth who most need help-and who become society's most serious problems. Thus, we often see evaluations of youth programs that say the programs do not show much impact. One reason for this might be that the youth that needed to change were either not engaged, or not engaged sufficiently. At an aggregate level, the result is that the public and policymakers never see the kinds of significant improvement they want to see in the things that concern them: juvenile crime, school performance, drug use, and so forth. [8]

Public money should be spent on public problems [9] and targeted to where it is most needed, rather than on all youth, most of whom will grow up just fine without government help. With this view in mind, we begin a discussion that shall continue beyond the life of the Task Force regarding the identification of "special target populations" of youth. These special target populations would be those who represent areas of serious concern, and who carry disproportionately negative consequences for youth and their communities if not addressed. It is these groups named below, as well as others who will be identified in the future, who should be the primary targets of relevant disadvantaged youth programs.

Target Youth in Public Care

The Task Force recommends that the first designated special target pic expense. These are foster care youth (particularly those aging out of foster care) and juvenile justice youth. For both these groups, the Federal government and governments at other levels are serving in loco parentis, in place of the parents.

Target Kids at High Risk

The Task Force recommends that a second group of youth also be considered among the special target populations. This subgroup includes youth with a high number of factors putting them at risk for unproductive or publicly costly lives, such as children of incarcerated parents and migrant youth.

The following recommendations represent a case study showing how the problems of a "special target population" could begin to be addressed. We emphasize that the recommendations below represent merely the first, early steps of this type of effort. Much more remains to be done, but we are excited about the possibilities that future cross-agency collaborative efforts hold for these groups of particularly needy young people. We also note that we anticipate that other special target populations will be identified in the future.

Education of Foster Youth Demonstration Program

The Task Force recommends the creation of a program designed to improve the quality of education for school-age youth in foster care. The program would be established at three levels: Federal, state, and local. It would involve the appointment of a point of contact at the Department of Education to assist in providing awareness of the barriers faced by foster care youth to improving their educational success, and a plan to encourage that state and local school districts establish a similar position in their education departments. Funding for this program could come from the existing sources available to State Education Agencies and Local Education Agencies for disadvantaged youth. [10]

Federal Interagency Committee to Focus on Education Needs of Foster Youth

The Task Force recommends the establishment of a new, ongoing interagency committee which help improve Federal efforts to address the educational needs of youth in foster care. The committee should involve the appropriate representatives from the Department of Health and Human Services, the Department of Labor (Employment and Training Administration), and the Department of Education, and should plan to meet at least on a quarterly basis to ensure that the best efforts are put forth on the Federal level on behalf of these children.

Workforce Training and Education Services for Migrant Youth

The Task Force recommends the creation of a joint venture between the Department of Labor, the Department of Education, and the Department of Agriculture to develop a model program to provide workforce training and basic education services to out-of-school migrant youth ages 16-21.

This model would combine workforce development services, including job training activities, with basic education services designed for individuals with Limited English Proficiency, and would provide these youth with an integrated plan of services and activities designed to raise their educational skills and increase their employment opportunities.

Expand Mentoring Programs to Special Target Groups

The Task Force recommends that the newly-created Interagency Working Group on Mentoring seek opportunities to expand mentoring programs to provide support to young people in foster care and migrant youth.

Chapter 1:
Introduction: Our Aspirations for Disadvantaged Youth

The complexity of the problems faced by disadvantaged youth is matched only by the complexity of the traditional Federal response to those problems. Both are confusing, complicated, and costly.

Ideally, we want the families and communities of young people to be able to supply all that they need-love, a secure childhood, housing, health care, a good education, discipline of character, a sense of personal responsibility, and a commitment to their communities and their country. Most young Americans are raised in this kind of environment, and they grow up to be healthy, responsible and productive citizens.

But there are children who have the same dreams for their own futures, but whose daily realities make those dreams seem forever out of reach. The Federal government plays a significant role in helping to make up for those daily deficits for millions of disadvantaged youth. [11]

"Our goal must be to make sure that all children have the opportunity to learn and succeed," President Bush said in his National Child's Day Proclamation for 2002. "To achieve this, we must use the resources of our families, communities, schools and government to ensure that no child is left behind. [12]"

Because of his commitment to the Nation's youth and to improving the effectiveness of Federal programs in general, the President created the White House Task Force for Disadvantaged Youth on December 23, 2002. [13] He directed the Task Force to develop for his consideration a comprehensive Federal response to the problems of youth failure, under existing authorities and programs, with a focus on enhanced agency accountability and effectiveness.

The Task Force begins this report with our vision for a national youth policy framework. This is an outcome-focused approach designed to proclaim those things that we as a country want for disadvantaged youth and for all children. Namely, we want them to grow up:

Healthy and Safe
Ready for Work, College, and Military Service
Ready for Marriage, Family, and Parenting
Ready for Civic Engagement and Service

Our national youth policy framework is designed to ensure that programs we invest in meet one or more of these four goals.

Our charge was to speak specifically to the Federal government's role in helping disadvantaged youth achieve these outcomes. Thus, for each of these four goals, we have laid out objectives, followed by some examples of the ways the Federal government plays a role in addressing these objectives. These are the basic objectives that should guide Federal programs when they address any or all of these goals.

Goal 1: Grow Up Healthy and Safe

While many trends in adolescent risk-taking are going in the right direction, risk-taking behaviors (illicit substance and tobacco use, violence and premature sexual relations) are still among the top causes of adolescent morbidity and mortality. Teens today are still taking far too many risks with their health and well-being. More than 2.6 million teens use illicit substances each month. More than 3 million youth, ages 12 to 17, are cigarette smokers, and everyday, more than 6,000 try smoking for the first time. About 14 percent of high school students smoke regularly. 29 percent of youth, ages 12 to 20, reported drinking alcohol in the month prior to the survey. In addition, one-third of high school students report having sexual relations in the previous three months, while 46 percent of high-schoolers have experienced sexual relations, putting themselves at risk for pregnancy, sexually transmitted diseases, and emotional trauma. Some 135,132 unmarried girls under the age of 18 gave birth in 2001, long before they were ready to be responsible, married parents. [14]

Whatever their reasons for doing so, adolescents who engage in these behaviors suffer. Some of these behaviors are against the law and have criminal justice consequences, and all have negative health consequences. The end result is the same: children who engage in high-risk behaviors place themselves at long-term risk of having a variety of chronic illnesses, many of which can seriously impair quality of life, and can even end life prematurely.

Federal youth risk prevention and treatment programs should:

Acknowledge that risk behaviors are interconnected- Risk behaviors are highly interconnected. First, the same primary risk factors lead adolescents to engage in a variety of negative behaviors, and conversely the same protective factors provide young people with the resolve to avoid risk-taking. Second, taking part in one risky behavior increases the likelihood that an adolescent will engage in another. For example, if a young person abuses alcohol, he or she is more likely to engage in other forms of risky behavior, such as sexual relations or delinquency.

Strengthen protective factors (such as the moral and social support of the family, school, faith-based and community resources)-Because of the related nature of risk factors, it is important that Federal programs address protective factors that help inoculate young people against all risk behaviors. Research has shown that the more children are connected to family, school, and community, the less likely they will engage in risk behaviors. Parents are especially important, as parent-child relations and parental supervision have been found to help children make wise choices and avoid risk-taking behavior. Federal programs should seek to strengthen these supports for youth.

Strengthen skills and competencies that promote responsible decision making, and improve academic, social and emotional outcomes-In addition to the importance of the outside supports, which some call "external assets", building youth's "internal assets" or strengths also provides them with protection against risk taking. Internal assets are such things as an adolescent's desire to learn, his or her social, emotional, moral and cognitive competencies, and his or her value system. Adults must provide support and guidance to young people, but when it is time to make decisions, each adolescent must have the inner strength and character to make the responsible choices. Federal programs must include skill-building components.

Provide youth with consistent messages about the legal, emotional, and health consequences of engaging in high-risk behaviors-To set a clear example and provide leadership, Federal programs need to communicate a consistent message that highlights the consequences of risk behaviors and avoids encouraging young people to even consider engaging in them. In the past, the messages about some behaviors have been mixed.

Support youth and families with a range of integrated service options that address youth needs-Because every child is different, a range of integrated service options to address youth needs is not only prudent, but cost effective. For most children, existing family and community resources, along with prevention programs, are sufficient. For children who are more engaged in problem behaviors, treatment programs are needed, and they, too, need to vary based upon the needs of each child.

The President has adopted initiatives that reflect these principles, especially the emphasis on sexual abstinence before marriage. As a part of the pending 1996 welfare reform reauthorization, the Administration is recommending that the Abstinence Education program, a feature of the original 1996 legislation, be reauthorized at the same level of funding. In addition, the President's budget for the Department of Health and Human Services Community-Based abstinence education grants increased funding from $40 million to 2002 to $73 million in 2003, an 83 percent increase. This increase ensures that more communities across the country are able to deliver the consistent message that abstinence is the surest way to avoid out-of-wedlock pregnancy and STDs. In addition, the President has proposed continuing the current funding of $12 million to the Adolescent Family Life abstinence program, which reaches more than 112,000 adolescents. [15]

Goal 2: Ready for Work, College, and Military Service

All American youth, regardless of whether or not they are disadvantaged, need to have a strong academic foundation upon which to build their future as they go on to college, work and military service. The Task Force believes that the elements of this foundation need to be provided throughout their academic careers, at all levels of schooling.

Education in the United States is primarily a state and local responsibility. Local and state governments, not the Federal government, establish the Nation's schools, develop curricula, and determine enrollment and graduation requirements. Through the Elementary and Secondary Education Act, the Federal role is to support state and local efforts in embracing major reforms that help to set high expectations, raise academic standards for students, and ensure accountability for results. One example of that role is the Title I formula grant to the states, an $11.7 billion Federal grant program which helps disadvantaged youth by assisting their school districts in low income urban and rural areas.

When President Bush reauthorized the Elementary and Secondary Act by signing the No Child Left Behind Act of 2001, he sought greater efficiency and effectiveness in meeting the needs of all children, including disadvantaged youth. "Too many of our neediest children," he has repeated many times, "are being left behind." The bipartisan legislation increases accountability for states, school districts, and schools; promises greater choice for parents and students, particularly those attending low-performing schools; allows more flexibility for state and local educational agencies in the use of Federal education dollars; and places a stronger emphasis on reading, especially for the youngest children.

Because the President is committed to improving the performance of elementary and secondary schools, his Task Force on Disadvantaged Youth includes "ready for work, college, and military service" as a part of its vision for American young people. The Task Force believes, like the President, that all young people, especially those who are considered disadvantaged, need a core academic foundation that will lead to meaningful skills development and life-long employment.

All youth in America will receive a core academic foundation leading to meaningful skills development and life-long employment opportunities through a continuum of services that improve:

Core academic skills at the kindergarten to eighth grade level-All students should leave eighth grade equipped with the knowledge and skills to take them to the next level. Acquired proficiency in skills associated with the core academic subjects will assist them. The core academic subjects include English, Reading, Math, Science, Foreign languages, Civics and Government, Economics, Arts, History and Geography. No Child Left Behind has reinforced this commitment to knowledge in these areas by requiring local school districts in Title I programs to ensure their teachers hired to teach these subjects are highly qualified. Since President Bush took office, there has been a 39 percent increase in funding for the Improving Teacher Quality State Grant program along with increased funding in other areas to ensure students have the best opportunities to improve these skills.

Core academic skills at the high school level with appropriate non-school workplace preparation-The progression of knowledge attainment in these subjects should continue into the secondary level of education. We should build on the foundation that began at the elementary level and improve adolescents' skills. As a complement to these skills, youth need appropriate introductions to occupational skills training to keep up with the changing nature of the 21^st Century economy. The business community has embraced No Child Left Behind as an economic and workforce development imperative. Graduates with improved academic and technical skills training have more opportunities in the workforce and cost businesses less money on training in basic skills such as reading and math.

These dynamics are also shaping much of the President's push for the reauthorization of the Workforce Investment Act (WIA). The President's reauthorization initiative includes reforms to make the workforce investment system more flexible, accountable, and focused on results to ensure that only programs that are most effective at helping Americans find work are funded.

The President's changes to the WIA also target disadvantaged youth, especially out-of-school youth, with a Targeted State Formula program and Challenge Grants. The Targeted State Formula program would be used at the local level. Challenge Grants to cities and rural areas would be awarded on a competitive basis, with funds going to programs proven effective at serving youth. Grants would also be awarded on a discretionary basis to high-quality programs that lead to high academic achievement.

Post-secondary skills attainment through an array of options that includes higher education or occupational skills training. Most youth will continue on to higher education but a quarter of our youth seek to enter the workforce in jobs that need special skills training. All youth need to be prepared for the technological advances and global changes that pose challenges to the competitiveness of the American workforce. These advances call for our institutions of higher education to better prepare our youth to succeed after graduation. [16] It also calls for better opportunities for those youth that are looking for alternative possibilities. Programs that train youth to be successful in the workforce need to connect with the corporate world to provide effective training and academic skills that make them marketable in the workplace.
Youths' connection to their schools-A key to the success of students' experiences in school is their sense of connection to their school. School connection is the belief by students that adults in the school care about their learning as well as about them as individuals. Research shows us that the critical requirements for feeling connected include students' experiencing: high academic expectations and rigor coupled with support for learning; positive adult-student relationships; and safety, both physical and emotional. [17]

Increasing the number of students connected to school is likely to impact critical accountability measures, such as academic performance; incidents of fighting, bullying, or vandalism; and absenteeism/school completion rates.

Strong scientific evidence demonstrates that increased student connection to school promotes educational motivation, classroom engagement, and improved school attendance. These three factors in turn increase academic achievement. Likewise, there is strong evidence that a student who feels connected to school is less likely to exhibit: disruptive behavior, school violence, substance and tobacco use, emotional distress, and premature initiation of sexual activity. These findings apply across racial, ethnic, and income groups [18].

Based on current research evidence, the most effective strategies for increasing the likelihood that students will be connected to school include:

Implementing high standards and expectations, and providing academic support to all students;
Applying fair and consistent disciplinary policies that are collectively agreed upon and fairly enforced;
Creating trusting relationships among students, teachers, staff, administrators, and families;
Hiring and supporting capable teachers skilled in content, teaching techniques, and classroom management to meet each learner's needs;
Fostering high parent/family expectations for school performance and school completion; and
Ensuring that every student feels close to at least one supportive adult at school.

Goal 3: Ready for Marriage, Family, and Parenting

Part of growing up is learning to be a grown-up, preparing to be a good spouse and a good parent. Generally this preparation is the responsibility of parents. Most often they do it well, but sometimes they will need support in providing this preparation. In the occasional instances where parents are unable to prepare youth for these roles, the Federal government must take steps to ensure that young people have caring adults in their lives to provide nurturing and role modeling. This task also should be borne by other levels of government, as well as community and faith-based organizations.

When Federal programs interact with youth, they should:

Seek to acknowledge, strengthen, and reinforce parental engagement in youths' development and improve parent-child relationships-Research has shown that healthy families with strong marriages and close parent-child relationships provide the strongest protection against adolescent risk-taking behaviors. According to "Reducing the Risk," [19] a publication from the National Longitudinal Study on Adolescent Health (Add Health), parents are central in shaping outcomes for young people. Controlling for the number of parents in a household, race, and economic status, the study found that children who report feeling connected to a parent are protected against many different kinds of risks: emotional distress and suicidal thoughts; cigarette, alcohol, and marijuana use; violent behavior; and early sexual activity. This correlation holds true for both older and younger adolescents.
Specifically, the study found that the keys to the healthy growth and development of teens include 1) parents' physical presence in the home at key times during the day, 2) parental emotional connectedness to their adolescent children; and 3) parental expectations of high academic performance of their children. These factors especially protected teens from premature sexual activity, substance use, and pregnancy.

The President's vision of strengthening marriages and families in America seeks to improve child well-being by increasing the number of two-parent families. The Administration's proposal for TANF reauthorization encourages the formation and maintenance of healthy marriages, promotes responsible fatherhood, and strengthens the bonds between parents and their children. The fatherhood dimension of this proposal is significant. It encourages fathers to assume greater responsibility for their children, by supporting the activities of public and nonprofit community entities, including faith-based organizations, to support fathers as husbands, parents and breadwinners. Fatherhood grants will accomplish a variety of objectives including promoting responsible, caring, and effective parenting through counseling, mentoring, and parenting education, dissemination of educational materials and information on parenting skills, and the encouragement of positive father involvement, including the positive involvement of nonresident fathers.
Seek responsible, caring adults to support young people in need-Unfortunately, not all young people have the support of loving parents. For this reason, Federal programs-to the degree that they address the problems of disadvantaged youth-need to help young people in need link up with responsible, caring adults. Research has shown that frequent and sustained contact with a trained mentor can significantly improve adolescent outcomes.
The President has made clear his commitment to mentoring through his Mentoring Initiative, announced in the 2003 State of the Union address. [20] This three year initiative has two components. The first seeks to link mentors with children less than 18 years of age with a parent in prison. The second provides mentors to middle school youth to improve their academic outcomes.
Prepare youths to pursue the ideal of a healthy, two-parent family through the development of character needed to become loving and responsible spouses and parents-Because healthy parent-child relations are important for every generation, young people who are navigating the transition to adulthood need to develop the personal qualities of heart and character so that they can become loving and responsible spouses and parents themselves. Federal programs should encourage character development and respectful adolescent relationships characterized by abstinence.

Goal 4: Ready for Civic Engagement and Service

From its earliest origins, America has flourished because of the willingness of its citizens to participate in democratic processes and institutions. It has been blessed again and again because citizens have stepped forward and served not only their communities but their Nation, without considering material reward. The values embedded in that spirit, and the benefits that flow from it have made us stronger, healthier, and more prosperous as a Nation. However, these qualities do not automatically appear and must be reinforced among ourselves and cultivated in our young people by teaching and by the examples set by every volunteer, every mentor, and everyone who cares enough about their community to give deeply of themselves. The Federal role in this involves the following goals:

Federal programs that address youth civic engagement and service should:

Foster youths' development into caring adults who have a clear sense of belonging and responsibility to their communities and the Nation through engagement in citizen service-Responsible citizenship in the United States begins with civic engagement at the local level and well-informed voting for public officials at all levels of government. This is the very heart of what it means to be an American. Yet, in the 2000 election, less than one-third of the population between the ages of 18 and 24 reported that they had voted. One key reason cited for not voting: people don't believe their vote makes a difference. In addition, only 9 percent of high school seniors could identify how a democratic society benefits from the active participation of citizens. [21] These sobering statistics demonstrate the imperative of a greater focus on readying youth, especially disadvantaged youth, for civic engagement.
President Bush's "We the People" initiative aims to reverse this troubling pattern. This initiative, run by the National Endowment for the Humanities, aims to cultivate an ethic of citizenship through enhanced civic education in the schools. The theory is that better history and civics education will lead naturally to higher levels of civic engagement, enabling young people to guard, nurture, and care for the Nation's guiding principles and values.

Of course, responsible American citizenship means more than just voting or speaking at a town meeting. It means giving back to society through volunteerism and social service. In his 2003 State of the Union address, the President encouraged all Americans "to apply the compassion of America to the deepest problems of America." Believing in the "wonder-working power, in the goodness and idealism of the American people," the President has encouraged civic engagement and volunteerism. [22]

The good news is that, unlike voting, volunteering has become more common among college-bound high school students since 1984. More than 80 percent now say they volunteer; 53 percent on a regular basis. [23] Such service among young people should be encouraged at an early age. The sooner young people become involved in addressing the needs of others, the sooner they develop a sense of civic responsibility and civic belonging, reinforcing the sense that they belong to something larger than themselves.

To what degree disadvantaged youth fit the pattern among college-bound youngsters is not clear. Rates of volunteerism may be lower among disadvantaged young people. Nevertheless, disadvantaged youth can and do volunteer, often finding that community service allows them an opportunity to give back what others have given to them. They feel better about themselves-less like a victim and more like someone who has something of value to offer.

There are many examples of Administration initiatives to foster greater levels of volunteerism by the general public. These include the Faith-based and Community Initiative and USA Freedom Corps, which includes:

Volunteer Network, a clearinghouse of volunteer opportunities;
Council on Service and Civic Participation, to encourage and recognize outstanding and dedicated volunteer activity (awards are given to individuals, families, schools, businesses, and community and faith-based organizations);
Citizen Corps, supporting local efforts to helping communities prevent, prepare for, and respond to crime, natural disasters, and terrorism.
The President has also expanded Learn and Serve-which integrates community service with academic and extracurricular activities of young people-to include a wider range of public agencies and nonprofit organizations, including faith-based groups. Additionally, he has called for amending the Higher Education Act to require every college and university to increase the percentage of Federal Work-Study funds devoted to community service to 50 percent.

The Task Force has identified 105 Federal programs in 10 Federal agencies or departments that involve young people, including disadvantaged youth, in programs that include community service goals or activities. [24] These programs are wide ranging. The Corporation for National and Community Service, for example, sponsors a grant program under Learn and Serve America that encourages elementary and secondary schools and community organizations to create, develop, and offer service-learning opportunities for their students.

Another program, in the Department of Education, specifically targets the involvement of disadvantaged youth through the 21^st Century Community Learning Centers. This program enables rural and inner city public elementary and secondary schools (or consortia of schools) to plan, implement, or expand projects that benefit the educational, health, social service, cultural, and recreational needs of communities.
Support youths' connection to their communities by building on their internal assets and creating opportunities-Federally funded state and community based programs should not only treat youth as recipients of services, but reach out to them as partners in program planning, design, and leadership. Youth have important talents to share and skills to learn that often fall outside of academic curricula. Many youth know well what will work best for them. When they feel ownership for a program, they are more likely to give commitment and energy.
The experience of being listened to and respected as partners can build a young person's self-efficacy and help develop the characteristics of responsible adult citizenship. Some Federal programs already support this approach. For example, Federal program announcements for funding runaway and homeless youth programs include program performance standards that encourage youth involvement in service design, program planning and organizational management. [25]

Youth engagement and leadership need to be developed carefully-with training, preparation, and realistic expectations. Youth involvement must be genuine, not token. A detailed guide to successful youth involvement has been published for the runaway and homeless youth programs but is applicable to virtually every program serving young people. [26]

Chapter 2:
Better Management: Streamlining the Federal Response to Disadvantaged Youth

The Task Force developed a series of additional recommendations to address problematic management issues regarding youth programs. The recommendations in this section address some important issues that we discussed in our April, 2003 preliminary report, including problems with overlap and duplication, as well as mission fragmentation.

In our April preliminary report, we reported that:

Most of the 72 million American youth ages birth to 17 are doing fine, but a number suffer from a variety of factors that place them at a disadvantage.

About 15 percent of American children live below the poverty level.

For some of them, the government serves in loco parentis, in the place of family. More than half a million children are living in foster care. Some have families who are either barely functioning or nonexistent; about 3 in 10 live with just one parent. About 1.5 million children had parents in state and Federal prisons. In 2000, 879,000 children were abused or neglected.

Some face difficult health problems. About 12 percent live in families whose dire circumstances qualify them to receive nutrition or food assistance. Some 15 percent of school-age children are seriously overweight. Some suffer from chronic illnesses.

Some young people engage in risk-taking behaviors. More than 2.6 million teens use illicit substances each month. More than 3 million youth ages 12 to 17 are current smokers, and everyday, more than 6,000 try smoking for the first time. About 14 percent of high school students smoke frequently. One-third of high school students reported having sex in the previous three months, while 46 percent of high school students have had sexual intercourse, putting themselves at risk for pregnancy and sexually transmitted diseases. Some 153,437 girls under the age of 18 gave birth in 2001, long before they were ready to be a responsible parent.

Some live in downtrodden or violent neighborhoods, leaving them vulnerable to crime. About 400,000 teens themselves commit violent crimes each year. Juveniles were involved in 16 percent of all violent crime index arrests and one-third of all property crime index arrests in 2000. One-quarter of all persons arrested for robbery that same year were under age 18.

Many are being left behind in school. Nearly 70 percent of inner city and rural 4^th graders cannot read at even a basic level. Some 13 percent of students are considered learning disabled. Almost 11 percent drop out of school entirely. More than 5.5 million children received special education services, 2.7 million for a specific learning disability and almost half a million for emotional disturbance. Many speak limited or no English. [27]

From the Federal Youth Programs Survey [28] we developed, we also found that:

There are 339 Federal programs serving disadvantaged youth in 12 departments. [29]
Federal involvement in issues surrounding disadvantaged youth has expanded significantly in the last four decades.
The Office of Management and Budget reports that Federal funding for disadvantaged youth in Fiscal Year 2003 totals $223.5 billion (see Appendix D).

Overall management and coordination of disadvantaged youth programs is poor:

Many of the 339 programs are offering multiple kinds of services to a wide variety of youth subgroups. These services and target populations frequently overlap.
The current Federal response to disadvantaged youth is a perfect example of "mission fragmentation." In situations like this, GAO recommends that programs with similar goals, target populations and services be coordinated, consolidated or streamlined as appropriate, to ensure that goals are consistent and that program efforts are mutually reinforcing.
Youth program statutes are often written quite broadly, allowing agencies to have considerable discretion in the activities they conduct and populations they serve. Agencies exercised that discretion aggressively and widely. As time goes on, agencies often expand their programs to add in the "issue du jour." This type of "mission creep" leads to a haphazard response and a lack of the rationality that these serious and complex problems demand. These problems require that all youth-serving agencies have a clear and focused mission and a plan to ensure collaboration among Federal programs involved in addressing the same issue.

Accountability and Research Pose Problems

We have very little information to show for certain that the billions of dollars for youth programs are being spent wisely and effectively. Our review of youth-serving programs shows that a high percentage of Federal youth programs fared poorly in the three critical areas of evaluation and research:

OMB's PART process: Only one of 28 youth-serving programs was rated "effective" by OMB during the Fiscal Year 2004 Performance Assessment Rating Tool (PART) process. Three were "moderately effective," five were "adequate" and the remaining 68 percent were rated either "results not demonstrated" or "ineffective."
The GPRA process: Less than half of the identified programs indicated that they were included in their Department's Government Performance Results Act (GPRA) plans. This is potentially problematic because the purpose of GPRA is to provide objective information about the effectiveness and efficiency of Federal programs and spending, and thus increase the level of accountability to Congress and the American people. Thus, with no related goals, and more importantly, no performance measures, there can be no accountability under GPRA.
Program Evaluations: More than half of all of the 339 youth-related programs had not been evaluated within the last five years. Of the smaller number that was evaluated, 75 percent were evaluated independently, while the remainder were done mostly by the grantees themselves. The quality of the evaluations was low: only 27 programs have been evaluated using the more scientifically reliable random assignment method. Only 70 programs reported using some form of "outcome" evaluation, rather than a process evaluation.

To address these problems, we begin with a proposal for a Disadvantaged Youth Initiative, followed by recommendations on mission alignment, interagency coordination and improving the Federal grants system.

Create a Disadvantaged Youth Initiative

The Task Force identified the following issues which need to be addressed properly and as comprehensively as possible in order to increase the quality of Federal disadvantaged youth programs:

Reduce/eliminate overlap and duplication of services;
Focus policies and resources on Administration priorities;
Maximize the use of expertise that agencies already have;
Increase collaborative efforts;
Keep public health messages to youth consistent across agencies and programs;
Bring programs into the agency with whose mission they are most closely aligned;
Improve the quantity and quality of program evaluations;
Have a unified definition of "Best Practices;"
Develop a unified research agenda to identify best practices;
Encourage the development and use of similar performance measures for similar programs.

To help ensure that disadvantaged youth grow up to be healthy, productive adults, and to maximize the return on the Federal investment, the Task Force proposes the creation of a Disadvantaged Youth Policy Initiative, to be coordinated through the Executive Office of the President, to do the following:

Develop and coordinate policy, within existing policy structures, to address the needs of disadvantaged youth;

Maximize interagency collaborations to utilize the significant expertise within specific Federal agencies;

Coordinate Federal research so the government can fund programs that produce results that help disadvantaged youth;

Find and elevate models of "what works" and help replicate them nationwide.

MISSION ALIGNMENT

Through the process of identifying and reviewing the 339 Federal youth programs, we encountered several programs which were located in departments whose mission did not provide a clear and compelling reason for locating them within that agency. Clearly, they belong in an agency whose mission more closely matches theirs, and our recommendations below reflect that sentiment.

Move YouthBuild to the Department of Labor and better align Youth ChalleNGe with the Department of Labor

The Task Force recommends that the Department of Labor (DOL) assume administrative responsibility for the YouthBuild program, currently administered by the Department of Housing and Urban Development (HUD), and work with the Department of Defense on better aligning the Youth ChalleNGe program with other youth programs funded by DOL. Each program is, at its core, an employment and training program for disadvantaged youth, and will benefit from administrative oversight in DOL within the Employment and Training Administration (ETA), since the strategic goals and mission of that department and agency align directly with the goals and mission of each program.

The Department of Labor has more than thirty years of experience working with employment and training programs. Importantly, DOL currently administers the Nation's One-Stop Career Center system, established under the Workforce Investment Act of 1998 (WIA). This system stands as a solid service delivery infrastructure to provide employment and training services to youth and adults. The One-Stop system has and continues to develop strong connections to education and training institutions and employers and is charged with being "in tune" with employer needs-understanding the skills employers seek, and knowing the high-demand occupations in local areas.

DOL also currently administers the WIA formula grant youth programs and the Job Corps program; employment and training programs specifically targeting young people. Under WIA youth programs, youth receive assessments and services that provide them with the education, skills training and supportive services needed to successfully undertake postsecondary education or unsubsidized employment. These programs are also directly connected to the job market and employers. In addition, the Job Corps program has helped more than 1.5 million severely disadvantaged young people, ages 16-24, who have either dropped out of school or are in need of additional education and training. The program is a full-time, year-round residential program where youth split their time between the classroom, earning their GED or diploma, and learning a technical skill or trade. Much of the program's success lies in its strict discipline and behavioral standards.

YouthBuild is currently administered by HUD, and provides unemployed young people, ages 16-24, with work experience while they build affordable housing for homeless and low-income families in their own neighborhood. Young participants split their time between the construction site and the classroom, where they earn their GED or high school diploma, learn to be community leaders, and prepare for college. The program provides economically challenged young adults with the skills necessary to achieve economic self-sufficiency, leadership and commitment to community. Under WIA, YouthBuild is a mandatory partner in the One-Stop system. After over five years of operation under WIA, however, it is evident that a more direct linkage is needed between the YouthBuild program and the activities of the One-Stop Career Center system.

The following story illustrates the value of this approach. Recently, a Rockford, Illinois, Job Corps program helped a local YouthBuild program that stood to lose 40 volunteers and as much as 20 percent of its funding due to cuts in the national AmeriCorps program. As reported in the August 5, 2003, Rockford Register Star, the local Job Corps volunteered to "pick up the slack" and share resources with the struggling YouthBuild project. At the same time, the programs would work "hand-in-hand," with Job Corps benefiting when its graduating students could gain additional hands-on job skills volunteering for community projects with YouthBuild. The YouthBuild director praised Job Corps for its willingness to share "turf."

This collaboration made news because it is all too rare. Situating similar programs in a single department permits department leadership to override issues of turf forcing collaborations that sustain programs in crisis when program operators do not display the spirit embodied in the Rockford community. [30]

Youth ChalleNGe is currently administered by DOD/Army National Guard and targets 16-18-year-old high school dropouts. The goals of this program are to enhance the life skills, educational levels and employment potential of at-risk youth through structured, quasi-military training. The first five months are a residential phase, which is followed by a year-long mentoring relationship with a trained mentor from the youth's community. Youth ChalleNGe is not a mandatory partner in the One-Stop system. In many cases it operates completely separately from the One-Stop system. The approach of the Youth ChalleNGe program is similar to the Job Corps program currently administered by DOL. The Youth ChalleNGe program and participants can benefit from stronger ties with the One-Stop system and Job Corps.

DOL should assume administrative responsibility for the YouthBuild program, currently administered by HUD, and administrative oversight of the Youth ChalleNGe program, currently administered by DOD. The programs would retain their core mission and service delivery model while being merged administratively with other ETA-funded youth programs. In addition, a strong partnership between DOL and HUD would be maintained, and DOL and DOD will finalize details of how Youth ChalleNGe can be better aligned with DOL programs while maintaining the strong link with the National Guard through contracting or other means. This recommendation will result in improved services to young people and enhanced program outcomes by:

Administering each program in the agency where the strategic goals of the department and agency align directly with the core goals of each program as discussed below;

Effectively using the One-Stop system's specialized resources, expertise and market knowledge, particularly in connecting individuals to supportive services necessary to complete the program, and in today's tight labor market, placing individuals in education, training or jobs with a market demand; and

Enhancing the One-Stop system through drawing on these two unique program models and integrating best practices from these models throughout the One-Stop system.

While YouthBuild and Youth ChalleNGe have unique program components, the programs align with and can be enhanced by the DOL strategic goals of:

A Prepared Workforce: DOL's outcome goals of increased employment, earnings and retention, as well as assistance to youth in the transition to work, are clearly in line with the programmatic goals of these programs; and

A Competitive Workforce: The outcome goal of anticipating and addressing workforce gaps-including demand for new workers, replacement workers, and highly skilled workers. [31]

The program goals also align with and can be enhanced by ETA's goals of:

Maximizing Partnerships with Other Programs: Using Federal dollars efficiently by maximizing collaboration with One-Stop partner programs and reducing duplication of effort across funding streams; and

Targeting Resources on Out-of-School Youth. [32]

Integrating these funding streams into the one Federal agency that is responsible for delivering youth employment services will benefit the programs and their participants. Policies can be streamlined, and services can be expanded by leveraging program funds. More individuals can be served in a more effective and comprehensive manner, boosting program performance, and supporting a competitive and prepared workforce.

Move the Gang Resistance, Education and Training Program to OJJDP

The Task Force recommends that the Gang Resistance Education and Training Program, or "G.R.E.A.T.," currently housed in the Bureau of Alcohol, Tobacco, Firearms, and Explosives (BATF), be transferred within the Department of Justice (USDOJ) to the Office of Juvenile Justice and Delinquency Prevention (OJJDP) in the Office of Justice Programs (OJP). This transfer would include technical assistance and other support resources, as well as the grants budget, and the basic content of the strategic plan. It will link G.R.E.A.T. into the many OJJDP programs to prevent gangs and violence and promote constructive behavior among young people.

G.R.E.A.T. is a $16 million life skills competence program designed to provide students with the skills they need to avoid gang pressure. The program began in 1991 with the goal of helping prevent youth crime, violence and gang association while developing a positive relationship among law enforcement, families, and young people to create safer communities. It is authorized through the Treasury and General Government Appropriations Act of 2002. However, the agency that houses this program, BATF, is fundamentally a regulatory and law enforcement organization "dedicated to reducing violent crime and protecting the public," and charged with enforcing the Federal laws and regulations relating to alcohol, tobacco, firearms, explosives and arson.

On the other hand, OJJDP's mission is fully in tune with that of the G.R.E.A.T. program. It is dedicated to delinquency prevention and youth development promotion, with programs such as Drug-Free Communities Support, Juvenile Mentoring Program (JUMP), National Youth Network, SafeFutures, Safe Kids/Safe Streets, Strengthening America's Families, Gang-Free Schools and Communities, Safe Start, and Boys and Girls Clubs.

OJJDP also maintains a strong focus on evaluation of programs and research into effective practices, for example, by funding the Center for the Study and Prevention of Violence. The Center provides technical assistance to a number of violence prevention programs across the Nation and publishes the widely recognized Blueprints for Violence Prevention, a set of model prevention and intervention programs that meet a strict scientific standard of program effectiveness. [33] OJJDP addresses many areas relevant to G.R.E.A.T. and provides resources for substance abuse prevention, public education/information, community re-entry support for juvenile offenders, mentoring, and many other services.

OJJDP is led by an Administrator, under the leadership of the Assistant Attorney General, (OJP), who reports to the Associate Attorney General, under the Attorney General and the Deputy. The other OJP agencies share many of the prevention and community strengthening objectives of OJJDP, such as victim assistance, prevention of violence against women, in-school conflict resolution, community development, research, etc. In turn, OJP is alongside the Office of Community Oriented Policing (Cops in Schools), and the Community Relations Service (reduction of racial tension), which also report to the Associate Attorney General.

With G.R.E.A.T. relocated, BATF can focus on its counter-terrorism and regulatory functions, many of which are particularly crucial post-9/11. From its new home in OJJDP, G.R.E.A.T. can connect with the larger world of youth development and risk prevention activities in which OJJDP has traditionally engaged. This program clearly shows signs of success. As such, it deserves careful attention and possibly expansion and replication of its methods in other risk prevention, character education, and youth development areas. Synergy with other OJJDP programs could assure more students are reached and underserved communities included.

INTERAGENCY COORDINATION

Interagency coordination should be accomplished around topic areas or special target populations. Where issue areas warrant the attention of multiple agencies, we recommend that an interagency group be created to ensure communication, coordination and collaboration. The Federal government should also help facilitate interagency collaboration at the state and local levels as well, particularly since these levels of government receive the bulk of the Federal funds for youth-serving programs. The following are two examples of recommendations which support this approach. This report includes additional recommendations regarding interagency coordination in Chapter 5.

Improve Coordination of Mentoring Programs

The Task Force recommends the creation of a Federal Interagency Workgroup on Mentoring to engage in a variety of activities identified below that would aid the coordination and collaboration of all mentoring programs and activities supported by Federal agencies.

Research has shown that an ongoing relationship with a caring adult is a primary component of healthy adolescent development. Mentoring is a powerful tool that, when done properly, connects vulnerable youth with a responsible advocate who can help navigate safe pathways to adulthood. Research also shows that when at-risk youth are linked with a well-matched, screened, and trained mentor, they are likely to improve their academic achievement while decreasing their involvement with the juvenile justice system. [34]

The President, therefore, has a comprehensive agenda of using mentoring as an important tool to ensure that no child is left behind. He has proposed, as part of his USA Freedom Corps, the placement of mentors in the lives of more than one million disadvantaged children who are transitioning from childhood to adolescence. His initiative, working in tandem with national youth service organizations as well as with local community and faith-based organizations, has two components:

A proposed $100 million per year program, administered by the Department of Education, to provide one million new mentors to disadvantaged middle school students. [35] This component builds on the Mentoring for Success Act passed as an amendment to the Elementary and Secondary Act in 2001, which provided $17.5 million to fund local mentoring programs in 2002 and 2003. It also reinforces No Child Left Behind by supporting strategies proven to enhance the academic performance of disadvantaged children. The strengths of this component are many: it is directed at school-based mentoring; it focuses on youth who are most at risk of educational failure; and it brings adults into the schools.

A $50 million per year program, administered by the Department of Health and Human Services, to provide 10,000 mentors for children whose parents are in prison. This furthers the goals of the Safe and Stable Families Amendment of 2001, which called for the expansion of services to strengthen families using community and faith-based groups. It is also intended to strengthen healthy and positive bonds between children and incarcerated parents and to cultivate mentors from within the child's own extended family and community. A first round of grants is already underway based on an initial appropriation of nearly $10 million for Fiscal Year 2003 [36].

The President's initiatives build upon existing mentoring activities already currently supported by 13 different Federal agencies. These agencies administer more than 120 different programs that provide support for a variety of mentoring activities. In fact, mentoring is the 8^th most frequent activity identified among the 41 types of activity identified in the 339 Federal programs that serve youth. However, currently there is little coordination or collaboration among these various Federal programs, nor does any single agency have the responsibility of cultivating and advancing the expertise and knowledge of "best practices" with regards to mentoring. As a result:

Agencies are often unaware of the mentoring activities being engaged in by other agencies or within their own agency. This often results in duplication of efforts in some areas while other areas receive little or no services.

Research and evaluation of the same or similar programs is not coordinated, thereby resulting in costly duplication of effort and inconsistent findings.

The new Interagency Work Group on Mentoring would include representatives from the departments of Justice, HHS, Education, Labor and the Corporation for National and Community Service. Functioning under the leadership of the USA Freedom Corps and the Disadvantaged Youth Initiative, the work group would:

Identify all of the mentoring programs and activities currently being engaged in and being planned for;

Map grantees receiving mentoring funds;

Identify areas where additional mentoring resources are needed;

Assess the current knowledge about what works in mentoring;

Propose a common definition of effectiveness;

Propose common data collection elements;

Identify gaps in current research;

Design an Interagency Mentoring Research Strategy;

Develop a process for identifying proven and promising practices;

Develop mechanisms for making the public and grantees in particular aware of best practices, research and evaluation findings.

Support State and Local Community Planning Process

States and local communities have bureaucratic infrastructures that parallel those of Federal agencies. Each of these governmental entities has its own mission and budget that guide the allocation of funds to communities. The fragmentation of Federal youth policy is compounded by fragmented state spending which leaves local communities piecing together program dollars from a wide variety of funding streams, each with its own regulatory and reporting requirements.

Over the years, states and local communities have tackled this issue by developing coordinating bodies (task forces, collaboratives, councils, etc.). Often, the creation of these groups was required in order to comply with the Federal statute. The majority of these groups tend to be focused on a specific outcome (e.g., a Governor's Council on Literacy), a specific population (e.g., an Attorney General's task force on children of incarcerated parents), or a specific type of service (such as substance abuse prevention services in the case of SAMHSA's state incentive grants). While efforts such as these are often quite successful, states and localities must then struggle to align these coordinating bodies with each other. States and localities may be able to increase the efficiency and effectiveness of their service delivery by incorporating a range of outcomes, populations and services into an overarching coordinated plan. Efforts to coordinate along specific lines such as those above could be handled as subsets of the overarching effort. This would ensure that these narrower coordinating efforts are aligned with each other and feed into an overarching state or local plan. Over the last five years, the HHS Administration for Children and Families has funded work in 13 states to ensure planning, innovation, and better collaboration at the state level and between the state and community levels, around issues and programs for youth development. Evaluation data and case study analysis are in the final stages for these projects and could provide useful lessons and models for state and local collaboration to improve outcomes for disadvantaged youth. [37]

Using existing resources, the multiple Federal agencies that support youth programs would jointly fund demonstration grants to support state and local efforts to develop and implement a strategic prevention framework to guide future investments in disadvantaged youth. Participating states would receive a grant to support a state-level coordinator and a strategic mapping and plan development process. Federal staff would also provide technical assistance and guidance related to the availability of funding resources (grants.gov) and evidence-based practices (the What Works Clearinghouses).

States would develop plans using a theoretical model designed to include: (1) the full age range from 0-17; (2) the full range of goals: physical health and safety, academic achievement and workforce preparation, healthy family/social relationships, and civic engagement and service; and (3) science-based models of public health and safety (e.g., universal, indicated and targeted interventions or an ecological framework that proposes individual, family school, community level interventions).

Each state would assign a coordinator who would be responsible for regularly convening representatives from all relevant departments and agencies, as well as other key stakeholders, and helping these representatives to craft and implement a plan to be jointly adopted by the Secretaries or equivalent. The coordinator would also ensure that drafts of the plan are regularly reviewed by service providers, local leaders (including members of the faith community), parents and residents. When possible, this work would be undertaken by an existing collaborative body (council, task force, etc.) Existing collaborative bodies would be encouraged to pursue ways to expand their mandate and membership to allow them to tackle the full strategic framework articulated above.

A report on the state mapping process would consist of the following elements:

Youth Indicators Report-how well youth are achieving in each area outlined in the framework above;
Youth Services, Supports and Opportunities Report-an assessment of the quantity and quality of the services, supports and opportunities available to young people in each area of the framework;
Youth Budget-mapping existing Federal, state and local dollars and programs to determine what resources are being allocated to each area of the framework;

The state strategic plan would then include recommendations on how to reduce fragmentation and improve the effectiveness of efforts for disadvantaged youth.

Recommendations would include:

Coordination strategies (to improve this and other collaborative efforts);

Communications strategies (to promote a positive vision of and for disadvantaged youth, and to link together into coherent messages the array of things young people need to know);

Funding strategies (to maximize Federal, state and local investments);

Capacity building strategies (to support practitioners and volunteers, organizations, and communities through technical assistance and other means aligned across departments and agencies);

Processes for selection and implementation of Evidence-Based Prevention programs;

Strategies for maintaining accountability for outcomes.

In addition, each state would select one or two target pilot communities to conduct parallel planning processes, and provide them with the technical assistance, incentives, and leadership necessary to spark and sustain their efforts. The state coordinator would be responsible for convening these communities that are engaged in parallel planning processes, and for encouraging the alignment of frameworks and efforts between the state and local levels.

Once the state and local plans are completed, the Federal government could convene Federal officials from across departments and agencies to review the plans, meet with the state and local representatives and entertain suggestions for ways the Federal government could streamline its work and increase flexibility as needed for states to fully implement their plans.

Subsequent Federal community prevention grant announcements could give priority to states that have established such coalitions and which have developed viable strategic planning frameworks and can show how additional funding will be used to support evidence-based practices that further these plans. Federal grant review processes might also reward applicants that demonstrate coordination and collaboration with other Federal grant programs and existing resources in their communities.

IMPROVE THE FEDERAL GRANTS SYSTEM

The current Federal grants process is in need of specific improvements to increase its value to these specific stakeholders: potential grantees, Federal program officers, and policymakers within the executive and legislative branches. This issue is critically important to those who care about disadvantaged youth, for the more they can maximize their effectiveness in using the resources of the Federal government, the more likely it is that there are going to be better outcomes for the youth who need help the most.

The Task Force has developed a series of recommendations to improve this current system. Ideally, a searchable database of all past, present and future grant activity would be created, which could then become the comprehensive database on discretionary grant spending in the United States. This will require a series of steps, which should include the following:

Modernize the Catalog of Federal Domestic Assistance

The Task Force recommends that the Catalog of Federal Domestic Assistance format be revised and updated to fit the way grants are currently administered. This would enable potential grantees as well as policymakers to better track grant opportunities and related activities within youth-serving agencies.

The Catalog of Federal Domestic Assistance Programs (CFDA) is a government-wide compendium of all 1,499 Federal domestic assistance programs. [38] This number includes the youth grants described in our April, 2003 preliminary report. The CFDA is maintained by the General Services Administration (GSA), while OMB serves as an intermediary agency between the Federal agencies and GSA. [39] The Catalog is published in hard copy and available in an online searchable database maintained by GSA.

The Catalog is considered the basic reference source of Federal programs. It is intended to improve coordination and communication between the Federal government and state and local governments. [40] Even with the e-grants initiative website (located at www.Grants.gov), the Catalog continues to be utilized. Thus, it plays a critically important role in helping all potential grantees, including state and local governments, as well as nonprofits, universities, hospitals, faith-based and community groups, to identify Federal programs which meet specific objectives.

Unfortunately, the CFDA system does not operate as well as it could today. The structure of the Catalog is partly out of date. While the system remains sufficient for regular, large block grants, the integrity of the system falls apart particularly under the weight of the hundreds of discretionary programs.

Here is why: the Catalog is designed using a five digit numeric system, where each agency has its own first two digits, and then all the programs they operate are assigned the remaining three digits. For example, HHS has all the grants from 93.001 to 93.999, Education has all the grants from 84.001 to 84.999, and so on. The problem with this system is that in many cases the departments have, over time, turned to an alphanumeric system in order to administrate their grants. There is nothing wrong with that, except the Catalog itself does not use the same numbers, and thus the numbering system has become obsolete. For example, the HHS Health Resources and Services Administration (HRSA) uses an alphanumeric system that can be problematic for those who are relying on the Catalog. Every year under their program entitled "93.110 Maternal and Child Health Federal Consolidated Programs" (popularly known as the SPRANS grants), HRSA offers multiple grant announcements under this single program, sometimes as many as several dozen. Thus, for example, the "Bright Futures Pediatric Implementation Cooperative Agreement Grant" used an alphanumeric marker, in this case, 93.110BI. However, that particular grant announcement name is not specifically mentioned anywhere in the Catalog's description for 93.110, and the number "93.110BI" does not exist in the Catalog. This situation is repeated numerous times for several different agencies. This is particularly important because many grant programs offer numerous grant announcements which are issued in a single year. Thus, this problem places constant barriers in front of grant seekers.

This happens every year with dozens of grant programs, most particularly in HHS and Education, which handle large numbers of grants. This can cause problems. Suppose a potential grantee or a policymaker has heard something about the "Bright Futures" program but does not know in which agency it is located. If they did a search on "Bright Futures" in the CFDA online, they would come up empty-handed. Likewise, a similar search at www.grants.gov would also yield no information. They would find that information if they knew enough to search at the grants information page on the HRSA website, but that would require a level of knowledge about the Federal grants process and far corners of the bureaucracy that the grant seeker might not have, and it certainly does not remove all the barriers to the grants process that can and should be removed.

In order for it to function well, the Catalog system also needs the uniform and constant cooperation of the grant making agencies, but currently participation is spotty. Some agencies have decided to mostly ignore the CFDA system and to collapse almost all their grants into a single CFDA number. However, this information is not reflected in the CFDA description for that program number. Some agencies run literally dozens of grant programs through a single CFDA number, but by doing so they make it impossible for those who rely on the CFDA system to have an accurate picture of the grants for which they can apply in a location that is central for all Federal agencies.

The CFDA remains a necessary tool for grant seekers and policymakers because, unlike other tools available, it provides the only ongoing directory of all grants, year round. Other sites list grants only during the application period, or for a limited time. In contrast. the CFDA allows grantees the often vitally useful time needed to plan ahead to respond when the grant announcement is issued by making the grant program information always available. Even though the grant announcement provides additional details that the CFDA might not offer, at least the basic information about that grant program is always available. In addition, while the individual departments and the components have by and large provided readily available grants data at their individual agency's websites, there is still the need for the Catalog to provide a central system with a uniform numbering system in a central location. [41] Finally, the Catalog also provides a greater picture of the activities of the myriad agencies in the Federal government.

Create a Centrally Available and Improved Grants Database

The Federal Assistance Awards Data System (FAADS) and the Consolidated Federal Funds Report (CFFR). Both contain very useful information concerning current and past grant activity that can orient a potential grantee as to what the government has been funding, in what amounts, and to whom the funds went. This information would enable a potential grantee to evaluate its competitive position vis-à-vis others. Unfortunately, FAADS is not fully searchable online and is often difficult to interpret, even for experienced users. [42] In addition, FAADS does not take advantage of all it could offer. It does not provide uniform coding of grant recipients. Federal Award Identifier Numbers (FAIN) that are unique to each agency are provided instead.

Improve the FAADS Database

The Task Force recommends the following ways to improve FAADS:

Include the EIN (employer identification number) as well as the DUNS [43] (Dun & Bradstreet, Inc., number) of each grant recipient. Both are unique identifiers and are required by statute and OMB policy to be submitted in applications for funding.

Provide the key to the Federal Award Identifier Numbers provided by each agency so that the code can be understood by all.

Convert the FAADS database from its current "flat-file" architecture to a modern, relational structure, which will then facilitate ad hoc query capability.

This inclusion of these three features would allow all users to identify specific grantees and determine which agencies and programs provided them with funds, and for what purposes. Users would also be able to look within specific agencies and determine all the groups to whom they gave out their grants and for what purposes.

The most important features of the new grants database would be:

A sophisticated search engine that enables text searches across all of the five levels and according to specific criteria about administrative agency, performer, and award types.

Criteria broken down into useful sub-criteria such as budget function, type of recipient, location of recipient, place of performance, etc. Most specifically, the place of performance must refer not to the headquarters of the grantee, but to the actual geographic locations where the services will be offered, eventually including zip codes (see next recommendation).

Inclusion of all five administrative levels of government: from departments, offices, and programs, to individual projects and awards.

These features would provide a user with a realistic snapshot of government grant activity in a given fiscal year. The database would provide the user with the actual structuring of all grant expenditures under a given theme across all levels of government (vertically-from an appropriation bill to an individual award, and horizontally-across all agencies of the Federal government that administer a given theme-based activity). It does not require the user to know the appropriate agency, office, CFDA category or budget code to retrieve data. Such a grant search system would enable effective, tailored queries about specific issue areas such as disadvantaged youth, because the system would text search across all possible dimensions of government activity and include detailed descriptions of particular awards that have been administered in a given period.

As soon as the significant and useful grants.gov initiative becomes fully operable, the last obstacle to creating a government-wide grant database system will disappear. This initiative aims to centralize grant reporting and administration across all agencies-the most important step in creating the database described in a previous section. With centralized FAIN coding, a user will be able to search to the lowest level of detail. A detailed search on "Bright Futures" or a broad issue search on "disadvantaged youth" has the maximum chance of successful retrieval of relevant past, present and future grant opportunities.

Create a Resource Mapping Function for the Database

The Task Force recommends that grantees of all Federal youth-serving programs be required to provide the zip codes or Geographic Information System (GIS) codes for all areas where they are providing services.

Federal agencies have a strong need to know where the funds from all their programs, as well as similar or compatible programs in other departments, are being spent and for what purposes. This would involve requiring all youth-serving grantees from all agencies to report in an identical manner as to exactly where they are spending the Federal funds they received. One simple way to accomplish this might be through the requirement that all grantees report where they are offering services by zip code. We note that we are not referring to the zip code of the headquarters of the grantee (which is data that is already available), but rather the zip code(s) where exactly the services are being delivered. This requirement could be waived for those few grantees who are offering services statewide or nationwide, as long as the database makes that clear to the user.

This resource-mapping process, done across all agencies, would be enormously useful to policymakers by allowing them to identify those communities where there is a dearth of needed services, as well as those locations that might be receiving overlapping funding for identical services. Nonprofit service providers who wish to apply for funds could use such a database to inform themselves about the services already available or lacking in their target area. This would help them design a better program of services, and would help those who decide which applicants receive the grant funds where the best places would be for the use of limited funds. The foundation community would find this useful because it would show them where they could help leverage their funds alongside or in place of government funds. State and local governments would have an easier way to determine where discretionary Federal funds are being spent within their area, which would improve their funding decision-making process as well. This last benefit is significant since the majority of funds coming from the Federal government to help disadvantaged youth go directly to the states via block or formula grants. The states then make funding decisions within their own jurisdictions.

Research Eligibility of Faith-Based Grant Applicants

For our April preliminary report, the Task Force developed the Federal Youth Programs Survey, which was designed to identify all Federal programs affecting youth ages 5 to 17. [44] In the survey, agency program managers were asked to identify which types of organizations were eligible to apply for these youth grants. The possible responses to this question were and the frequency of their responses were reported as shown in Figure 1.

Surprisingly, faith-based organizations were reported to be eligible to apply for only approximately half the number of programs for which all nonprofits were eligible. Normally, faith-based organizations should be eligible to apply for the same grant programs for which other nonprofit groups can apply.

Figure 1: Eligible Grantees for Youth-Serving Programs

Type of Organization	Number of Programs
Nonprofit organization	201
State government agencies	189
Local government agencies	157
Tribal organizations	155
Institutions of higher education	138
Territories	133
Faith-based organizations	106
Local education agencies (LEAs)	104
State education agencies (SEAs)	97
Individuals	31

Source: White House Task Force for Disadvantaged Youth,
Federal Youth Programs Survey, Fiscal Year 2002 programs.

On January 29, 2001, President Bush issued Executive Order 13198, which established a national effort designed to empower and expand the ability of faith-based entities to deliver social services in order to better meet the social needs in their communities. The initiative was intended to guarantee that faith-based social service providers be able to compete for Federal grants on a level playing field with other organizations. At the same time, the President also ordered the creation of Centers for Faith-Based and Community Initiatives in five Federal departments: Justice, Labor, Education, Health & Human Services, and Housing & Urban Development. A subsequent Executive Order issued December 12, 2002, added offices in the Department of Agriculture and in the Agency for International Development. Each of these seven department offices, working closely with the White House Office of Faith-Based and Community Initiatives, is charged with eliminating regulatory, contracting, and other programmatic obstacles to the participation of faith-based and other community organizations in the provision of social services. They are to identify barriers and propose initiatives to remove those barriers. They are also to propose demonstration programs with the aim of coordinating a comprehensive department effort to incorporate faith-based groups in department programs and initiatives "to the greatest extent possible."

In a second Executive Order also issued on December 12, 2002, the President reaffirmed his commitment to distributing Federal grants in the "most effective and efficient manner" possible so that no religious organization that is otherwise qualified to deliver social services be discriminated against in the awarding of Federal grants and programs. [45]

With this in mind, the Task Force believes that the Department Centers for Faith-Based and Community Initiatives are in the best position to evaluate the information provided by the agency program officers and to determine what steps need to be taken to follow-up.

Chapter 3:
Better Accountability: Producing Results, Not Just Promises

Government likes to begin things-to declare grand new programs and causes. But good beginnings are not the measure of success. What matters in the end is completion. Performance. Results. Not just making promises, but making good on promises. [46]

President George W. Bush

The public policy world of youth programs suffers from a credibility gap. While the will exists among the public to help young people address the many difficult problems they face, there is a lack of consensus as to how to do it successfully. Unfortunately, the Federal government has been ineffective in closing that gap in the public's mind. [47]

Through the last four decades, there has been growing Federal involvement and a rapidly increasing infusion of funds to address numerous problems of youth, including substance abuse, violence, teen pregnancy, hunger and nutrition, school failure, and workforce preparation. In Fiscal Year 2003 alone, the Federal government will spend $223.5 billion to help needy children and their families, focusing on these and related issues (see Appendix D). State and local governments and private groups will contribute billions more.

As the President has noted, the focus needs to be on achieving results. Part of the responsibility for this lack of focus on results lies in the fact that the Federal government has often made funding decisions without clear evidence that what it is attempting to do will actually work. Thus, public faith in the efficacy of social programs to successfully address youth failure has eroded. They wonder, what really works? How can we know? This is important because, as Gary Walker of Public/Private Ventures has noted, "even the most perfect solution, if there were such a creature, needs to be recognized and believed in, in order to be adopted as durable policy." [48]

This section of our report includes recommendations for how the Federal government can strengthen its role in research and evaluation of youth programs. With these recommendations, we insist that, from this point on, the Federal government commit to a coherent and comprehensive plan designed to identify with confidence and adopt those practices that will successfully help youth.

One of the most significant roles the Federal government plays in the social services arena is in supporting research and program evaluation. These, along with performance measurement, are central to any effort designed to strengthen and improve Federally funded youth programs, while infusing a culture of accountability and responsibility throughout the administration and management of critical Federal investments.

In general, evaluations estimate the impacts of programs by comparing the difference between the outcomes for individuals receiving a service or participating in a program to the outcomes for similar individuals who do not. In contrast, performance measures capture ongoing progress towards meeting program objectives. Ongoing (at least annual) program performance assessments should be supplemented with more in-depth, rigorous evaluation studies that measure the particular impact of a youth serving program and can attribute the impact to the program.

While evaluation and performance measurement both include data collection and measurement of progress, evaluations use scientifically-based research methods to systematically investigate the effectiveness of programs by comparing the observed program outcomes with what would have happened in the absence of the program. Together, high-quality evaluation and performance measurement can provide Federal agencies that oversee youth-serving programs with the tools to determine whether or not the intended intervention worked (evaluation) and whether the intervention is achieving its desired objectives on an ongoing basis (performance measurement) .

The focus of the Task Force here was in two areas. First, we had broad consensus on the need to improve the Federal role in helping to understand "what works." Second, we also recognized that, with a Federal investment in youth-serving programs of hundreds of billions of dollars annually, we needed to firmly hold programs accountable for achieving the goals and objectives they have set out for themselves. The Task Force recommendations reflect those two beliefs.

UNDERSTANDING WHAT WORKS

The Task Force has developed several recommendations aimed at improving the quality of the information we have about what works to improve youth outcomes. The first recommendation (in several parts) addresses our concern that we create a more consistent set of guidelines for assessing the quality of program evaluations across agencies. The second puts forth a suggested road map for guiding the direction of future Federally supported research on youth programs, and the third offers a suggestion on improving national survey data collection.

Develop a Unified Protocol for Federal "What Works" Clearinghouses

The Task Force recommends that a committee of the relevant Federal agencies develop a consistent approach to the assessment of youth program and policy evaluations, including the development of protocols. Random assignment experiments are considered the "gold standard" of evaluation because they can most clearly attribute outcomes to interventions. The Task Force strongly recognizes this gold standard and believes those evaluations should be given the greatest weight in shaping what we know about what works. Because individual agencies have different needs, the protocols need not be identical, just sufficiently consistent so that materials and findings can be shared among agencies with relative ease.

Many Federal agencies have developed research-based efforts to identify youth programs that "work," broadly called "'What Works' Clearinghouses." For example, the Department of Education maintains the "What Works Clearinghouse," [49] the Department of Justice the "Blueprints for Violence Prevention" program, [50] the Substance Abuse and Mental Health Services Administration (SAMHSA) has the "National Registry of Effective Programs" (NREP) [51] and the CDC supports the "Community Guide to Preventive Services" [52] and the Prevention Evaluation Research Registry for Youth (PERRY) [53]

Whatever the name of such efforts, the idea is the same: social science findings should guide government decisions about which programs to support and at what funding levels, the content of technical assistance, and additional research that is needed. In fact, making government decisions more evidence-based should be a major priority in this area. Relatively few youth programs supported by the Federal government meet this test. [54]

Unfortunately for most "what works" efforts, many believe there is not a sufficiently large body of research (or evaluation) that has established the effectiveness of a relatively broad set of youth programs (or approaches) upon which important program choices can be based. Some studies of youth programs are not of sufficient rigor, and many of those that are have limited applicability that cannot be the sole bases of broad policy planning. Broad generalizations going beyond the findings on a particular program are unwarranted, but revision and redirection within that program may be supported by a strong study. The problem is that too few promising programs have been independently evaluated using rigorous designs.

Jodie Roth and her colleagues describe the limitations of this research: "The review of the evaluation literature highlights the paucity of high quality outcome evaluations of programs fitting the youth development framework. As noted previously, little improvement in the state of program evaluation has occurred since the 1992 Carnegie Report... Nationally, strong interest in expanding adolescents' access to youth development programs exists. However, the current mismatch between the enthusiasm for these programmatic efforts and the empirical evidence calls into question the effectiveness of such efforts." [55] Rob Hollister adds, "what do we know about what works-our answer has to be: not much." [56]

As a result, the Task Force believes we must commit to increasing the number of high quality evaluation studies that employ random assignment, the most reliable evaluation method available. One way to encourage more attention to the most credible evidence is to provide guidance on criteria that determines what is "most credible" and why some studies do not rank as high-quality. The criteria need to be used in agency assessments of evidence concerning particular programs. The assessments should have fully and clearly articulated criteria, and the overall review process should be transparent and open so that others may understand judgments made in the assessment. The results of these efforts must be provided in a format that is accessible to program staff and policy makers.

Federal agencies are beginning to develop such "what works" efforts. Reflecting their recent origins, most of these "what works" efforts are still in their formative stages, with individual agencies now grappling with how best to proceed. Reflecting the fact that they have been developed largely from within specific Federal agencies, they often lack common terminology and methodologies. To some extent, of course, such diversity is needed to reflect each agency's specific needs, disciplinary framework, and statutory and programmatic context. Nevertheless, greater commonality would facilitate the sharing of information among Federal agencies and with the public, and would allow Federal agencies to build on each other's efforts.

Below, we outline two steps in establishing a more unified Federal protocol: (1) identifying what types of programs and evaluations should be included in Federal "what works" clearinghouses (2) developing the process for assessing evaluations.

Step 1. Agency protocols should identify the types of programs whose evaluations should be assessed and specify the evaluation methodologies that are acceptable.

A first order question is what kinds of evaluations or research studies should be included in a "what works clearinghouse." The short answer is: all the evaluations that would give a full picture of what is known about the effectiveness of youth programs. That means evaluations with sufficient scientific rigor, whether or not they show that a program or policy "works" or "does not work."

Each agency should develop a system for assessing the evaluations of the programs and policies under its jurisdiction and establishing a priority-setting system for deciding which evaluations to assess. For programs or policies that cross agencies, joint or coordinated efforts should be considered.

Below we provide an overview of the types of evaluation methodologies that might be included in a protocol.

Random assignment experiments. Unfortunately, many studies that seek to evaluate the effectiveness of youth programs do not have causal validity, that is, their design does not support causal inferences.

Many social welfare programs look successful-to their own staffs as well as to outsiders-because their clients seem to be doing so well. A substantial proportion of trainees, for example, may have found jobs after having gone through a particular program. But did they get their jobs because of the program, or would they have gotten them anyway?

In actuality, any number of factors, however, could have caused the observed result: The economy may have improved, making more jobs available (and perhaps increasing the demand for low-skilled workers); the participants may have been especially amenable to help (or not amenable); or they may have gotten their jobs because of the passage of time (and perhaps the normal process of maturation). Determining what would have happened in the absence of the program or policy is the central task of impact evaluation. To do so, researchers try to establish the "counterfactual," that is, they try to see what happened to a similar group that was not subject to the program or policy.

Most social scientists believe that experimental designs are the best way to measure a program or policy's impact-because they can have high causal validity. Classical experimental design typically involves random assignment of individuals, families, or other units of analysis to a specific condition, i.e., either (1) a "program group," [57] whose members can take part in the program or are subject to the policy, or (2) a "control group," [58] whose members do not. The experience of the control group, thus, is meant to represent what would have happened absent the program or policy, that is, the counterfactual.

If properly conducted, and with sample of sufficient size, random assignment should result in statistically comparable program and control groups, that is, groups whose aggregate characteristics (measurable and immeasurable) are comparable (within the limits of chance variation). This similarity means that the two groups are likely to be exposed to the same outside forces and to respond to those forces in similar ways, so that any subsequent differences in average outcomes can be attributed to the program or policy-to a known degree of statistical precision. This ability to rule out other causes gives randomized experiments a high degree of causal validity. [59]

Because experimental designs ordinarily do not require complex statistical adjustments to eliminate differences between program and control groups, they gain a credibility (and accessibility) that gives their results substantial influence over policy. Policymakers can then focus on the implications of findings, rather than "become entangled in a protracted and often inconclusive scientific debate about whether the findings of a particular study are statistically valid." [60] For example, the evaluations of welfare-to-work programs conducted by Manpower Demonstration Research Corporation (MDRC) in the 1980s-which used experimental designs-are widely credited with having shaped the Family Support Act of 1988. [61] So, too, in the 1990s, for the Abt Associates evaluation of the Job Training Partnership Act (JTPA) program. [62] As a result, experimental designs are increasingly used to evaluate a wide range of social programs and policies. [63]

The superior causal validity of randomized experiments has led most experts in program evaluation to call them the "gold standard" of evaluation. The Task Force strongly recognizes this gold standard and believes those evaluations should be given the greatest weight in shaping what we know about what works.

Of course, the credibility of a particular randomized experiment depends on its being well implemented. Hence, the Task Force recommends that protocols examine (1) the quality of the randomization to ascertain the credibility of the impact estimates and (2) the evaluation's generalizability to other circumstances.

Nonexperimental evaluations. Another kind of evaluation is called "nonexperimental," and sometimes "quasi-experiment." [64] In non-experimental evaluations, the counterfactual is established by identifying a "comparison" [65] group (for example, persons not participating in the program or from another site, another time, or a data set) whose members are not subject to the program or policy but are nevertheless thought to be similar to those in the program group.

The most common nonexperimental designs compare program participants before and after a program or policy change (pre-post comparison) or program participants to nonparticipants, to individuals from other geographic areas (comparison sites), [66] to individuals from different time periods, and to individuals drawn from secondary data sets. Aggregate data is often used to compare changes in outcomes over time or across geographic areas.

The major disadvantage of nonexperimental evaluations is that they have uncertain causal validity, at best. Put simply, the members of the comparison groups may differ substantially in some unmeasured or undetectable ways from those who have been exposed to the particular program or policy. Typically, nonexperimental designs employ statistical analyses to control for such differences, but how well they do so is open to debate. As Rob Hollister cautions: "Without random assignment there is always the chance that there will be a concentration within the program participant group of those with characteristics that affect the outcome (e.g., the program participants may be more motivated than those who are in the comparison group). To the extent that those characteristics are measured it is possible to control for their effects with statistical models. It is the unmeasured, or unmeasurable, characteristics (like motivation) which create the bias problem."(italics provided by the author) [67]

Researchers deal with selection bias through careful regression analysis that statistically controls for a variety of background variables. Examples of common background characteristics include age, sex, income, education, and family status. But even in the best circumstances, such methods cannot capture all of the differences between the two groups, because no data set has information on all the characteristics that may affect the outcomes being examined. All studies are missing some variables of interest. As a result, there is always some uncertainty regarding the causal validity of nonexperimental evaluations.

Selection is an especially difficult problem in the evaluation of youth programs. The decision a young person makes to participate is voluntary. So by definition, only the motivated apply, and that motivation cannot be reliably modeled by nonexperimental techniques. This voluntary selection by the participant is further complicated by selection of participants by program operators creating a process that cannot be reliably duplicated at this time using nonexperimental methods.

Consequently, some literature reviews, meta-analyses, and "what works" efforts completely exclude nonexperimental evaluations from their assessments. [68] This is problematic for several reasons. First, there are relatively few experimental evaluations of youth programs, [69] so that relying solely on them would provide less information about promising programs or approaches. In addition, there are cases in which studies labeled as randomized experiments often, on closer examination, turn out to be something less. [70] Second, because of ethical issues, randomized experiments ordinarily cannot be used to evaluate full-coverage programs, while nonexperimental ones can. Randomized experiments that involve denying a service to someone who is otherwise entitled to it cannot be conducted. [71] Third, when coupled with randomized experiments, nonexperimental evaluations can provide supportive and enriching information about the program or policy being evaluated.

Therefore, the protocol should explicitly address whether nonexperimental evaluations will be assessed-and under what conditions and with what limitations.

Programs or policies that "do not work." Some literature reviews and "what works" efforts appear to include only those programs or policies that seem to work, and they exclude evaluations that show no statistically significant effects, small effects, negative effects, or effects that do not last.

For example, one assessment of the evidence on the impact of youth development programming included only those evaluations that showed "evidence of behavioral outcomes." [72] Rob Hollister laments that "this means that well designed evaluations that found no statistically significant impact were not reported. I believe the exclusion of evaluations where there was no statistically significant impact was a mistake, as it is important for us to learn what doesn't work as well as what does work." [73] The Task Force agrees.

For example, if there were ten studies of a particular approach to youth training and only one study found that the program "worked" (based on a 10 percent level of statistical significance), including only that finding in a review of "what works" would give the approach too much credibility because just by chance, one in ten studies are likely to show significant effects. [74] In other words, it is important to know whether a particular program or policy that has been tested "worked" in ten studies out of ten, or just one out of ten. Including just those studies with positive effects obscures this point.

Even evaluations that find no effect or a negative effect can offer important lessons. Was the program poorly implemented? Did control/comparison group members have easy access to similar services? Were there defects in the evaluation that could affect the findings? Were there effects for some subgroups, even if there are no statistically significant effects overall? Answers to these questions can lead to refinements and further testing. Before ruling out a particular approach, it is also important to determine whether similar findings have come up in replications. If so, the research can point to program approaches or policies that should not be replicated. This information can be critically important to policymakers and practitioners, as they examine ways to improve their policies.

Meta-analyses. Another means of assessing the effects of a program or policy is a meta-analysis. A meta-analysis is a statistical procedure for combining the results from individual studies, even those with conflicting findings and different evaluation approaches, into a single study with an integrated set of findings. [75] It can lead to stronger findings of effects, because it often combines evaluations with small samples and thereby increases the statistical power of the analysis.

Meta-analysis is frequently used in medical research, where interventions and outcome variables tend to be clearly defined, but where clinical trials often involve relatively small samples. As a result, some successful treatments may not appear to be effective, because their evaluation samples were too small to detect anything but the biggest impacts. (This is known as a Type II error, erroneously accepting a finding of no effect.) Even if they have statistically significant effects, they may have such wide confidence intervals, that the possible range of effectiveness is very large and different studies may have very different findings. The added statistical power that comes with a meta-analysis can transform a series of evaluations with no statistically significant effects into an overall finding with a statistically significant effect and, because the confidence intervals become smaller, the overall result trend tends to look more precise.

A meta-analysis involves several steps. First, the purpose of the analysis and the questions to be addressed are determined. Second, the evaluations that address the purpose are identified. Third, the data from each evaluation are collected and coded. This includes information on the outcomes to be examined, as well as the characteristics of the evaluations and programs themselves. Fourth, the outcomes are transformed into a common metric-an effect size-so that they can be compared across evaluations. (An effect size is the standardized difference between program and control/comparison group mean outcomes.) Finally, the data are statistically combined to determine overall program effects.

Second, the programs in a meta-analysis often have different design features, often are aimed at different target groups, and often are carried out in different environments, making it more difficult to discern which aspects of the programs studied are responsible for their effects. This may be particularly true with most social programs, where the implementation, services provided, and other program characteristics can vary tremendously from program to program. Proponents of the meta-analysis approach argue that the analysis can take these differences into account by including them in the statistical model, but this adds another layer of uncertainty and subjectivity to the process.

Despite these shortcomings, meta-analysis is very important tool when done carefully. As described above, randomized trials seldom are strictly statistically representative, and thus large numbers of trials are necessary to achieve reasonably generalizable results. Meta-analysis is an important method for synthesizing trials, because, if done with care and transparency, it represents a more objective way of summarizing studies than more qualitative approaches.

Step 2. Agency protocols should establish a formal process of evaluation that specifies the criteria for assessment and the levels of evidence available. The process should have written guidelines and data collection instruments, and it should be open, transparent and subject to outside review.

Evaluations can go wrong in many ways. Some have such obvious faults that almost anyone can detect them. Other flaws can be detected-and properly assessed-only by experts with long experience and high levels of judgment. Hence, the proposed protocol recommends an intensive inquiry into the quality of the evaluation.

Criteria for assessments. In recent years, various evaluations have sought to determine the effectiveness of particular youth programs. Many of these evaluations provide important information about the impact of such programs, but most also have serious flaws that sharply limit their usefulness. Hence, the proper use of these evaluations requires distinguishing relevant and valid findings from those that are not.

Whether an evaluation uses an experimental or nonexperimental design, a host of questions must be answered before deciding that its findings should be accepted. This inquiry should be based on the generally accepted criteria for judging evaluations. [76] See Appendix A for suggested criteria.

Specified levels of evidence. Given the wide range in the quality of evaluations, and the limited number in many important areas of youth policy, the assessment process needs to distinguish among levels of evidence and, based on the level of evidence, indicate the appropriate use of the findings. The categories might be, for example, "strongly supported by research," "somewhat supported by the research," "no research on the subject," "somewhat negated by the research," and "strongly negated by the research." There might also be additional categories like "supported by theory and available data" and "negated by theory and available data."

A formal process. The assessment process needs to be institutionalized, with standard procedures to encourage unbiased treatment. This includes clear guidelines about collecting and interpreting data concerning the research, ordinarily pursuant to written instruments. (To the extent feasible, the information or documents to be collected should include the evaluation's sampling plan, data collection plan, evaluation plan, all interim and final reports (including appendices), and any other publications by the evaluators themselves or by other commentators writing about the evaluation.) There should also be provision for the systematic sharing of information among agencies.

The Department of Education's "What Works" Clearinghouse, for example, follows a standardized process. Each year, it selects general "topic areas," or categories that it would like to evaluate during the coming year (e.g., "Programs for Increasing Adult Literacy" or "Curriculum-Based Interventions for Increasing K-12 Math Achievement"). [77] It then accepts nominations of interventions and evaluations to be reviewed that fall under the topic areas. The Evidence Report Team then evaluates each intervention using the Study Design and Implementation Assessment Device (Study DIAD) and then the Cumulative Research Assessment Device (CREAD), and writes an Evidence Report synthesizing this information. [78]

Open and transparent. The assessment process needs to be open and transparent to outsiders, with the presentation of detailed data about the research and the reasons for the assessment given.

Thus, the National Institutes of Health's (NIH) Consensus Development Conference (CDC) Program convenes conferences to discuss "controversial issues in medicine important to health care providers, patients, and the general public." [79] These conferences include a two-day session, open to the public, with presentations of research by scientific experts and a period of questions and comments by the attendees. The end result of the conference is a "consensus statement that advances understanding of the technology or issue in question... and that will be useful to health professionals and the public." [80] The results of this report are released to the public during a press conference, and are then made available in web and print versions.

Outside review. The process should be subject to review by outside experts, post-publication debate, and revision. The principal investigator of any evaluation assessed should have the opportunity to submit materials or comments.

As part of NIH's Consensus Development Conference, the information presented by scientific experts is reviewed and evaluated by a panel of between nine and sixteen members from outside NIH, ranging from other scientists in the field to health professionals. [81] The panel then drafts the consensus statement based on the research presented, presents it to conference attendees for comment, and revises the statement prior to release of the findings. Final panel revisions continue following the conference, and the statement is published by NIH's Office of Medical Applications of Research (OMAR) and often by a medical journal. [82]

Build a Rigorous and Unified Disadvantaged Youth Research Agenda

Although some programs for disadvantaged youth have been evaluated, the evaluations have been conducted quite independently with little attempt to draw more general lessons about what makes a program effective or not. As mentioned earlier, many programs have not been the subjects of high quality evaluations, with too many evaluations being under-funded, poorly designed and rarely based on random assignment. A comprehensive and unified research agenda that can yield credible knowledge about how to improve outcomes for disadvantaged youth must begin with a thorough and systematic assessment of the programs that have been evaluated with random assignment

The following describes a series of actions that should be taken to pursue the goal of a unified Federal research agenda:

Create a comprehensive catalog of high quality randomized trials that have been conducted for disadvantaged youth. In order for the catalog to be comprehensive, not only programs like Job Corps whose sole target population is disadvantaged youth should be included, but also many other programs where disadvantaged youth have been a large enough sub-population to analyze separately. For example, David Olds' Nurse Home Visiting program in Elmira had many large significant effects, but only for young, never married low-income mothers, i.e., disadvantaged youth [83]. In addition, determine what trials are currently underway to avoid duplicating their efforts.

Catalog what systematic reviews have already been conducted to synthesize subsets of these trials. Systematic reviews should be based on a well-specified protocol that includes a thorough and comprehensive search for all randomized trials, clear standards for inclusion and exclusion of studies, and transparent methods for coding and synthesizing the included studies. Where possible, meta-analysis or other formal, quantitative methods should be used to conduct the synthesis, but where this is not possible, transparent and objective narrative methods should be employed.

Determine what further systematic reviews need to be conducted in order to maximize the knowledge we can extract from existing randomized trials. After conducting these reviews, a comprehensive effort should be made to identify gaps in what is known. In addition, hypotheses should be generated not only through examining the systematic reviews of experimental findings, but also by looking at other research and theory. For example, one could examine the success/failure of more targeted versus more comprehensive interventions, or whether programs that create relationships between disadvantaged youth and significant adults (through formal mentors or otherwise) are likely to be effective.

Use the above results and other information to formulate key questions to be addressed in further experiments. These experiments could include replication of approaches that the evidence suggested were effective, program and research designs specifically formulated to answer unanswered questions, or other approaches that could shed light on broad issues of what makes a program for disadvantaged youth effective.

As part of these experiments, an effort should be made to develop a reasonably uniform set of research sample characteristic measurements and outcome measures that could be used across the experiments. These would augment measures that were needed to assess the particular intervention, e.g., in a drug treatment program there would presumably be more intensive measurement of participation in the treatment and assessment of subsequent substance abuse. It would be important to develop a broad set of measures given that past experiments suggest that critical impacts may occur in areas not directly related to the treatment. Both strengths and risk factors should be measured.

Develop a set of strategies to engage program partners in conducting the experiments. These could range from tying Federal program funds to cooperation with experimental evaluation to setting requirements for experimental evaluation into statute and providing funding for them as part of program authorizations. The Chaffee Independent Living evaluation would be one place to look [84].

Do all of the above in the context of a cross-agency team, all of whom are individuals committed to conducting rigorous evaluations, including experimental evaluations.

Improve Data Collected on the Well-being of Families

In the past several years, national survey data and research have reinforced the now popular adage that "parents matter." Children who are raised in two-parent families are at less risk for poor health and economic outcomes. Beyond family structure, nurturing parent-child relationships have been shown to be tremendously protective against adolescent risk-taking behaviors. While we are increasing our understanding of the roles parents play in their children's lives, we have limited nationally representative data documenting family structure and functioning.

National indicator data is important if we are to assess the well-being of our country's population. Indicators distill large amounts of information and data into usable statistical values that, taken together, provide information on the population of interest. As a result, indicators are a useful way to track problems, identify how populations are changing and affected and provide direction with agenda-setting. This is true with all kinds of indicators, economic as well as social. We gather a number of economic indicators, but we have few indicators that monitor the well-being of our families.

In following the development of a family, we first should seek to improve the data on couple formation. In December 2001, the Federal Interagency Forum on Child and Family Statistics sponsored a conference entitled Counting Couples to explore improving the measurement and collection of information about couples. This conference resulted in numerous recommendations for data on marriage, divorce, remarriage and cohabitation. Several of these recommendations are being pursued by divisions within HHS. Currently, the HHS Assistant Secretary for Planning and Evaluation (ASPE), the Administration for Children and Families and the National Center for Health Statistics have a joint effort to analyze alternatives for developing a complete and systematic approach to generating data on marriage and divorce.

Having documented national rates of marriage, the next step is to better understand the internal and external functioning of families once they have been formed. ASPE is supporting work to improve indicators of family connection. This project is expanding current work in the area of indicator development to consider measures such as family living arrangements, as well as measures that provide greater social context such as religiosity, family functioning, community interaction and volunteerism. The result of this project will be a chart book of 25 data elements currently being collected on the family, as well as papers exploring options for new measures. Examples of papers are analyses of measures of family time and cultural variations in family connection.

The Task Force recommends that the Federal statistical agencies seek to explore ways to integrate new family measures (such as family structure, formation and functioning) in their national surveys.

HOLDING PROGRAMS ACCOUNTABLE FOR RESULTS

In this section we present recommendations for developing and implementing common youth program performance standards and measures. We suggest that these serve as a starting point for discussion and consensus-building among various stakeholders. This process, once completed, will allow policymakers to compare the outcomes of similar programs, no matter which agency they are in. It would also facilitate considerations of program consolidation, redirection of resources, and elimination of ineffective programs, where appropriate. In the interest of improving our ability to document the results of Federal investments, we also offer recommendations on addressing earmarks, and implementing the principles of the No Child Left Behind Act in Department of Defense schools.

Develop Standards for Measuring Grantee Performance

In developing examples of common GPRA-type performance measures, the Task Force focused on the following ten service areas:

Alcohol abuse
Drug abuse
Tobacco use
Violence/crime (e.g., juvenile delinquency and school violence)
Sexual risk behavior
Academic performance
Community service
Substance abuse treatment
Self-sufficiency skills
Mental health

The Task Force sought to identify key performance measures applicable to multiple Federal programs, including programs in different Federal agencies. This recommendation does not identify nationwide measures for use in assessing and tracking the overall condition of disadvantaged youth in the United States. However, many of the performance measures that appear relevant to comparing outcomes of individual programs are variations of measures appropriate for tracking nationwide status. In such instances, the national data can be used as benchmarks against which to compare the findings from individual programs. However, when national measures are considered, it is important to remember that questions about risk-taking behaviors that are asked anonymously in national surveys may be inappropriate in a program setting.

The President's Fiscal Year 2004 budget adopted a set of common performance measures for the major Federal job training programs. Seven programs identified the following measures for job training programs for youth and lifelong learning: (1) placement in employment or education, (2) attainment of a degree or certificate, (3) literacy and numeracy gains, and (4) program efficiency. With common terms and definitions across programs, states and local areas managing multiple programs will be able to use a simplified format and measurement system for reporting performance.

We provide recommendations for establishing a process to identify additional common performance measures, while achieving agreement among various Federal agencies on the resulting set of common measures.

It is also important to note that many central areas of concern for disadvantaged youth are not included in the areas addressed below. These include such areas as family financial well-being, physical health, parental guidance/family support, abuse, foster care and living/housing conditions and employment. Instead, our emphasis has been on outcome measures, but we have also included a small number of output measures, particularly "number of clients served," and efficiency measures (defined as the ratio of cost in dollars or employee-time to amount of output-or the much truer measure of efficiency, the ratio of cost to the amount of outcome). Outputs, however, it should be noted, provide little, if any, information on the results of the service.

Figure 2: Definitions Used in This Report

Input: Resources (expenditures or employee time) used to produce outputs and outcomes.

Output: Products and services delivered. Outputs are completed products of internal activity: the amount of work done within the agency or program or by its grantees or contractors (such as number of prevention classes or counseling sessions provided).

Outcome: An event, occurrence, or condition that is outside the activity or program itself and is of direct importance to program customers or the public generally. Service quality is also included under this category.

Intermediate Outcome: An outcome that is expected to lead to a desired end but is not an end in itself (such as service response time, which is of concern to the customer making a call or requesting a service, but does not tell anything directly about the success of the call or request). A program may have multiple intermediate outcomes.

End Outcome: The end result that is sought (such as reduced incidence of disease or substance abuse, or improved academic performance). A program may have more than one end outcome.

Outcome Indicator: An outcome indicator is a numerical measure of the amount or frequency of a particular outcome.

Performance Indicator: A quantifiable measurement for one aspect of performance (e.g., output or outcome) under consideration.

The next section discusses the potential major uses for performance measures. The following section identifies a number of problems and issues in selecting common measures. We then present suggestions for specific measures as a starting point for debate among the various stakeholders. The final section provides recommendations on a process for obtaining agreement on a common set of performance measures, including the identification of additional measures and long-run institutionalization of the process.

Potential Uses for Common Performance Measures

The overall purpose is to use the information obtained on the common set of performance measures to help improve the quality of life for disadvantaged children. More specifically, comparative data on the measures will hopefully help both program managers and Federal officials with broader responsibilities do the following:

Identify which programs are being successful in improving the lives of disadvantaged youth, encouraging those types of programs;

Provide information on service delivery characteristics associated with greater levels of success, or conversely associated with low success;

Identify specific youth problem areas that need special attention;

Identify less successful programs that warrant attention, such as program modifications, reduced funding, etc.;

Provide an improved basis for setting Federal targets for the measures;

Help make budget choices among programs. When costs can be related to outcomes, the relative cost effectiveness of programs can be better assessed. However, performance and cost data by themselves are seldom sufficient for making budget decisions. Depending on the circumstances, for example, the best decision on a program with unsuccessful outcomes might be to add funds. Similarly, the best decision on a program with highly successful outcomes might be to reduce its funding.

Identify the need for coordination and cooperation among certain programs and help encourage such cooperation;

Identify instances where consolidation of programs is appropriate;

Motivate individual programs and their staffs to improve, whether to catch up or stay ahead of other programs;

Provide more information to the public regarding progress in these youth programs.

Potential Limitations and Issues with Common Measures

The concept of regularly reporting data on a common set of measures that provides accurate and fair comparisons across programs that have similar objectives is very attractive. While the concept also appears at first to be easy to implement, many problems and issues exist that can make effective implementation quite difficult. Following are a variety of problems and issues, with suggestions for how to alleviate or address them:

The definition of a "program." Which "programs" should be asked to use the common measures? No formal definition exists to our knowledge. The set of programs identified in Appendix III of the Task Force's April, 2003 Preliminary Report could be used as a good a starting point.

Different programs will frequently have at least somewhat different client groups, making accurate comparisons difficult. Program clients may differ on such characteristics as age, gender, race/ethnicity, problem type and history. One possible way to alleviate this problem is to group such programs by client characteristics to allow more valid comparisons. Ask programs to provide breakouts by such characteristics, in addition to providing aggregate data.

The term "disadvantaged youth" will need to be defined more precisely, not an easy task.

Programs already collecting data on performance measures may be measuring the same basic outcome in different ways. Requiring these programs to change their measurements will cause a break in their time series for the data, as well as add cost to conducting the new measurement. The selection of common measures should not constrain programs from using whatever measures they find useful. The need is only to identify a small set of measures for which the definitions and data collection are reasonably standardized.

Some of the most important measures may require new data collection procedures. In particular, to assess the success of many programs seeking to help disadvantaged youth, these programs are likely to need to follow up on former clients at some period of time, such as 12 months later. Some precedent exists for such follow-ups, such as the Department of Labor's requirements for such follow-ups on youth and adult clients of job training programs and HHS requirement for post-service follow ups for some substance abuse treatment programs. (Obtaining such feedback should become a regular part of program management.)

The data to be collected often will need to come from state or local government agencies. This complicates implementation consider- ably. In these instances, state and local governments should be asked for their input, and, preferably, be included as a partner in the data collection, analysis and dissemination process.

Much of the data on youth outcomes are "national" in scope rather than being linked to specific Federal programs. These national data (even if broken out by state) provide very important national aggregates. However, they say little if anything about the effectiveness of individual Federal programs. Seek performance measures that provide feedback on the particular clients of individual programs. Where this is not practical, emphasize the joint ownership by many programs of the aggregate outcome measure. Encourage the notion that programs sharing the same outcome measures are partners in a performance partnership.

The tendency to want to separate one's own agency from results that do not look good, rather than considering common outcomes measures as a joint partnership among those agencies that contribute to the outcome. Outcome data uses that encourage constructive program improvement should be emphasized. Avoid assigning blame to programs or individuals, at least until explanations have been sought, including identification of contributory external factors, as well as internal factors.

Candidates for Common Core Measures

This section presents a number of suggested cross-cutting performance measures applicable to most, if not all, service areas. The first set of these measures addresses output and efficiency; the other addresses outcomes.

Output and Efficiency Measures

The following are a small number of suggested cross-cutting performance measures that measure outputs and efficiency but not outcomes: Such output and efficiency measures are useful but say nothing about the results of the services. They should not be used in place of outcome measures.

Number of clients served. This is an output measure. This measure is common to most, if not all, youth programs. How many youth did the program attempt to help?
Although this measure says nothing about the outcomes of the service, it has other important uses. The data provides perspective about the number of clients that individual programs are attempting to help. This measurement is also needed as the denominator for both efficiency measures and for outcome measures that are expressed as percent of clients with successful outcomes. This measure will likely be widely acceptable to programs providing services, since it is familiar and many, if not most, programs are likely to have data on the number of clients served.

If this measure and the following two measures are used, ground rules are desirable as to the minimum amount of service provided to an individual client before that client should be included in the measurement. (For example, should youth who came in the door of a program only to pick up a pamphlet, or who only called to learn the location of the program but never came in, or who only participated in one event, be included in the counts?) Another issue is determining whether the measurements should be of the number of different clients rather than the number of visits or services received.

The caution in the use of the measure "number of clients served" is that too much emphasis on it as a performance measure can lead programs to focus on quantity rather than quality of programs' service.
Cost per client served. This measures "efficiency." Efficiency measures are generally defined as the ratio of the amount of input used to achieve a particular amount of output. Because of the limitation (noted above) in the meaningfulness of the denominator of this ratio (number of clients served), this measure should be used cautiously.
Another caution: both of these first two measures can encourage programs to inflate the number of clients served in order to show up better on these two measures. This underscores the need to establish ground rules as to minimum amounts of services before the client is counted, as discussed above
Cost per client that improved. This is also an efficiency measure. However, this one provides a considerably better perspective on true efficiency. This measure, in fact, can also be called a "cost effectiveness" measure.
To implement this potentially highly desirable measure, however, the program needs to be able to measure in a reasonably reliable way both the number of clients that improved and the costs of producing that result. Measures that attempt to assess the number of clients improved are included in the next section on outcome measures.

Outcome Measures

Outcome measures can be categorized as to whether they are "intermediate" or "end" outcomes. End outcomes are related to the primary objectives of the Federal programs and are the most important. In effect, they indicate the extent to which the condition of disadvantaged clients has improved or reached a desired level within the constraints of the program. Intermediate outcomes measure the occurrence of events that are expected to lead to the end outcomes. For example, an HIV-prevention program for youth might show such intermediate outcomes as improvements in knowledge and attitudes. Such improvements are expected to lead to such end outcomes as being free of HIV/AIDS infection.

Note that for the following outcome measures, when a measure contains the word "youth," the intention is that the measure covers only, or at least primarily, disadvantaged youth.

For the disadvantaged youth programs, a few outcome measures apply to many programs and across most, if not all, service areas, and are not unique to a single service area. These are the following:

Percent of clients satisfied with the service provided. This is probably best considered an intermediate outcome measure. While the youth may be satisfied with the service they received, this does not indicate whether the youth's condition improved. This measure applies to programs that have specific identifiable clients.
If this measure is included, for comparison purposes, it is necessary that the programs to be compared use the same, or at least similar, wording in the satisfaction question and administer the survey at the same approximate time relative to the time the client left service. (For example, clients might be asked a month after they completed the service, in a representative survey of clients, "Overall, how would you rate the service you received: excellent, good, fair, or poor?" The common core measure might be the percent that responded "excellent" or "good.") The question wording might need some tailoring to particular programs.
Percent of clients who completed the program. This is a low-level intermediate outcome measure. This measure is applicable to programs that are structured to "enroll" youth for a specified time period or for a particular number of sessions-such as prevention or treatment programs. Completion of voluntary programs reflects the program's success in retaining clients through the end of the program, increasing the likelihood that clients will subsequently improve their condition. One advantage of this as an intermediate outcome measure is that it is likely to be perceived as non-controversial to most programs.

Percent of disadvantaged youth with a need for help that received help. This would appear to be a useful intermediate outcome measure. However, it may be particularly difficult to measure for some programs.

Percent of clients whose condition improved, or reached a pre-specified level. This is a generic measure. It applies to many, probably most, programs for disadvantaged youth
There are two measurements that may be used together or separately. Each of these versions can be measured at various points in time after service has been provided. Determining the variations to be used as common measures is a major and controversial decision. This is due to the difficulties in tracking the clients after they leave the service.

a.    Measure the after-service condition level of the client. This provides the performance measure "Percent of clients whose condition has reached a pre-specified level." The condition might be measured right at the end of the service or after a given interval of time after service, such as one year after completion of service. (The client's condition after a reasonable period of time after service has been completed is highly preferable, particularly for programs that seeks to improve the condition of its clients in a sustained way, and not only initially. However, obtaining this post-service data is more difficult.)

b.    Measure the amount of change from when the client came in for service until the follow-up time after service has been completed. This provides the performance measure "Percent of clients whose condition improved by at least a pre-specified amount from the time the client began receiving the program's services."

c.     Measure both of the above. Each of the above options measures something of importance that the other does not. For example, the client might have improved significantly but the client's final condition might not have been up to the target level. Conversely, the client may not have improved much but just enough to reach the targeted level.

Specific End Outcome Measures: Appendix B identifies candidate performance measures in selected service areas. For each measure, it provides the original source of the data for those measures for which the source was able to be identified. It also provides additional measures for consideration. Appendix C lists the source materials and other materials that were reviewed in developing Appendix B.

Develop Additional Common Performance Measures and Institutionalize the Process

Below, the Task Force identifies a set of proposed steps for selecting measures, and outlines a process for regularly collecting, analyzing, and reporting on a set of common performance measures for disadvantaged youth. The Task Force believes this should be a sustained effort, and not merely a one-time activity to identify common measures and then left up to current regular processes for the performance measurement data to be obtained and reported.

The following basic steps appear to be key elements, each of which is discussed below:

Coordinate and oversee this effort;

Select an initial set of common performance measures; and
Develop and maintain a process for sustaining the effort.

Coordination and Oversight of the Common Measures Effort

Major choices of approach to coordination and oversight appear to be between a centralized (more top-down) and more decentralized (more grass roots) approach. The centralized approach would involve assembling a relatively small number of persons, such as Federal employees and experts in disadvantaged youth, who would decide, in a relatively small number of meetings and other interactions, on the common performance measures. The centralized approach has the advantage of producing the set of performance measures much sooner. The primary disadvantage is that it may be perceived as not being sufficiently inclusive of various stakeholders.

It is possible, due to the scope and complexity of programs seeking to help disadvantaged youth, that an approach more related to the decentralized approach could be used. This would involve the creation of an overall coordinating group to establish and coordinate the work of multiple working groups, one for each category of youth services. In addition to representatives from key Federal agencies, each working group could include representatives of stakeholders representing the levels at which programs are delivered-in effect, representatives of the grantees and subgrantees who provide services. These include representatives from state and local government agencies, national associations of professionals in the service area, nonprofit organizations, and faith-based organizations.

The decentralized approach would involve bringing together many people from many perspectives, probably using a number of working groups (such as one for each program area)-to hammer out the common measures. The term "working group" is used here, although terms such as committee or task force might be used.

At least two states (New Mexico and Minnesota) have been or are planning to, use some form of what they are calling "interagency coordinating groups" to identify common measures. (In New Mexico this effort is being encouraged by the Legislature, which is considering legislation for them.)

Selecting Initial Set of Common Performance Measures

Once the working group has been formed, its initial task should be to reach agreement on an initial set of program areas and common, cross-cutting, performance measures. The common measures identified in this report (see Appendix B) could be a starting point for this set, at least for the service areas examined for this report. However, even for these areas the group undoubtedly will want to delete some measures, modify some, and add others.

Another key consideration is the number of measures to include in the common set. Multiple performance measures for a given service area are inevitable and desirable in order to cover the complexities and different perspectives provide by the multiple measures. Agencies and their programs will be motivated to focus on those measures that have the greatest visibility. Service performance characteristics not included among the common measures may be neglected.

At the same time, it is desirable to keep the set of common measures manageable in size, to avoid overwhelming those using the cross-cutting performance reports and also the service providers with excessive burden of collecting and reporting data. Clearly, a balance must be drawn, based on the judgment of participants.

The working group should also categorize candidate measures as to whether each is an output, efficiency, intermediate outcome, or end outcome measure. While different people can disagree over the categories for some measures, this is a useful step in order to help place into perspective the relative importance of each measure. The groups might find it useful to develop "logic models" ("outcome sequence" charts) which ask the program to identify the result to which each measure is expected to lead.

Outcome-Data Breakouts. Although the main focus of the group's efforts will be on developing common measures, the group should also select key breakout categories by which the performance data should be reported. Performance information for disadvantaged youth will be considerably more useful if the data are broken out by key client characteristics, such as age group, gender, race/ethnicity, and physical or mental disability type and severity. Such breakouts are helpful in identifying youth groups that need special attention. Breakouts can help agencies identify which other agencies and programs are serving similar client groups. Breakouts also will enable fairer comparisons, since comparisons based on aggregate data may obscure differences in client groups served. For example, some programs may have served client groups that had considerably more difficulties than others. Inevitably, programs whose outcomes appear worse than other programs will say "But my clients are different." The use of breakouts can help reduce this concern.

The groups should identify the breakouts, if any, that would be useful for each measure. For each breakout, the groups will need to develop specific definitions for each breakout, such as the age group ranges or race/ethnicity categories, to enable comparisons across agencies and to permit aggregations of the data.

Obtain feedback. Once the group develops a candidate list of initial common measures and breakout categories, the list could be circulated for feedback to each affected agency. Each agency would be responsible for developing its own process for obtaining feedback, particularly in terms of identifying whose feedback should be sought.

The group also should seek input from key stakeholders or groups representing key stakeholders who are not part of the process before finalizing the initial set of measures. It is beyond the scope of this report to identify such groups, but some candidates to consider include Congressional committees, representatives of state and local service agencies that are likely to be involved in the collection of the data, and organizations with an interest in disadvantaged youth services. [86]

Develop and Maintain a Process for Sustaining the Effort

Once the initial indicators have been selected, a number of additional major steps are needed:

Collecting, analyzing, and reporting of data on common measures;

Setting targets for common measures;

Encouraging coordination and cooperation (or even "partnerships") among agencies with similar programs or those serving roughly the same clients-including, where appropriate, development of action plans;

Encouraging use of common measures for performance management;

Reviewing and improving the measures/process on a regular basis (preferably at least annually).

These tasks assume require identifying a more or less permanent entity (within existing structures) has been selected to coordinate, provide continuity, and sustain the effort.

Implement Grantee-Level Performance Measurement Guidelines

The Task Force recommends launching a major effort to work with applicants and programs over the next several years to strengthen the accountability and performance of organizations receiving Federal funds to operate disadvantaged youth programs. The Task Force believes an increased emphasis on the importance of performance measurement as both a program management tool and a means by which to communicate program impact will improve the effectiveness of youth-serving programs, while providing Federal agencies the necessary information to hold grantees accountable for results.

Grantee-level performance measurement is central to the effectiveness and accountability of Federally funded disadvantaged youth programs. These efforts help articulate program goals and results, while providing program managers information to help assess the performance of grantees.

The approach outlined below is intended to be a tool to determine the need the program will address and the impact the program will have. The measures then provide indicators as to whether the program is having the intended effect.

A Federally funded disadvantaged youth program will:

Develop output, intermediate-outcome, and end-outcome measures;

Propose measures that capture the primary human service activities of Federally funded youth serving programs,

Participate in negotiating these measures as part of any grant award;

Develop a system for collecting and organizing this performance data on an ongoing basis; and,

Include the results in progress and final reports.

Grantees will likely have a variety of performance measures that reflect unique characteristics of their program and target populations. With respect to those reported to the appropriate Federal agency, grantees should be sure to specify performance measures that provide adequate coverage for all primary human service activities. Each primary service activity should include several output, intermediate-outcome, and end-outcome indicators.

Grantees seeking continuation grants, that is, those seeking funds for the second or third year of a multi-year grant, would be expected to provide performance measures, unless the program model makes it unable to do so. In such cases, the grantee should provide alternative measures of performance consistent with these guidelines.

The Task Force fully recognizes that these performance measures will reflect individual program goals and circumstances, and we expect that there will be a wide spectrum of different performance measures across all grantees. Moreover, we recognize that performance measures alone do not reflect the full scope and impact of disadvantaged youth programs. As a result, these measures (and reported results) should be considered to be one of several important elements taken under consideration when making funding decisions.

During implementation of a grant, we anticipate that performance measures may need to be adjusted based on grantee experience. As a general philosophy, changes to negotiated performance measures should be infrequent. Grantees and the appropriate Federal agency will negotiate measures that permit organizations to report actual results against established measures at the appropriate time. A process for changing performance measures is not intended to become a vehicle for altering measures that would not otherwise be met. In general, grantees will not request changes to performance measures when they should simply report on actual results when compared to an original target. However, it is also understood that, after grant award, organizations may experience significant changes in organizational or program circumstances that may require changes to performance measures. A grantee's record of meeting targets established in their performance measures should be a determining factor in future funding.

In general, performance measures are intended to be a useful tool for program managers to define and communicate the need that their program will address and the impact that the program will have. The measures provide indicators as to whether the program is having the intended effect.

Defining performance measures is just one step in the program design phase. Youth serving grantees should be required to utilize a standard framework (called a "logic model") to help think through each of the different primary human service activities, identify the likely result of those activities, and determine how to measure those results.

For each major youth-serving activity, grantees need to think through what the likely outputs, intermediate-outcomes, and end-outcomes might be. It is important to note that assigning a measure to these categories is not a science. Youth serving grantees should use this categorization as a way to help organize their thinking and recognize that there is not a "right" or "wrong" answer as each grantee will have its own set of circumstances.

Youth-serving grantees should be required to select several outputs, intermediate- and end-outcomes for each primary human service activity. Each performance measure should include the following information:

The result the grantee expects to achieve;

How the grantee plans to achieve this result;

The data and instruments the grantee will use to measure the result; and

A target (over time) the grantee expects to meet, including a baseline figure for every performance measure from prior years.

Youth-serving grantees have a variety of high-quality public resources available to help them meet this new requirement. We recommend that current and prospective youth serving grantees use any of the following resources in developing their performance measures:

W. W. Kellogg Foundation Evaluation Handbook, www.wkkf.org/Pubs/Tools/Evaluation/Pub770.pdf

Measuring Volunteering: A Practical Toolkit, www.independentsector.org/programs/research/toolkit/IYVToolkit.PDF

Online Evaluation Resource Library, oerl.sri.com

Evaluation Toolkit: A User's Guide to Evaluation for National Service Program, www.projectstar.org/star

American Society for Public Administration, Center for Accountability and Performance www.aspanet.org

Outcome Measurement Resource Toolkit, United Way of America, http://www.liveunited.org/outcomes

The Results & Performance Accountability Implementation Guide, www.raguide.org

We also encourage grantees, when needed, to seek the help of local or regional professionals to perform and oversee performance measurement activities, including the development of a plan that encompasses the collection of and reporting on outcome data that will be used to improve program quality.

Conduct Rigorous Oversight of Earmarked Grantees

In our preliminary report of April, 2003 we stated:

Congressional earmarking of funds for disadvantaged youth programs creates an especially problematic situation. It eliminates what linkages there should be between accountability measures and funding decisions. Earmarked programs do not receive the oversight that enables agencies to make sure they are actually helping youth, achieving their goals, and making wise use of limited funds. The earmark process also keeps Federal agencies, charged with implementing the statutes, from making funding decisions based on a coordinated, identified need to address a specific problem. [88]

To give us a complete picture of the earmark situation with regard to youth programs, we asked five of the main youth-serving agencies to provide us with a list of all the earmarks that appear in their youth programs for Fiscal Year 2003. We asked for the name of the recipient organization, the amount earmarked, the source of the earmark (e.g., appropriations committee report, etc.), as well as a description of how the funds will be used. The agencies reported a total of 304 earmarks for a total of $206.2 million in Fiscal Year 2003 (Figure 3; see Appendix H for details).

Figure 3: Youth Program Earmarks, Fiscal Year 2003

Agency	Number of earmarks	Amount, Fiscal Year 2003
Department of Justice	160	$147,340,711
Department of Health and Human Services	88	$29,764,447
Corporation for National and Community Service	2	$15,000,000
Department of Education	42	$10,385,000
Department of Labor	12	$3,775,000
Totals	304	$206,265,158

There are a number of reasons why the Task Force strongly recommends against the use of earmarks in youth programs:

Earmarks might fund programs that sound good but actually do harm to youth. For example, in the 1970s, inmates serving life sentences at a New Jersey prison began a program to "scare" or deter at-risk or delinquent children from a future life of crime. The program, known as "Scared Straight," featured as its main component an aggressive presentation by inmates to juveniles visiting the prison facility. A TV documentary on the program aired in 1979 provided evidence that 16 of the 17 delinquents remained law-abiding for three months after attending Scared Straight, and claimed a 94 percent success rate. [89] Other data provided in the film indicated success rates that varied between 80 and 90 percent. The program received considerable and favorable media attention and was soon replicated in over 30 jurisdictions nationwide, resulting in special Congressional hearings on the program.
There was one major problem with this widely hailed program: it actually harmed the youth who participated in it. A recent review of nine randomized controlled trials (the highest level of program evaluation) of Scared Straight and related programs found that these programs either did not affect, or in some cases actually caused a small increase in, subsequent criminal activity by program participants. [90]
Earmarks might fund programs that duplicate services that are already available.

Earmarks might leave some especially needy communities and successful programs-that are less politically powerful-out in the cold.

Earmarks undermine the promise that taxpayer dollars go to the best programs because earmark grantees do not compete for their funds.

Earmarks completely tie the hands of experts whom the Federal government pays to determine the greatest needs and the most potentially successful methods to meet those needs. For example. the DOJ Office of Juvenile Justice and Delinquency Prevention operates an important juvenile delinquency prevention program which receives millions of dollars annually. The staff in that agency are knowledgeable about this field, including knowing where the needs are great, what the research in the field is showing works well, and what the future research needs are. Unfortunately, all the funds that they would normally be able to use to further knowledge in this area is tied up in congressional earmarks. In fact, Congress earmarks more funds for that program than it actually provides through the appropriations process. (The agency adjusts for this by slight reductions in some of the earmarks, in order to provide funds to all groups named in the legislation.)

The system of having a large number of earmarks turns state and local governments into bystanders to the public policies implemented in their communities. [91]

Figure 4 highlights significant and problematic differences between regular grants and earmarked grants:

Figure 4: Comparison of Regular Grants and Earmarked Grants

Issue	Regular grants	Earmarked Grants
Competitive grant review process	Grantees must participate in a rigorous, competitive grant application process, which is carefully designed and often peer/expert reviewed.	Grantees bypass the competitive grant review process altogether. Many such grantees are advantaged year after year. For instance, 41% of DOJ earmarks and 20% of HHS earmarks received funding both in 2002 and again in 2003. (These do not represent multi-year grants but are repeating single year awards.)
Alignment with the agencies' statutory and strategic mission	They are the only means for carrying out some agencies' primary missions, and the criteria for selection reflects both the judgment of political leadership and the experience of professional civil servants, all of whom are mindful of Congressionally-legislated requirements for achieving results, such as the Government Performance and Results Act.	They may be incidental to strategic goals or even represent a poor match with an agency's capacities and experience, not to mention the intent of the statute. They are sometimes shifted to another bureau, but a proper fit may not always be achieved. The Department of Labor requires a fundable proposal that meets the legal requirements of the funding source.
A key aspect of grant-making is available to public view	Criteria for grants competitions are published as program announcements for all to see. Some programs are required to publish annual program priorities in the Federal Register for public comment.	There is very little transparency in the earmarking process, particularly since many are inserted "at the last minute" in conference reports rather than in statutes passed by Congressional vote.
Responsive-ness to requests and oversight requirements of Federal program staff.	Because they are multi-year grants (usually 3-5 years), they are more attentive to the information requests and other accountability-related demands made on them by the Federal program staff, since they want to be eligible to receive more funds in the future. Moreover, multi-year grants are more efficient than single year funding because the application and funding process is labor-intensive for both grantee and government. It is also likely that projects need more than one or two years to be effectively established and produce results, not to mention undergo evaluations and follow-up studies.	Though some earmarks are advantaged again and again (see "competitive process" issue above), there is also a problem with one-timers. Because they are often for a single year (e.g., 80% of HHS 2003 earmarks), compliance and follow-up pressure is often difficult to maintain because the Federal money is no longer flowing as an incentive to follow the reporting rules. In any case, the remaining earmarks, those funded for more than one year, are really getting a series of single year grants, which is administratively inefficient (see statement at left). Moreover, some of these are being newly funded for different, often short-term projects, thus bringing in a whole new cast of characters.
Reporting requirements	They are required to regularly report their progress and other data throughout the life of the grant as a condition of receiving funds.	Reports may be requested, and earmark grantees usually cooperate, but there is little incentive or leverage for compliance. Some simply dismiss Federal oversight with impunity
Account-ability for results	Many agencies are requiring applicants to propose specific, measurable outcomes and report on their achievement. Such proposals become part of the grounds for winning the award and their achievement becomes part of the grantee's track record. When significant shortfalls occur or threaten, agencies may devote considerable energy to technical assistance to help projects recover. Agencies are also working to build their (regular) grantees' capacities for data collection and performance reporting.	The agencies which have made the most progress in holding regular grantees to outcome achievement are trying to do the same with earmarks, but face the barriers of willingness and capacity. Because earmarks may be small and lack management skills, as well as program experience, they may not have access to relevant data or the skills to collect or analyze such data, not to mention lacking key capacities for good management and programmatic success.

Clearly, as Figure 4 shows, there are real and important differences between regular and earmarked grantees on key accountability issues. Neither system is perfect, but even the current system, with all its flaws, is an improvement over one where earmarked grantees are able to bypass careful selection and accountability processes.

In expressing these strong concerns about the widespread use of earmarks for youth programs, the Task Force does not presume that the current process functions administratively as well as it should. This is particularly true in the critically important area of program evaluation and accountability. We have pointed out many of the current system's weaknesses in this report and made recommendations to repair and improve it. But in comparing the two approaches, even this current system is better than the earmark process of doling out funds for youth programs to well-connected organizations. In that system, all the necessary information about that program is not widely available. Even the current practices of oversight and accountability, as weak as they are prior to the implementation of our recommendations, are tossed aside completely.

In one way, the widespread use of funding home district projects via the earmark process has enabled many Members of Congress to express their support for the youth in their communities. This is laudable. Members of Congress certainly know their districts, but their wisdom is often restricted because they do not have access to the larger picture of the needs of disadvantaged youth in communities all across this country. We are strongly concerned that this system has so many inherent weaknesses that it has the potential to do more harm than good.

Addressing the difficulties that earmarks cause now has a sense of urgency, because of their growth. A study by the Heritage Foundation found that:

Between Fiscal Year 1985 and Fiscal Year 1999, the growth in annual earmarks increased substantially faster-between 25 to 1,000 times faster in most cases-than inflation-adjusted Federal domestic discretionary spending. Moreover, although the increase in the number of earmarks has risen since 1985, the growth appears to be accelerating: The number of earmarks in five of the 13 appropriations bills doubled between Fiscal Year 1998 and Fiscal Year 1999. Although project-specific earmarks represent only a fraction of the spending in most domestic discretionary programs, if their use continues to grow at the current rate, congressionally-mandated earmarks could account for a majority of the spending in several budgetary accounts. [92]

Very few of the major youth-related organizations with offices here in the Nation's Capital complain loudly about the earmark process, since a many of them have figured out how to take advantage of that process for themselves and/or their affiliates.

Recognizing that it will likely take several years to phase-out the habit of earmarked youth programs, the Task Force recommends the use of these Guidelines for Oversight and Accountability of Earmarked Grants:

Within the annual budget and performance plan, each department/agency shall describe its policies for management, oversight, audit, and reporting requirements for all earmarked grants. This plan should include a summary of earmark grantee self-nominated performance measures and the results for each grantee's progress towards meeting the numerical targets established in the performance measures.

Agencies should be particularly careful to note when necessary data is missing or when the grantee's conclusions are not supported by the data.

Agencies should provide their own evaluation of the quality and value of the performance measures.

Agencies should provide their own evaluation of any other aspect of the program (management, budget, outcomes) that will be useful for policymakers.

Implement No Child Left Behind in Department of Defense Schools

The Task Force recommends that the Department of Defense consider implementing select relevant provisions of the No Child Left Behind Act in Department of Defense Education Activity (DoDEA) schools, in conjunction with the Department of Education. Specific recommendations for consideration are delineated below.

The No Child Left Behind Act of 2001 is the most sweeping reform of Federal education policy in a generation. It is designed to implement the President's agenda to improve America's public schools by 1) ensuring accountability for results, 2) providing flexibility in the use of Federal funds, 3) focusing on proven educational methods, and 4) expanding education choice for parents. This legislation is intended to address the serious problems facing our Nation's schools in adequately preparing young people for postsecondary education and meaningful adult lives.

While it does appear that the DoDEA schools are moving in the right direction in improving student achievement, there is cause for any optimism to be guarded. Data from the 1998 and 2002 National Assessments of Education Progress (NAEP) show mixed results:

Students in DoDEA schools performed at or slightly above the national average in reading in grades 4 and 8.

However, in 2002, 66 percent of 4^th graders and 63 percent of 8^th graders did not achieve proficiency in reading.

Between 1998 and 2002, the percentage of 8^th graders achieving proficiency in reading dropped slightly from 39 percent to 37 percent.

We can do better, both as a Nation and for the children of our military families around the world. While NCLB does not require DoDEA schools to implement the law, it does provide a mechanism for the secretaries of Education and Defense to jointly determine the extent to which NCLB should apply to DoDEA schools. DoDEA schools face the same problems related to student achievement as their colleagues across the country. Therefore, there is reason to implement select provisions of NCLB in DoDEA schools.

Which provisions of NCLB are currently being implemented in DoDEA schools?

DoDEA schools have developed a set of system-wide curriculum standards based on the content standards produced by the National Council of Teachers of Mathematics, the National Council of Teachers of English, the International Reading Association, the National Research Council's National Science Education Standards and the National Council for Teachers of Social Studies. However, it is not clear whether the standards-based curriculum is being implemented consistently across all DoDEA schools.

NCLB calls for a qualified teacher in every classroom. A 2001 GAO study found that virtually all teachers in DoDEA schools are certified in the subjects or grades they teach. In addition, approximately two-thirds of teachers in DoDEA schools have advanced degrees compared to about 46 percent of public school teachers generally.

DoDEA schools currently test their students annually, using the TerraNova exam, a nationally normed test, in grades 3-11. On this measure, students in DoDEA schools have consistently scored above the national average in every subject and at every grade level [93].

Preliminary data from several external studies confirm that DoDEA students are at or near the top of all states in achievement on a range of standardized assessments. [94] [95] [96]

In order to compare DoDEA schools with other schools in the Nation, a sample of students are also tested using the National Assessment of Educational Progress (NAEP). As the scores in Figure 5 indicate, both NAEP scores and proficiency levels for DoDEA exceed the national average. Furthermore, as NCLB demands, the DoDEA schools are also making annual progress:

Figure 5: NAEP scores in reading for 1998 and 2002 (scale score 0-500).

Grade year	Year of test	DDESS school average	DoDDS school average	National average	DDESS percent at or above proficient	DoDDS percent at or above proficient	National percent at or above proficient
4	1998	219	221	213	32	33	28
4	2002	225	224	217	34	33	30
8	1998	268	269	261	39	37	30
8	2002	272	273	263	37	40	31

Source: National Center for Education Statistics, 2003 [97]
Note: DDESS denotes scores for students in the United States; DoDDS denotes scores for students overseas.

Although the NAEP scores above indicate that DoDEA schools are close to national averages, no school or nation should be satisfied with a 30 percent to 40 percent level of proficiency in its students. DoDEA students could achieve at a higher level, and implementation of select provisions of NCLB will ensure that DoDEA schools continue to improve on their current levels of academic achievement. At the same time, NCLB implementation should support the unique DoDEA practices that have led to its success thus far.

How will further implementation of NCLB improve these existing levels of achievement?

The Department of Education should assist DoDEA administrators in assessing the applicability of each specific NCLB provision and develop an implementation plan based on this assessment.

An assessment of the extent to which DoDEA practices are consistent with NCLB requirements should be conducted that focuses on each of the four NCLB "pillars": standards and accountability; flexibility in the use of Federal dollars; increased parental choice; and a stronger emphasis on effective practices that are supported by research. This document covers the broad points, but the fine details of implementation deserve a closer look.

At a minimum, DoDEA schools should bring their system into line with NCLB provisions in terms of the pillars of standards and accountability and a focus on "what works." DoDEA schools should:
- Clearly define a measure and standard of Adequate Yearly Progress as NCLB requires. As NAEP scores indicate that DoDEA schools are achieving at levels of proficiency that are between 30-40 percent, an adequate standard will be critical to ensuring that all students are proficient within 12 years as NCLB requires.
- Continue with the practices they have already implemented including: a system-wide standards-based curriculum, annual standardized testing, commitment to teacher quality, use of NAEP scores, and evidence of annual progress.
- In concert with the Department of Education, examine the system-wide content standards for rigor and ensure that assessment tools are aligned to these standards.
The Department of Defense should ensure that its strategic plan focuses on improving student achievement, particularly in the core subjects of reading and language arts and math.

While students in Department of Defense schools are performing at levels consistent with student performance across the Nation, there is significant room for improvement, particularly in the areas of reading and math where far too many students are below grade level.

DoDEA schools should become a model of international excellence by implementing instructional practices that are based on rigorous scientific research.

DoDEA has potentially effective practices that may be useful for dissemination to the states. However, it is a relatively isolated system that has limited communication with other systems, and its practices may or may not be validated by the rigorous scientific-research standard that is a hallmark of NCLB. In this sense, implementation of this pillar of NCLB has a clear benefit: it will hold success measures of DoDEA programs to the same rigorous, scientific standards to which programs in other systems are held, and allow for a two-way dissemination of effective practices that will be beneficial for DoDEA and all other systems.

The Department of Education should consider initiating formal research on the distinctive effective practices of DoDEA schools, particularly with respect to the achievement of students of color, and disseminate them to the states.

In two external studies, the authors found that African-American and Latino students from DoDEA schools ranked first in the Nation on disaggregated NAEP scores and that, in general, minority students in DoDEA schools are performing at a significantly higher level than are minority students elsewhere in the Nation. [98] This is critical because a central component of the NCLB legislation is to narrow the achievement gap between white and minority students. As the fourth pillar of the NCLB legislation focuses on "doing "what works"," DoDEA schools should serve as a national model for effective practices that promote the academic success of minority students.
The DoDEA administration should assess current levels of parental input and design mechanisms to further increase parental participation within the context of the "Parent Empowerment" pillar of NCLB.

As a result of the fact that school choice is limited at schools on military bases, a large part of this pillar is inapplicable to DoDEA schools. One way in which DoDEA schools could implement this pillar is to augment their existing efforts to involve parents in school decision-making and in their students' academic careers. Some of the DoDEA's major successes are built on its emphasis on parent and family involvement in every student's education. One area in which the DoDEA schools could benefit from NCLB is by providing parents and students with information about how their school and district are performing in improving student achievement. This could be done by providing parents with annual information, either through a "report card" or some other vehicle, about the achievement levels at the school, the rates of annual progress on standardized measures, and efforts to ensure that there is a qualified teacher in every classroom.

Chapter 4:
Better Connections: Engaging Youth and Families

Research has shown that in order to ensure their healthy development, adolescents need caring adults in their lives; opportunities to learn marketable skills and maintain good health; and opportunities to contribute meaningfully to their communities and society. [99] Generally, American families and communities are doing a good job of addressing these youth needs and opportunities. In the next chapter we will make recommendations for improving the supports for youth who, for a variety of reasons, are not in nurturing families or communities. In this chapter, however, we make several recommendations aimed simply at validating and building on the strengths that exist in most families and communities. The first recommendation is born from the knowledge that parents play a pivotal role in guiding their children's development and should be supported in that role. The second two recommendations provide opportunities for young people to contribute through service, recognizing the value of the assets they bring to their communities and to the Nation.

Increase Parent Involvement in Youth Programs

The Task Force recommends that any Federal program that serves youth should endeavor, when appropriate, to involve parents as much as possible in the program. This means including parents in the planning stages and in any advisory groups, as well as in the program itself.

Conventional wisdom has held that one of the hallmarks of adolescence is the decline of parental influence as peer influence increases. Whether this was ever the case is unclear. Current research has shown that parental influence remains both a strong and central influence in the lives of young people, even as peer influences increase.

In August, 2003, USA Today reported that American teens preferentially choose spending time with their parents over others. [100] In the same month's edition of the Journal of Adolescent Health, [101] the editorial and a number of articles about empirical studies reaffirm the centrality of parents in the lives of teens. And, repeatedly, polls show that parents are the adults young people admire most. Likewise, research has shown that when young people feel connected to at least one parent, they are less likely to report emotional distress, violence involvement, suicidal thoughts or attempts, cigarette/alcohol/other drug use, and early sexual debut. Clearly, parents count. But what is it that leads to young people experiencing this connection with parents?

One element of parent-child connectedness relates to parental characteristics. Parents who are warm, have an open communication style, can listen as well as share their thoughts and beliefs, create an environment that fosters connection; but parent-teen connections are built on more than an affective style. Parental monitoring is key-knowing their children's friends, their friends' parents, and their teachers. Also critical are setting clear expectations, such as behavioral and school expectations; and monitoring both behavior and performance. Parents need to know honestly what their children are doing socially and how they are doing academically.

While behavioral monitoring matters, so too does psychological autonomy. Parents need to realize that their children are not an extension of themselves, but rather separate individuals with thoughts, opinions, and beliefs. Children need to be accorded the respect parents grant to others.

So, too, there is emerging evidence that certain parenting styles create a more positive relationship between parents and their children than do others. Specifically, an authoritative parenting style sets clear boundaries for behavior, is respectful of the child, and negotiates rules (but not at the time of an infraction). In such an environment, children have a voice, but do not control the family.

The evidence is clear that young people who report feeling connected to at least one parent do better across every outcome studied. Parents do matter in the lives of teens; and when positive connections are fostered, young people flourish.

Federal programs should acknowledge the vital role parents play in their children's lives by encouraging their participation whenever possible in programs and services their children receive. Schools, nonprofit agencies and other entities that provide services to children and youth should welcome parents and seek opportunities to strengthen their roles in their children's lives.

The Task Force recommends that a continuum of parental involvement components be established as guidance for new programs. This could be a menu of parenting components grantees can choose to implement as part of their program development. These components should include, but not be limited to the following:

a parent advisory committee to help in program development and parent recruitment;

dissemination of proven public education messages on the importance of parents knowing what their children are doing socially and how they are doing academically;

parent/child counseling and/or support groups to enhance communication skills;

parent/child recreational and civic activities;

parent support groups built around particular development issues children may be facing, e.g., Attention Deficit Hyperactivity Disorder (ADHD), other health-related issues, school performance, risk taking behaviors;

parenting classes and support groups for parents whose children are in foster care or the juvenile justice system (a possible component of mentoring programs).

Design a Youth Service Initiative

The Task Force recommends that a youth service initiative be designed which would allow older youth (college age) to display leadership by providing opportunities for them to serve children living in high poverty areas of the United States.

Through campus organizations, local community initiatives, and national programs such as Habitat for Humanity and AmeriCorps, college students and other young Americans are responding in numerous ways to President Bush's call to build a culture of service. These young people are broadening their perspective and transforming their careers through living for the sake of their fellow citizens and investing their time and talents to revitalize lower-income communities.

AmeriCorps*VISTA has created an intensive, ten-week Summer Associates program for students who work with VISTA nonprofit sponsors to organize and recruit youth volunteers. This program could provide the infrastructure for this initiative. Full-time members and Summer Associates over 18 years of age, recruited through VISTA nonprofit sponsors, provide program development and volunteer coordination support to local community leaders and volunteers in the program (approximately 6,000 members nationwide). The student volunteers assist local residents in implementing local self-help projects ranging from senior citizen initiatives to housing rehabilitation and provision of indoor plumbing to remote areas of Appalachia and the rural south.

The Volunteers in Service to America (AmeriCorps*VISTA) program, which has nearly four decades of local capacity-building experience, could provide member resources and administrative leadership. Localized projects could be developed by a consortium of nonprofit organizations that could host youth service participants as part of their AmeriCorps*VISTA Summer Associates program. Youth could be offered an eight to ten week structured experience in service through existing community based organizations that could provide local host sponsorship for housing and project leadership.

The Youth Service Corps experience can also be promoted in universities by AmeriCorps Recruitment to encourage field experiences in education. Recruitment also assists by receiving and forwarding online student applications to the appropriate Corporation for National and Community Service (CNCS) state offices and VISTA project sponsors, or through another application process developed specifically for this program.

Recruit Youth for Federal Grant Review Panels

The Task Force recommends that college youth be recruited and included as participants on Federal panels that review youth program grants, where feasible.

Federal social programs designed to serve youth often require the use of a competitive grant process to review and rate the many applications from groups seeking Federal funding. The review process of these discretionary grant applications involves multidisciplinary teams of reviewers meeting together to receive training, work in teams and review anywhere from dozens to hundreds of applications.

If monitored closely and supplied with the proper training, the grant review process can be very rewarding for the reviewers in terms of instruction, exposure to innovative, cutting edge programs, and seeing the difference between applications prepared with due diligence and those that fall short. Reviewers gain skills in connecting theoretical research with practice, teamwork and consensus decision making. Due to the dynamics of academic, program and policy orientations, team members are often engaged in discussions at the program level that are analogous to larger policy questions and are at times described as the cross-walk between policy, research and practice. Grant reviewer panels typically consist of from four to ten individuals, depending on the type of review being conducted. [102] Participants are usually from a variety of disciplines and backgrounds relevant to the program area, and this diversity of experiences enriches the process and final outcome.

This exposure and its instructive value are precisely why youth should be involved in this process. Youth would benefit from learning about the reviews and programs, while the adults would benefit from having an individual from the proposed targeted age group involved in the decision-making process. Federal program staff closely oversee this process and are available throughout the review process to offer youth and adult panelists additional advice and guidance as needed.

Obviously, because of the differences in age and experience, the inclusion of youth along with adults on Federal grant review panels presents some challenges. We believe these include: placing youth in a situation that results in a successful experience; training youth reviewers to be competent and confident about the subject matter; keeping their interest; and breaking down inhibitions of the professional adult co-reviewers. Fortunately, these challenges also have solutions:

Recruitment and Selection of Youth as Panelists:

Youth with at least sophomore standing and up through graduate studies, and who are enrolled in a related field of study, should be recruited from American colleges and universities. College students who are serving internships in Federal executive branch offices would be a useful source of participants, particularly since the bulk of grant reviews take place during late spring and summer, when the number of interns is at its highest.

Students can come from a wide variety of academic disciplines. These include, but are not limited to, public policy, social work, political science, psychology, sociology, education, criminal justice, philosophy, public administration, business, and journalism. Youth who have completed programs that address the subject of the grant and then attended college in a related field could make especially valuable contributions.

Generally desired qualifications include the ability to:
-       Evaluate and apply criteria related to program requirements.

-       Read and analyze applications (e.g., for strengths and weaknesses).

-       Write clearly, accurately, concisely, and effectively.

-       Communicate effectively.

-       Contribute to an effective group process by being cooperative, constructive, and flexible.

- Maintain strict confidentiality.

In the same manner as adults who apply to be considered as grant reviewers, youth shall be required to apply via the Internet and complete the same application process required of all reviewers.

Each of the four main youth-serving agencies (departments of Education, Health and Human Services, Labor and Justice) should include "college youth" in the official call for qualified grant reviewers, where feasible.

Ensuring a Successful Experience

Interested youth should be made aware of the rigors involved in the grant review process, which involves a rigorous reading schedule of an estimated 500 pages. This requires a significant degree of effort on their part in order to ensure a successful experience, and that should be made clear.

Quality grant reviewer training shall be conducted for all reviewers, including youth.

Youth shall be provided with logistical information prior to the review, including a copy of the program announcement.

Youth shall be compensated at the same rate as adults.

The HHS Family and Youth Services Bureau conducted a pilot effort of this recommendation to test its feasibility during its summer 2003 grant reviews. This agency oversees the various programs that serve runaway and homeless youth. In this pilot, a youth was placed on each team of four reviewers. The Bureau reported that everyone was surprised as to the comfort level achieved through training and support. They noted that the adult grant reviewers were very impressed with the inclusion of youth on each review team. They commented that they appreciated most the unique views and the energy that youth brought to the process. There were no negative evaluations of the experience received from either youth or adults. [103]

Chapter 5:
Give Priority to the Neediest Youth: Caring for Special Target Populations

While the Federal government is spending billions of dollars to address the problems of youth, the problem is often that too many of these dollars are spread out among too many youth who already have access to the resources they need to grow up successfully. Although these actions may appear to be preventive, in fact they typically lead to under-serving or never even engaging the youth who most need help-and who become society's most serious problems. Thus, we often see evaluations of youth programs do not show much impact. One reason for this might be that the youth that needed to change were either not engaged, or not engaged sufficiently. At an aggregate level, the result is that the public and policymakers never see the kinds of significant change they want to see in the things that concern them: juvenile crime, school performance, drug use, and so forth. [104]

Public money should be spent on public problems, rather than on grand plans to benefit all youth, most of whom will grow up just fine without government help. [105] Thus, we begin a discussion which shall continue beyond the life of the Task Force regarding the identification of "special target populations" of youth. These special target populations would be those who represent areas of serious concern, and who carry disproportionately negative consequences for youth and their communities if not addressed. It is these groups named below, as well as others who will be identified in the future, who should be the primary targets of relevant disadvantaged youth programs.

Target Youth in Public Care

The Task Force recommends that the first designated special target populations represent youth who are already in public institutions, and who create public expense. These are youth in foster care (particularly those aging out of foster care), and juvenile justice youth. For both these groups, the Federal government and governments at other levels are serving in loco parentis, in place of the parents.

There are about 542,000 youth in foster care. While about half will be in the foster care system for about 11 months (a long time for a child), one-third will remain for more than three years. The median age of these children is 10.6 years old, with 38 percent African American, 37 percent white, and 17 percent Hispanic. Half will be placed in foster family homes with non-relatives, one quarter with relatives, 18 percent in group homes and institutions, and 10 percent in other forms of placement. [106]

Foster youth face a number of problems and often without responsible, caring adults looking out for them and their best interests on a consistent basis. In school, they are among those most likely to be left behind. About 70 percent of them are school age, [107] and their school work often suffers for a whole range of reasons, including unstable, often violent homes, frequent placement changes while in foster care, a very high rate of disabilities, and existing difficulties with school work, for which they receive insufficient help. They score lower on standardized tests, [108] have higher absentee and tardy rates, are more likely to drop out of school, and are three times more likely to be referred for special education and related services. [109]

Foster youth are also at greater risk for health problems and risk behaviors. Living in foster care before the age of 15 increases the odds of juvenile delinquency. [110] They have a high level of disability. They also initiate sexual intercourse at an earlier age, and report a greater number of partners, both of which place them at greater risk for sexually transmitted diseases and pregnancy. [111]

The outlook can be particularly dim for those youth who age out of foster care. Somewhere between 18,000 and 20,000 youth age 16 and older transition out of the foster care system annually. [112] They need help with finishing high school or applying to college, getting the health care they need, finding new housing on their own, getting a job, and more. The Federal Foster Care Independence Act offers some resources to help with these problems, and the data from those efforts is beginning to come in. A recent report from Child Trends, Inc., notes that without the extended support most families provide young adults, youth leaving foster care face enormous challenges in building successful lives. "They are less well prepared educationally, have a harder time embarking on a productive career, are more likely to abuse drugs and alcohol, and are more likely to be involved in the legal system." [113] Studies of youth who have left foster care have shown they are more likely than those in the general population to not finish high school, be unemployed, and be dependent on public assistance. Many end up in prison, homeless, or as parents at an early age. [114]

Sadly, there are about 106,000 American young people in the juvenile justice system, either in detention, correctional or shelter facilities. Risk behaviors such as drug use and violence, which often are the reason they are in "the system," are rampant. Studies show that juvenile delinquency is also highly correlated with sexual promiscuity, which is further correlated with substance abuse. ^[115]

Illiteracy and school failure are serious and widespread: detained and committed youth score below their expected grade levels across subject areas. Such youth, when of high school age, typically score between grades 5-7 and 5-6 in reading and written language, respectively. Similarly, these youth score between grades 5-9 in math. An estimated 38 percent of incarcerated youth also qualify for speech and language services. [116] Approximately 75 percent of these youth fail one or more courses, while 40 to 50 percent are retained in grade. In one study, over a three-year period, 40 percent of youth who entered correctional facilities had earned no high school credits. In one examination of incarcerated female teenagers, nearly half had been expelled from school and a disproportionate number had learning disabilities. [117] Other studies also demonstrate correlations between delinquency and lower levels of academic aspirations, [118] lower levels of academic achievement, [119] with higher rates of dropping out of high school, [120] and with higher levels of aggressive behavior on the school playground. [121] Despite these unsuccessful histories, many youth report having a positive attitude toward school and realize its importance. [122] This offers some hope of potentially averting a disaster, since about 75 percent of offenders at the adult level are illiterate.

Target Youth at High Risk

The Task Force recommends that a second group of youth also be considered among the special target populations. This subgroup includes youth with a high number of factors putting them at risk for unproductive and publicly costly lives, [123] such as children of incarcerated parents and migrant youth.

The President has already taken the lead on addressing the needs of children of incarcerated parents, announcing a three-year, $150 million initiative during his 2003 State of the Union address to bring mentors to the children of prisoners. Government would help by funding the enormous task of supporting the training and recruiting of mentors.

The Department of Health and Human Services notes that children of incarcerated parents are seven times more likely to become involved in the juvenile and adult criminal justice systems. [124] Parental arrest and confinement often lead to stress, trauma, stigmatization, and separation problems which may be compounded by existing poverty, violence, substance abuse, high-crime environments, child abuse and neglect, multiple caregivers and/or prior separations. These children are more likely to develop attachment disorders and often exhibit broad varieties of behavioral, emotional, health, and educational difficulties. Many children of incarcerated parents are angry and lash out at others, resulting in confrontations with law enforcement. Lacking the support of families, schools, and other community institutions, they often do not develop values and social skills leading to the formation of successful relationships.

HHS reports that between 1991 and 1999, the number of children with a parent in a Federal or state correctional facility increased by more than 100 percent, from approximately 900,000 to approximately 2,000,000. [125] According to the national data from the Bureau of Justice Statistics, in 2001, 3.5 million parents were supervised by the correctional system. Prior to incarceration, 64 percent of female prisoners and 44 percent of male prisoners in state facilities lived with their children. With the parent gone, these children often have to leave their home as well. During incarceration, nearly 90 percent of children of incarcerated fathers lived with their mothers and 79 percent of children of incarcerated mothers lived with a grandparent or other relative, according to HHS.

Although research has indicated that parents in prison and children should visit one another, less than 50 percent of prisoners receive visits from their children. In a number of cases, the caregiver may not want the child to visit the inmate, and prisons are often located far away from the urban areas where most children of prisoners live. According to the Bureau of Prisons, there is evidence to suggest that inmates who are connected to their children and families are more likely to avoid negative incidents and have reduced sentences.

The migrant youth population cannot be described by a single profile or description. There are a number of youth that work in the fields to raise money for families that are located in the United States or abroad. Of these youth, there are approximately 126,000 children between the ages of 14 and 17 that work on America's farms. [126] These adolescents comprise about seven percent of all farm workers. [127]

Another group within this population does not work in the fields but instead tries to benefit from the education system in the United States. However, they have a difficult time due to many different barriers, including mobility, language, and culture. The U.S. Department of Education estimates that the school completion rate for migrant students is approximately 50 percent. Once these youth reach secondary school age, they drop out for a multitude of reasons. Some need to stay home to take care of younger siblings while their parents are in the fields, many find it too difficult to maintain the credits necessary to graduate because they are moving from state to state where requirements are different, and some need to contribute to their families' income. The different migrant populations face different problems, but many are caused by the same factors.

There is also a population of migrant youth that work in the fields and have been described as de facto emancipated minors. [128] They lack adult guardianship and live in households with individuals other than family members. [129] Most of these youth are foreign-born and recent arrivals to the United States. Seventy-five percent of these foreign-born youth arrived in the United States between the ages of 14 and 17, and 58 percent were between the ages of 16 and 17. [130] On average they work fewer weeks (14 weeks a year) compared to their adult counterparts, who average 25 weeks a year. They make minimum wage or sometimes less and live below the poverty level. Their English skills are limited and, for some, this is their first experience working in agriculture. Due to their age they are vulnerable to hazardous work conditions and inappropriate work assignments. These youth live in crowded housing, usually with many adult males (80 percent of farm workers are men), who tend to introduce these youth to unhealthy habits, such as drinking and smoking.

Migrant youth who do attend school have difficulties staying at their grade level for many reasons, one of which is because they follow their parents to the next employment location. Many were born in the United States but their parents were not; 81 percent of farm workers were foreign-born, and 95 percent of those were from Mexico. [131] Language and culture barriers make it difficult for parents to be involved in ensuring their children are succeeding in school and avoiding risk behaviors. The families of these youth live below the poverty level and are frequently facing periods of unemployment. For the past decade, the median income of individual farm workers has remained less than $7,500 per year, while that of farm worker families has remained less than $10,000. [132] As these youth reach the secondary level, it becomes more difficult for them to remain in school. Many times they need to contribute to their family incomes, and most have had so many interruptions to their schooling that keeping track of their records has made school completion increasingly difficult.

Due to their poor living conditions and the level of poverty, many migrant youth face health issues. Health insurance is not typically provided for farm workers, as only five percent report they are covered. [133] Even though needs-based services may be available, especially since the majority of migrant farm workers live below the poverty line, only 17 percent reported using them. [134] Migrant youth have difficulties receiving health services for accidents that happen off-site from the fields as well as for preventive measures. The infant mortality rate is 25 percent higher than the national average, and poor nutrition causes health issues and poor physical development. [135] Childcare is difficult to manage for the migrant farm workers, and unfortunately, young children at times must accompany their parents in the field. Seven percent of parents with children from birth to five reported bringing their children into the field with them while they worked. [136] This potentially exposes these children to pesticides and other dangerous conditions. One report conducted by a Migrant Health Program found that 48 percent of children had worked in fields still wet with pesticides; 36 percent had been sprayed either directly or by drift; and 34 percent of the children's homes had been sprayed by pesticides in the process of spraying nearby fields. [137] This potential health risk, accompanied with living in unsanitary overcrowded housing with no access to health benefits, creates another difficult barrier for these vulnerable youth

Special Target Populations: A Case Study

The concept behind special target populations is that these groups would be considered high priority groups when determining where to target resources within relevant discretionary programs. In addition, and most importantly, these groups would be the subject of interagency working groups (IWGs) that would assemble the relevant Federal agencies that currently address each population. The goal of these newly-created IWGs would be to find ways to identify and successfully address the most pressing needs of each group through new collaborative efforts, as well as through research (as needed), the targeting of available resources, and the pursuit of additional funds and/or legislation for new initiatives as they are identified. This notion of targeting would also help us define a more manageable problem, for which concrete outcomes could be established, monitored and reported.

The Task Force took advantage of the assembled agencies involved and chose to focus our ideas for new initiatives around foster youth and migrant youth as a type of "case study" of what can be done using the special target population model. We emphasize that the recommendations below represent merely the first, early steps of this type of effort. Much more remains to be done, but we are excited about the possibilities that future collaborative efforts hold for these groups of particularly needy young people. We also note that we anticipate that other special target populations will be identified in the future.

Education of Foster Youth Demonstration Program

The Task Force recommends the creation of a program designed to improve the quality of education for school-age youth in foster care. The program would be established at three levels: Federal, state, and local. It would involve the appointment of a point of contact at the Department of Education to assist in providing awareness of the barriers faced by foster care youth to improving their educational success, and a plan to encourage state and local school districts to establish a similar position in their education departments. Funding for this program could come from the existing sources available to State Education Agencies (SEAs) and Local Education Agencies (LEAs) for disadvantaged youth. [138]

Change is a difficult fact of life for the more than 500,000 children in foster care across the country. [139] Its impact on their personal lives is often difficult, and that impact is felt in their school lives for the approximately 70 percent of foster children who are school age (350,000 youth, 6 to 18 years of age). [140] 43 percent will remain in the child welfare system for two or more years. [141] Many will not stay in the same placement; they must move frequently and usually with only a few hours notice. Foster care youth repeatedly face disruptions in their life which affects the continuity of their education. The trauma caused by factors that remove these children from their families, coupled with their high mobility and inconsistent education, creates an educational nightmare for this population of youth. These experiences cause many factors leading to educational failure. [142]

Studies have shown that in general, children in foster care are doing poorly in our educational system and are too often not provided the opportunity to reach their full potential. The Casey Family Programs reported in Improving Special Education for Children with Disabilities in Foster Care that foster children have:

Higher rates of grade retention;

Lower academic skills as measured by standardized tests;

Higher absentee and tardy rates; and

Higher dropout rates. [143]

Changing schools and frequent absences due to court dates and mandatory meetings make it difficult for these students to stay at grade level. It is reported that more than 30 percent of foster care students are below grade level in reading and math. [144] Many of these students may be eligible for supplemental educational services provided under the Elementary and Secondary Education Act as reauthorized by the No Child Left Behind Act of 2001 (NCLB). [145] However, parents are most often the catalysts for finding the right provider and ensuring their children are receiving these services. Yet without the support of a parent or other educational advocate, children in foster care may miss their opportunity to receive these services.

Special education presents yet another set of problematic issues for foster care youth. They are three times more likely to be referred for special education and related services. [146] Many of these children are over-identified for services due to behavioral problems (often not surprising, given their problems at home), but equally disturbing is the fact that many of these children are under-identified or not receiving services due to many factors, including their mobility.

The governing legislation for special education issues, the Individuals with Disabilities in Education Act (IDEA), grants every child the right to a free, appropriate public education in the least restrictive environment. [147] The special education system provides parents the opportunity to advocate for their children. In cases where the parents have lost their rights, the law allows for surrogate parents to advocate on behalf of children. The surrogate parent represents the child in all matters relating to the identification, evaluation, and educational placement of the child. The criteria for selecting a surrogate parent is determined by state law. However, the regulations establish that a surrogate needs to have "knowledge and skills that ensure adequate representation," cannot have a conflict of interest with their representation of the child in education issues, and cannot be an employee of the State Educational Agency or Local Educational Agency. [148]

The role of foster parents in this process is tricky because many foster parents are a short term placement, and it is important for advocates to follow the child through high school and ensure IDEA services are being provided at each stage of their education. This is a difficult issue for states because the alternative may be an individual chosen to represent the child that may not know all of the history and special needs of that particular child. Another problem with foster care youth receiving special education services is that when children move to new schools their records tend to get lost or the transfer is delayed, which prolongs the time before they can begin to receive the necessary services to help them succeed in school. Finally, the lack of knowledge regarding IDEA by those individuals making educational decisions, such as caseworkers, foster parents, and the courts, also prevents them from taking advantage of special education services.

As the No Child Left Behind Act begins to improve the education of children within our country and uses student test scores to assess whether schools are educating our youth, the importance of addressing the education needs of children within the foster care system becomes increasingly more apparent.

Federal advocate for the education of foster youth: Currently, there is no point of contact at the U.S. Department of Education charged with looking specifically at these issues and coordinating with other Federal agencies to provide assistance to the states and localities. The Administration for Children and Families at the Department of Health and Human Services provides many of the programs that support foster care youth, but when dealing with problems in the education system, there is no one assigned to assist them with their questions.

The establishment of a position at the Department of Education with the responsibility to represent the needs of children in foster care and look for ways to address their barriers would be an asset to states, agencies, school districts, and the children in foster care they are serving. The purpose of this position would be to coordinate with other offices in the Department of Education to address the needs of foster care youth and provide more awareness to assist with their educational success. This officer would also serve as a liaison to other agencies, such as HHS. The officer in this position could work with other agencies to discuss these issues in providing better services to this community.

Encourage SEAs and LEAs to significantly improve the quality of education received by the foster care youth within their communities. States and local communities bear the primary responsibility of addressing children's educational needs. Some states have recognized this need and have begun the process of trying to improve the services provided to youth in care. For example, the Foster Youth Services program is a California State Initiative created to address the needs of foster children residing in Licensed Children Institutes or group homes. Some of the program's components provide for an Educational Liaison to assist group homes and local school districts in meeting the needs of foster youth and an inter-agency collaboration effort. Their system encompasses many more aspects that are improving outcomes for their youth in care. In Florida, the School Board of Broward County created an interagency agreement to improve information sharing between systems, improving services through collaboration. One of the methods included in the agreement is trying to ensure transportation issues do not prevent a child from staying in their school, even though their placement may be in another district. These are just some examples of states and local communities trying to improve the educational opportunities for these youth. Although steps are being taken, more can be done to find the best approaches for providing better and more comprehensive services for our youth in care.

States and local education agencies would be encouraged to use practices that are already improving outcomes for youth in care and to incorporate policies that would address their needs and provide more support for implementing effective programs. States and local education agencies would be given an opportunity to improve the services offered and evaluate those methods to ensure that the most effective practices are available for other states and local communities to model.

State Education Agency Demonstration Program: State demonstration projects would allow a state to assign an officer at the State Education Agency to help develop policies and programs to address the education needs of children in foster care. These demonstration programs would include an evaluation component to ensure that effective practices serve as models that other states throughout the Nation might modify and implement.

Program Elements:

Staffing: Identify a staff position in the education agency with full-time responsibility and organizational authority for coordinating interagency meetings between agencies that deal with issues of youth in foster care. The individual should have credentials, training, and experience needed for leadership and coordination of proposed activities; knowledge and experience in working with schools and child welfare personnel; and communication skills necessary to effectively promote and facilitate proposed plans and activities.

Partnerships and Planning: Develop and implement a state plan that addresses the education barriers (stated under proposal objectives) facing youth in foster care. The plan should be developed in collaboration with the state child welfare agency and any other agency with policies that affect this population, including but not limited to non-governmental agencies, teachers, foster parents, Court Appointed Special Advocates (CASAs), judicial staff, institutions of higher education, and other coalitions or groups. This plan should emphasize safety and education as priorities for these youth. Finally, the plan should leverage resources already available at the state and local level and avoid duplication.

Program activities should:

Coordinate interagency collaboration and facilitate training between the education, child welfare, and judicial systems.

Review the state and local policies for providing placement and school continuity for children in foster care. Work to develop a policy that meets this need.

Review state and local policies for identifying youth in foster care who need special education services. Develop a plan for better coordination between child advocates during IEP meetings and for transition plans and independent living plans. Also review the state and local policies with regard to surrogate parents. Incorporate practices that would allow for a new approach to providing surrogates for children in foster care.

Develop training opportunities and technical assistance to be provided for LEAs, teachers, foster parents, child welfare workers and judicial staff on the education needs of foster youth.

Develop a program to provide youth in foster care with a better awareness of their post-secondary options or vocational education opportunities.
Develop a system for securing, maintaining, and transferring education records of youth in foster care. This system should look at current successful projects being implemented in states and expand on their progress.

Provide an evaluation by State Education Agencies that includes but is not limited to: the strengths and weaknesses of the coordination efforts; how their policies have assisted local education agencies in providing better special education services to youth in care; how many foster care youth were transitioning out of care and seeking opportunities for higher education or vocational training based on programs implemented to provide better awareness and assistance to pursue those goals; providing information as to whether there is a correlation between the improved programs offered through the state and improved test scores; and a thorough explanation and critique of the system implemented for tracking student's records and credits.

Local Education Agency Demonstration Program: The LEA Foster Youth demonstration project should leverage resources and avoid duplication at the local levels. It would also include an evaluation component to ensure that effective practices and policies can be models for other LEAs.

Program Elements:

Staffing: Identify and establish a position in the local education agency with full-time responsibility and organizational authority for management and supervision of proposed activities. The individual should have the necessary credentials, training, and experience needed for leadership and coordination of proposed activities; knowledge and experience working with the school, child welfare and judicial systems; and communication skills necessary to effectively promote and facilitate proposed plans and activities. This individual should also have knowledge and experience dealing with IDEA and the development and coordination of IEP plans for youth with disabilities.

Monitoring of local school district and relevant state policies and programs and developing an action plan that would coordinate the efforts of all three systems. The action plan would establish a system for that school district to improve the educational outcomes for their students that are part of the child welfare system.
Partnerships and Planning: Grantees would be expected to develop and implement a district-wide plan to address the educational needs of children in foster care, and:

1. Work with the child welfare system and the education system to make education a priority for youth in foster care.

2. Develop a program for assuring that the educational needs of children are addressed and made a priority during court hearings for students within their school district.

3. Create a system for allowing students in the child welfare system to remain in the same school-or establishing a protocol for determining transfers on a case-by-case basis.

4. Review the policy for transferring records and implement a system to ensure records are transferred expeditiously.

5. Develop a system that better tracks the records and credits obtained by foster care youth.

6. Develop requirements for an IEP meeting and the participation of the child welfare worker, surrogate parent, foster parent, CASA worker (if applicable), student, and school personnel.

7. Facilitate training opportunities for teachers, administrators, case workers, foster parents, surrogate parents and judicial personnel to educate them on the educational needs of youth in foster care.

8. Develop a program for welcoming and providing support for students within the school district that have recently been placed in foster care or have transferred to their school because of their foster care situation. This plan may incorporate a new mentoring program.

9. Develop a program to introduce and encourage high school students in the foster care program to pursue higher education objectives or vocational education goals.

Evaluation: Local Education Agencies that receive funding would provide an evaluation that includes but is not limited to: the strengths and weaknesses of the coordination efforts; providing information as whether these programs improved student's test scores, their abilities to receive special education services and the LEAs ability to maintain school continuity for students; and a thorough explanation and critique of the system implemented for tracking student's records and credits.

Federal Interagency Committee to Focus on Education Needs of Foster Youth

The Task Force recommends the establishment of a new, ongoing interagency committee which will help improve Federal efforts to address the educational needs of youth in foster care. The committee should involve the appropriate representatives from the HHS Administration for Children and Families, the Department of Labor, and the Department of Education, and should plan to meet at least on a quarterly basis to ensure that the best efforts are made at the Federal level on behalf of these children.

The lives of children in foster care intersect many different systems: the child welfare agency, the courts, and schools. Two of these three systems, the courts and child welfare agencies, work together on a regular basis to address the needs of youth in foster care and to make the systems more successful in providing services. For example, caseworkers have a mandatory obligation to inform judges as to the placement of the child, efforts made by the parents, and details of the reunification or adoption process. An important aspect of helping these children overcome a life that has been so disrupted is to guarantee that their educational needs are being met, along with trying to provide them with a stable home and family.

But caseworkers and courts already have numerous responsibilities placed on them, making it difficult to focus effectively on the child's education. Caseworkers and the courts have little training with respect to the education system. The Individuals with Disabilities in Education Act (IDEA) and the No Child Left Behind Act (NCLB) provide many services that affect children in foster care, who tend to perform lower than their peers in math and reading and either need special education services or are placed in special education classes unnecessarily. When children change schools, health and education records often are not transferred expeditiously and sometimes are even lost. Individual education plans may be stalled in their implementation or not recognized at all. [149] This and many other problems cause additional barriers to providing these children with a quality education. However, educational outcomes for foster care youth could improve if all three systems collaborate to find better ways to support each other and the services they provide to foster care youth.

The Secretaries of Education, Labor, and Health and Human Services should appoint representatives from their respective departments to a Federal interagency committee. Representatives should be from the HHS Administration for Children and Families, the Office of Elementary and Secondary Education, the Office of Post Secondary Education, the Office of Special Education and Rehabilitative Services, the Office of Vocational and Adult Education, the Employment and Training Administration, the Office of Disability Employment Policy, and any other offices deemed appropriate. This interagency committee would be coordinated by the new officer appointed at the Department of Education to address education issues facing foster care youth. The purpose of this group would be to discuss the barriers preventing foster care youth from receiving a quality education and make education a priority in their lives. The following steps should be taken:

Address disabilities and special education needs: Discuss how to better identify those that need services; who should be included in creating the Individual Education Plan; who should be the advocate to ensure the plans are being implemented. Provide better guidance to the surrogate parent as an advocate; and address any other issues that may be presented through the reauthorization of the IDEA.
Provide technical assistance and training regarding the needs of foster care youth to those serving this population through the child welfare, court, and education system. Identify states and localities that have implemented systems that are models for other localities and are making improvements in how foster care youth are being educated.
Develop ideas for states and localities for a more efficient method for transferring student records.
Develop a plan to assist foster care youth in their transition into post-secondary education, vocational, and job training programs. Disseminate information on current Federal programs that would assist in this transition and develop resources that would address the gaps in their transition planning.
Conduct periodic meetings with youth in the foster care system to help the interagency committee get firsthand views of the difficulties they are facing within the system.
Address the mobility problem among youth in foster care, and discuss how provisions within the McKinney-Vento Homeless Assistance Act or other methods to ensure educational continuity could help these youth to stay in their original school even after new placements.

Workforce Training and Education Services for Migrant Youth

The Task Force recommends the creation of a joint venture between the Department of Labor, the Department of Education, and the Department of Agriculture to develop a model program to provide workforce training and placement services, and basic education services for high school completion to out-of-school migrant and seasonal farm worker youth ages 16-21.

There are many different groups of migrant youth with differing social and educational needs. The exact numbers of the adolescent migrant population is very difficult to track. A 2001 study on migrant adolescents conducted for the National Institute for Occupational Safety and Health estimated that of the nearly two million migrants and seasonal farm laborers, about 7 percent are between the ages of 14 and 17. Among these adolescents is a large and growing proportion of single males. Some are recent immigrants, while others may be school dropouts It is difficult to count dropout migrant youth due to their continuous mobility. [150]

Currently, there is a gap in the services provided to assist out-of-school migrant youth, ages 16-21. The Employment and Training Administration (ETA) at the Department of Labor, the Office of Migrant Education at the Department of Education, and the Cooperative State Research, Education, and Extension Service at the Department of Agriculture have an interest in working together in order to address the needs of this population. To fill in this gap in services to migrant out-of-school youth, these departments would work together to develop a pilot program. The model would combine workforce development services, including job training activities, with basic education services designed for out-of-school migrant youth, including whose with Limited English Proficiency (LEP), and would provide these youth with an integrated plan of services and activities designed to raise their educational skills and increase their employment opportunities.

Expand Mentoring Programs to Special Target Groups

The Federal government offers a wide array of mentoring services to disadvantaged youth. However, these services do not always reach the most vulnerable youth. As mentioned previously, migrant youth and youth in the foster care systems all may be in need of particular services but may not have the adult support that is necessary to successfully to pursue such assistance.

In the case of mentoring programs, it is particularly important to target young people who for one reason or another may need an additional caring adult in their lives. As disadvantaged youth navigate the various public systems that impact their lives (e.g., child welfare, juvenile justice, education), they can be greatly helped by an adult who really knows them and their needs and who can give them guidance while advocating on their behalf.

Mentors for Foster Care Youth. Children in the foster care system have suffered significant emotional losses as a result of the circumstances that lead to their removal from their families of origin. Whether it be through abuse or neglect, these children have often been betrayed by the adults they counted on the most. Yet in spite of the maltreatment they may have endured, many are even more traumatized by the separation from the only family they have ever known. This complex emotional dilemma can be further compounded by numerous moves within the child welfare system that do not allow them to form the long-term attachments necessary for healthy emotional development. Such life experiences make children in foster care especially good candidates for caring and committed mentors. In particular, they would benefit from mentors who have received the specialized training and support necessary to provide for consistent, long-term relationships for children in care.

Mentors could be assigned to foster care children at two different developmental points. First, mentors might be assigned to children (ages 5-17) when they first enter the child welfare system. These mentors might play the role of big brother/big sister and provide a continuous caring presence to bridge any changes in foster homes and caseworkers.

A second, particularly vulnerable time for children in the child welfare system, is when they are aging out of care. The transition to adulthood (age 18 for foster youth) is a difficult period for most adolescents. While adolescence is a time of separation and individuation, teens are tremendously dependent on the adults in their lives to help make successful transitions to adulthood. For foster children, becoming an adult may reflect a permanent separation from the only source of support they have known: the child welfare system. In some cases, children with foster families are able to maintain this support, but often aging out of foster care at 18 means aging out of a family.

A mentor might help a young person make this transition by helping sort through the variety of decisions they face as they move to independence. For some, college may be an option. If so, they will need help with numerous obstacles: applications are complicated; the choice of the right school can be overwhelming; and even getting situated on campus can be disorienting. There are opportunities for financial assistance, but many youth are not aware of this assistance, and the financial aid packets can be daunting even for adults to complete.

For the foster care youth for whom college is not an immediate option, a mentor can help with finding appropriate vocational training, housing, and jobs. An adult with a strong employment history of his or her own can help provide work readiness skills and provide the support that is needed as a young person experiences the ups and downs of entering the labor market.

Mentors for Migrant Youth. The migrant youth population cannot be characterized as one homogenous group. They arrive in this country many different ways and stay for many different reasons. Some move north from Mexico at the very young age of 13 to find work as farm workers harvesting whichever crop is in season. Others come with family and move as crop seasons end.

Both of these groups of youth face different types of barriers. Some need the guidance of an older adult to assist them through their pursuit of a better life and to steer them away from risk behaviors. Others have family and support but need a responsible peer or adult to assist them with things such as educational goals and career paths. No matter what the situation is for the migrant youth, this population of youth would benefit from a specially-trained mentor.

Bi-national migrant youth face many pressures. They range in age from 13 to 22 and have entered the country in the search for money to support themselves and their families. Approximately 7 percent of all Seasonal Agriculture Services farm workers are between the ages of 14 and 17. [151] A majority of these youth come to this country as a group. They find themselves living in small quarters with many people and being introduced to destructive behavior. They are not in school and are vulnerable youth being introduced to a difficult way of life. A responsible adult mentor could provide the necessary guidance to assist them through their situation and onto a productive path.

Many migrant youth live with their families who are farm laborers and move frequently to harvest the next crop. These youth have supportive parents who want the best educational opportunities for their children. However, there are still many obstacles to their educational success. They frequently change schools and are at a higher risk of school failure. They tend to enter in the middle of a semester or quarter, subjecting them to new hallways, subjects, teaching styles, and peers. Many times the obstacles to satisfying requirements for graduation become too challenging, and their need to make financial contributions to their families diverts them from education to seeking employment.

Migrant parents often have limited knowledge of formal schooling and the intricacies of our educational system. Migrant students find themselves trying to navigate the system on their own without advocates to assist in acclimating them to their new school settings and the unfamiliar culture. These obstacles are even more difficult to overcome for the limited English proficiency students. Providing a mentor could be the little extra assistance they need to be successful. A mentor could be an older migrant student who is now familiar with the system, a teacher, or a community resident. Mentors could provide the encouragement, tutoring, and assistance these students need to complete their education and pursue additional educational goals and aspirations.

ENDNOTES

EXECUTIVE SUMMARY

[1] Statement by the President/Executive Memorandum, "White House Task Force for Disadvantaged Youth," December 23, 2002. www.whitehouse.gov /news/releases/2002/12/20021223.html

[2]A D&B® D-U-N-S® Number is a unique nine-digit sequence and is used as the standard for all Federal government electronic commerce transactions to help streamline and reduce Federal procurement costs.

[3]"The President's Management Agenda, Fiscal Year 2002," page 3, www.whitehouse.gov/omb/budget/fy2002/mgmt.pdf

[4]Gary Walker, "The Policy Climate for Early Adolescent Initiatives," Public/Private Ventures, January 2001, found at http://www.ppv.org/ppv/publications/assets/73_publication.pdf.

[5] Ibid.

[6]White House Task Force for Disadvantaged Youth, Preliminary Report on Findings for the Federal Response to Disadvantaged Youth, April 30, 2003.

[7]Eric Erikson, Childhood and Society, New York, N.Y., Norton, 1963, and Identity: Youth and Crisis, (same publisher), 1968, cited by National Research Council, Board on Children, Youth, and Families, in "Community Programs to Promote Youth Development", 2002, page 70.

[8]Gary Walker, President, Public/Private Ventures, letter to Margaret Spellings, Assistant to the President for Domestic Policy, March 4, 2003.

[9] Ibid.

[10] SEAs and LEAs can use Title I, IDEA and often, Education for Homeless Children and Youth funds for these kinds of activities.

CHAPTER 1

[11] For the purposes of the work of the Task Force, the following definition of "disadvantaged" was developed: "Youth who, because of certain characteristics, circumstances, experiences or insufficiencies, encounter financial, legal, social, educational, emotional and/or health problems and may have significant difficulties growing into adults who are responsible citizens, productive workers, involved members of communities, and good parents."

[12]President George W. Bush, "National Child's Day 2002, A Proclamation," June 5, 2002. www.whitehouse.gov/news/releases/2002/06/ 20020605-3.html

[13]"Statement by the President," December 23, 2002, op cit.

[14]Data sources include various publications of the U.S. Census Bureau, the National Center for Health Statistics (Centers for Disease Control and Prevention), "2002 National Survey on Drug Use and Health" (Substance Abuse and Mental Health Services Administration) and "Trends in the Well-Being of America's Children and Youth, 2002," (Office of the Assistant Secretary for Planning and Evaluation). The latter three agencies are part of the U.S. Department of Health and Human Services).

[15]Office of Management and Budget, "The President's Budget, Fiscal Year 2003." NOTE: References throughout this report to the President's Budget or the Administration's legislative proposals, may be reviewed at www.whitehouse.gov/omb. Specific program legislation may be reviewed at http://thomas.loc.gov.

[16]Institutions of higher education include community colleges.

[17]The section on school connections reflects the consensus statement from the Wingspread Conference on School Connections, June 2003, sponsored by the Centers for Disease Control and Prevention, and the University of Minnesota, Division of General Pediatrics and Adolescent Health. "The Wingspread Declaration on School Connections" is based on the research emanating from the "National Longitudinal Study on Adolescent Health," (ongoing).

[18]Ibid.

[19]Robert W. Blum and Peggy Mann Rinehart, "Reducing the Risks: Connections That Make a Difference in the Lives of Youth," University of Minnesota, Division of General Pediatrics and Adolescent Health, a monograph based on the first analysis of Add Health data, "Protecting adolescents from harm: Findings from the National Longitudinal Study on Adolescent Health," published in the September 10, 1997, issue of the Journal of the American Medical Association, written by: Michael D. Resnick, Ph.D., Peters S. Bearman, Ph.D., Robert Wm. Blum, M.D., Ph.D., Add Health Project Group: Karl E. Bauman, Ph.D., Kathleen M. Harris, Ph.D., Jo Jones, Ph.D., Joyce Tabor; Minnesota Analysis Group: Trish Beuhring, Ph.D., Renee E. Sieving, Ph.D., Marcia Shew, M.D., M.P.H., Marjorie Ireland, Ph.D., Linda H. Bearinger, Ph.D., M.S., J. Richard Udry, Ph.D., Principal Investigator.

[20]President George W. Bush, State of the Union, 2003, www.whitehouse.gov/news/releases/2003/01/20030128-19.html.

[21]The Center for Information and Research on Civic Learning and Engagement, "Youth Civic Engagement: Basic Facts and Trends," January 9, 2002.

[22]President George W. Bush, op cit.

[23]Ibid.

[24]White House Task Force for Disadvantaged Youth, op cit., unpublished analysis of survey data.

[25]U.S. Department of Health and Human Services, Administration for Children and Families, "FY 2003 Runaway and Homeless Youth Program Funding Announcement," page 75, found at http://www.acf.hhs.gov/ programs/fysb/rhyfund-anncmt.htm.

[26]U.S. Department of Health and Human Services, Administration for Children and Families, National Clearinghouse on Families and Youth, "Putting Real Youth Participation Into Practice," The Exchange, July, 2002.

CHAPTER 2

[27]See list of data sources at endnote 12, Chapter One.

[28]White House Task Force for Disadvantaged Youth, op cit., Appendix I.

[29] In our April, 2003 preliminary report, the Task Force identified 335 youth-serving programs. In the interim period, we identified four additional programs, all within the Department of Defense. For a listing of all youth-serving programs we identified, see the Appendices of that report.

[30]Tina Shah, Rockford Register Star, August 5, 2002

[31]U.S. Department of Labor, "Strategic Plan, Fiscal Years 1999 2004," http://www.dol.gov/_sec/stratplan/main.htm.

[32]Ibid.

[33]Center for the Study and Prevention of Violence, "Blueprints for Violence Prevention," University of Colorado, http://www.colorado.edu/cspv/ blueprints/index.html.

[34]McGill, D.E., Mihalic, S.F., & Grotpeter, J. K. (1998), "Blueprints for Violence Prevention, Book Two: Big Brothers Big Sisters of America," Boulder, CO, Center for the Study and Prevention of Violence. www.colorado.edu/ cspv/blueprints/model/programs/BBBS.html

[35]The proposal in the President's Budget is for this program to be $100 million, which does not reflect the final appropriation.

[36]The President's Budget for Fiscal Year 2003.

[37]Family and Youth Services Bureau, U.S. Department of Health and Human Services, State Youth Development Collaboration Projects, www.acf.hhs.gov/programs/fysb/content/youthdivision/initiatives/statecollabfactsheet.htm.

[38]The number of grant programs often changes annually.

[39]31 U.S.C. Section 6104.

[40] www.cfda.gov, also http://aspe.hhs.gov/cfda/intro.htm#overview.

[41]The agencies' grant websites normally list information about specific grants for only a limited time period.

[42]FAADS is available at http://www.census.gov/govs/www/faads.html. It is searchable only by fiscal quarter. The data can be searched for a full fiscal year only through the CD-ROM version, available for sale.

[43]See endnote 2, Executive Summary.

[44]White House Task Force for Disadvantaged Youth, loc. cit.

[45]Executive Orders are available online by date at www.whitehouse.gov/ news/orders/.

CHAPTER 3

[46]"The President's Management Agenda, Fiscal Year 2002," page 3, www.whitehouse.gov/omb/budget/fy2002/mgmt.pdf

[47] Gary Walker, op. cit., January 2001.

[48] Ibid.

[49]U.S. Department of Education, What Works Clearinghouse, "About the WWC," available from: http://www.w-w-c.org/about.html, accessed August 5, 2003.

[50]Center for the Study and Prevention of Violence, "Blueprints for Violence Prevention Overview," University of Colorado, Boulder, CO, available from: www.colorado.edu/cspv/blueprints/index.html, accessed June 30, 2003.

[51]U.S. Department of Health and Human Services, Substance Abuse and Mental Health Services Administration, "SAMHSA Model Programs: Effective Substance Abuse and Mental Health Programs for Every Community," available from: http://www.modelprograms.samhsa.gov, accessed August 19, 2003.

[52]The Community Guide is being developed by a, non-Federal Task Force on Community Preventive Services (Task Force), appointed by the Director of the Centers for Disease Control and Prevention (CDC). This group was convened in 1996 by the Department of Health and Human Services to provide leadership in the evaluation of community, population, and health care system strategies to address a variety of public health and health promotion topics such as physical activity. See http://www.thecommunityguide.org

[53]Prevention Evaluation Research Registry for Youth (PERRY)
A comprehensive database of research on adolescent risk behaviors, PERRY includes both school-based and community-based studies and organizes them according to topic and research methods. PERRY currently contains citations on sexual risk reduction and violence prevention. In the future, other topical areas will be added, and PERRY will be made available to the public. PERRY supports searches on topics of interest, and provides a foundation for conducting summaries of the research literature. Such summaries are useful in informing current policy and practice and in identifying gaps in the research literature. One such summary is "School-based Programs to Reduce Sexual Risk Behaviors: A Review of Effectiveness," Public Health Reports, May-June 1994, Vol . 109, No. 3, 339-359.

[54]The White House Task Force for Disadvantaged Youth, "The Federal Response to Disadvantaged Youth: April, 2003 Preliminary Report" (Washing-ton, D.C.: April, 2003), pp. 32-36, stating: "6 of the 28 youth programs (21 percent) rated were scored as 'ineffective' by the OMB PART system. Thirteen youth-related programs were rated as 'results not demonstrated.' Five were ranked as 'adequate.' Three were rated as 'moderately effective.' Only a single youth program (Consolidated Health Centers, which addresses disadvantaged youth as only one part of the population it serves) was given the highest rating, 'effective.'"

[55]Quoted in Rob Hollister, "The Growth in After‑School Programs and Their Impact" (Washington, D.C.: The Brookings Institution, February 2003), p. 12, http://www.brookings.edu/views/papers/sawhill/20030225.pdf, accessed August 21, 2003.

[56]Hollister, op. cit., p. 12.

[57]We use the term "program group" because it seems to encompass all the other variations on the same concept, including "experimental," "treatment," and "intervention" groups-as long as they have been randomly assigned to the group. We do not use "experimental" group because in an evaluation of an ongoing program, the term would erroneously suggest that something new is being tested. In addition, the term would not apply in the case of a nonexperimental evaluation. Similarly, we do not use the terms "treatment" or "intervention" group, because the program or policy being tested may not be perceived as a "treatment" or "intervention" by those participating in an ongoing program.

[58]The terms "control" group and "comparison" group are sometimes used interchangeably. However, to emphasize the difference between randomized experiments and nonexperimental evaluations, we limit the use of "control" group to the nonprogram groups created by random assignment. We are careful to call the nonprogram groups in nonexperimental evaluations "comparison" groups.

[59]The ability of an evaluation to establish a cause-and-effect relationship between an outcome and a program or policy being evaluated is usually called "internal validity." This paper, however, adopts a variant of a revised term proposed by Donald Campbell: "causal validity." Campbell's full description of term was "local molar causal validity." See Donald T. Campbell, "Relabeling internal and external validity for applied social scientists," in W. M. K. Trochim (editor), Advances in quasi-experimental design: New directions for program evaluation (San Francisco, CA: Jossey-Bass, 1986), pp. 66-77.

[60]Gary Burtless, "The Case for Randomized Field Trials in Economic and Policy Research," Journal of Economic Perspectives vol. 9, no. 2, Spring 1995, p. 69.

[61]Erica Baum, "When the Witch Doctors Agree: The Family Support Act and Social Science Research," Journal of Policy Analysis and Management vol. 10, Fall 1991, pp. 603-615.

[62]Larry L. Orr, Howard S. Bloom, Stephen H. Bell, Winston Lin, George Cave, Fred Doolittle, "The National JTPA Study: Impacts, Benefits, and Costs of Title II-A," (Bethesda, MD: Abt Associates Inc., Mar. 1994).

[63]David Greenberg and Mark Shroder, Digest of Social Experiments (Washington, D.C.: The Urban Institute Press, 1997).

[64]Evaluations that do not involve random assignment are often called "quasi-experiments" since they often involve the comparison of the outcomes of program participants to those of a similar "comparison" group of nonparticipants. Unfortunately, using the term "quasi-experiment" tends to obscure the problem of uncertain causal validity inherent in all nonexperimental approaches. Hence, like an increasing number of commentators, we do not use the term "quasi-experiment" but rather use the term "nonexperimental" to cover both micro- and macro-econometric evaluations.

[65]See footnote 9, supra.

[66]Sometimes the comparisons across geographic areas also follow individuals over time.

[67]Hollister, op. cit., p. 8.

[68]See, e.g., Matthew Stagner, Jennifer Ehrle, and Jane Reardon-Anderson, "Systematic Review of the Impact of Mandatory Work Policies on Family Structure," (Washington, D.C.,: The Urban Institute, February 24, 2003).

[69]For a summary of the major evaluations of youth programs, see Rob Hollister, op. cit., p. 10. See also Peter Benson and Rebecca N. Saito, "The Scientific Foundations of Youth Development," Youth Development: Issues, Challenges and Directions, Fall 2000, available from: http://www.ppv.org/ppv/publications/assets/74_publication.pdf, accessed August 25, 2003; Richard M. Catalano, Lisa Berglund, Jeanne A. M. Ryan, Heather S. Lonczak, and J. David Hawkins, Positive Youth Development in the United States: Research Findings on Evaluations of Positive Youth Development Programs, Social Development Research Group (Seattle, WA: Social Development Research Group, University of Washington, November 13, 1998), available from: http://aspe.hhs.gov/hsp/PositiveYouthDev99/, accessed August 25, 2003; Mark Dynarski, Suzanne James-Burdumy, Wendy Mansfield, Daniel Mayer, Mary Moore, John Mullens, and Tim Silva, A Broader View: The National Evaluation of the 21^st Century Learning Centers Program, Design Report: Volume 1 (Washington, D.C.: Mathematica Policy Research, Inc., March 2, 2001, available from: http://www.mathematica-mpr.com/PDFs/broadviewvol1.pdf, accessed August 25, 2003; Jacquelynne S. Eccles and Janice Templeton, "Community-Based Programs for Youth: Lessons Learned from General Developmental Research and From Experimental and Quasi-experimental Evaluations," Urban Seminar on Children's Health and Safety, John F. Kennedy School of Government, Harvard University, 2001; Olatokunbo S. Fashola, "Review of Extended-Day and After-School Programs and Their Effectiveness," Report 24, Center for Research on the Education of Students Placed at Risk, Johns Hopkins University, 1998; and Jodie Roth, Jeanne Brooks-Gunn, Lawrence Murray, and William Foster, "Promoting Healthy Adolescents: Synthesis of Youth Development Program Evaluations," Journal of Research on Adolescence, vol. 8, no. 4, 1998, pp. 423‑459. See generally Peter H. Rossi, Howard E. Freeman, and Mark W. Lipsey, Evaluation: A Systematic Approach 6, 6th ed. (Newbury, CA: SAGE Publications, Inc., 1998).

[70]For example, a random assignment evaluation of the Memphis Extended-Day Tutoring Program was apparently subverted when children assigned to the program group who did not attend the program (or had low attendance rates) were added to the control group. Rob Hollister observes, "Adding to the control group members of the group initially assigned to the program group but selected out because of non-attendance, or some other reason, seriously undermines the strength of the initial random assignment in avoiding selection bias."

[71] In the late 1980s, for example, the state of Texas implemented a random assignment evaluation to test the impact of twelve-month transitional child care and Medicaid benefits. When the study began, the program group was receiving a benefit (the transitional services) that was otherwise unavailable. Hence, denying the same benefit to the control group did not raise an ethical issue. But a year later, nearly identical transition benefits became mandatory under the Family Support Act. At that point, the control group was to be denied what had become part of the national, legally guaranteed benefit package. Thus, the Secretary of Health and Human Services required the control group to receive the benefits, and the experimental design was terminated.

[72]Quoted in Rob Hollister, op. cit., p. 12. Hollister was referring to Richard M. Catalano, Lisa Berglund, Jeanne A. M. Ryan, Heather S. Lonczak, and J. David Hawkins, Positive Youth Development in the United States: Research Findings on Evaluations of Positive Youth Development Programs, Social Development Research Group (Seattle, WA: Social Development Research Group, University of Washington, November, 1998), available from: http://aspe.hhs.gov/hsp/PositiveYouthDev99/, accessed August 25, 2003.

[73] Hollister, op. cit., p. 8.

[74]This problem is exacerbated by what is commonly referred to as the "file-drawer problem," the publication bias of journals for studies that show significant results can offer a one-sided view of the evidence. The idea is that for every study with significant results that is published, there may be many more with insignificant results languishing in file drawers unpublished.

[75]Morton Hunt, How Science Takes Stock: The Story of Meta-Analysis (New York, NY: Russell Sage Foundation, 1997). See, e.g., Mark W. Lipsey, "Juvenile Delinquency Treatment: A Meta-Analytic Inquiry into the Variability of Effects, in Thomas D. Cook, Harris Cooper, David S. Cordray, Heidi Hartmann, Larry V. Hedges, Richard J. Light, Thomas A. Louis, and Frederick Mosteller (editors), Meta‑Analysis for Explanation: A Casebook (New York, NY: Russell Sage Foundation, 1992), pp. 83-127.

[76]Douglas J. Besharov, Peter Germanis, and Peter H. Rossi, Evaluating Welfare Reform: A Guide for Scholars and Practitioners (College Park, MD: University of Maryland, School of Public Affairs, 1997).

[77]U.S. Department of Education, "What Works Clearinghouse: Evidence Report Topics," available from http://www.w-w-c.org/topicnom.html, accessed June 30, 2003.

[78]U.S. Department of Education, "Introduction to the What Works Clearinghouse Evidence Report Process and the Role of Scientific Standards," March 5, 2003.

[79]National Institutes of Health, "About the Consensus Program: Frequently Asked Questions," available from: http://consensus.nih.gov/FAQs.htm, accessed August 5, 2003.

[80]National Institutes of Health, "Guidelines for the Planning and Management of NIH Consensus Development Conferences," available from http://consensus.nih.gov/ABOUTCDP.htm, accessed August 5, 2003.

[81]National Institutes of Health, "Guidelines for the Planning and Management of NIH Consensus Development Conferences," available from http://consensus.nih.gov/ABOUTCDP.htm, accessed August 5, 2003.

[82]National Institutes of Health, "Guidelines for the Planning and Management of NIH Consensus Development Conferences," available from http://consensus.nih.gov/ABOUTCDP.htm, accessed August 5, 2003.

[83]Olds, D., Hill, P., Mihalic, S., & O'Brien, R. (1998). Blueprints for Violence Prevention, Book Seven: Prenatal and Infancy Home Visitation by Nurses. Boulder, CO: Center for the Study and Prevention of Violence.

[84]Solicitation Notice, "Evaluation of Independent Living Programs Funded Under the Chaffee Foster Care Independence Program," May 7, 2002, Department of Health and Human Services, (Contracting Office), Rockville, MD. www.fbodaily.com/archive/2002/05-May/09-May-2002/FBO-00072724.htm.

[85]White House Task Force for Disadvantaged Youth, Preliminary Report on Findings for the Federal Response to Disadvantaged Youth, April 30, 2003.

[86]The Office of National Drug Control Policy sought input from numerous stakeholders in developing and revising its cross-agency Performance Measures of Effectiveness. See the list of individuals and organizations whose input was sought in National Drug Control Strategy. (Washington, DC: Office of National Drug Control Policy, February 2002).

[87]For the purposes of this discussion, a youth "earmark" is defined as funds designated by Congress to go to a specific group for a purpose relating to youth, allowing that organization to bypass the normal competitive grants process. There are two kinds of earmarks. Hard earmarks are where Congress directs funds to a specific program it has identified. Soft earmarks are where Congress identifies a program and directs the Federal agency to evaluate the program and fund it, if warranted.

[88]White House Task Force for Disadvantaged Youth, op. cit., page 4.

[89]James O. Finckenauer, Scared Straight! and the Panacea Phenomenon, Englewood Cliffs, N.J.: Prentice-Hall, Inc., 1982.

[90]Anthony Petrosino, Carolyn Turpin-Petrosino, and James O. Finckenauer, "Well-Meaning Programs Can Have Harmful Effects! Lessons From Experiments of Programs Such as Scared Straight," Crime and Delinquency, vol. 46, no. 3, July 2000, pp. 354-379.

[91]"How Congressional Earmarks and Pork-Barrel Spending Undermine State and Local Decision-making," by Dr. Ronald D. Utt, Backgrounder #1266 , April 2, 1999. Executive Order 12372 "Intergovernmental Review of Federal Programs," structures the Federal government's system of consultation with state and local governments on its decisions involving grants, other forms of financial assistance, and direct development. Under E.O. 12372, States, in consultation with their local governments, design their own review processes and select those Federal financial assistance and direct development activities they wish to review. Earmarked programs are able to bypass this system.

[92]Ibid.

[93]Department of Defense Education Activity: An Overview. (2003, April). Retrieved from: http://www.dodea.edu/aar/2003/pdf/overview2003.pdf.

[94]Smrekar, C., Guthrie, J.W., Owens, D.E., and Sims, P.G. (2001, September). March Toward Excellence: School Success and Minority Student Achievement in Department of Defense Schools. A Report to the National Education Goals Panel. Retrieved from: http://govinfo.library.unt.edu/negp/reports/DoDFinal921.pdf.

[95]United States General Accounting Office. (2001, September). BIA and DOD Schools: Student Achievement and Other Characteristics Often Differ from Public Schools'. Retrieved from: http://www.gao.gov.

[96] Anderson, L.B., Bracken, J. and Bracken, M.C. (2000, October). Review of Department of Defense Education Activity (DODEA) Schools. Volume II: Quantitative Analysis of Educational Quality. Institute for Defense Analyses Paper P-3544, handle.dtic.mil/100.2/ADA394234.

[97]National Center for Education Statistics. (2003). Nation's Report Card. Retrieved from: http://nces.ed.gov/nationsreportcard/states/

[98]Smrekar, op. cit., Anderson, op. cit.

CHAPTER 4

[99] Eric Erikson, op cit.

[100]Greg Toppo, "Teens with multiple choices make surprising one; Survey has good news for parents," USA Today, McLean, Va., August 6, 2003, page D.05.

[101]Iris F. Litt, "Parents are "in" again," Journal of Adolescent Health, August, 2003, (editorial) and additional articles in this issue.

[102]There are three types of grant application reviews generally used for funding social programs. On-site grant reviews take place over about 5 days, usually in a hotel in the Washington, D.C., area, and involve chaired groups of about 4 (often non Federal) panelists in each group, coordinated by Federal managers. All panel members in a group review the same applications, and each group writes a consensus opinion. Field reviews mean that reviewers do most of their review work at their home or office, and then gather with other panelists (often a larger group, about 10 persons) at a hotel often in the D.C. area. The third type of review involves contact with fellow panelists either via the Internet or telephone conference calls.

[103]As reported by Harry Wilson, Associate Commissioner, Family and Youth Services Bureau, U.S. Department of Health and Human Services, and other staff of the bureau, summer, 2003.

CHAPTER 5

[104]Gary Walker, op cit., January 2001.

[105]Ibid.

[106]Data sources on foster care youth include "Foster Care National Statistics, June, 2003," National Clearinghouse on Child Abuse and Neglect Information, www.childwelfare.gov/pubs/factsheets/foster.pdf.

[107]Adoption and Foster Care Analysis and Reporting System (AFCARS) Report, Preliminary Fiscal Year 2001 estimates as of March 2003.

[108]"Improving Special Education for Children with Disabilities in Foster Care," (Education Issue Brief), Casey Family Programs, June, 2002.

[109]Ibid.

[110]R. Jean Haurin, "Patterns of Childhood Residence and the Relationship to Young Adult Outcomes," Journal of Marriage and the Family 54 (1992): 846-860.

[111]Leighton Ku et al, "Factors Influencing First Intercourse for Teenage Men," Public Health Reports 108 (1993): 680-694; Sara C. Carpenter, "The Association of Foster Care or Kinship Care with Adolescent Sexual Behavior and First Pregnancy, " Pediatrics 108, No. 3 (2001): e46.

[112]See http://www.aecf.org/upload/publicationfiles/youth who age out.pdf

[113]Richard Wertheimer, "Youth who 'Age Out' of Foster Care: Troubled Lives, Troubling Prospects," Child Trends, Inc., December, 2002, www.childtrends.org/Files/FosterCareRB.pdf.

[114]Ibid.

[115]See Judith S. Brook et al., "Sequence of Sexual Behavior and Its Relationship to Other Problem Behaviors in African-American and Puerto Rican Adolescents," Journal of Genetic Psychology 1 (1994): 107-114; David Huizinga et al., "Longitudinal Study of Delinquency, Drug Use, Sexual Activity, and Pregnancy Among Children and Youth in Three Cities," Public Health Records 108 (1993): 90-96; Jeffrey Fagan et al., "Delinquency and Substance Use Among Inner-City Students," Journal of Drug Issues 20 (1990): 351-402; and Anne C. Petersen et al., "Social Changes Among Youth," Journal of Adolescence Health 14 (1993): 632-637.

[116]Wertheimer, op. cit.

[117]Loper, "Female Juvenile Delinquency Rates."

[118]Bill McCarthy, "Mean Streets: The Theoretical Significance of Situational Delinquency Among Homeless Youths," American Journal of Sociology 98 (1992): 597-628.

[119]Robert E. Larzelere et al., Parental Management: Mediator of the Effect of Socioeconomic Status on Early Delinquency," Criminology 28 (1990): 301-323.

[120]Jeffrey Fagan et al., "Contributions of Delinquency and Substance Use to School Dropout Among Inner-City Youths," Youth and Society 21 (1990): 306-354.

[121]E. R. Temblay et al., "Disruptive Boys with Stable and Unstable High Fighting Behavior Patterns During Junior Elementary School," Journal of Abnormal Child Psychology 19 (1991): 285-300.

[122]National Center on Education, Disability, and Juvenile Justice, "Characteristics of Youth and Educational Programming in Juvenile Corrections;" EDJJ Notes, January, 2002, www.edjj.org.

[123]Gary Walker, op cit., January, 2001.

[124]Federal Register: May 16, 2003 (Vol. 68, No. 95), pp. 26622-26628.

[125] Ibid.

[126]U.S. Department of Labor, "Report on the Youth Labor Force," June, 2000, revised November, 2000.

[127] Ibid.

[128] Ibid.

[129]National Adolescent Farmworker Occupational Health and Safety Advisory Committee, "Migrant and Seasonal Hired Adolescent Farmworkers: A Plan to Improve Working Conditions," 2001.

[130] U.S. Department of Labor, op. cit.

[131] Findings from the "National Agriculture Workers Survey," 2000.

[132] Ibid.

[133] Ibid.

[134] Ibid. NOTE: Needs-based services include financial aid through programs such as TANF, general assistance or welfare, and publicly provided housing and medical and nutritional assistance such as WIC, Food Stamps, and Medicaid.

[135] Migrant & Seasonal Head Start Quality Improvement Center, "Special Needs of Migrant Farmworker Population."

[136] U.S. Department of Labor, op cit.

[137] Migrant & Seasonal Head Start Quality Improvement Center, op. cit.

[138] SEAs and LEAs can use Title I, IDEA and often, Education for Homeless Children and Youth funds for these kinds of activities.

[139]Children's Bureau, U.S. Department of Health and Human Services, Adoption and Foster Care Analysis and Reporting System (AFCARS) report from Fiscal Year 2001.

[140]Ibid.

[141]Ibid.

[142]Casey Family Programs, "Improving Special Education for Children with Disabilities in Foster Care," Education Issue Brief, June, 2002.

[143]Ibid.

[144]Child Welfare League of America, "Improving Educational Outcomes for Youth in Care," Symposium Summary Report.

[145]20 U.S.C. 7101

[146]"Caught Between Two Systems: How Exceptional Children in Out-of-Home Care are Denied Equality in Education," Yale Law and Policy Review, 10 (2000).

[147]Individuals with Disabilities in Education Act, 20 U.S.C. 1472.

[148]Education Department, General Administrative Regulations, 34 C.F.R. 300.515.

[149]"IEP" means a written statement for a child with a disability that is developed, reviewed, and revised in a meeting in accordance with §§ 300.341-300.350. It is a plan that creates expected goals and outcomes for the child with a disability. A complete listing of the content found in an IEP can be found at 34 CFR 300.347.

[150]National Adolescent Farmworker Occupational Health and Safety Advisory Committee, op.cit.

[151]Kissam, Edward; "No Longer Children," Aguirre International, 2001.