To address the diversity of these student problems and behaviors, school systems are attempting to implement multifaceted and coordinated approaches in collaboration with public and private community agencies and parents. Long-standing mandates for parental and community involvement exist in the most prominent federal compensatory education programs such as Head Start, Title I, and Chapter 1. However, "the shared responsibilities of families, schools, and communities are not well understood nor well-developed in family practice, school practice, or community practice" (Center on Families, Schools, Communities, and Children's Learning, 1990, p. 1).
Two distinct but compatible perspectives have emerged regarding how to deal with the deterioration within and among schools, communities, and families (Natriello, McDill, and Pallas, 1990, pp. 194-197). Wilson's (1987) prescription advocates reindustrializing and economically revitalizing inner cities. A second perspective, which is our concern here, argues for a strengthening of the bonds among the key educating institutions -- the family, community, and school -- to educate students at risk more successfully. As noted by Natriello, McDill, and Pallas (1990, p. 196), the most advocated strategies for strengthening these bonds include greater family involvement in the school and making the school more prominent in the lives of the family and community. In terms of family-school partnerships, collaboration, or "connections," Epstein (1992, in press) has developed a sixfold typology to codify research and link research to practice:
A specific way in which some schools are addressing the personal problems that impede students' learning is by integrating and coordinating the social services many students at risk need. "Joining Forces" (Levy, 1989) is a national effort to help education and human services professionals at both the state and local levels collaborate in aiding children and families at risk. The initiative was undertaken in 1987 by the National Association of State Boards of Education with an initial grant from the Ford Foundation. In 1989, sponsorship was transferred jointly to the Council of Chief State School Officers and the American Public Welfare Association with financial support from several private foundations. Specific objectives are to promote dialogue among education and human service agencies; to collect, collate, and disseminate information on successful examples of collaboration; to assist states in the development and evaluation of collaborative approaches through techniques such as on- site technical assistance; and to foster supportive action at the national level.
A more recent national programmatic effort is the National Center for Service Integration, established in 1991 with support from the U.S. Department of Health and Human Services. "The broad purpose of the Center is to stimulate, guide, and actively support service integration efforts across the country" (Mathtech, Inc., 1991, p. 1). The Center is engaged in two broad programmatic activities: operating an information clearinghouse on services integration and providing technical assistance to field-based services integration initiatives. The initial attention of the Center has focused on integration of educational, health, and related human services directed to families and their children.
The most ambitious integrated services program at the state level is the School Based Youth Services Program (SBYSP), funded at $6 million annually and launched in 1988 by the New Jersey Department of Human Services in collaboration with the departments of labor, education, and health (New Jersey Department of Human Resources, 1988). This program offers a wide array of human and educational resources to youth at risk between the ages of 13 and 19, including employment counseling, job training and placement, summer and part-time job development, drug and alcohol abuse counseling, family crisis counseling, academic counseling, primary and preventive health services, recreation, and referral to health and social services. All state funds support local initiatives, and all applications for state funding must be submitted jointly by the school district and at least one public or private service agency. The Department of Human Services requires demonstrations of broad community involvement, and local advisory boards are required to have multisector representation (Levy, 1989).
To maximize accessibility of sites to clients, local centers are located at or near schools. There are 29 program sites throughout the state, with at least one site located in each of the state's 21 counties. SBYSP does not impose a single model or programmatic approach on any community. Rather, the constellation of staff and services represents an adaptation to local needs (Levy, 1989, p. 12). However, all project sites are required to provide core services such as counseling, health and substance abuse services, employment counseling and training services, information and referral services, and recreational opportunities.
Unfortunately, SBYSP conforms to the norm recently observed by Dryfoos (1991) in her extensive critique of integrated services programs: "Few of these efforts in or out of schools have been adequately or systematically evaluated" (p. 123). The state has not undertaken a systematic evaluation of the program to monitor its long-term effects. Instead, it appears that the cognizant agency is relying on "anecdotal" and "baseline" information to support its position that "they are on the right track" (Cohen, 1989, p. 12).
The New Jersey model has been adapted by Kentucky as part of its Kentucky Integrated Delivery System (KIDS) program, a collaborative effort between the State Department of Education and the Cabinet for Human Resources to meet the personal, social, and educational needs of students. Implementation is under way at 125 local centers around the state. Staff from human services agencies have been placed in schools to provide them access to children and to facilitate communication between school and agency staff persons (Levy, 1989, p. 12). At the middle and high school levels, the focus is on youth service centers modeled after the New Jersey initiative, while at the elementary school level family resource centers constitute the major effort.
At the local level, San Diego's New Beginnings program represents a prototypical effort to design and implement an interagency collaboration to improve the lives of poor families and their children through the creation of a new system concentrating on integrated services and prevention efforts. The program began in 1988 as an outgrowth of discussions among top agency executives in San Diego county, the superintendent of San Diego City Schools, and officials from other public agencies. The discussions quickly focused on the problems of fragmentation of services in the inner-city population characterized by high density, ethnic diversity, transience, high crime rates, and child abuse (Deputy Superintendent of San Diego City Schools, 1990). A feasibility study, undertaken in 1989 at Hamilton Elementary School, examined the potential for an integrated services initiative to improve the welfare of its students and their families. The study concluded that families have difficulty in "traversing" the complex of human services without additional support and information; services are fragmented and do not emanate from a common theoretical or philosophical base; too few resources are dedicated to preventive efforts, with most being focused on crisis intervention; and although the school is the logical base for providing services to both families and children, collaborative efforts should not necessarily be governed by the school.
Implementation planning for the New Beginnings program began in October 1990, with support from two private foundations. The program is most appropriately described as a strategy for coordinating educational, social, and health services with foci such as prevention and early intervention; a holistic focus on the family as an interactive unit, not on any single member; reallocation of existing services from participating agencies; and an emphasis on adaption in several settings rather than a project focus (Joining Forces, no date, p. 2).
A demonstration center was opened at Hamilton Elementary School in the fall of 1991. In Phase One, the center is serving families in grades K-5, with expansion to preschool children in a planned second phase. Services provided include the following: expanded school registration and preliminary assessment for all families, parent education and adult education classes, expanded health services, a team of family services professionals to provide continuous planning and counseling and to direct services to families in need, and links to an extended team to provide support services from the participating agencies.
New Beginnings appears to be an exemplary case of an integrated services approach involving a careful design and implementation of interagency collaboration. Clearly, its most problematic aspect is the lack of a careful evaluation of service outcomes. Funders of the program, including private foundations and the U.S. Department of Health and Human Services, pressed program officials for a "tight outcome evaluation" (Cohen, 1991, p. 19). A tentative framework for evaluation describes a variety of outcome measures to be assessed, such as parents' involvement, student achievement, infant morbidity and mortality, and welfare dependency. However, as Cohen (1991) notes, "... the specifics are still sketchy, and some proposed indicators -- especially those relating to health -- need refinement" (p. 19).
The integrated services model clearly represents an advancement over earlier compensatory education models that often ignored the complex of recent demographic, economic, and social changes that interfered with schools' abilities to educate a growing population of disadvantaged students. The success of this approach most likely depends on the ability and willingness of school and human services agencies to develop and implement a comprehensive plan to link the school restructuring movement with health and social services programmatic initiatives (Dryfoos, 1991). Recent experience indicates that achieving such a coordinated, multifaceted effort is a daunting challenge.
For example, New York City's Dropout Prevention Initiative, which operated in the New York City school system from 1985-86 through 1987-88, fell far short of meeting its primary goal of reducing dropout rates among students at risk. The evaluation of the program (Grannis, Riehl, Pallas, Lever, Randolph, and Jewell, 1988) clearly revealed that a substantial proportion of targeted students received few or none of the services the program was designed to deliver.
Data from the Teachers College evaluation show that services fell most conspicuously short of program expectations in areas that required crossing organizational boundaries: health, which required collaboration between the board of education and the health department; and school linkage, which required collaboration between the middle schools and the high schools, at both the central office and the district levels (Grannis, 1991, p. 149).