Since the mid-1980's, the State of Michigan has experienced a dramatic expansion of its prison populations of more than three-fold. The imprisoned population rose from a level of one percent of the total state population to over four percent between 1985 and 1995 (Maguire & Pastore 1997). Subsequently, there has been a substantial increase in the fiscal budget of the State's Corrections Department. In 1999, the State will spend nearly 1.5 billion dollars on corrections, five percent more than last year's spending, and nearly twice the 1990 spending level (Department of Management and Budget 1998). Besides the substantial economic cost to the state, more than 40,000 state residents will be incarcerated. These persons typically come from poor inner city neighborhoods and are now recognized as a leading cause of the continued urban plight. Some scholars are beginning to argue that the United States' high rate of incarceration, which has resulted in nearly one in four African-American males spending time under criminal justice supervision is a major factor in the frequency and lethality of violence among inner city minority youths (Clear 1994).

During the Blanchard administration, the Director of Corrections recommended putting more dollars into early childhood prevention programs rather than prisons. In support, a Michigan psychiatrist organized a group that prompted the Michigan Department of Mental Health to organize a steering committee to implement such a prevention program. The Mental Health Department awarded funding to the Mott Children's Health Center in conjunction with the local mental health agency. The Departments of Public Health, Corrections, and Social Services co-sponsored their efforts which resulted in Project SEEK (Services to Enable and Empower Kids).

The purpose of this research is to establish a base line of information of whether this unique prevention program is reaching its intended objectives of: (1) reducing risk factors and increasing protective domains over the multiple measurement periods; and, (2) decreasing the rate of onset into delinquency and antisocial behaviors among the adolescents. Two general research tasks are planned: (a) the coding and verification of existing data; and, (c) the scaling and modeling of these data. These tasks are: 1)Coding and verifying existing data. The Mott Children's Health Center and the Michigan Department of Community Health have already brought several boxes of data to the Institute for Children, Youth, and Families. These data were placed in secure University storage until sufficient funding is secured and human subjects are cleared. As the project is ongoing, continued data collection will occur on site in Genessee County. 2) Scaling and Modeling. Once the necessary data collection and coding are completed, steps will be taken to combine these data for analysis and modeling. The following are the planned analytical steps. First, we will develop scales and subscales for all of the instruments and prepare reports on their validity and reliability. These latter tests are essential given the special population targeted by SEEK. Second, we will compare the service and control groups using a series of simple bivariate and multivariate models that will enable visual and quantitative assessments of differences by treatment group across the different stages of intervention on measures of risk and protection factors, behavior and performance in the school, and other measures of interest such as the likelihood of inmate visitations. ^[2] The final step is to develop models addressing the longitudinal nature of the data. A common problem encountered in prevention research is the need to examine growth and differences in growth across treatment groups over time rather than simply estimating mean differences during a fixed period. (Curran & Muthen 1996). By recasting the analysis in terms of growth, we will look for differences in degree to which Project SEEK altered the normative growth trajectory, that is, the trajectory without exposure to treatment. The analysis will therefore estimate individual patterns of change over time on several theoretically interesting characteristics by treatment group. ^[3]

The implications for public policy on this project are tremendous. Large amounts of data have been collected on this well designed study, however, no long term evaluations have been completed for the policy development. This proposed research would determine whether the past seven years of intervention / prevention services are making a significant difference in the experimental population and whether Project SEEK should receive continued funding. Genessee County has been treated as a pilot study in the hopes this program would be spread to nine other urban centers in Michigan. Policy makers are currently waiting to see the benefits of prevention, details of the cost and savings of this funding program and a suggested course of action.

The children and families involved in this program are the largest benefactors due to the quality and quantity of intervention services. Policy makers will benefit by knowing where to concentrate their funding in order to prevent future criminality. Society benefits in that it may have less recidivism. For example, it costs approximately $1400 per child to participate in Project SEEK while it costs the state approximately $25,000 per year to keep one person in prison. The public policy derived from this research may determine whether nine other urban centers will implement Project SEEK. Finally, Michigan State University will benefit tremendously by playing the major role as the evaluator and the public policy advisor. The directors of the Family Independence Agency, Community Health, Corrections, Mott Children's Health Center and many local agencies are supportive of this research and the assistance of the land-grant University. In addition, there is significant attention received by universities that conduct longitudinal studies on significant social problems. Already, Project SEEK has received large amounts of attention in the way of national media coverage and awards. More than 2.3 million dollars have already been invested into Project SEEK (does not include funding used thus far for evaluation research). However, no significant research or evaluation has been conducted on this project. Michigan State University would not only have the opportunity to cross new frontiers in this area of research and publications, but also has the opportunity to push the general body of knowledge forward on violence and crime prevention for very little cost. The brunt of the cost has already been incurred by multiple state agencies. This project would allow Michigan State University to strengthen its relationship with the multiple state agencies and legislatures and provide the university partnership for an important community partner - - Mott Children's Health Center. Furthermore, Flint is an urban center much in need of the outreach by this university. Michigan State University would also benefit from the databases from this longitudinal study that will be made available to all MSU researchers and students for research purposes on the completion of this study.

This research will produce reports for legislatures, state and local government agencies, and professionals working in the area of prevention. Also, reports will be prepared to help with developing guidelines for professionals such as prisons, schools, and state and local agencies. In addition, scholarly articles will be prepared and submitted to peer-reviewed journals and national conferences. The data from this longitudinal study will be made available after the completion of this project for research purposes. The reports and databases will be made available through a server, which allow for file-transfer-protocol.

• Bloom, Barbara. 1993. "Incarcerated Mothers and Their Children: Maintaining Family Ties." Pp. 60-68 in Female Offenders: Meeting Needs of a Neglected Population. Laurel, MD: American Correctional Association.
  
• Cairns, Robert B., et al. 1989. "Growth and Aggression: 1. Childhood to Early Adolescence." Developmental Psychology 25(2):320-30, 2.
  
• Center for Children of Incarcerated Parents. 1992. Children of Offenders. Report No. 6. Pasadena, CA: Pacific Oaks College and Children's Program.
  
• Clear, Todd R. 1994. Harm in American Penology: Offenders, Victims, and Their Communites. SUNY Series in New Directions in Criminal Justice Studies. Albany, NY: State University of New York Press.
  
• Curran, Patrick J. and Bengt O. Muthen. 1996. Testing Developmental Theories in Intervention Research: Latent Growth Analysis and Power Estimation. Los Angeles, CA: Graduate School of Education & Information Studies, University of California, Los Angeles.
  
• Gabel, Stewart. 1992. "Behavioral Problems in Sons of Incarcerated or Otherwise Absent Father: The Issue of Separation." Family Process 31(3):303-14, 3.
  
• ________. 1993. "Characteristics of Children Whose Parents Have Been Incarcerated." Hospital & Community Psychiatry 44(7):656-60, 7.
  
• Goldstein, Harvey. 1995. Multilevel Statistical Models. 2d ed. Kendall's Advanced Theory of Statistics and Kendall's Library of Statistics, vol. 3, no. 2nd Edition. New York, NY: Halsted Press.
  
• Hostetter, Edwin C. and Dorothea T. Jinnah. 1993. "Families of Adult Prisonners." Http://info.rutgers.edu/newark/wcjlen/family/Family.Corrections.Report. Prison Fellowship Ministries. 6/15/97.
  
• Kellam, Sheppard G., et al. 1994. "The Course and Alleability of Aggressive Behavior from Early First Grade Into Middle School: Results of a Developmental Epidemiologically-Based Preventive Trial." Journal of Child Psychology & Psychiatry & Allied Disciplines.
  
• Loeber, Rolf and Magda Stouthamer-Loeber. 1986. "Models and Meta-Analysis of the Relationship Between Family Variables and Juvenile Conduct Problems and Delinquency." In Crime and Justice: An Annual Review of Research, vol. 7, edited by Norval. Morris and Michael Tonry. Chicago: University of Chicago Press.
  
• Maguire, Kathleen and Ann L. Pastore. 1997. "Sourcebook of Criminal Justice Statistics [Online]." .
  
• Maxwell, Sheila Royo and Christopher D. Maxwell. 1997. Report to Project SEEK: Data Assessment and Preliminary Analysis. Report prepared for the Mott Children's Center, Flint MI. Williamston, MI: Joint Centers for Justice Studies, Inc. 27.
  
• Michigan State Department of Management and Budget. 1998. "Fiscal Year 1999 Budget." .
  
• Moses, Marilyn C. 1995a. "Familes in Peril: "Girl Scots Beyond Bars": A Synergistic Solution for Children of Incarcerated Parents." Corrections Today, December, 124-42.
  
• ________. 1995b. Keeping Incarcerated Mothers and Their Daughters Togeather: Girl Scouts Beyond Bars. Program Focus. Washington, D.C.: U.S. Department of Justice, Office of Justice Programs, National Institute of Justice.
  
• Muthen, Bengt O. and Patrick J. Curran. 1996. General Growth Modeling in Experimental Designs: A Latent Variable Framework for Analysis and Power Estimation. Los Angeles, CA: Graduate School of Education & Information Studies, University of California, Los Angeles. 32.
  
• Plomin, Robert. 1989. "Environment and Genes: Determinants of Behavior." American Psychologist 44:105-11.
  
• ________. 1990. Nature and Nurture: An Introduction to Human Behavioral Genetics. Pacific Grove: CA: Brooks/Cole Publishing Company.
  
• Rowe, David C. and David P. Farrington. 1997. "The Familial Transmission of Criminal Convictions." Criminology 35(1):177-201, 1.
  
• Sampson, Robert J. and John H. Laub. 1993. Crime in the Making: Pathways and Turning Points Through Life. Cambridge, MA: Harvard University Press.
  
• Sampson, Robert J. and Janet Lauritsen. 1994. "Violent Victimization and Offending: Individual-,Situational, and Community-Level Risk Factors." Pp. 1-115 in Understanding and Preventing Violence, vol. 3, edited by Albert J. Reiss and Jeffrey A. Roth. Washington, D.C.: National Academy of Science.
  
• Stoolmiller, Mike. 1994. "Antisocial Behavior, Delinquent Peer Association, and Unsupervised Wandering for Boys: Growth and Change from Childhood to Early Adolescents." Multivariate Behavioral Research 29(3):263-88, 3.
  
• Tolan, Patrick and Nancy Guerra. 1994. What Works in Reducing Adolescent Violence: An Empirical Review of the Field. F-888. Boulder Colorado: Center for the Study and Prevention of Violence.
  
• Tremblay, Richard E. and Wendy M. Craig. 1995. "Developmental Crime Prevention." Pp. 151-236 in Building a Safer Society: Strategic Approaches to Crime Prevention, edited by Micheal Tonry and David P. Farrington. Crime and Justice: A Review of Research, vol. 19. Chicago: The University of Chicago Press.
  
• Weiss, Healther Bastow. 1987. "Family Support and Education in Early Childhood Programs." Pp. 133-60 in American's Family Support Programs: Perspectives and Prospects, edited by Sharon L. Kagan and . New Haven, CT: Yale University Press.
  
• ________. 1993. "Home Vistis Necessary but not Sufficient." In The Future of Children, vol. 3, 3. Lost Altos, CA: The David & Lucillie Packard Foundation.
  
• West, Donald and David P. Farrington. 1977. Who Becomes Delinquent? London: Heinemann.
  
• Willett, John B. and Aline G. Sayer. 1994. "Using Covariance Structures Analysis to Detect Corrlelates and Predictors of Individual Change Over Time." Psychological Bulletin 116(2):363-81, 2.
  
  

1. Because of the extended recruitment period, the data available for each family and child varies. For example, some families may have been participating in the evaluation for seven years, while other have only been part of the study for four years. The number of years of assessments for each child within a family may also vary because some children were born after the family had been participating for sometime.[back]
2. These analyses will address the specific questions raised by the program staff and planners, which include: (a) do the children assigned to the program have higher cognitive development and social competence; (b) do the children assigned to the program have better physical and emotional well-being; (c) do the families assigned to the program have more family stability and less familial stress; (d) were the caregivers assigned to the program better able to access services, meet basic needs; etc.[back]
3. Growth curve models can better address the methodological problems inherent in longitudinal prevention research (Muthen & Curran 1996; Curran & Muthen 1996). For example, a potential problem is lapses in assessments, where not everyone has had the same number of follow-up assessments, or, not everyone was assessed at every period. These issues are commonly faced by prevention scientists analyzing longitudinal prevention trials. Growth curve models are particularly unique in their ability to address these problems (Goldstein 1995), which is why they have quickly become the standard technique for assessing programs designed to prevent antisocial behaviors, among others (see: (Stoolmiller 1994; Willett & Sayer 1994; Cairns, et al. 1989; Kellam, et al. 1994).[back]