The High/Scope Perry Preschool Project

U.S. Department of Justice Office of Justice Programs Office of Juvenile Justice and Delinquency Prevention

John J. Wilson, Acting Administrator October 2000

The High/Scope Perry From the Administrator Preschool Project The more we learn about risk factors for delinquency, the more obvious it is that effective prevention programs targeting children at risk can provide benefits far beyond their cost. This Bulletin revisits a time-tested early childhood education program and Greg Parks looks at the results to date from an ongoing, well-designed study of the The Office of Juvenile Justice and Delin- appear to vary according to a child’s stage of program. quency Prevention (OJJDP) recently pub- development and may be reduced with ap- lished Costs and Benefits of Early Childhood propriate preventive measures. These devel- Nearly 40 years ago, the High/Scope Intervention (Greenwood, 1999), a Fact opmental differences for risk factors indicate Perry Preschool Project developed a Sheet reviewing the benefits of early child- the need for targeted interventions that ad- high-quality educational approach hood intervention in the prevention of dress specific age-related factors (Wasserman focusing on 3- and 4-year-olds at risk later delinquency. Among the most notable and Miller, 1998). Given this link between for school failure. The longitudinal and longstanding secondary prevention early risk factors and later delinquency, it is study has found that not only was the programs considered was the High/Scope important for practitioners to plan interven- project effective as an educational Perry Preschool Project of Ypsilanti, MI.1 tion programs for high-risk youth early in a intervention, it also demonstrated This Bulletin examines this successful youth’s life so that he or she can develop other positive outcomes, including a program model, which demonstrates a a strong foundation for later development. significantly lower rate of crime and potential link between early childhood in- delinquency and a lower incidence tervention and delinquency prevention. of teenage pregnancy and welfare Background dependency. By the age of 27, pro- The High/Scope Perry Preschool Project is a The High/Scope Perry Preschool Project, gram participants were nearly three well-established early childhood interven- which began in 1962, is the focus of an on- times as likely to own their own tion that has been in operation for almost going longitudinal study—conducted by homes than the control group and 40 years. A review of the program’s findings the High/Scope Educational Research less than half as likely to be receiving is useful at this time in light of the field’s Foundation—of 123 high-risk African public assistance. growing knowledge of risk factors associated American children.2 Participants were of with juvenile delinquency, including early This Bulletin explains how and why childhood risk factors that may be dimin- the Perry Preschool Project was ished by secondary prevention programs 2 The original Perry Preschool no longer exists, but the successful, presents two positive targeted at high-risk populations. Juvenile High/Scope Educational Research Foundation—founded cost-benefit analyses, and examines justice research has made great strides in in 1970 by Perry Preschool researcher David Weikart— the implications for future policy identifying risk factors that may be precur- continues to collect followup data from the participants of decisions. One conclusion is that an sors to delinquency. Although the prob- the 1962 study. The foundation is an independent organi- effective prevention strategy requires zation dedicated to nonprofit research, development, both quality programming and an ability of delinquency increases with the training, and public advocacy. Its principal goals are to number of risk factors, specific risk factors promote the learning and development of children world- adequate commitment of resources. wide from infancy through adolescence and to support John J. Wilson and train educators and parents as they help children Acting Administrator 1 Unlike primary prevention programs, which are learn. In a High/Scope program, students should learn directed at the general population, secondary preven- through active involvement with materials, events, and tion programs target children at risk for school failure ideas. The Foundation disseminates the High/Scope Pre- or delinquency. school model worldwide. low socioeconomic status, had low IQ scores (between 70 and 85, the range for borderline mental impairment) with no organic deficiencies (i.e., biologically based mental impairment), and were at high risk of failing school. Fifty-eight of these 3- and 4-year-old children were assigned to the program group, and 65 of these children were assigned to a control group that did not go through the program. The groups were matched according to age, IQ, socioeconomic status, and gender. There were no differences between the groups with regard to father absence, parent education level, family size, household density, or birth order. Researchers collected followup data annually when the children were between ages 4 and 11 and at ages 14, 15, and 19 and collected age 27 data from 1986 to 1991 (Schweinhart, Barnes, and Weikart, 1993; Schweinhart and Weikart, 1995).3 research. This Bulletin reviews the program ences between the program and control outcomes, describes the early childhood groups. The program group had fewer ar- The High/Scope Perry Preschool Project’s risk factors that can be targeted with inter- rests overall than the control group (aver- high-quality educational approach is based vention, and explores the relationship be- ages of 1.3 versus 2.3 arrests per person), on an active learning model that empha- tween program components and risk factors. fewer felony arrests (averages of 0.7 ver- sizes participants’ intellectual and social sus 2.0 arrests per person), and fewer ju- development. Children attended the pre- venile court petitions filed (averages of 0.2 school Monday through Friday for 2.5 Program Outcomes versus 0.4 petitions per person). hours per day over a 2-year period. During Outcomes of the High/Scope Perry Pre- Like the criminal record data, a miscon- that same period, a staff-to-child ratio of school longitudinal study can be divided duct scale based on teacher-report data one adult for every five or six children en- into three major categories: social respon- and self-report data from the 19-year-old abled teachers to visit each child’s family sibility, scholastic success, and socioeco- respondents demonstrates a significant in their home for 1.5 hours each week. In nomic success (Schweinhart et al., 1985). difference between the program and con- addition, parents participated in monthly Social responsibility variables include de- trol groups, as reflected by the following small group meetings with other parents, linquency, marital status, and pregnancy. results for the program group: facilitated by program staff. Scholastic success is determined by a number of factors including graduation ◆ Although it was initiated as an educational Lower overall scores for total miscon- intervention, the High/Scope Perry Pre- rate, grade point average, and postsecond- duct and serious misconduct at ages ary education, whereas socioeconomic school Project has demonstrated a number 15 and 19. success is measured in terms of employ- of other positive outcomes, including a sig- ◆ Lower incidence of fighting and other ment, earnings, and welfare assistance. nificantly lower rate of crime and delin- violent behavior. quency and lower incidence of teenage Cost-benefit is included as an additional outcome because of the long-term savings ◆ Lower incidence of property damage. pregnancy and welfare dependency. Over- all, the program group has demonstrated to society as a result of program success. ◆ Fewer police contacts. significantly higher rates of prosocial be- Data collected from respondents at age havior, academic achievement, employ- Social Responsibility 27 indicate significant differences be- ment, income, and family stability as com- Delinquency. Data collected from police tween the program group and control pared with the control group. The success and court records show that juvenile de- group for adult arrests: the control group of this and similar programs demonstrates linquency was significantly lower for the underwent more than twice as many ar- intervention and delinquency prevention High/Scope Perry Preschool program group rests as the program group (averages of in terms of both social outcome and cost- as compared with the control group, in- 4.0 versus 1.8 arrests per person). Thirty- effectiveness and has a number of useful cluding fewer arrests and fewer juvenile six percent of the control group accounted implications for policy, practice, and ongoing court petitions (Schweinhart, Barnes, and for 98 felony arrests between ages 19 and Weikart, 1993; Schweinhart and Weikart, 27, while 27 percent of the program group 1995). Only 31 percent of the program accounted for 40 felony arrests during the 3 Researchers are currently collecting followup data group had ever been arrested, compared same period. Thirty-five percent of the from the original program participants. Called the with 51 percent of the control group. In control group were considered frequent High/Scope Perry Preschool Midlife Study, researchers have already interviewed 30 of the 39- to 41-year-old addition to police and court records, data offenders (defined as five or more arrests), participants. The interview emphasizes health and the collected from respondents at age 19 were compared with only 7 percent of the pro- performance of the program participants’ children. used as an overall indicator of delinquency. gram group. In addition, 25 percent of the The researchers expect to complete the data collec- When study participants were 19 years control group had been arrested for drug- tion by the end of 2001. This study is funded by the old, researchers found significant differ- related offenses, versus 7 percent of the McCormick Tribune Foundation in Chicago, IL.

2 program group (see figure 1). The control (an average difference of 16 percent). also reflect that fewer program group group also averaged more months on pro- The difference in the final achievement members received welfare assistance (18 bation (6.6 versus 3.2 months) and had test scores of the two groups at age 14 percent versus 32 percent). The data for more than twice as many of its members was particularly significant: the program age 27 reveal a continuation of significant placed on probation or parole for longer group’s scores were 29 percent higher economic differences characterized by than 18 months (20 percent versus 9 than those of the control group. more economic stability among the pro- percent). ◆ Grade point average. The mean high gram group members, as measured by the following indicators: Marital status and pregnancy. Marital school grade point average of the pro- status among the males was the same for gram group was higher than that of the ◆ Public assistance. Fifteen percent of the both groups, with 26 percent married at control group (2.09 versus 1.68). program group were receiving public age 27, although program group males, on ◆ Graduation from high school. Seventy- assistance, versus 32 percent of the average, had been married for a longer one percent of the program group gradu- control group. period (6.2 versus 3.3 years). Marital sta- ated from high school, compared with ◆ Monthly earnings. Twenty-nine percent tus among the females differed signifi- 54 percent of the control group. The of the program group had monthly earn- cantly, with 40 percent of program group difference was largely accounted for by ings of $2,000 or more, versus 7 percent females married, compared with 8 percent graduation rates among females (84 of the control group (36 percent versus of the control group females. Although percent and 35 percent, respectively). 11 percent, respectively, when compar- fewer females in the program group were ing only employed members in each parents (64 percent versus 75 percent), Socioeconomic Success group). significantly more of them were married, Data collected at ages 19 and 27 indicate cohabiting parents (28 percent versus 8 ◆ Household earnings. When the income that the program group has been more percent). Fifty-seven percent of mothers of the spouses of the study participants successful socioeconomically than the in the program group gave birth out of was taken into account, 47 percent of control group. The data for age 19 reveal wedlock, compared with 83 percent of the program group had household in- that significantly more program group mothers in the control group. In measures come earnings of $3,000 or more per members were employed (50 percent ver- related to family stability, the program month, versus 17 percent of the control sus 32 percent) and self-supporting (45 group scored significantly higher on a group. percent versus 25 percent). These data measure of closeness to family and friends (66 percent versus 48 percent) and the ability to maintain persistence at tasks (i.e., work or study hard all day) (47 per- Figure 1: Major Findings of the High/Scope Perry Preschool Project cent versus 33 percent). for Participants Followed Up at Age 27

Scholastic Success Five or More 7% Participants in the High/Scope Perry Pre- Arrests 35% school study were characterized by better academic performance than those in the Arrests for 7% control group, as measured by higher grad- Dealing Drugs 25% uation rates, better grades, higher standardized test scores, and fewer instances Public Assistance 15% of placement in special education classes. as Adult 32% In addition, the program group spent more time on homework and demonstrated Births Out of 57% more positive attitudes toward school at Wedlock 83% ages 15 and 19. More parents of program Earnings group members had positive attitudes 29% of $2,000+ 7% regarding their children’s educational ex- per Month periences and were hopeful that their children would obtain college degrees. The 36% Home Ownership 13% program group demonstrated significant academic differences in the following areas: 12th-Grade 71% ◆ Education 54% Special education for mental impair- or Higher ment. Only 15 percent of the program group had been placed in special edu- 0% 20% 40% 60% 80% 100% cation programs for mental impairment, compared with 34 percent of Program Group Control Group the control group. Source: High/Scope Educational Research Foundation. 1999. High-Quality Preschool ◆ Test scores. Each year from ages 7 to Program Found To Improve Adult Status. Ypsilanti, MI: High/Scope Educational Research 14, the mean achievement test scores Foundation. Retrieved March 13, 2000, from the World Wide Web: http://www.highscope.org/ of the program group were noticeably research/Perry%20fact%20sheet.htm. Reprinted with the permission of the High/Scope Educational Research Foundation. higher than those of the control group

3 intelligence and delinquency. Additional Figure 2: Public Costs/Benefits per Participant in the High/Scope early risk factors include poor attachment Perry Preschool Study to caregivers (Egeland and Farber, 1984; Shaw and Bell, 1993), poor parenting Preschool skills (Hawkins et al., 1998; Loeber and $12,356 Program Stouthamer-Loeber, 1986), and multiple family stressors (Fergusson and Lynskey, Welfare $2,918 1996; Shaw et al., 1998). These risk factors may not only directly affect delinquency

Schooling $6,287 but may also indirectly influence other factors that interact with delinquency, Taxes on such as school- and community-related Earnings $8,847 risk factors. As demonstrated in the Prenatal and Justice $12,796 System Early Childhood Nurse Home Visitation Program supported by OJJDP (Olds, Hill, Crime $57,585 and Rumsey, 1998), prenatal and early Victims postnatal prevention are shown to reduce $0 $20,000 $40,000 $60,000 risk factors that contribute to the development of antisocial behavior in child- Cost Benefit hood. Early childhood intervention during the preschool years also offers an Source: High/Scope Educational Research Foundation. 1999. High-Quality Preschool opportunity to halt the developmental Program Found To Improve Adult Status. Ypsilanti, MI: High/Scope Educational Research trajectory toward delinquency and related Foundation. Retrieved March 13, 2000, from the World Wide Web: http://www.highscope.org/ behavioral disorders. Family support ser- research/Perry%20fact%20sheet.htm. Reprinted with the permission of the High/Scope Educational Research Foundation. vices help develop parenting skills, attachment, and coping mechanisms that have a positive effect on family stressors. A multicomponent approach to enhancing ◆ Home ownership. Thirty-six percent least reliable savings category (savings to child development promotes protective of the program group owned a home, crime victims) still left a return of more factors and reduces risk factors by ad- versus 13 percent of the control group. than twice the initial investment. Savings dressing the many systems and influ- ◆ Automobile ownership. Thirty percent to crime victims make up 65 percent of the ences that affect a child’s development. of the program group owned a second total investment return in the earlier analy- car, versus 13 percent of the control sis (Barnett, 1993). Although victim sav- group. ings should be considered a significant Program Components outcome and societal benefit of early and Related Risk Cost-Benefit Analysis childhood intervention, this factor is also distinct from the other factors that can be Factors for A cost-benefit analysis of the High/Scope estimated based on direct governmental Delinquency Perry Preschool study indicates a savings costs and savings. With victim savings The components of the High/Scope Perry to the public of more than seven times the factored out of the analysis, the largest Preschool Project affect a number of the initial investment per child, with a return savings category is in criminal justice early childhood risk factors associated of $7.16 for every dollar spent (Barnett, costs (40 percent), followed by increased with later delinquency and other behav- 1993; see figure 2). When adjusted for in- taxable revenue (26 percent), reduced ioral problems. In addition to directly re- flation and a 3-percent discount rate, the educational services (25 percent), and inforcing early developmental processes investment in early childhood prevention reduced welfare costs (9 percent). resulted in a taxpayer return of $88,433 in the educational setting, the program per child from the following sources: strengthens positive parenting skills. Early Childhood ◆ Savings in welfare assistance (prior to The High/Scope Educational Research welfare reform). Risk Factors for Foundation explains the effectiveness of the High/Scope Perry Preschool model in ◆ Delinquency Savings in special education. terms of empowerment, which includes ◆ Savings to the criminal justice system. An understanding of early childhood risk developing skills for success by enabling factors for delinquency is helpful to inter- ◆ Savings to crime victims. children to be active and independent preting the success of the High/Scope learners, helping parents to support the ◆ Increased tax revenue from higher Perry Preschool Project. One factor identi- development of their children, and pro- earnings. fied with risk for delinquency is poor lan- viding teachers with effective training and guage skills (Stattin and Klackenberg- An independent reanalysis is provided in support (Schweinhart and Weikart, 1995). Larsson, 1993). As a component of overall a recent RAND Corporation report (Karoly mental development, language functions Because an ongoing home-school relation- et al., 1998; see figure 3). This report as an indicator of later intelligence and is a ship enhances socialization, involving found that eliminating the largest and critical factor in the relationship between parents early in the educational process

4 is critical to the later success of partici- delinquency, school success (as indicated that address only educational factors, pants in an early childhood intervention by higher standardized test scores, school have not been shown to demonstrate sig- such as High/Scope Perry Preschool (Seitz, commitment, attachment to teachers, col- nificant results. 1990). Weekly home visits by teachers and lege aspirations, and parent expectations) regular parent group meetings promote appears to be a protective factor against In addition to the need to target appropri- ate populations and address multiple risk the strengthening of parent-child relation- delinquency (Smith et al., 1995). Academic ships and increase parent involvement in achievement outcomes of the High/Scope factors in multiple domains, program quality is essential to success. The High/Scope the educational process. A more recent Perry Preschool study indicate that the OJJDP longitudinal study, the Rochester program group was more successful than Perry Preschool model is based on a high- quality educational approach that assumes Youth Development Study (Thornberry et the control group in school-related factors al., 1998), confirmed a significant relation- that appear to protect against delinquency. a low staff-to-child ratio, an active learning curriculum, and a home visitation com- ship between parents’ involvement in their children’s lives and reduced delinquency. The positive outcomes of the High/Scope ponent that engages parents in the educa- Perry Preschool study are the result of a tional process. Furthermore, teachers are In addition to enhancing parent attach- cumulative effect that begins with increased well educated, adequately compensated, ment, parent involvement, and parenting school readiness (Berrueta-Clement et and well supported in their tasks. skills, early childhood intervention aimed al., 1987; Zigler, Taussig, and Black, 1992). at both parents and children influences a School readiness results in positive rein- Head Start, perhaps the largest and best- known early childhood intervention pro- child’s attachment to school and later com- forcement from teachers in the early grades mitment to school success (Thornberry followed by enhanced academic perfor- gram, has recently made efforts to expand and improve its effectiveness by empha- et al., 1998). Findings from the Rochester mance in subsequent grades and an over- study confirm earlier research linking poor all stronger commitment to school. A cor- sizing family support, staff training, and performance standards (U.S. Department school attachment, commitment, and relational analysis of the High/Scope Perry achievement to delinquent behavior and Preschool data reveals a strong associa- of Health and Human Services, 1999). The 1994 legislation reauthorizing Head Start drug use (Krohn et al., 1995; Smith et al., tion between school motivation in the 1995). Another OJJDP study, the Seattle early years and literacy scores at age 19 incorporated a number of recommenda- tions from the Advisory Committee on Social Development Project (Hill et al., (Schweinhart, Barnes, and Weikart, 1993). 1999), found that a lack of success in el- School motivation is also highly correlated Head Start Quality and Expansion (1993), including increased parent involvement, ementary school was linked to later gang with the highest year of schooling com- membership. Even in the midst of multiple pleted, which is associated with higher a lower staff-to-child ratio, and increased mental health services.4 Head Start has other factors placing youth at high risk for monthly earnings in adulthood and fewer lifetime arrests. increased the emphasis on curriculum and child outcomes as a result of this re- authorization and has formed Head Start Figure 3: Costs, Savings, and Program and Policy Quality Research Centers to respond to Benefits of the High/ Implications the need for additional research in the Scope Perry Preschool area of early childhood intervention. Fur- The outcomes of the High/Scope Perry Project ther research is clearly needed to build Preschool study demonstrate the value of on the limited existing knowledge base prevention and early intervention efforts Reduction and assess the effectiveness of programs in Welfare in promoting protective factors that re- across various demographic groups, risk Cost duce delinquency. The program was de- factors, and co-occurring factors that are Reduction veloped for high-risk children who stood related to delinquency, such as mental 9% in Criminal to benefit the most from such an interven- Reduction in health issues and substance abuse Justice tion. The intervention also affected mul- Education Cost (Yoshikawa, 1995). Services 25% 40% tiple risk factors and was carried out in multiple domains (i.e., home and school). Although the High/Scope Perry Preschool In an extensive review of early childhood study’s sample size was small in propor- interventions, Yoshikawa (1995) con- tion to its eventual influence, its strong 26% cluded that the combination of an early experimental design has contributed to its Increase educational component with family sup- prominence in the field of early childhood in Taxes port, as exemplified by the High/Scope education. Subsequent early childhood on Greater Perry Preschool Project, is a determining research that is carefully controlled and Income factor in long-term effects on antisocial longitudinal in design remains limited. The behavior. Other combination programs limited research involving similar models Source: Karoly, L.A., Greenwood, P.W., that combine educational and family sup- Everingham, S.S., Hoube, J., Kilburn, M.R., that have demonstrated long-term effects Rydell, C.P., Sanders, M., and Chiesa, J. on delinquency include the Yale Child port components, however, supports the 1998. Investing in Our Children: What We Welfare Project (Seitz and Apfel, 1994), positive outcomes of the High/Scope Perry Know and Don’t Know About the Costs and Houston Parent Child Development Cen- Preschool model. Subsequent independent Benefits of Early Childhood Interventions. ter (Johnson and Walker, 1987), and Syra- evaluations of the programs that have Washington, DC: RAND Corporation. cuse Family Development Research Pro- Reprinted with the permission of RAND gram (Lally, Mangione, and Honig, 1988). Corporation. 4 Single-component models, such as those Head Start Act Amendments of 1994, Pub. L. No. 103– 252, tit. I, § 108, Stat. 624 (1994).

5 implemented the High/Scope model have References Howell, J.C., ed. 1995. Guide for Imple- rated those programs significantly higher menting the Comprehensive Strategy for than comparison programs, with 58 per- Advisory Committee on Head Start Qual- Serious, Violent, and Chronic Juvenile Of- cent of High/Scope programs versus 40 ity and Expansion. 1993. Creating a 21st fenders. Washington, DC: U.S. Department percent of comparison programs being Century Head Start. Final Report of the of Justice, Office of Justice Programs, Of- rated as high quality (Epstein, 1993). In Advisory Committee on Head Start Qual- fice of Juvenile Justice and Delinquency addition, 72 percent of children in High/ ity and Expansion. Washington, DC: U.S. Prevention. Scope programs versus 57 percent of Government Printing Office. children in comparison programs scored Johnson, D.L., and Walker, T. 1987. Pri- Barnett, W.S. 1993. Benefit-cost analysis of mary prevention of behavior problems high on measures of emotional, social, preschool education: Findings from a 25- American cognitive, and motor development. in Mexican-American children. year follow-up. American Journal of Ortho- Journal of Community Psychology 15(4): Some targeted, multicomponent early psychiatry 63(4):25–50. 375–385. childhood interventions have been dem- Berrueta-Clement, J.R., Schweinhart, L.J., Karoly, L.A., Greenwood, P.W., Everingham, onstrated to exceed their costs in even- Barnett, W.S., and Weikart, D.P. 1987. The tual savings and benefit to the public. S.S., Hoube, J., Kilburn, M.R., Rydell, C.P., effects of early educational intervention on Sanders, M., and Chiesa, J. 1998. Investing However, implementing an effective pre- crime and delinquency in adolescence and in Our Children: What We Know and Don’t vention strategy requires a commitment early adulthood. In Primary Prevention of Know About the Costs and Benefits of Early to provide empirically based quality pro- Psychopathology: Vol. 10, Prevention of De- Childhood Interventions. gramming and to invest the up-front re- Washington, DC: linquent Behavior, edited by J.D. Burchard RAND Corporation. sources that will result in long-term sav- and S.N. Burchard. Newbury Park, CA: Sage ings and positive social change in the Publications, Inc., pp. 220–240. Krohn, M.D., Thornberry, T.P., Collins-Hall, lives of children and families. The High/ L., and Lizotte, A.J. 1995. School dropout, Scope Perry Preschool Project provides Egeland, B., and Farber, E.A. 1984. Infant- delinquent behavior, and drug use: An ex- one such model for early childhood inter- mother attachment: Factors related to amination of the causes and consequences vention that has proven successful when its development and changes over time. of dropping out of school. In Drugs, Crime, executed with quality and commitment to Child Development 55(3):753–771. and Other Deviant Adaptations: Longitudi- long-term results. The complexity of juve- Epstein, A.S. 1993. Training for Quality: nal Studies, edited by H.B. Kaplan. New nile delinquency requires multiple strate- Improving Early Childhood Programs York, NY: Plenum Press, pp. 163–183. gies that address the problem at various Through Systematic Inservice Training. Lally, J.R., Mangione, P.L., and Honig, A.S. stages of development; early childhood Monographs of the High/Scope Educa- intervention is one promising component 1988. The Syracuse University Family De- tional Research Foundation, Number 9. velopment Research Project: Long-range in the context of a more comprehensive Ypsilanti, MI: High/Scope Press. approach, as recommended in OJJDP’s impact of an early intervention with low- Comprehensive Strategy for Serious, Vio- Fergusson, D.M., and Lynskey, M.T. 1996. income children and their families. In lent, and Chronic Juvenile Offenders (Wil- Adolescent resiliency to family adversity. Parent Education as Early Childhood Inter- son and Howell, 1993). The High/Scope Journal of Child Psychology and Psychiatry vention: Emerging Directions in Theory, Perry Preschool model is worthy of con- and Allied Disciplines 37(3):281–292. Research, and Practice, vol. 3, edited by D.R. sideration as an effective early childhood Powell. Norwood, NJ: Ablex, pp. 79–104. Greenwood, P.W. 1999. Costs and Benefits intervention as communities attempt to of Early Childhood Intervention. Fact Loeber, R., and Stouthamer-Loeber, M. implement a comprehensive strategy that Sheet #94. Washington, DC: U.S. Depart- 1986. Family factors as correlates and includes prevention, intervention, and ment of Justice, Office of Justice Programs, predictors of juvenile conduct problems graduated sanctions (Howell, 1995; Wil- Office of Juvenile Justice and Delinquency and delinquency. In Crime and Justice, vol. son and Howell, 1993). Prevention. 7, edited by M. Tonry and N. Morris. Chi- cago, IL: University of Chicago Press, pp. Hawkins, J.D., Herrenkol, T., Farrington, 219–339. For Further Information D.P., Brewer, D., Catalano, R.F., and Harachi, For more information about the High/ T.W. 1998. A review of predictors of youth Olds, D., Hill, P., and Rumsey, E. 1998. Pre- Scope Perry Preschool Project, contact: violence. In Serious and Violent Juvenile natal and Early Childhood Nurse Home Offenders: Risk Factors and Successful In- Visitation. Bulletin. Washington, DC: U.S. High/Scope Educational Research terventions, edited by R. Loeber and D.F. Department of Justice, Office of Justice Foundation Farrington. Thousand Oaks, CA: Sage Pub- Programs, Office of Juvenile Justice and 600 North River Street lications, Inc., pp. 106–146. Delinquency Prevention. Ypsilanti, MI 48198–2898 734–485–2000 Hill, K.G., Howell, J.C., Hawkins, J.D., and Schweinhart, L.J., Barnes, H.V., and Weikart, 734–485–0704 (fax) Battin-Pearson, S.R. 1999. Childhood risk D.P. 1993. Significant Benefits: The High/ E-mail: [email protected] factors for adolescent gang membership: Scope Perry Preschool Study Through Age Internet: www.highscope.org Results from the Seattle Social Develop- 27. Monographs of the High/Scope Educa- ment Project. Journal of Research in Crime tional Research Foundation, Number 10. and Delinquency 36(3):300–322. Ypsilanti, MI: High/Scope Press.

6 Schweinhart, L.J., Berrueta-Clement, J.R., criminal behavior. Journal of Abnormal of Justice Programs, Office of Juvenile Barnett, W.S., Epstein, A.S., and Weikart, Psychology 102(5):369–378. Justice and Delinquency Prevention. D.P. 1985. Effects of the Perry Preschool Thornberry, T.P., Krohn, M.D., Lizotte, Yoshikawa, H. 1995. Long-term effects of program on youths through age 19: A summary. Topics in Early Childhood Spe- A.J., Smith, C.A., and Porter, P.K. 1998. early childhood programs on social out- Taking stock: An overview of findings comes and delinquency. Future of Children cial Education 5(2):26–35. from the Rochester Youth Development 5(3):51–75. Schweinhart, L.J., and Weikart, D.P. 1995. Study. Paper presented at the American Zigler, E., Taussig, C., and Black, K. 1992. The High/Scope Perry Preschool study Society of Criminology meeting, Washing- through age 27. In Going Straight: Effective ton, DC, November 11, 1998. Early childhood intervention: A promising Delinquency Prevention and Offender Re- preventative for juvenile delinquency. American Psychologist habilitation, edited by R.R. Ross, D.H. U.S. Department of Health and Human Ser- 47(8):997–1006. vices. 1999. Head Start: Administration for Antonowicz, and G.K. Dhaliwal. Ottawa, Children and Families. Canada: Air Training and Publications, Fact Sheet. Wash- This Bulletin was prepared by Greg Parks, ington, DC: U.S. Department of Health and pp. 57–75. Intern Program Specialist, Research and Human Services. Program Development Division, Office of Seitz, V. 1990. Intervention programs for Wasserman, G.A., and Miller, L.S. 1998. Juvenile Justice and Delinquency Prevention. impoverished children: A comparison of educational and family support models. In The prevention of serious and violent ju- Photograph copyright © 1999 PhotoDisc, Inc. Serious and Violent Annals of Child Development: A Research venile offending. In Juvenile Offenders: Risk Factors and Suc- Annual, vol. 7, edited by R. Vasta. London, cessful Interventions, edited by R. Loeber The Office of Juvenile Justice and Delin- England: Jessica Kingsley Publishers, pp. quency Prevention is a component of the Of- 73–103. and D.P. Farrington. Thousand Oaks, CA: Sage Publications, Inc., pp. 197–247. fice of Justice Programs, which also includes Seitz, V., and Apfel, N. 1994. Parent-focused the Bureau of Justice Assistance, the Bureau A Com- of Justice Statistics, the National Institute of intervention: Diffusion effects on siblings. Wilson, J.J., and Howell, J.C. 1993. prehensive Strategy for Serious, Violent, Justice, and the Office for Victims of Crime. Child Development 65(2):677–683. and Chronic Juvenile Offenders. Washing- Shaw, D.S., and Bell, R.Q. 1993. Develop- ton, DC: U.S. Department of Justice, Office mental theories of parental contributors to antisocial behavior. Journal of Abnormal Psychology 21(5):493–518. Share With Your Colleagues Shaw, D.S., Winslow, E.B., Owens, E.B., and Unless otherwise noted, OJJDP publications are not copyright protected. We Hood, N. 1998. Young children’s adjustment encourage you to reproduce this document, share it with your colleagues, and to chronic family adversity: A longitudinal reprint it in your newsletter or journal. However, if you reprint, please cite OJJDP study of low-income families. Journal of and the authors of this Bulletin. We are also interested in your feedback, such as the American Academy of Child and Ado- how you received a copy, how you intend to use the information, and how OJJDP lescent Psychiatry 37(5):545–553. materials meet your individual or agency needs. Please direct your comments and questions to: Smith, C., Lizotte, A.J., Thornberry, T.P., and Krohn, M.D. 1995. Resilient youth: Juvenile Justice Clearinghouse Identifying factors that prevent high-risk Publication Reprint/Feedback youth from engaging in delinquency and P.O. Box 6000 drug use. In Delinquency and Disrepute in Rockville, MD 20849–6000 the Life Course, edited by J. Hagan. Green- 800–638–8736 wich, CT: JAI Press. 301–519–5600 (fax) E-mail: [email protected] Stattin, H., and Klackenberg-Larsson, I. 1993. Early language and intelligence development and their relationship to future

7 U.S. Department of Justice PRESORTED STANDARD Office of Justice Programs POSTAGE & FEES PAID DOJ/OJJDP Office of Juvenile Justice and Delinquency Prevention PERMIT NO. G–91

Washington, DC 20531 Official Business Penalty for Private Use $300

Bulletin NCJ 181725