Effects of Extended Early Childhood Intervention: Early Adult Findings from the Child-Parent Centers Arthur J. Reynolds, Judy A. Temple, & Suh-Ruu Ou,

Effects of Extended Early Childhood Intervention: Early Adult Findings from the Child-Parent Centers

Arthur J. Reynolds, Judy A. Temple, & Suh-Ruu Ou, University of Minnesota

Society for Research in Child DevelopmentMarch 30, 2007, Boston, MA.

What is PK-3 Education?

ProgramsPlanned interventions and services

beginning during any of the first 5 years of life and continue up to third grade

Practices Elements of PK-3 programs such as

preschool, full-day kindergarten, class sizes, parent involvement.

Rationale for PK-3

Help sustain effects of preschool and kindergarten programsPromote better early transitions Improve continuity of development Greater “dosage” will help children at risk

Key Principles of PK-3 Programs

Continuity: Promoting consistencyOrganization:Staffing, leadership, services Instruction: Aligning curriculum, encouraging communication Family support services

PK-3 Programs

Head Start/Follow Through Key: Home-school resource teacherChild-Parent CentersKey: organization and services Abecedarian Project Key: Home-school resource teacher Head Start-Public School TransitionKey: Family service coordinator, services

Child-Parent Centers Foci

Basic Skills: Language, Numeracy, ListeningStructured Learning ActivitiesLanguage and literacy emphasis

Field Trips: Zoos, Museums, Libraries

Parent Involvement:Classroom volunteeringParent room activitiesEducational workshops and trainingHome visits and activities

Organizational StructureStaffing patterns and Coordination

Class size reductions

Staffing at Each SiteHead Teacher Parent Resource TeacherSchool-Community RepresentativeTeachers and aidesSchool nurse, psychologist, social workerPreschool class size was 17 to 2Kindergarten, school age was 25 to 2School-age program had coordinator called curriculum-parent resource teacher

Child-Parent Centers

PrincipalChild-Parent Center

Preschool/Kindergarten(Wing or Building)

Child-Parent CenterPreschool/Kindergarten

(Wing or Building)

Elementary SchoolGrades 1 to 3

Elementary SchoolGrades 1 to 3

Curriculum Parent-Resources TeacherCurriculum Parent-Resources TeacherHead TeacherHead Teacher

OutreachServices

OutreachServices

ParentComponent

ParentComponent

CurriculumComponent

CurriculumComponent

HealthServices

HealthServices

ParentComponent

ParentComponent

CurriculumComponent

CurriculumComponent

School-WideServices

School-WideServices

School-Community RepresentativeResource MobilizationHome VisitationParent Conferences

Parent Resource TeacherParent Room ActivitiesClassroom VolunteeringSchool ActivitiesHome Support

Language FocusSmall Class SizesInservice Training

Health ScreeningNursing ServicesFree + Reduced- Price meals

Parent Room ActivitiesClassroom VolunteeringSchool ActivitiesHome Support

Reduced Class SizeTeacher AidesInstructional Materials Individualized InstructionInservice Training

Health ServicesSchool-Community RepresentativeFree + Reduced- Price mealsResource Mobilization

Age 3 To Age 9

Johnson Child-Parent Center

Characteristics of Study Groups

CPC Intervention Comparison

Sample Complete cohort Random sample of K sites

Recovery, by age 24

514 of 553 (93%) 858 of 986 (87%)

Key attributes Reside in highest poverty areasOver 80% of children enrollMean no. of risks = 4.4; 72% with 4 or more risksParent ed > than in c-group

Reside in high poverty areasHad school-based enrichmentMean no. of risks = 4.3; 71% with 4 or more risksArea poverty > than in p-group

Intervention levels

Preschool 100% 1 or 2 years 15% in Head Start

Kindergarten 57% full day 84% full day

School age 100% 2 or 3 years 30% 1-3 years

Reading Achievement

0

20

40

60

80

100

120

5 6 7 8 9 10

Ages

Te

st

Sc

ore

s

P+K

P+K+SA

Remedial and Child Welfare Services for Extended Intervention Groups

32.3

21.9 20.7

13.511.3

5.8

0

5

10

15

20

25

30

35

40

Per

cen

tag

e o

f S

amp

le

Grade retention Special education Child maltreatment

Research QuestionsIs Participation in the Child-Parent Centers Extended Intervention Program Associated with Greater Adult Well-Being?

Do the Estimated Effects Vary by Child, Family, and Program Characteristics?

Major Data SourcesTeacher ratings yearly from K-Grade 7. Parent ratings, Grades 2, 6, 11. Child ratings yearly from Grades 3-6, 10, 12, and at age 22-24.Interviews/essays with selected children at Grades 6 and 10.ITBS test scores yearly starting in K.Administrative records from schools, social services, child welfare, public health, justice system, employers, and higher ed.Community attributes from Census data.

Major Outcome Measures

Educational Attainment (by age 23)High school completion, years of education, college attendance

Economic Well Being (22-24)Quarterly income, employment, in school or working, public aid

Criminal Behavior (18-24)Arrests, Convictions, Incarceration

Health Status & Behavior (up to 24)Substance use, teen parenthood, health insurance, child abuse/neglect, health insurance

Mental health (22-24) Depressive symptoms

Sample Sizes and Recovery Rates for Adult Follow Up

Outcome measure assessed at ages 22-24

Numberof cases

Recovery rate (%)

Educational attainment 1,368 88.9

Employment status 1,249 81.2

Arrest status/Incarceration history 1,418 92.1

Substance use/Mental health (self report)

1,142 76.4

Health insurance (self report/admin records)

1,277 85.5

Public aid history 1,315 88.0

Basic Impact Model1. 4 to 6 years of CPC starting in preschool

vs2a. All others with 0 to 4 years (Contrast 1)

or2b. CPC preschool/kindergarten only

(Contrast 2) Covariates (age 0 to 3)A. Sex of child, race/ethnicity, 8 family risk

indicators, child maltreatment reportB. Kindergarten word analysis (Contrast 2)

Educational AttainmentE-Prog Comp

Diff

HS completion 73.9% 65.5%8.4%*

Years of ed 11.82 11.510.31*

Attend 4-year 16.7% 13.1% 3.6%College

No Group Differences using Contrast 2 Model

Economic StatusE-Prog Comp

Diff

Q income > $3000 38.3% 34.6% 3.7

Full-time employment 42.7% 36.4% 6.3*

Any public aid 58.8% 64.1% -5.3t

Contrast 2: Program group had > Full-time employment and < Months of Public Aid

Alternative SES IndicatorsE-Prog Comp

Diff

Occu. prestige 30.3% 25.9% 4.4%(4+ on scale)

Occu. prestige 2.87 2.61 0.26*(1-8 scale)

SES index 5.10 4.75 0.35*(2-10 scale)

Adult CrimeE-Prog Comp

Diff

Felony arrest 18.4% 20.6% -2.2%

Violence arrest 13.9% 17.9% -4.0%*

Violence convict. 5.5% 8.% -2.5%*

Incarceration/jail 21.7% 24.7% -3.0

No Group Differences using Contrast 2 Model

Health Status/BehaviorE-Prog Comp

Diff

Daily Tobacco 18.4% 20.0% -1.6%

Health insurance 69.7% 65.8% -3.9%

Depressive symp. 12.4% 15.6% -3.2%

Disability status 4.4% 7.0% -2.6%*

Contrast 2: Program group showed < disability

PK-3 Effects from ECLS-K Cohort

Selected children with these attributesPreschoolFull-day kindergartenSchool stable K-3High parent involvementInstructional time in readingTeacher certification

Rates of Grade Retention by Third Grade in ECLS-K Cohort

Total Low-sample income

1. No PK-3 13% 22%2. Pre-K+ADK 7% 11%

+Stable3. +HiPI+Cert 4% 9%

+HiInst

PK-3 Practices (Elements): Hierarchy of Effects on Child Development

Preschool participationSchool mobility/stabilityInstructional practices/Teacher qualityParent involvementReduced class sizesFull-day kindergarten

Benefit-Cost Ratios for CPC Program Components

10.15

2.12

9.05

-1

1

3

5

7

9

11

13

15

Pre

sen

t V

alu

e o

f B

en

efi

ts p

er

Do

llar

Investe

d (

$)

CPC Preschool Program CPC School age CPC extended

Conclusions and Implications

1. CPC PK-3 Model Demonstrated Positive Impacts on Some Indicators of Adult Well-Being.

2. Findings Varied by Comparison Group Approach.

3. Impact on Full-Time Employment, Public Aid, and Disability Status were most Consistent.

4. The CPC Model of PK-3 has Demonstrated relatively High Economic Returns.

Recommendations

1. Disseminate PK-3 Programs and Practices based on Key Principles of Effectiveness.

2. Use Evidence on Cost-effectiveness to Better Prioritize Funding of PK-3 Programs & Practices.

3. Different Models of PK-3 should be Investigated across Contexts.

4. More Evidence is Needed on the Relative Contributions of PK-3 Practices/Elements.

For more information about the Chicago Longitudinal Study, contact:

Arthur Reynolds, Institute of Child Development ([email protected])

Judy Temple, Humphrey Institute and Dept. of Applied Economics ([email protected])

Suh-Ruu Ou, Institute of Child Development ([email protected])

Funding support provided by:Foundation for Child DevelopmentNICHDDoris Duke Charitable Foundation

Web Site: www.education.umn.edu/icd/reynoldslab/ www.waisman.wisc.edu/cls/