Programs and Policies to Reduce School Readiness Gaps Jeanne Brooks-Gunn Virginia and Leonard Marx...

Programs and Policies to Programs and Policies to Reduce School Readiness Reduce School Readiness

GapsGapsJeanne Brooks-Gunn

Virginia and Leonard Marx Professor of Child Development and EducationTeachers College and the College of Physicians and Surgeons

Columbia University

March 2, 2009

National Center for Children & FamiliesNational Center for Children & FamiliesTeachers College, Columbia UniversityTeachers College, Columbia University

www.policyforchildren.orgwww.policyforchildren.org

National Center for Children and Families (NCCF) - Teachers College, Columbia University National Center for Children and Families (NCCF) - Teachers College, Columbia University www.policyforchildren.org www.policyforchildren.org

NAEP Reading Assessment for NAEP Reading Assessment for Grade 4 (1994) & Grade 12 (2002) ― Grade 4 (1994) & Grade 12 (2002) ― Percentage of Students At or Above Percentage of Students At or Above

Proficient by Race/EthnicityProficient by Race/Ethnicity43%

16%

22%

34%37%

9%

13%

44%

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

White Black Hispanic AsianAmerican

Race/Ethnicity of Student

Per

cen

tag

e o

f S

tud

ents

Grade 12 Grade 4

Rouse, Brooks-Gunn & McLanahan, 2005


Gaps in Kindergartners’ Achievement: Gaps in Kindergartners’ Achievement: Ethnicity and Social ClassEthnicity and Social Class

Duncan & Magnuson, 2005

-0.61

-0.39

-0.71

-0.45

-0.9

-0.8

-0.7

-0.6

-0.5

-0.4

-0.3

-0.2

-0.1

0S

tan

da

rd d

ev

iati

on

fro

m w

hit

es

Blacks Hispanics

ECLS-K Math ECLS-K Reading


What defines “school readiness”?What defines “school readiness”?

A child’s academic skills (e.g., vocabulary size, complexity of spoken language, basic counting, “general knowledge”);

A child’s social and emotional skills (e.g., following directions, working in a group, impulse control) – which are as important as academic skills but not well studied.


Children who are not “ready” Children who are not “ready” for school are more likely to…for school are more likely to…

Perform less well in elementary & secondary school;Become teen parents;Engage in criminal activities;Suffer from depression.


(From Jencks & Phillips, 1998)

0

5

10

15

20

25

30

20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100

P P V T -R S co re (B lack M ed ian =40; W h ite M ed ian =52)

Per

cen

t of P

op

ula

tion

B lackW hite

Vocabulary Scores for Black and Vocabulary Scores for Black and White Three and Four-year OldsWhite Three and Four-year Olds


Standard Deviation UnitsStandard Deviation Units

If the standard deviation = 15 and white students score 12 points more than black students on average (a white-black gap of 12), then…

12/15 points = .80 standard deviation units(80% of the standard deviation)


If the white-minority gap in test If the white-minority gap in test scores is 1 standard deviation scores is 1 standard deviation then…then…84% of white children will perform better than the average minority child

16% percent of minority children will perform better than the average white child

Whites students are 13 times more likely than minority students to score in the top 5% and enroll in “gifted” class

The average textbook is written so that the average white student understands 75% of it

The average minority student will understand 53% of the same textbook, virtually guaranteeing that such a reader will not engage with the text

(From Rock & Stenner, 2005)


Racial / Ethnic Gaps in Selected Test Racial / Ethnic Gaps in Selected Test Scores and in Family Socio-economic Scores and in Family Socio-economic

Status for KindergartnersStatus for Kindergartners

Duncan & Magnuson, 2005

-0.61

-0.39

-0.70-0.71

-0.45

-0.78

-0.9

-0.8

-0.7

-0.6

-0.5

-0.4

-0.3

-0.2

-0.1

0

Sta

nd

ard

de

via

tio

n f

rom

wh

ite

s

B lacks H ispanics

EC L S-K M ath EC L S-K Re ad in g So cio e co n o m ic In d e x

Growth Trajectories of Two Groups of Growth Trajectories of Two Groups of ChildrenChildren

*Children whose families are in repeated poverty (poverty at the time of the Fall K test and poverty at one or more subsequent measurement points).

National Center forNational Center for Children and Families (NCCF) - Teachers College, Columbia UniversityChildren and Families (NCCF) - Teachers College, Columbia Universitywww.policyforchildren.orgwww.policyforchildren.org

Layzer, 2008

Full sample

Repeat poverty

Poverty Affects DevelopmentPoverty Affects Development

16 mos. 24 mos. 36 mos.

Cu

mu

lati

ve V

ocab

ula

ry (

Word

s) College Educated Parents

Working Class Parents

Welfare Parents

Child’s Age (Months)

200

600

1200

Hart & Risley, 1995


Language Gaps Increase During Language Gaps Increase During Preschool YearsPreschool Years

Age span of children Median deficit Sample size

CCDP

3 years to 3 years,11 months 8.7 months 2541

4 years to 4 years,11 months 13.8 months 2360

5 years to 5 years, 9 months 14.8 months 2878

Even Start

3 years to 3 years,11 months 8 months 2187

4 years to 4 years,11 months 12 months 2805

Layzer, 2008


• Parenting– Home Visiting Programs– Center-based Programs with Parenting

Component– Parental Language and Literacy

Programs– Parent Behavior Training Programs

Parenting Programs as a StrategyParenting Programs as a Strategy



Early Childhood Education as a Early Childhood Education as a StrategyStrategy

Access and attendance• Low access for specific groups• Head Start and access• Most common child care arrangements• Access and school readinessQuality• Quality indicators• High quality programs• Quality and school readiness


Preschool Enrollment for Four Preschool Enrollment for Four Year-olds, from 1968 to 2002 Year-olds, from 1968 to 2002

0

10

20

30

40

50

60

Year

Black, Non-Hispanic

Hispanic

White, Non-Hispanic

Magnuson & Waldfogel, 2005

National Center for Children and Families (NCCF) - Teachers College, Columbia University National Center for Children and Families (NCCF) - Teachers College, Columbia University www.policyforchildren.orgwww.policyforchildren.org

State Pre-K and Head Start Enrollment State Pre-K and Head Start Enrollment as Percentage of Total Populationas Percentage of Total Population

3-Year-Olds 4-Year-Olds

Pre-K 2.7 17.3

Head Start

7.3 11.3

Special Ed

3.9 6.2

Other 24.8 33.6

None 61.3 31.6

The National Institute for Early Education Research, The State of Pre-School 2005


Head Start EnrollmentHead Start Enrollment

About 10% of all of all 3 and 4 year-olds are in Head Start– 20% of black children– 15% of Hispanic children– 4% of white children

If Head Start did not exist, gaps in preschool enrollment would increase (compared to white children)– 9 percentage points for black children– 31 percentage points for Hispanic children

Gaps in school readiness would increase if Head Start did not exist

School Readiness: Comparing Head Start School Readiness: Comparing Head Start and Other Care Arrangementsand Other Care Arrangements

OLS Prop

PPVT-III 0.08 0.19

WJ-R Letter-Word Identification 0.11 0.16

Social Competence 0.10 0.14

Attention Problems -0.11 -0.16

Zhai, Brooks-Gunn, & Waldfogel, 2009

N=2803; OLS includes city- & child-fixed effects; Prop. indicates propensity matchingBold indicated significance. Fragile Families Study

Access and School ReadinessAccess and School Readiness

If all children whose families were in poverty were in a preschool, reductions in test gaps would be:

• 2 to 12 percent for black children• 4 to 16 percent for Hispanic children

If all children whose families were at or below 200% of the poverty threshold, were in a preschool, reductions in test gaps would be possibly double


Reading Scores By Preschool Reading Scores By Preschool Attendance & Kindergarten Reading Attendance & Kindergarten Reading

InstructionInstruction

N=7748; ECLS-K


Magnuson, Ruhm, & Waldfogel, 2007


Indicators of High Quality Indicators of High Quality PreschoolPreschool

• Teacher training• Teacher education• Teacher to child ratios• Class size• Language and conversation• Organization of classroom• Continued training


Child Care Global Quality Child Care Global Quality MeasuresMeasures

Early Childhood Environment Rating Scale-ECERS (37 items)• Space and furnishings• Personal care routines• Language-Reasoning• Activities• Interaction• Program Structure

Each item is scored 1-7 with average of all items used as quality score:• 1 = inadequate• 3= minimal• 5= good• 7= excellent


Distribution of Quality in Fragile Distribution of Quality in Fragile Families and Child Well-being StudyFamilies and Child Well-being Study

Minimal37%

Good33%

Excellent10%

Inadequate20%

Rigby, Ryan & Brooks-Gunn, 2006.

Data based on 786 three-year-olds in 13 Cities from the Fragile Families Study


Quality of Care in Different Quality of Care in Different SettingsSettings

Care ratings:• Kith and kin = 3.5• Family child care = 4. 0• For-profit centers = 4.2• Non-profit centers = 5.0

1=inadequate, 7=excellent

Rigby, Ryan, and Brooks-Gunn, 2007


High Quality Preschool Education High Quality Preschool Education and School Readinessand School Readiness

1. Large effects for small-scale experiments (one-half to three-quarters of a SD)

2. Small to modest effects for federal programs (one-sixth to one-quarter of a SD)

3. Effects for mothers with a high school education or less, but not for mothers with BA

4. Effects sometimes larger for black than white or Hispanic children

Small-scale Early Childhood Education Small-scale Early Childhood Education EvaluationsEvaluations

• Perry Preschool• Abecedarian• Infant Health & Development Program

• Parent-Child Development Centers• Tennessee Program• Planned Variation Program


Infant Health & Development Program Infant Health & Development Program Intervention ServicesIntervention Services

• Intervention Group– Home visiting in child’s 1st – 3rd years

of life– Day care center in child’s 2nd and 3rd

years of life• Transportation available• Center open all workday

• Intervention and Follow-up Only Groups– Free medical surveillance, referrals to

specialists in child’s 1st – 3rd years of life


Infant Health & Development Program:Infant Health & Development Program:Impacts from Age 3 to 18Impacts from Age 3 to 18

*Heavier low birth weight children only*All impacts were significant

IQ PPVT

3 Years 14.3 9.4

5 Years 3.7 6.0

8 Years 4.4 6.7

18 Years 3.3 5.1


Brooks-Gunn et al., 1994, JAMA; McCarton et al., 1997, JAMA; Table 3 & 4, McCormick et al., 2006, Pediatrics

Infant Health & Development Program:Infant Health & Development Program:Impacts on Math AchievementImpacts on Math Achievement

WJ Math

8 Years 4.9

18 Years 3.6


*Heavier low birth weight children only*All impacts were significant

Brooks-Gunn et al., 1994, JAMA; McCarton et al., 1997, JAMA; Table 3 & 4, McCormick et al., 2006, Pediatrics

Infant Health & Development Program: Infant Health & Development Program: IQ Impacts Based on Likely Care SituationIQ Impacts Based on Likely Care Situation

Age 3 Age 5 Age 8

Maternal Care 20.2 12.3 8.2

Other Home Based Care 16.4 5.9 7.2

Center Based Care 11.6 2.0 1.7

Hill, Waldfogel, Brooks-Gunn, 2002, JPAM


*Heavier low birth weight children only

School Readiness: Comparing Head School Readiness: Comparing Head Start, Parent, and Other Non-center CareStart, Parent, and Other Non-center Care

Parent Non-Center

OLS Prop. OLS Prop.

PPVT-III 0.21 0.33 0.21 0.32

WJ-R 0.39 0.46 0.32 0.41

Social Competence 0.17 0.24 0.03 0.05

Attention Problems -0.11 -0.10 -0.18 -0.19


N=2803; OLS includes city- & child-fixed effects; Prop. indicates propensity matchingBold indicated significance.Fragile Families Study


School Readiness: Comparing Head Start and School Readiness: Comparing Head Start and Other Center-based CareOther Center-based Care

Pre-K Other Center

OLS Prop. OLS Prop.

PPVT-III 0.06 0.09 0.04 0.09

WJ-R -0.02 0.01 0.03 0.05

Social Competence 0.13 0.15 0.09 0.17

Attention Problems -0.06 -0.05 -0.15 -0.18


N=2803; OLS includes city- & child-fixed effects; Prop. indicates propensity matchingBold indicated significance.Fragile Families Study


Infant Health & Development Program: Infant Health & Development Program: Impacts As A Function Of Number Of Impacts As A Function Of Number Of

Days Of Center-Based CareDays Of Center-Based Care

Hill, Brooks-Gunn, Waldfogel, 2003, DP


Total Sample

>350 days of treatment over 2 years

Age 3

IQ 14.3 16.7

Age 8

IQ 4.4 8.4

*Heavier low birth weight children only

Federal ECE Programs and School Federal ECE Programs and School ReadinessReadiness

• Head Start Impact Study• Early Head Start National Evaluation• Early Reading First• Preschool Curriculum Evaluation

Research Study (PCERS)

All randomized trials of center-based early childhood education programs

Layzer, 2008


Impacts of Federal ECE EffortsImpacts of Federal ECE Efforts• Head Start: modest effects on letter recognition,

none on language

• Early Head Start has modest effects across a range of outcomes

• Modest effect of Early Reading First

• Some impacts in PCERS of many different curricula

Layzer, 2008



Change in Policy Scores over TimeChange in Policy Scores over Time

-0.6

-0.4

-0.2

-0

0.2

0.4

0.6

1991 1994 1998 2002

ECE Preschool Subsidy Tax Regulation

Avera

ge P

olicy S

core

Notes: N= 50 at each time point. Subsidy policy levels in 1991 unknown. Thus, 1994 policy levels are used to proxy 1991 subsidy policy choices.

NCCF, State ECE policies


Five Different Policy ApproachesFive Different Policy Approaches

Policy Approaches

Number of States

States

All 1 MN

Preschool Investment

8 KY, IL, NJ, OK

Child Care Tax 8 AR, CO, OH, OR

Subsidy and/or Regulation

21 RI, VA, VT, WI

Limited Policies 12 AL, AZ, MS, SC



Improving Child Care Quality: Improving Child Care Quality: ProgramsPrograms

• Be of high quality (e.g., small classes with high teacher- pupil ratios, teachers with BA degrees and training in early childhood education).

• Train teachers to identify children with moderate to severe behavioral problems and to work with them to improve their emotional and social skills.

• Include a parent-training component for mothers of low-literacy.

• Include (and document) high attendance• Include well-defined and tested curricula

ChallengesChallenges

• Sites vs. populations• Dosage• Timing and length of early childhood

education• Immigrants• Curricula• Training

Heckman, J. “Investing in Disadvantaged Young Children Is Good Economics and Good Public Policy,” Testimony before the Joint Economic Committee, Washington D.C., June 27, 2007

Rates of Return to Human Capital Investment(Return to an extra dollar at various ages)

ExtraExtra


State Policies Affect All ChildrenState Policies Affect All Children

• Preschool investment

• Subsidy generosity

• Regulation stringency

• Tax generosity



States with the Lowest Scores (1990-2002)States with the Lowest Scores (1990-2002)

-2.5

-1.5

-0.5

0.5

1.5

2.5

Alabama Arizona Idaho North Dakota South Dakota

Preschool Investment Subsidy Generosity Tax Generosity Regulation Stringency

Sta

nd

ard

ized

Policy

Score



States with the Highest Scores (1990-States with the Highest Scores (1990-2002)2002)

-2.5

-1.5

-0.5

0.5

1.5

2.5

Alaska Connecticut New York Hawaii Minnesota

Preschool Investment Subsidy Generosity Tax Generosity Regulation Stringency

Sta

nd

ard

ized

Policy

Score


Sustained Impacts of ECE Programs: Sustained Impacts of ECE Programs: Life-time EarningsLife-time Earnings

• Outcome in adulthood– Life-time earnings based on completing

high school or some college

• Inputs in early childhood– Health (reduction of low birth weight)– Achievement (increase in achievement

test scores)– Parenting (increase in HOME scores)

Magnusson, Brooks-Gunn & Waldfogel, 2009


Reducing Low Birth Weight Rate: Reducing Low Birth Weight Rate: Life-time EarningsLife-time Earnings

• If LBW, 10 % to 15% reduction in life time earnings

• Value of $41,156 (10% effect of LBW on life time earnings)

• Value of $61,733(15% effect of LBW on life time earnings)



Increasing Early Math and Reading Skills:Increasing Early Math and Reading Skills:Life-time EarningsLife-time Earnings

• One SD increase in academic skills in adolescent is associated with 15% to 20% increase in LTE

• One SD increase in academic skills at ages 5-6 is associated with .16 (reading) & .22 (math) SD increase in youth achievement

• One SD increase in math and reading at ages 5-6 is associated with a .38 SD increase in youth achievement



Increasing Early Academic Skills: Increasing Early Academic Skills: Life Time EarningsLife Time Earnings

If one SD increase in early skills, assuming a 15% or 20% effect of youth achievement on LTE:

• $33,138 if math & reading skills at 15% effect

• $44,184 if reading & math skills & 20% effect



Increasing Quality of the HOME: Increasing Quality of the HOME: Life-time EarningsLife-time Earnings

One SD increase in HOME in early childhood is associated with .10 to .28 increase in youth achievement:

• $ 9,941 if .12 improvement in youth achievement & 15% effect

• $13,255 if .12 improvement in youth achievement & 20% effect

• $21,125 if .255 improvement in youth achievement & 15% effect



Programs and Policies to Reduce School Readiness Gaps Jeanne Brooks-Gunn Virginia and Leonard Marx...

Documents

Transcript of Programs and Policies to Reduce School Readiness Gaps Jeanne Brooks-Gunn Virginia and Leonard Marx...