1

The Early Head Start Research

and Evaluation Project and Beyond:

Lessons Learned and Implications

for Longitudinal ECD Research

Kimberly Boller

The Maria Cecilia Souto Vidigal Foundation’s

Second International Workshop on ECD

Sao Paulo, October 3, 2008

2

Overview

Review of Early Head Start, the national

evaluation findings, the Survey of EHS

programs, and Baby FACES

Impacts of EHS on pre-k program participation

and quality

Impacts of EHS on school readiness and family

functioning when children were at pre-k

The special stories of subgroups

How programs evolved

The newest study: Baby FACES

Lessons learned

3

The Early Head Start Program

More than 650 programs serving more

than 65,000 children

Various originating agencies

About 10% of the Head Start budget

Follows the Head Start Program

Performance Standards

4

Early Head Start Is an Intensive,

Two-Generation Program

Self-Sufficiency and

Healthy Families

Parenting

Child

Development

5

Services Among Early Head Start

Programs Are Diverse

FOUR PROGRAM MODELS:

Home-only: programs provide home based services to

all families

Center-only: programs provide center-based services to

all families

Multiple: programs provide center-based services to

some families and home-based to others, some may get

both

Combination: programs provide center-based services

and frequent home visits to all families

6

7

Readiness for School and Life

Ed Zigler’s definition of social competence,

“the child’s everyday effectiveness in dealing with his

environment and later responsibilities in school and life.

Social competence takes into account the interrelatedness of

cognitive and intellectual development, physical and mental

health, nutritional needs, and other factors that enable a

child to function optimally.”

Quoted in Raizen and Bobrow (1974)

8

Policy and Economic

Context

Welfare reform enacted July 1996, as

research programs began enrolling

families

Strong economy with low

unemployment rates

Increase in needs for child care

Some programs changed their

approach based on family child care

needs

9

The Early Head Start Research

and Evaluation Study

♦ Began in 1995

♦ 3,001 children and families randomly assigned

to EHS program or control group in impact

study with experimental design:

♦ Followup: (1) age 14, 24, and 36 months; (2)

when children in prekindergarten

♦ Fifth grade followup in progress

♦ Response rates are a challenge

Early Head Start Group Control Group

10

Early Head Start Research

Sites

Brattleboro,

VT

New

York,

NY

Pittsburgh,

PA

Sumter, SC

Jackson,

MI

Russellville,

AR

Kansas City,

MO

Marshalltown,

IA

Kansas

City, KS

Logan,

UTDenver, CO

(2 programs)

Kent, WA

Sunnyside,

WA

Alexandria,

VA

Venice,

CA

McKenzie,

TN

11

Research Conducted by Early

Head Start Research Consortium

Representatives from 17 programs participating

in the evaluation, 15 local research teams, the

evaluation contractors, and ACF/ACYF

Research institutions in the Consortium (and principal researchers) include ACF (Rachel Chazan

Cohen, Judith Jerald, Esther Kresh, Helen Raikes, and Louisa Tarullo); Catholic University of America

(Michaela Farber, Lynn Milgram Mayer, Harriet Liebow, Christine Sabatino, Nancy Taylor, Elizabeth

Timberlake, and Shavaun Wall); Columbia University (Lisa Berlin, Christy Brady-Smith, Jeanne

Brooks-Gunn, and Alison Sidle Fuligni); Harvard University (Catherine Ayoub, Barbara Alexander Pan,

and Catherine Snow); Iowa State University (Dee Draper, Gayle Luze, Susan McBride, Carla

Peterson); Mathematica Policy Research (Kimberly Boller, Ellen Eliason Kisker, John M. Love, Diane

Paulsell, Christine Ross, Peter Schochet, Susan Sprachman, Cheri Vogel, and Welmoet van

Kammen); Medical University of South Carolina (Richard Faldowski, Gui-Young Hong, and Susan

Pickrel); Michigan State University (Hiram Fitzgerald, Tom Reischl, and Rachel Schiffman); New York

University (Mark Spellmann and Catherine Tamis-LeMonda); University of Arkansas (Robert Bradley,

Mark Swanson, and Leanne Whiteside-Mansell); University of California, Los Angeles (Carollee

Howes and Claire Hamilton); University of Colorado Health Sciences Center (Robert Emde, Jon

Korfmacher, JoAnn Robinson, Paul Spicer, and Norman Watt); University of Kansas (Jane Atwater,

Judith Carta, and Jean Ann Summers); University of Missouri-Columbia (Mark Fine, Jean Ispa, and

Kathy Thornburg); University of Pittsburgh (Carol McAllister, Beth Green, and Robert McCall);

University of Washington School of Education (Eduardo Armijo and Joseph Stowitschek); University of

Washington School of Nursing (Kathryn Barnard and Susan Spieker); and Utah State University (Lisa

Boyce and Lori Roggman).

12

Families of All Types and

Backgrounds

13

Positive, Modest Impacts for

Children at Age 3

♦ Higher immunization rate

♦ Fewer emergency room visits for

accidents and injuries

♦ Cognitive development (higher Bayley

scores & fewer in low-functioning group)

♦ Larger receptive vocabularies

♦ Lower levels of aggressive behavior

♦ Greater sustained attention with objects,

engagement of parent, and less

negativity

14

But, Disparities Begin Early

90.2 88.0 91.4 89.9

83.3 81.1 82.9 81.4

0

10

20

30

40

50

60

70

80

90

100

EHS Bayley Age 2 EHS Bayley Age 3 EHS PPVT Age 3 HSIS PPVT Age 3

Treatment Control

15

Positive, Modest Impacts for

Parents

♦ More positive (and less negative)

parenting observed in parent-child play:

both mothers and fathers

♦ More stimulating home environments,

support for learning

♦ More daily reading

♦ Less spanking: both mother and father

report

♦ More hours in education and job training

♦ Probability of being employed (trend)

16

Parenting Example: More EHS than

Control Group Parents Read to Children

Daily

5752

0

10

20

30

40

50

60

70

80

90

100

Read Daily

Program Control

17

Parenting Example: EHS Parents Were

Less Likely to Spank Their Children

4754

0

20

40

60

80

100

Program Control

18

Important Subgroup Stories:

Programs

All program approaches had favorable

impacts. Mixed approach had strongest.

Well-implemented programs affected more

outcomes when children were 3:

♦ Child outcomes

♦ Parent-child interactions

♦ Parenting

♦ Mental health

♦ Progress toward economic self-sufficiency

Strongest impacts found in early-

implemented mixed programs

19

Important Subgroup Stories:

Families

Positive Impacts in 28 of 29 subgroups

Larger impacts in important subgroups:

♦ African American

♦ Enrolled during pregnancy

♦ Moderate demographic risk

No impacts: highest level of demographic

risk

Promising effects: families affected by

maternal depression

20

Examining EHS Children’s

Experiences After Age 3

♦ 1994 Advisory Committee: continuity

and smooth transitions

♦ While in EHS, families got help

accessing quality child care.

♦ At age 3, 52% of EHS children were in

center child care at least 30 hours per

week.

♦ EHS provided transition services as

children left their program.

♦ Control group families were on their

own.

21

Three Main Hypotheses

♦ EHS will increase children's enrollment in

formal care and education programs (FPs)

after EHS.

♦ Children’s program participation after EHS will

be moderated by:

♦ Characteristics of the programs they were in

♦ Community characteristics

♦ Family demographics

♦ EHS participation will increase the likelihood

that children will attend good-quality formal

programs after EHS.

22

What Difference Did Being in

EHS Make on Program

Participation After EHS?

• More likely to be in Head Start at some

time, 3-5 (55% vs. 49%)

• More likely to be in a formal ECE

program both 3-4 and 4-5 (47% vs. 42%)

23

What Difference Did it Make

for Particular Children?

• White children: more likely ever to be in

Head Start (46% vs. 35%)

• Moderate-risk children: more likely ever

to be in Head Start (60 vs. 51%)

• Low-risk children: more likely to be in

any formal ECE both periods (48 vs. 36%)

24

What Difference Did EHS make

on Enrollment in Good-Quality

Programs

♦ 60% of EHS children were in a formal

pre-k program with ECERS-R of 5 or

better.

61% of control group children were.

♦ Average quality of centers we observed

was good: ECERS-R of 5.25.

No difference in average quality of programs EHS and control children were in.

♦ Quality of Head Start centers children

attended was higher than quality of

other formal programs (5.6 vs. 5.0).

25

Context for Understanding

Early Head Start’s

Prekindergarten Impacts

♦ Impacts assessed 2 years after children

and families left the program.

♦ Control group had considerable program

experience after age 3.

♦ Quality of control group’s pre-k program

experience was same as for EHS

children.

♦ Quality of all observed ECE settings

apparently good.

26

EHS Impacts on Children Two

Years After Early Head Start

♦ Decreased behavior problems

♦ Higher level of positive approaches to

learning

♦ Larger receptive vocabularies for

Spanish-speaking children, but not for

English speakers

♦ No impacts on achievement-related

outcomes: letter-word identification,

applied problem solving

27

EHS Impacts on Parents and

Home Environment 2 Years

After Early Head Start

♦ Higher percentage read to child daily

♦ Higher scores on HOME total scale and

warmth scale

♦ Higher on summary of 8 teaching

activities

♦ Lower risk for maternal depression

♦ Parent more likely to attend meetings or

open houses at child’s program (if child

was in a program)

28

Association of Children’s Participation

in Formal Programs from Birth to

Age 5 with Age 5 Outcomes

Formal Programs, 3-5

More-aggressive behavior (negative association)

Higher pre-academic skills (letter-word

identification)

More IEPs

More parent reading to child

Ever in Head Start

Higher pre-academic skills (letter-word

identification and applied problems)

More likely to have an IEP

Reduction in negative parental regard (interaction)

No association with aggressive behavior

29

Highest-Risk Families Appear to

Benefit from 5 Years of

Comprehensive Services

No positive impacts at 3

Some positive impacts at 5

♦ Improved approaches to learning

♦ Reduced living with someone using drugs

♦ Reduced neighborhood exposure to violence

♦ Reduced parent experiencing abuse

♦ But reduced letter-word identification (negative impact)

Ages 3-5: Least likely to be in formal

programs

♦ 39% for high risk vs. 47% for low and 43% for moderate risk.

♦ EHS did not increase in enrollment in formal programs.

♦ EHS did not increase use of Head Start but 57% of highest risk were in Head Start at some time 3-5

30

So far, we’ve seen:

♦ Impacts of EHS observed at 36 months

♦ Impacts of EHS on program participation after EHS

♦ Impacts of EHS observed at prekindergarten

Next: which 36-month outcomes mediate

pre-k outcomes?

31

Do EHS Child Outcomes at 36

Months Mediate

Prekindergarten Outcomes?

EHS children’s developmental outcomes at 36

months, particularly their cognitive abilities and

their ability to engage their mothers in play, were

important mediators.

32

Overall Summary

EHS has impacts at age 5.

0-3 and 3-5 experiences contribute to

child and parent pre-k outcomes in

complementary ways. Greatest

benefits when 0-3 program is followed

by 3-5 experiences.

EHS serves as protective factor for

negative social-emotional outcomes.

For families at highest risk, may need

comprehensive services 0-5 to see

benefits.

Many lessons learned…Research to

Practice materials.

33

Putting it All Together: At Age 5

Children with EHS and 3-5 fared the

best, followed by those with EHS only

(for child social-emotional and parent

outcomes) or HS/formal program only

(for child school-related outcomes).

Important for 0-3 services to be

supported by 3-5 services

A National Perspective on

Early Head Start

35

More Than Half of Early Head Start

Programs Provide Multiple Options

17%

23%

51%

9%

Home-Based

Center-Based

Multiple

Combination

Source: Survey of Early Head Start Programs, 2006

Sample size: 660 Programs.

36

Highest Degree Held by

Front Line Staff

0

20

40

60

80

Primary

Caregivers

Home Visitors

Graduate

BA/BS

AA

CDA/Equivalent

Percentage of Programs

Note: Indicates percentage of programs who employ 50 percent

or more staff holding each credential.

37

High Risk Families in

Early Head Start

Demographic Risks

1. Teen mother

2. Single parent

3. Lack HS/GED

credential

4. Unemployed/Not in

school

5. More than 3 of above

Psychological Risks

1. Mental health problems

2. Substance abuse

problems

3. Unsafe neighborhood

4. Family violence

5. More than 2 of above

risks

Risk factors present challenges for serving families

38

Concentration of High Demographic

Risk Families

0%

20%

40%

60%

80%

100%

Teen Mother Single

Parent

No HS/GED Welfare Unemployed >3 Risks

Very

High

High

Med.

Low

Percentage of Programs

39

Concentration of High Psychological

Risk Families

0%

20%

40%

60%

80%

100%

Mental Health Substance

Abuse

Unsafe

Neighborhood

Family

Violence

>2 Risks

Very

High

High

Med.

Low

Proportion of programs

40

Race/Ethnicity of

Early Head Start Families

26%

25%

4%33%

6%

6%White

Black/Af. Am.

Hispanic/Latino

AmericanIndian/Alaska Native

Biracial/Multiracial

Asian/Hawaiian/PacificIslander/Other

Source: Survey of Early Head Start Programs N = 55,611

families.

41

Primary Non-English Language

of Early Head Start Families

81%

3%

3%

13%

Spanish

Asian

European

Other

Source: Survey of Early Head Start Programs, 2006 Sample size:

12,930 families speaking language other than English.

42

Percentage of Early Head Start

Children with Any Suspected or

Diagnosed Disability

0

20

40

60

80

None 1-10 11-20 21-30 30+

Percentage of Enrolled Children

Percentage of Programs

43

Early Head Start

Community Partnerships

0

20

40

60

80

100

Child Care Health Mental

Health

Part C

Percentage of Programs

Percentage of programs with a formal written partnership with each type of

community partner.

44

Early Head Start Family and

Child Experiences Survey

Baby FACES

• Nationally representative sample of 90

programs

• 2,000 children in two age cohorts

♦ Perinatal

♦ Age 1

• Longitudinal follow-up until children 3

years old

♦ Supplemental survey of age 1 cohort when they

are 3.5 to learn about transitions

45

Baby FACES Data and

Respondents

Ongoing

• Family service use information

Annual 0 to 3

• Parent reports

• Home visitor/Teacher reports

• Classroom and home visit

observations

• Program director interviews

Information at 2 and 3

• Direct child assessments

• Parent/child interactions

46

For More Information on All EHS

Research

http://www.acf.hhs.gov/programs/opre

47

Learn More About US EC Policy

and Evaluation Projects

OPRE

http://www.acf.hhs.gov/programs/opre/project/projectIndex.

jsp#hs

Child Care & Early Education Research Connections

http://www.childcareresearch.org/discover/index.jsp

NIEER

http://nieer.org/

NCEDL

http://www.fpg.unc.edu/ncedl/

ECLKC

http://eclkc.ohs.acf.hhs.gov/hslc