Should the United States

Have Pre-K for All? Duke University, Durham NC

October 22, 2013

Steve Barnett, PhD

Why Offer Universal High-Quality Public Pre-K?

1. High-quality pre-K can produce large

benefits to society

2. Positive benefits can be produced on a large

scale--depends crucially on implementation

3. Targeting is ineffective and inefficient

4. Need not limited to those in or near poverty

5. UPK can be effective and benefits poor most

6. Pre-K for all has a larger net benefit

Potential Gains from Pre-K Investments

Educational Success and Economic Productivity

Achievement test scores

Special education and grade repetition

High school graduation

Behavior problems, delinquency, and crime

Employment, earnings, and welfare dependency

Smoking, drug use, depression

Decreased Costs to Government

Schooling costs

Social services costs

Crime costs

Health care costs (teen pregnancy and smoking)

Preschool programs 0-5 in the US:

Impacts in 123 studies since 1960

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Treatment End Ages 5-10 Age >10

Eff

ects

(1

sd

)= p

erc

etn

of

ac

h. g

ap

Age at Follow-Up

All Designs HQ Designs HQ Programs

What did we learn from meta-analysis besides

average effect sizes?

1. Intentional teaching matters

2. Individualization & small groups increase

gains

3. Comprehensive services decrease effects

– Why?

4. No other robust relationships except time

and quality

Perry Preschool Education Effects

60%

45%

66%

15%

40%

28%

77%

66%

85%

49%

68%

67%

0% 20% 40% 60% 80% 100%

HS Graduate

Graduated HS on Time

No Special Education

(Cog.)

Age 14 Achievement at

10th %ile +

Age 14 Homework

Age 5 IQ>90

Program group No-program group

Perry Preschool Crime Effects

34%

48%

55%

29%

27%

14%

33%

36%

7%

14%

0% 10% 20% 30% 40% 50% 60%

Drug Crime by 40

Violent Crime by 40

Arrested >5X by 40

Arrested >5X by 27

Discipline Problems ages 6-12

Program No Program

Perry Preschool Economic Effects

50%

60%

13%

62%

40%

7%

76%

82%

36%

76%

60%

29%

0% 10% 20% 30% 40% 50% 60% 70% 80% 90%

Had savings Account at 40

Own Car at 40

Own Home at 27

Employed at 40

Earned >$20K at 40

Earned >$20K at 27

No Program Program

Chicago CPC: Academic and

Social Benefits at School Exit

Temple, J. A., & Reynolds, A. J. (2007). Benefits and costs of investments in preschool education: Evidence from the Child-Parent Centers and related programs.

Economics of Education Review, 26(1), 126-144

Economic Returns to Pre-K

for Disadvantaged Children

(In 2006 dollars, 3% discount rate) Cost Benefits B/C

Perry Pre-K $17,599 $284,086 16

Abecedarian $70,697 $176,284 2.5

Chicago $ 8,224 $ 83,511 10

Barnett, W. S., & Masse, L. N. (2007). Early childhood program design and economic returns: Comparative benefit-cost analysis of the Abecedarian program and

policy implications, Economics of Education Review, 26, 113-125; Belfield, C., Nores, M., Barnett, W.S., & Schweinhart, L.J. (2006). The High/Scope Perry

Preschool Program. Journal of Human Resources, 41(1), 162-190; Temple, J. A., & Reynolds, A. J. (2007). Benefits and costs of investments in preschool

education: Evidence from the Child-Parent Centers and related programs. Economics of Education Review, 26(1), 126-144.

But don’t the effects fade-out

in large scale public programs?

Sometimes yes

Head Start and other “comprehensive programs”

Other “weak” public programs

Weak initial results

Few, if any, detectable lasting gains

These are intent-to-treat (under)estimates

They do not follow the models found to produce

large gains

Perry Preschool Cognitive Effects over Time

Good Preschool is Rare (ECLS-B)

Program Effectiveness Varies Greatly

(effect of 1 year at age 4)

CPC Tulsa 8 St Head St

Language NA NA .26 .13

Math .33 .36 .32 .18

Literacy NA .99 .80 .34

Effects in standard deviations. Head Start adjusted for crossovers in randomized trial.

• Teacher with BA & Cert. + asst. in each class;

• Full-day (6 hour educational day), 180-day program, plus extended day/full year;

• Access to all 3 and 4 yr. olds in 31 school systems

• Maximum class size of 15 students;

• Evidence-based curricula;

• Early learning standards and program guidelines;

• Support for potential learning difficulties;

• Professional development for key staff;

• Part of systemic reform P-12

NJ’s Counterexample

3.9

19.9

34.6

27.7

12.1

1.70.0 0.24.2

32.2

47.4

16.0

0

10

20

30

40

50

60

1.00-1.99 2.00-2.99 3.00-3.99 4.00-4.99 5.00-5.99 6.00-7.00

Perc

enta

ge o

f C

lassro

om

s

ECERS-R Score (1=minimal, 3=poor 5= good 7=excellent)

00 Total (N = 232) 08 Total (N = 407)

NJ Raised Quality in Public and Private

Abbott Pre-K Effects on NJASK by Years of Participation

.12

.18 .17

.14

.17

.26

.22

.37

.29

.37

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

LAL 4th LAL 5th Math 4th Math 5th Science 4 th

1 year Abbott pre-k 2 year Abbott pre-k

12% 12%

19% 17%

0%

2%

4%

6%

8%

10%

12%

14%

16%

18%

20%

Retention Special edcuation

Abbott pre-K no Abbott pre-K

Abbott Pre-K Effects on Retention and Special Education

• The US still does not serve most children in poverty

– Capacity is too limited

– Some parents do not want to send their children to a program that is just for the poor

• Major programs provided to children in poverty have smaller effects than “expected”

• Many children in targeted programs are not poor

– Poverty is a moving target

– There is great incentive to “cheat” for both families and programs

Targeting: 50 years of evidence

Cognitive Development Gap

Lost Potential Growth

Median Abilities of Entering Kindergarteners by Family Income

40.00

45.00

50.00

55.00

60.00

Lowest 20% 4th Quintile Middle 20% 2nd Quintile Highest 20%

Reading

Math

General

Know ledge

Lost Potential Growth

Barnett, W. S. (2007). Original analysis of data from the US Department of Education, National Center for Educational Statistics, ECLS-K Base Year Data files

and Electronic Codebook (2002).

Social Skills Gap Median Social Skills of Entering Kindergarteners by Income

8.40

8.60

8.80

9.00

9.20

9.40

9.60

Lowest 20% 4th Quintile Middle 20% 2nd Quintile Highest 20%

Lost Potential Growth

Barnett, W. S. (2007). Original analysis of data from the US Department of Education, National Center for Educational Statistics, ECLS-K Base Year Data files

and Electronic Codebook (2002).

Effects of UPK in the US

• Rhode Island Randomized Trial

– Positive gains for all, larger gains for low income children

• Oklahoma (multiple studies)

– Gains for all, larger gains for the lowest income children

– Grade 3 gains on attention and academic achievement

• Georgia, West Virginia, New Jersey all have studies with

positive effects

• Long-term positive effects in GA and NJ

• BCA in GA, earnings gains alone may exceed cost

Effects of UPK Globally

OECD test scores higher and more equal as

access approaches 100%

France: Ecole Maternelle increased income

Norway: universal child care increased earnings

and employment

Arg. Uru. and UK: universal pre-K raised long-

term achievement

Denmark, Quebec: universal child care null or

negative effects on children--quality matters

CBA of Targeting vs. Universal

Targeted: Lower cost, fewer benefits Screening and identification is costly and imperfect

Still the population served has greater need on avg.

Many who need the program are not served

Pre-K for all: More benefits but higher cost Reaches all disadvantaged children

Larger gains for disadvantaged children

Benefits children from middle income families too

Smaller average benefit

Sliding fee scale can reduce public cost Source: Barnett (2004). Maximizing returns from pre-kindergarten education. Federal Reserve Bank of Cleveland

Research Conference.

Simulated Economic Returns

Cost Benefits NPV

Targeted

50% correct $12.5 $ 79.9 $ 67.4

80% correct $12.5 $ 96.0 $ 83.5

Universal

80% enroll $62.4 $213.2 $150.8

Source: Barnett (2004). Maximizing returns from pre-kindergarten education. Federal Reserve Bank of Cleveland Research Conference.

Conclusions 50 years of means-testing has largely failed

Inadequate coverage, quality, and results

This is not necessary--programs can succeed at scale

UPK has achieved full coverage and higher quality

UPK permits positive peer effects & systemic change

Benefits are not limited to the disadvantaged

UPK benefits all children while reducing inequality

UPK plausibly yields a larger net economic benefit with an adequate (if lower) rate of return

References

1. Barnett, W. S. (2011). Effectiveness of early educational intervention. Science, 333, 975-978.

2. Barnett, W. S. (2011). Four reasons the United States should offer every child a preschool education. In E. Zigler, W. Gilliam, & W.

S. Barnett (Eds.), The pre-k debates: current controversies and issues (pp. 34-39). Baltimore: Brookes Publishing.

3. Barnett, W. S., & Frede, E. C. (2010). The promise of preschool: Why we need early education for all. American Educator, 34(1),

21-40.

4. Barnett, W. S. (2013). Getting facts right on Pre-K and the President’s Proposal. nieer.og

5. Burger, K. (2010). How does early childhood care and education affect cognitive development? An international review of the

effects of early interventions for children from different social backgrounds. Early Childhood Research Quarterly, 25, 140-165.

6. Camilli, G., Vargas, S., Ryan, S., & Barnett, W.S. (2010). Meta-analysis of the effects of early education interventions on cognitive

and social development. Teachers College Record, 112(3), 579-620.

7. Dumas C. & Lefranc, A. (2010). Early schooling and later outcomes: Evidence from preschool extension in France. THEMA

Working Paper 2010-07. Université de Cergy-Pontoise.

8. Frede, E. C., & Barnett, W. S. (2011). New Jersey’s Abbott pre-k program: A model for the nation. In E. Zigler, W. Gilliam, & W. S.

Barnett (Eds.), The pre-k debates: current controversies and issues (pp. 191-196). Baltimore: Brookes Publishing.

9. Gormley, W., Phillips, D., Newmark, K., Welti, K., & Adelstein, S. (2012). Social-emotional effects of early childhood education

programs in Tulsa. Child Development, 82, 2095-2109.

10. Havnes, T. & Mogstad, M. (2011). No Child Left Behind: Subsidized Child Care and Children's Long-Run Outcomes. American

Economic Journal: Economic Policy, 3(2): 97–129.

11. Hill, C., Gormley, W., & Adelstein, S. (2012). Do the short-term effects of a strong preschool program persist? Working Paper No.

18. Washington, DC: Georgetown Unversity, CROCUS.

12. Lipsey, M., Farrran, D., Bilbrey, C., Hofer, K., & Dong, N. (2011). Initial results of the evaluation of the Tennessee Voluntary Pre-K

Program. Nashville: Vanderbilt University.

13. Neidell, M., & Waldfogel, J. (2010). Cognitive and noncognitive peer effects in early education. The Review of Economics and

Statistics, 92(3), 562-576.

14. Sylva, K., Melhuish, E., Sammons, P., Siraj-Blatchford, I., Taggart, B. (2004). The final report: Effective pre-school education.

Technical paper 12. London: Institute of Education, University of London.

15. Waldfogel, J., & Zhai, F. (2008). Effects of public preschool expenditures on the test scores of fourth graders: Evidence from

TIMMS. Educational Research and Evaluation, 14, 9–28.