IMPROVING ACHIEVEMENT AND CLOSING GAPS BETWEEN GROUPS: Lessons from Schools and Districts on the Performance Frontier 22 nd Annual Washington State Assessment

IMPROVING ACHIEVEMENT AND CLOSING GAPS BETWEEN GROUPS: Lessons from Schools and Districts on the Performance Frontier22nd Annual Washington State Assessment Conference

Seattle, WA December, 2006

First, some good news.

After more than a decade of fairly flat achievement and stagnant or growing gaps, we appear to be

turning the corner.

NAEP Reading, 9 Year-Olds:Record Performance for All

Groups

150

170

190

210

230

250

1971 1975 1980 1984 1988 1990 1992 1994 1996 1999 2004

Ave

rag

e S

cale

Sco

re

African American Latino White

Source: National Center for Education Statistics, NAEP 2004 Trends in Academic Progress

Note: Long-Term Trends NAEP

African American-White Gap Narrows to Smallest Size in History

NAEP Reading, 9 Year-Olds

150

170

190

210

230

250

1971 1975 1980 1984 1988 1990 1992 1994 1996 1999 2004

Avera

ge S

cale

Sco

re

African American White

29 3526



Latino-White Gap Narrows to Smallest Size in History


150

170

190

210

230

250

1971 1975 1980 1984 1988 1990 1992 1994 1996 1999 2004

Ave

rag

e S

cale

Sco

re

Latino White

24 2821



NAEP Math, 9 Year-Olds: Record Performance for All

Groups

150

170

190

210

230

250

1973 1978 1982 1986 1990 1992 1994 1996 1999 2004

Ave

rag

e S

cale

Sco

re




African American-White Gap Narrows to Smallest Size in History

NAEP Math, 9 Year-Olds

150

170

190

210

230

250

1973 1978 1982 1986 1990 1992 1994 1996 1999 2004

Avera

ge S

cale

Sco

re

African American White

25

2823



Latino-White Gap Narrows to Smallest Size in HistoryNAEP Math, 9 Year-Olds

150

170

190

210

230

250

1973 1978 1982 1986 1990 1992 1994 1996 1999 2004

Ave

rag

e S

cale

Sco

re

Latino White

2126

17



Bottom Line:When We Really Focus on

Something, We Make Progress

Clearly, much more remains to be done in elementary and

middle schoolToo many youngsters still

enter high school way behind.

But the bigger problem is that we’re not really building

on these successes in the upper grades.

Achievement Flat in Reading 13 Year-Olds, NAEP

200

220

240

260

280

300

1971

1975

1980

1984

1988

1990

1992

1994

1996

1999

2004

Avera

ge S

cale

Sco

re

All

Source: US Department of Education, National Center for Education Statistics. NAEP 1999 Trends in Academic Progress (p. 107) Washington, DC: US Department of Education, August 2000

Achievement Flat or Declining in Reading,

17 year olds, NAEP

250255260265270275280285290295300

1984 1988 1990 1992 1994 1996 1999 2004

Sca

le S

core

Source: NAEP 2004 Trends in Academic Progress.


Math?

At first blush, appears to be trending upwards.

Achievement Up in Math, 13 Year-Olds, NAEP

220230240250260270280290300

1973

1978

1982

1986

1990

1992

1994

1996

1999

2004

Av

era

ge

Sc

ale

S

co

re

All

Source: US Department of Education, National Center for Education Statistics. NAEP 1999 Trends in Academic Progress (p. 108) Washington, DC: US Department of Education, August 2000

Achievement up in Math,17 year olds, NAEP

280

285

290

295

300

305

310

315

1986 1990 1992 1994 1996 1999 2004

Sca

le S

core

Source: NAEP 2004 Trends in Academic Progress and NAEP 1999 Trends in Academic Progress.


But Value Added in Middle and High School Math Actually Declined

During the Nineties

Value Added Declining in Middle School Math...

50 50 4945 45

30

35

40

45

50

55

60

Math

8th Graders '82 8th Graders '86 8th Graders '908th Graders '94 8th Graders '96

Age 9-13 Growth

Source: NAEP 1999 Trends in Academic Progress

Value Added Declining in High School Math...

36 3634

25

30

35

40

45

50

55

60

Scale

Sco

re G

ain

Age 17 in 1990 Age 17 in 1994 Age 17 in 1996

Scale Score Growth, From Age 13 to Age 17

Source: NCES, 1999. Trends in Academic Progress. Data from Long Term Trend NAEP

Note: Scale score gains reflect the difference between the scale scores of 17-year-olds and the scale scores of 13-year-olds four years prior.

... Still

36

29

25

30

35

40

45

50

55

60

Math

Scale

Sco

re G

ain

Grade 12 in 1996 Grade 12 in 2000

Scale Score Growth, From Grade 8 to Grade 12

Source: NAEP Data Explorer, http://nces.ed.gov/nationsreportcard/nde

Note: Scale score gains reflect the difference between the scale scores of 12 th Graders and the scale scores of 8th Graders four years prior.

Gaps between groups wider today than in 1990


220

240

260

280

300

320

1971 1975 1980 1984 1988 1990 1992 1994 1996 1999 2004

Avera

ge S

cale

Sco

re


21 29



NAEP Math, 17 Year-Olds

220

240

260

280

300

320

1973 1978 1982 1986 1990 1992 1994 1996 1999 2004

Avera

ge S

cale

Sco

re


20 28



Hormones?

If so, we’d see the same pattern in other countries.

And we don’t.

Looking across the Grades?

2003 TIMSS and PISA Math(US only compared with countries that participated in all

three assessments TIMSS 4+8 and PISA)

2003TIMSS Grade 4 Math

300

350

400

450

500

550

600

Hon

g K

ong

Japa

n

Bel

giu

m

Net

her

land

s

Latv

ia

Rus

sia

Hun

gar

y

Ave

rage

Uni

ted

Sta

tes

Italy

Aus

tral

ia

New

Ze

ala

nd

Nor

way

Av

era

ge

Sc

ale

Sc

ore

Source: American Institutes For Research, November 2005, Reassessing U.S. Mathematics Performance: New Findings from the 2003 TIMSS and PISA

2003TIMSS Grade 8 Math

300

350

400

450

500

550

600

Hon

g K

ong

Japa

n

Bel

giu

m

Net

her

land

s

Hun

gar

y

Ave

rage

Latv

ia

Rus

sia

Aus

tral

ia

Uni

ted

Sta

tes

New

Ze

ala

nd

Italy

Nor

way

Av

era

ge

Sc

ale

Sc

ore


PISA 2003Mathematics, 15-Year-Olds

300

350

400

450

500

550

600

Hon

g K

ong

Net

her

land

s

Japa

n

Bel

giu

m

Aus

tral

ia

New

Ze

ala

nd

Ave

rage

Nor

way

Hun

gar

y

Latv

ia

Uni

ted

Sta

tes

Rus

sia

Italy

Av

era

ge

Sc

ale

Sc

ore


The U.S. Ranks Low Among Participating Countries in Each of the

International Math Assessments Given in 2003

0

12

TIMSS 4 Math TIMSS 8 Math PISA Math

Pa

rtic

ipa

tin

g C

ou

ntr

ies

US

Average


Average

Average

US

US

Note: Countries in this analysis participated in all three of these assessments.

Let’s take a closer look at our 15 year olds.

A few years ago, we got a wake up call when the 1999 PISA results were published.

US 15 Year-Olds Rank Near Middle Of The Pack Among 32 Participating Countries:

1999

U.S. RANKREADING 15TH

MATH 19TH

SCIENCE 14TH

The new ones?

PISA 2003: US 15 Year-Olds Rank Near The End Of The

Pack Among 29 OECD Countries

U.S. RANK READING 20TH

MATH 24TH

SCIENCE 19TH

Source: NCES, 2005, International Outcomes of Learning in Mathematics, Literacy and Problem Solving: 2003 PISA Results.NCES 2005-003

A closer look at math?

2003: U.S. Ranked 24th out of 29 OECD Countries in Mathematics

300

350

400

450

500

550

Fin

lan

dK

ore

aN

eth

erla

nds

Japa

nC

ana

daB

elgi

um

Sw

itzer

land

New

Ze

ala

ndA

ustr

alia

Cze

ch R

epu

blic

Icel

and

Den

mar

kF

ranc

eS

wed

en

Aus

tria

Ger

man

yIr

ela

ndO

EC

D A

vera

geS

lova

ck R

epu

blic

Nor

way

Luxe

mbo

urg

Pol

and

Hun

gar

yS

pain

Uni

ted

Sta

tes

Por

tuga

l

Italy

Gre

ece

Tu

rkey

Mex

ico

Ave

rag

e S

cale

Sco

re

Source: Organization for Economic Cooperation and Development (OECD), PISA 2003 Results , data available at http://www.oecd.org/

Problems are not limited to our high-poverty and high-

minority schools . . .

U.S. Ranks Low in the Percent of Students in the Highest Achievement Level (Level 6)

in Math

0

2

4

6

8

10

Bel

giu

m

Kor

ea

Japa

nF

inla

nd

Net

her

land

sN

ew Z

eal

and

Sw

itzer

land

Aus

tral

iaC

ana

daC

zech

Re

pub

licIc

ela

ndD

enm

ark

Sw

ede

nO

EC

D A

vera

geA

ustr

iaG

erm

any

Fra

nce

Slo

vak

Rep

ublic

Nor

way

Ire

land

Pol

and

Luxe

mbo

urg

Hun

gar

yU

nite

d S

tate

s

Italy

Tu

rkey

Spa

inP

ortu

gal

Gre

ece

Mex

ico

Per

cen

t o

f S

tud

ents

Source: Organization for Economic Cooperation and Development (OECD), PISA 2003 Results, data available at http://www.oecd.org/

U.S. Ranks 23rd out of 29 OECD Countries in the Math Achievement of the Highest-

Performing Students*

300

350

400

450

500

550

600

650

700

Bel

giu

mJa

pan

Kor

ea

Sw

itzer

land

Net

her

land

sN

ew Z

eal

and

Fin

lan

dA

ustr

alia

Can

ada

Cze

ch R

epu

blic

Den

mar

kS

wed

en

Ger

man

yO

EC

D A

VE

RA

GE

Aus

tria

Icel

and

Fra

nce

Slo

vak

Rep

ublic

Nor

way

Hun

gar

yLu

xem

bour

gIr

ela

ndP

olan

dU

nite

d S

tate

sS

pain

Italy

Tu

rkey

Por

tuga

lG

reec

eM

exic

o

Ave

rag

e S

cale

Sco

re

* Students at the 95th PercentileSource: Organization for Economic Cooperation and Development (OECD), PISA 2003 Results , data available at http://www.oecd.org/

U.S. Ranks 23rd out of 29OECD Countries in the Math

Achievement of High-SES Students

300

350

400

450

500

550

600

Bel

giu

mN

eth

erla

nds

Fin

lan

dC

zech

Re

pub

licC

ana

daJa

pan

Kor

ea

Sw

itzer

land

Aus

tral

iaG

erm

any

New

Ze

ala

ndF

ranc

eD

enm

ark

Sw

ede

nA

ustr

iaH

ung

ary

OE

CD

AV

ER

AG

ES

lova

k R

epub

licLu

xem

bour

gIr

ela

ndIc

ela

ndP

olan

dN

orw

ayU

nite

d S

tate

sS

pain

Por

tuga

lIta

lyG

reec

eT

urk

eyM

exic

o

Ave

rag

e S

cale

Sco

re


Problems not limited to math, either.

PISA 2003: Problem-Solving, US Ranks 24th Out of 29 OECD

Countries

300

350

400

450

500

550

600

Kor

ea

Fin

lan

dJa

pan

New

Ze

ala

ndA

ustr

alia

Can

ada

Bel

giu

mS

witz

erla

ndN

eth

erla

nds

Fra

nce

Den

mar

kC

zech

Re

pub

licG

erm

any

Sw

ede

nA

ustr

iaIc

ela

ndH

ung

ary

OE

CD

Ave

rage

Ire

land

Luxe

mbo

urg

Slo

vak

Rep

ublic

Nor

way

Pol

and

Spa

inU

nite

d S

tate

sP

ortu

gal

Italy

Gre

ece

Tu

rkey

Mex

ico

Ave

rag

e S

cale

Sco

re

Source: NCES, 2005, International Outcomes of Learning in Mathematics, Literacy and Problem Solving: 2003 PISA Results.NCES 2005-003

More than half of our 15 year olds at problem-solving level

1 or below.

Source: OECD Problem Solving for Tomorrow’s World. 2004

One measure on which we rank high?

Inequality!

*Of 29 OECD countries, based on scores of students at the 5th and 95th percentiles.

PISA 2003: Gaps in Performance Of U.S.15 Year-Olds Are Among the Largest of OECD Countries

Rank in Performance Gaps Between Highest and Lowest Achieving

Students *

Mathematical Literacy 8th

Problem Solving 6th


These gaps begin before children arrive at the

schoolhouse door.But, rather than organizing our

educational system to ameliorate this problem, we organize it to

exacerbate the problem.

How?

By giving students who arrive with less, less in school, too.

Some of these “lesses” are a result of choices that policymakers make.

Nation:Inequities in State and Local

Revenue Per StudentGap

High Poverty vs. Low Poverty Districts

-$907 per student

High Minority vs. Low Minority Districts

-$614 per student

Source: The Education Trust, The Funding Gap 2005. Data are for 2003

While many educators find these inequities offensive, they can be comforting, as

well. They make the achievement gap somehow

not about us.

In truth, though, some of the most devastating “lesses”

are a function of choices that we educators make.

Choices we make about what to expect of whom…

Source: Prospects (ABT Associates, 1993), in “Prospects: Final Report on Student Outcomes”, PES, DOE, 1997.

Students in Poor Schools Receive ‘A’s for Work That Would Earn ‘Cs’ in Affluent

Schools87

35

56

34 41

22 21

11

0

100

Per

cent

ile -

CT

BS

4

A B C DGrades

Seventh Grade Math

Low-poverty schools High-poverty schools

Choices we make about what to teach whom…

Source: CCSSO, State Indicators of Science and Mathematics Education, 2005

Fewer Latino students are enrolled

in Algebra 1 in Grade 8

2429

0

50

2003

Per

cen

t En

rolle

d

Latino

White

Source: CCSSO, State Indicators of Science and Mathematics Education, 2001

Fewer Latino students are enrolled

in Algebra 2

45

62

0

80

1998

Per

cen

t En

rolle

d

Latino

White

And choices we make about Who

teaches whom…

More Classes in High-Poverty, High-Minority Schools Taught By Out-of-

Field Teachers

34%

19%

29%

21%

0%

50%

Per

cen

t o

f C

lass

es T

aug

ht

by

Ou

t o

f F

ield

Tea

cher

s

*Teachers lacking a college major or minor in the field. Data for secondary-level core academic classes.Source: Richard M. Ingersoll, University of Pennsylvania. Original analysis for the Ed Trust of 1999-2000 Schools and Staffing Survey .

High poverty Low poverty High minority Low minorityNote: High Poverty school-50% or more of the students are eligible for free/reduced price lunch. Low-poverty school -15% or fewer of the students are eligible for free/reduced price lunch.

High-minority school - 50% or more of the students are nonwhite. Low-minority school- 15% or fewer of the students are nonwhite.

Poor and Minority Students Get More Inexperienced*

Teachers20%

11%

21%

10%

0%

25%

Perc

en

t o

f T

each

ers

Wh

o A

re

Inexp

eri

en

ced

Source: National Center for Education Statistics, “Monitoring Quality: An Indicators Report,” December 2000.

*Teachers with 3 or fewer years of experience.

High poverty Low poverty High minority Low minority

Note: High poverty refers to the top quartile of schools with students eligible for free/reduced price lunch. Low poverty-bottom quartile of schools with students eligible for free/reduced price lunch. High minority-top quartile; those schools with the highest concentrations of minority students. Low minority-bottom quartile of schools with the lowest concentrations of minority students

Results are devastating.

Kids who come in a little behind, leave a lot behind.

By the end of high school?

African American and Latino 17 Year-Olds Do Math at Same Levels

As White 13 Year-Olds

0%

100%

200 250 300 350

Average Scale Score

Per

cent

of

Stud

ents

White 13 Year-Olds African American 17 Year-Olds Latino 17-Year Olds



African American and Latino 17 Year-Olds Read at Same Levels As

White 13 Year-Olds

0%

100%

150 200 250 300 350

Average Scale Score

Per

cent

of

Stud

ents

White 13 Year-Olds African American 17 Year-Olds Latino 17 Year-Olds



And these are the students who remain in

high school.What do those numbers look

like?

Students Graduate From High School At Different Rates, 2001*

* 4-Year Graduation Rates

Source: Jay P. Greene and Greg Forster, “Public High School Graduation and College Readiness Rates in the United States,” Manhattan Institute for Policy Research, September 2003.

54%51%

79%

52%

72%

0%

100%

African American Asian Latino Native American White

ADD IT ALL UP...

Of Every 100 White Kindergartners:

94 Graduate from high school

66 Complete at least some college

34 Obtain at least a Bachelor’s Degree

(25-to 29-Year-Olds)

Source: US Department of Commerce, Bureau of the Census. March Current Population Surveys, 1971-2003, in The Condition of Education 2005. http://nces.ed.gov/programs/coe/2005/section3/indicator23.asp#info

Of Every 100 African American Kindergartners:

89 Graduate from High School

51 Complete at Least Some College

18 Obtain at Least a Bachelor’s Degree



Of Every 100 Latino Kindergartners:






Of Every 100 American Indian/Alaskan Native

Kindergartners:

(25 Years Old and Older)

Source: U.S. Census Bureau, We the People: American Indians and Alaska Natives in the United States. Data source: Census 2000, www.census.gov/population/www/socdemo/race/censr-28.pdf




College Graduates by Age 24

Young People From High Income Families

75%

Young People From Low Income Families

9%

Source: Tom Mortenson, Postsecondary Educational Opportunity..

What Can We Do?

An awful lot of educators have decided that we can’t

do much.

What We Hear Many Educators Say:

• They’re poor;• Their parents don’t care;• They come to schools without

breakfast; • Not enough books• Not enough parents . . .

But if they are right, why are low-income students and

students of color performing so high in some schools…

Capitol View ElementaryAtlanta, Georgia

Capitol View ElementaryAtlanta, Georgia

• 252 students in grades K-5• 95% African American• 88% Low-Income

Source: Georgia Governor’s Office of Student Achievement, http://reportcard2006.gaosa.org/

High Achievement at Capitol View

2006 Grade 5 Reading100 100 100

8172 73

0

20

40

60

80

100

Overall AfricanAmerican

Low-Income

Perc

en

t M

eets

or

Exceed

s

Capitol View

Georgia


High Achievement at Capitol View

2006 Grade 5 Math100 100 100

8983 83

0

20

40

60

80

100

Overall AfricanAmerican

Low-Income

Perc

en

t M

eets

or

Exceed

s

Capitol View

Georgia


M. Hall Stanton ElementaryPhiladelphia, Pennsylvania

M. Hall Stanton ElementaryPhiladelphia, Pennsylvania

• 487 students in grades K-6• 100% African American• 86% Low-Income

Source: Philadelphia School District, https://sdp-webprod.phila.k12.pa.us/school_profiles/servlet/

Rapid Improvement at StantonGrade 5 Reading Over Time

12

70

5761

0

20

40

60

80

100

2002 2006

Perc

en

t P

rofi

cie

nt

or

Ad

van

ced

Stanton

Pennsylvania

Source: School Information Partnership, http://www.schoolmatters.com Pennsylvania Department of Education, http://www.pde.state.pa.us

Rapid Improvement at StantonGrade 5 Math Over Time

21

83

53

67

0

20

40

60

80

100

2002 2006

Perc

en

t P

rofi

cie

nt

or

Ad

van

ced

Stanton

Pennsylvania

Source: School Information Partnership, http://www.schoolmatters.com Pennsylvania Department of Education, http://www.pde.state.pa.us

Frankford Elementary School

Frankford ElementaryFrankford, Delaware

• 449 Students in Grades PreK-5• 29% African American• 34% Latino• 34% White• 76% Low-Income

Source: Delaware Department of Education Online School Profiles, http://issm.doe.state.de.us/profiles/EntitySearch.ASPX

Frankford ElementaryClosing Gaps, Grade 5 Reading

78

100100 100

0

20

40

60

80

100

2003 2005

Pe

rce

nt

Me

eti

ng

or

Ex

ce

ed

ing

Sta

nd

ard

s

African American

White

Source: Delaware Department of Education, DSTP Online Reports, http://dstp.doe.k12.de.us/DSTPmart/default.asp

Elmont Memorial Junior-Senior High School


Elmont, New York

• 1,966 Students in Grades 7-12• 75% African American• 12% Latino• 24% Low-Income

Source: New York State School Report Card, http://www.emsc.nysed.gov/irts/reportcard/

Elmont MemorialHigher Percentage of Students Meeting

Graduation Requirements than the State, Class of 2004 Regents English

99 99 100 100 99

85

74 72 75

88

0

20

40

60

80

100

All AfricanAmerican

Latino Poor Non-Poor

Pe

rce

nt

Me

eti

ng

Gra

du

ati

on

Re

qu

ire

me

nts

Elmont

New York


Elmont MemorialHigher Percentage of Students Meeting

Graduation Requirements than the State, Class of 2004 Regents Math

96 95 94 94 96

83

68 6872

86

0

20

40

60

80

100

All AfricanAmerican

Latino Poor Non-Poor

Pe

rce

nt

Me

eti

ng

Gra

du

ati

on

Re

qu

ire

me

nts

Elmont

New York


University Park Campus School

University Park Campus School

Worcester, Massachusetts

• 220 Students in Grades 7-12• 9% African American• 18% Asian• 35% Latino• 39% White• 73% Low-Income

Source: Massachusetts Department of Education School Profile, http://profiles.doe.mass.edu/

University Park Results: 2004

• 100% of 10th graders passed MA high school exit exam on first attempt.

• 87% passed at advanced or proficient level.

• Fifth most successful school in the state, surpassing many schools serving wealthy students.

University ParkHigher Percentage of Students at Proficient and

Advanced than the State2005 Grade 10 Math

1511

2437

27

5335

0%

20%

40%

60%

80%

100%

University Park Massachusetts

Perc

en

t o

f S

tud

en

ts

Advanced

Proficient

Needs Improvement

Warning/Failing

Source: Massachusetts Department of Education School Profile, http://profiles.doe.mass.edu/

Very big differences at district and state levels,

too.

180

190

200

210

District ofColumbia

LosAngeles

Atlanta Chicago Cleveland NationalPublic

SanDiego

Charlotte Houston New YorkCity

Boston

Low-Income African American Students do Better in Some Districts (NAEP Reading 4th 2003)

SOURCE: U.S. Department of Education, Institute of Education Sciences, National Center for Education Statistics, National Assessment of Educational Progress (NAEP), 2003 Trial Urban District Reading Assessment.

* There is a 19 point gap between Poor African American 4th graders in the District of Columbia and Boston (roughly equivalent to 2 years’ worth of learning)

225

235

245

255

LosAngeles

District ofColumbia

Atlanta Chicago NationalPublic

SanDiego

Cleveland Boston Charlotte New YorkCity

Houston

Low-Income African American Students do Better in Some Districts

(NAEP Math 8th 2003)


* There is a 28 point gap between Poor African American 8th graders in Los Angeles and Houston (roughly equivalent to 3 years’ worth of learning)

180

190

200

210

Los Angeles Chicago District ofColumbia

National (Public) New York City Houston

Latino Students do Better in Some Districts (NAEP Reading 4th 2002, 6 Urban Districts)


* There is an 18 point gap between Los Angeles and Houston (equivalent to almost 2 years worth of learning)

Sca

le S

core

What about states?

Washington?

NAEP 2005 Grade 4 Reading, Overall Scale Scores

185

195

205

215

225

235

Mas

sach

uset

tsN

ew H

amps

hire

Del

awar

eN

orth

Dak

ota

Min

neso

taC

olor

ado

Was

hing

ton

Ohi

oN

ew J

erse

yId

aho

Uta

hM

isso

uri

Mar

ylan

dK

ansa

sFl

orid

aIn

dian

aN

orth

Car

olin

aA

rkan

sas

Illin

ois

Tenn

esse

eG

eorg

iaA

lask

aLo

uisi

ana

New

Mex

ico

Cal

iforn

iaM

issi

ssip

pi

Ave

rag

e S

cale

Sco

re

Source: National Center for Education Statistics, NAEP Data Explorer, http://nces.ed.gov/nationsreportcard/nde

Washington

NAEP 2005 Grade 4 Reading, African American Scale

Scores

185

195

205

215

225

235

Was

hing

ton

Ala

ska

Virg

inia

Col

orad

oN

ew M

exic

oK

entu

cky

Wes

t Virg

inia

Iow

aP

enns

ylva

nia

Nor

th C

arol

ina

New

Jer

sey

Geo

rgia

Rho

de Is

land

Ohi

oK

ansa

sLo

uisi

ana

Wis

cons

inIll

inoi

sA

rizon

aM

inne

sota

Mic

higa

n

Ave

rag

e S

cale

Sco

re


NAEP 2005 Grade 4 Reading, Latino Scale Scores

185

195

205

215

225

235

Mon

tana

Del

awar

eFl

orid

a

Ohi

o

Texa

sM

aryl

and

Wis

cons

inIn

dian

aC

olor

ado

Okl

ahom

aM

inne

sota

Mas

sach

uset

tsG

eorg

iaW

ashi

ngto

nN

atio

nal P

ublic

Uta

hN

ew M

exic

o

Idah

oN

evad

aR

hode

Isla

nd

Ave

rag

e S

cale

Sco

re


NAEP 2005 Grade 4 Reading, Low-Income Scale Scores

180

190

200

210

220

230

Wyo

min

gN

orth

Dak

ota

New

Ham

pshi

reK

entu

cky

Mas

sach

uset

tsN

ew Y

ork

Ver

mon

tM

inne

sota

Virg

inia

Iow

aTe

xas

Indi

ana

Ohi

oN

ebra

ska

Pen

nsyl

vani

aW

isco

nsin

New

Jer

sey

Nor

th C

arol

ina

Mic

higa

nS

outh

Car

olin

aN

ew M

exic

oM

aryl

and

Rho

de Is

land

Mis

siss

ippi

Cal

iforn

iaN

evad

a

Ave

rag

e S

cale

Sco

re


NAEP 2005 Grade 8 Reading, Overall Scale Scores

225

235

245

255

265

275

Mas

sach

uset

tsN

ew H

amps

hire

Mon

tana

Sou

th D

akot

aM

inne

sota

Wyo

min

gK

ansa

sO

hio

Del

awar

eC

olor

ado

New

Yor

kC

onne

ctic

utIll

inoi

sO

rego

nIn

dian

aM

ichi

gan

Nat

iona

l Pub

licA

lask

aA

rkan

sas

Texa

sS

outh

Car

olin

aA

rizon

aLo

uisi

ana

Ala

bam

aN

ew M

exic

oH

awai

i

Ave

rag

e S

cale

Sco

re


NAEP 2005 Grade 8 Reading, African American Scale

Scores

225

235

245

255

265

275

Was

hing

ton

Mas

sach

uset

tsN

ew J

erse

yA

lask

aK

ansa

s

Texa

s

Illin

ois

Neb

rask

aO

klah

oma

Nat

iona

l Pub

licM

isso

uri

Sou

th C

arol

ina

Indi

ana

Con

nect

icut

Nev

ada

Tenn

esse

eM

inne

sota

Flor

ida

Ark

ansa

sW

isco

nsin

Ave

rag

e S

cale

Sco

re


NAEP 2005 Grade 8 Reading, Latino Scale Scores

225

235

245

255

265

275

Virg

inia

Iow

aW

yom

ing

Del

awar

e

Flor

ida

Ark

ansa

sN

ew Y

ork

Nor

th C

arol

ina

Col

orad

o

Indi

ana

Wis

cons

inM

assa

chus

etts

Nat

iona

l Pub

licN

ebra

ska

Ohi

oW

ashi

ngto

n

Uta

h

Haw

aii

Cal

iforn

ia

Ave

rag

e S

cale

Sco

re


NAEP 2005 Grade 8 Reading, Low-Income Scale Scores

225

235

245

255

265

275

Mai

neM

onta

naW

yom

ing

Ken

tuck

yIo

wa

Ver

mon

tK

ansa

sM

isso

uri

New

Yor

kM

inne

sota

Okl

ahom

aO

hio

Indi

ana

Col

orad

oN

atio

nal P

ublic

Pen

nsyl

vani

aFl

orid

aS

outh

Car

olin

aW

est V

irgin

iaN

orth

Car

olin

aG

eorg

iaN

ew M

exic

oA

rizon

aM

issi

ssip

piA

laba

ma

Haw

aii

Ave

rag

e S

cale

Sco

re


NAEP 2005 Grade 8 Math, Overall Scale Scores

235

245

255

265

275

285

295

Mas

sach

uset

tsN

orth

Dak

ota

Ver

mon

tN

ew H

amps

hire

Wis

cons

inK

ansa

sN

ew J

erse

yO

hio

Nor

th C

arol

ina

Wyo

min

gC

onne

ctic

utId

aho

Pen

nsyl

vani

aTe

xas

Ala

ska

Nat

iona

l Pub

licM

aryl

and

Mis

sour

iFl

orid

aA

rkan

sas

Rho

de Is

land

Tenn

esse

eC

alifo

rnia

Loui

sian

aN

ew M

exic

oM

issi

ssip

pi

Ave

rag

e S

cale

Sco

re


NAEP 2005 Grade 8 Math, African American Scale

Scores

235

245

255

265

275

285

295

Was

hing

ton

Texa

sN

orth

Car

olin

aV

irgin

iaN

ew J

erse

yM

aryl

and

Indi

ana

Col

orad

oK

ansa

sK

entu

cky

Nat

iona

l Pub

licFl

orid

aW

est V

irgin

iaC

onne

ctic

utO

klah

oma

Cal

iforn

iaM

issi

ssip

piN

evad

aW

isco

nsin

Neb

rask

a

Ave

rag

e S

cale

Sco

re


NAEP 2005 Grade 8 Math, Latino Scale Scores

235

245

255

265

275

285

295

Ala

ska

Virg

inia

Del

awar

eA

rkan

sas

Flor

ida

Mas

sach

uset

tsN

orth

Car

olin

aW

yom

ing

New

Jer

sey

Mar

ylan

dW

ashi

ngto

n

Idah

oN

ebra

ska

Col

orad

oG

eorg

iaO

klah

oma

Nev

ada

Uta

hC

onne

ctic

ut

Ave

rag

e S

cale

Sco

re


NAEP 2005 Grade 8 Math, Low-Income Scale Scores

235

245

255

265

275

285

295

Sou

th D

akot

aM

assa

chus

etts

Mon

tana

Wyo

min

gK

ansa

sO

rego

nM

aine

Indi

ana

Texa

sN

ew Y

ork

Nor

th C

arol

ina

Ohi

oK

entu

cky

Wis

cons

inN

ew J

erse

yN

atio

nal P

ublic

Ariz

ona

Flor

ida

Wes

t Virg

inia

Loui

sian

aM

ichi

gan

Nev

ada

Con

nect

icut

New

Mex

ico

Rho

de Is

land

Ala

bam

a

Ave

rag

e S

cale

Sco

re


Bottom Line:At Every Level of Education, What We Do Matters A Lot!

What do we know about the anatomy of their

success?Eleven powerful lessons

#1. They focus on what they can do, rather than what

they can’t.

Some schools and districts get all caught up in

“correlations”.

Spend endless time tracking:

• Percent of babies born at low-birthweight;

• Percent of children born to single moms;

• Percent of children in families receiving government assistance;

• Education levels of mothers; and…

The leaders in high-performing high poverty

schools and districts don’t do that.

They focus on what they can do, not on what they can’t.

It’s not that they don’t understand the effects of

poverty, and many work hard on public policies that will

help. But…

“ Some of our children live in pretty dire circumstances. But we can’t dwell on that, because we can change it. So when we come here, we have to dwell on that which is going to move our kids.”

Barbara Adderly, Principal,M. Hall Stanton Elementary, Philadelphia

#2. They don’t leave anything about teaching and

learning to chance.

An awful lot of our teachers—even brand new ones—are

left to figure out on their own what to teach and what

constitutes “good enough” work.

Result? A System That:

• Doesn’t expect very much from MOST students; and,

• Expects much less from some types of students than others.

Source: Prospects (ABT Associates, 1993), in “Prospects: Final Report on Student Outcomes”, PES, DOE, 1997.

‘A’ Work in Poor Schools Would Earn ‘Cs’ in Affluent

Schools87

35

56

34 41

22 21

11

0

100

Per

cent

ile -

CT

BS

4

A B C DGrades

Seventh Grade Math

Low-poverty schools High-poverty schools

Students can do no better than

the assignments they are given...

Elementary School ExampleKindergarten Assignment

Based on our reading, draw a picture of an ocean animal that you would like to be.

Elementary School Example

Kindergarten Assignment

Based on our reading, choose an ocean animal you would like to be. Explain what you would look like, what you would eat, and what you would do. Why do you want to be this animal?

Middle School Example:

7th Grade Assignment Name and describe functions of the five body systems.

Middle School Example

7th Grade AssignmentExplain the difference between the systems of the body affected by an allergy to pollen and those affected by an allergy to food as well as the process by which different medicines reduce the symptoms of each allergy.

High School Example

10th Grade Assignment

Draw a map of the Caribbean, labeling major cities and geologic features.

High School Example

10th Grade AssignmentHow does Mercator’s 1633 map of the New World differ from Kircher’s 1678 map? If you were going to sail alone from Europe to the New World, which map would you use and why?

High Performing Schools and Districts

• Have clear and specific goals for what students should learn in every grade, including the order in which they should learn it;

• Provide teachers with common curriculum, assignments;

• Assess students every 4-8 weeks to measure progress;

• ACT immediately on the results of those assessments.

They’re methodical, in other words, but also

responsible.When teachers in these

schools say they “taught” it, that means their students

learned it.

#3. They set their goals high.

Elementary Version…

M. Hall Stanton Elementary:Percent of 5th Graders

ADVANCED

1

30

1

42

05

1015202530354045

2001 2005

Reading

Math

High School Version…

Even when they start with high drop out rates, high

impact high schools focus on preparing all kids for college

and careersEducation Trust 2005 study, “Gaining Traction,

Gaining Ground.”

Source: US bureau of Labor Statistics and Bureau of the Census, Current Population Survey, March 2002

That’s Good, Because Education Pays:

Annual Earnings of 25-34 yr-olds by Attainment, 2001

27831 2966334259 36135

49011

$10,000

$20,000

$30,000

$40,000

$50,000

$60,000

$70,000

HS, nodiploma

HS diploma Some coll,no deg

Assoc deg BA/BA

Growing Need for Higher Levels of Education:

Projections of Education Shortages and Surpluses in 2012

-4,000,000 -3,000,000 -2,000,000 -1,000,000 0 1,000,000 2,000,000 3,000,000 4,000,000

Less Than HighSchool

High School

Associates Degree

Some College

Shortage Surplus

Source: Analysis by Anthony Carnevale, 2006 of Current Population Survey (1992-2004) and Census Population Projection Estimates

Bachelor’s Degree

Even if you have your doubts, NEW STUDY FROM

ACT:College ready, workforce

training ready=same thing

#4. Higher performing secondary schools put all kids—not just some—in a

demanding high school core curriculum.

Single biggest predictor post-

high school success is

QUALITY AND INTENSITY OF

HIGH SCHOOL CURRICULUM

Cliff Adelman, Answers in the Tool Box, U.S. Department of Education.

But are most of our kids getting anything that even

remotely resemblesINTENSE?

Jake: Fall Schedule, Freshman Year

English

Health Ed/Academic Foundations (Required Course for all freshmen)

Conceptual Physics

Volleyball

Spring Schedule, Freshman Year

Algebra

Auto Shop

Auto Shop

Volleyball

Fall Schedule, Sophomore Year

English

Spanish

Chemistry

Open Period (required)

Spring Schedule, Sophomore Year

Geometry

W. History

Volleyball

Open Period (required)

Fall Schedule, Junior Year

Mythology

Algebra

Auto Shop

Career Choices

Spring Schedule, Junior Year

Algebra 2

American History

Arts Tech

English

Senior Year?

Too embarrassing to even show

Ed Trust Transcript Study: Our Current Favorites

• Pre-Spanish;• Future Studies;• Exploring;• Principles of PE;• Teen Living;• Life Management;• Food Fundamentals;• Winter Activities.

Source: Education Trust Analysis of High School Transcripts; 2005

But college prep curriculum has benefits far

beyond college.

Students of all sorts will learn more...

Source: USDOE, NCES, Vocational Education in the United States: Toward the Year 2000, in Issue Brief: Students Who Prepare for College and Vocation

*Grade 8-grade 12 test score gains based on 8th grade achievement.

Low Quartile Students Gain More From College

Prep Courses*

1916

28

20

0

30

Math Reading

NE

LS

Sco

re G

ain

Vocational College Prep

They will also fail less often...

Challenging Curriculum Results in Lower Failure Rates, Even for Lowest Achievers

16

23

47

31

0

50

Quartile I (Lowest) Quartile 2

Per

cen

t E

arn

ing

"D

" o

r "F

"

College Prep Low Level

Source: SREB, “Middle Grades to High School: Mending a Weak Link”. Unpublished Draft, 2002.

Ninth-grade English performance, by high/low level course, and eighth-grade reading achievement quartiles

And they’ll be better prepared for the

workplace.

Leading districts, states making college prep the

default curriculum.

Texas, Indiana, Arkansas, Michigan, Oklahoma,

Kentucky, Kansas.

#5. High performing schools are obsessive about time,

especially instructional time.

Scouring the schedule for minutes:

The case of the “pencil sharpener lady”.

High School?Take, for example, the

matter of reading.Kids who arrive behind in

reading…often simply assigned to courses that don’t demand much

reading.

Average High School: Percent of Instructional Time in Reading Intensive Courses

Below Grade Level Students

On Grade Level Students

Advanced Students

24% 29% 35%

Surprise: Gaps Grow.

– “Behind” students spend 60 additional hours (25% more time) over 1 year in reading related courses)

– “Behind” students get 240 additional hours over 4 years!

Higher Performing High Schools:

In other words, high performing schools both maximize time and don’t leave its use to chance.

There is also the matter of how we deploy our people.

9th Grade Bulge:Largely about “poor

preparation” and “difficult transitions?”

One Colorado High School: Student/Teacher Ratio by

GradeGrade 9th 10th 11th 12th

Average number of students per

teacher

30.3 16.7 11.6 12.1

Source: Jovenes Unidos & Padres Unidos; March, 2004.

Same Colorado High School:Counselor Deployment by

GradeGrade 9th 10th 11th 12th

Number of Counselors

1 1 1 1

Number of Students

572 366 309 213

Source: Jovenes Unidos and Padres Unidos; March, 2004

Is this school structured around student, or adult

needs?High performing schools are

driven by student needs.

#6. Principals are hugely important, ever present, but

NOTthe only leaders in the school


High performing schools…

• Teachers regularly observe other teachers;

• Teachers have time to plan and work collaboratively;

• New teachers get generous and careful support and acculturation;

• Teachers take on many other leadership tasks at the school

#7. Good schools know how much teachers matter, and they act on that knowledge.

1998 by The Education Trust, Inc.

Students in Dallas Gain More in Math with Effective Teachers: One Year Growth From 3rd-4th

Grade

7

16

0

24

6

810

12

1416

18

Students with Teachers inLowest Quintile of

Effectiveness

Students with Teachers inHighest Quintile of

Effectiveness

Ave

rag

e N

orm

al

Cu

rve

Eq

uiv

alen

ts

Source: Heather Jordan, Robert Mendro, and Dash Weerasinghe, The Effects of Teachers on Longitudinal Student Achievement, 1997.

LOW ACHIEVING STUDENTS IN TN GAIN MORE WITH EFFECTIVE TEACHERS: One Year Growth

14

53

0

10

20

30

40

50

60

low high

Sanders and Rivers, Cumulative and Residual Effects of Teachers on Future Academic Achievement, 1998.

Cumulative Teacher Effects On Students’ Math Scores in

Dallas (Grades 3-5)

76

27

0

10

20

30

40

50

60

70

80

90

100

Ave

rage P

erc

entile

Rank

Dallas StudentsAssigned to 3Highly EffectiveTeachers in aRow

Dallas StudentsAssigned to 3IneffectiveTeachers in aRow

Source: Heather Jordan, Robert Mendro, and Dash Weerasinghe, The Effects of Teachers on Longitudinal Student Achievement, 1997.

Beginning Grade 3Percentile Rank= 55

Beginning Grade 3Percentile Rank= 57

1998 by The Education Trust, Inc.

Good teachers matter a lot.

But some groups of kids don’t get their fair share of quality

teachers.

More Classes in High-Poverty, High-Minority Schools Taught By Out-of-

Field Teachers

34%

19%

29%

21%

0%

50%

Per

cen

t o

f C

lass

es T

aug

ht

by

Ou

t o

f F

ield

Tea

cher

s

*Teachers lacking a college major or minor in the field. Data for secondary-level core academic classes.Source: Richard M. Ingersoll, University of Pennsylvania. Original analysis for the Ed Trust of 1999-2000 Schools and Staffing Survey .

High poverty Low poverty High minority Low minorityNote: High Poverty school-50% or more of the students are eligible for free/reduced price lunch. Low-poverty school -15% or fewer of the students are eligible for free/reduced price lunch.

High-minority school - 50% or more of the students are nonwhite. Low-minority school- 15% or fewer of the students are nonwhite.

Poor and Minority Students Get More Inexperienced*

Teachers20%

11%

21%

10%

0%

25%

Perc

en

t o

f T

each

ers

Wh

o A

re

Inexp

eri

en

ced

Source: National Center for Education Statistics, “Monitoring Quality: An Indicators Report,” December 2000.

*Teachers with 3 or fewer years of experience.

High poverty Low poverty High minority Low minority

Note: High poverty refers to the top quartile of schools with students eligible for free/reduced price lunch. Low poverty-bottom quartile of schools with students eligible for free/reduced price lunch. High minority-top quartile; those schools with the highest concentrations of minority students. Low minority-bottom quartile of schools with the lowest concentrations of minority students

High performing schools don’t let that happen.

They work hard to attract and hold good teachers; they make sure that their best are assigned to the students who

most need them; and they chase out teachers who are not “good enough” for their

kids.

Why is this so important?

“By our estimates from Texas schools, having an above average teacher for five

years running can completely close the average gap between low-income students and others.”

John Kain and Eric Hanushek

#8. They are obsessive about data.

Charts and graphs decorate the school walls.

And every scrap of evidence is carefully scrutinized.

What does that mean? At every moment, they know which students are behind

and are intensely focused on bringing them up.

#9. They are nice places to work.

Not EASY places. And folks work really hard.

But there is lots of camaraderie, lots of stability,

and lots of support.

And when they have vacancies, get out of the

way.Elmont Memorial:

350 applications for every opening.

#10. They are very different places for

students, too.

Today, we adults make lots of assumptions about the

youth culture. And a lot of educators think that low-

income and minority youth are somehow inherently anti-

intellectual and anti-authority.

“At my old school, it was functional to act stupid. At this school, nobody lets me get away with that. Not my teachers. Not the students.”

---Elmont Student, 2005

#11. They never back down.

The Education TrustDownload this Presentation

www.edtrust.orgWashington, DC: 202-293-1217

Oakland, CA: 510-465-6444

Documents

IMPROVING ACHIEVEMENT AND CLOSING GAPS BETWEEN GROUPS: Lessons from Schools and Districts on the Performance Frontier 22 nd Annual Washington State Assessment