nclb under a microscope - accountabilityworks.org · aw believes that the next few years represent...

NCLB Under A Microscope

A Cost Analysis of the Fiscal Impact of the No Child Left Behind Act of 2001 on States and Local Education Agencies

Prepared by AccountabilityWorksJanuary 2004

ORGANIZATIONS:

Education Leaders CouncilEducation Leaders Council (ELC), a member-based organization comprised of policymakers,administrators, teachers and parents, advocates for a new day in education that focuses on makingeducation work better students – not the system. ELC advances policies and programs that supportacademic accountability, teacher quality, school choice and educational technology. Based in WashingtonDC, ELC acts aggressively at the grassroots and national levels to overcome the status quo, break downbarriers to change, and pave the way for all student to access and learn through quality education.

AccountabilityWorksAccountabilityWorks (AW), a non-profit research and consulting organization, is dedicated toassisting states in the design and implementation of high quality elementary and secondary educationreforms. AW works on such initiatives with individual states, consortia of states, and other reform-minded organizations. AW believes that the next few years represent a unique opportunity to ensurethat all students passing through elementary and secondary education receive a solid foundation ofskills and knowledge on which to build their futures.

CONTRIBUTORS:

Meave O’ MarahMs. O’Marah is a consultant to AccountabilityWorks with extensive experience in public policy,working on issues ranging from education to criminal justice to administration and finance. Whileworking for the Commonwealth of Massachusetts, Ms. O’Marah’s focused on fiscal policy and theclose tie between policy initiatives and their effect on state finances. After leaving the public sector,Ms. O’Marah served as an officer and senior manager in one of the largest charter school managementcompanies in the country. Recently, Ms. O’Marah founded her own consulting firm, providing servicesthat draw on her experience from the public and private sectors in fiscal policy, budgeting, start-upactivities, accounting systems design and client management.

Kenneth KlauMr. Klau is a consultant to AccountabilityWorks with substantial experience in education and policy.Recently, he completed a Masters degree in education and policy from Harvard University. Previously,he managed curriculum development for an organization providing comprehensive education reformservices to public schools. He has also taught at the secondary school level.

Theodor RebarberMr. Rebarber is president and founder of AccountabilityWorks. He has been co-founder and chiefeducation officer of a system of public charter schools serving predominantly disadvantaged studentsin ten states. Before that, he was senior staff author in Congress for federal charter legislation for theDistrict of Columbia. He worked as a special assistant to the assistant secretary for research andimprovement at the U.S. Department of Education. Prior to that experience, he was a researcher atVanderbilt Institute for Public Policy Studies (VIPPS).

t a b l e o f c o n t e n t s

2 Executive Summary

4 Introduction

6 Methodology

8 Results8 Federal Funding Calculations

8 Total Federal Funding Available Under NCLB is Sufficient for the Implementation on New Requirements

9 1 Accountability

10 Table 1

11 Table 2

12 Table 3

14 2 Personnel Cumulative Total

15 Table 4

19 3 Information Management Cumulative Total

20 4 School Improvement Cumulative Total

20 General School Reform and Improvement

21 Table 5

26 Brief Commentaries on Previous Studies of NCLB Funding8 New Hampshire School Administrators Association Study

8 Democratic Congressional Study

8 General Accounting Office (GAO) Study

8 Cost Studies of “Educational Adequacy”

20 Endnotes

NCLB UNDER A MICROSCOP E

1 ]

2[

e x e c u t i v e s u m m a r y

The No Child Left Behind Act of 2001 (NCLB) promotesimportant new reforms in public education, promising flex-ibility in exchange for greater accountability. Key NCLBreforms include requirements that states: develop rigoroustargets for determining if schools are succeeding in educatingall students; assess students grade-by-grade in reading andmathematics; implement corrective actions in failing schools;offer public school choice and supplemental education serv-ices for students in schools in need of improvement; andensure that all teachers are highly qualified, possessingknowledge of their subject matter.

Concerns have been raised that NCLB is an “unfundedmandate” and that insufficient federal dollars are providedto enable states to comply with new requirements. Thisreport reviews studies that have claimed that NCLB pro-vides insufficient funding – we find that all contain sig-nificant flaws and unsupported assumptions. Often, expen-ditures that are not required by NCLB are included in thecalculations. In other cases, such studies include expendi-tures that were required by prior federal law. In all cases,careful scrutiny results in little solid evidence that NCLBis insufficiently funded.

This study goes further, including a rigorous analysis ofthe “hard” costs of compliance with specific new elementsin NCLB. These hard costs are substantial in four areas:accountability, personnel, information management andcorrective action for schools in need of improvement.Though most of the assumptions in this analysis tend toerr on the high side for given expenditures, they are based

on states and districts selecting efficient, as well as effec-tive, options for meeting NCLB requirements. Some ofthe options assumed in this study would require state ordistrict leaders to demonstrate the vision – and politicalwill – to change well-established policies and practices. Inmany cases, states and districts will make very differentchoices in determining how to satisfy NCLB, includingselecting some options that are far more expensive, thoughnot necessarily more effective. NCLB has already provid-ed substantial additional funding to states and districts to

The No Child Left Behind Act of 2001(NCLB) promotes important newreforms in public education, promisingflexibility in exchange for greateraccountability.

assist them with reform efforts, and we project continuedincreases; however, NCLB is intended as a spur to reform,not as a spur to spend more on programs that do not pro-duce results. For the purpose of this analysis, it is reasonableto expect states and districts to be somewhat creative – insome cases even frugal – in how they respond to the chal-lenge of reform.

We find that, for every year studied, the additional revenuesprovided exceed the state and local “hard costs” resulting fromspecific NCLB requirements. Nationally, the money left overfor general school improvement and raising student achieve-ment levels ranges from a low of approximately $785 millionin the 2004-05 school year to a high of approximately $5 billionin the 2007-08 school year. These remaining funds must beapplied by states to additional reforms not specified inNCLB. We further review a range of research suggestingthat states and districts can achieve substantial improve-ments if they redirect current expenditures and practices.

Beyond specific new requirements, NCLB broadly calls onstates and districts to educate all students to state-estab-lished standards of proficiency. Some appear to argue that,as a result, this implies that broad responsibility for fund-ing an effective education for all children now rests withthe federal government. Given the historically limited roleof the federal government in elementary and secondaryeducation – consistently supporting less than a tenth of totalexpenditures – we see no reason to accept such a radicalassumption. Rather, we view NCLB as encouraging statesto honor their pre-existing obligation to ensure a qualityeducation for all students regardless of their race or back-ground, while also providing additional resources to achievesuccess.

NCLB UNDER A MICROSCOP E > EXECU T IVE SUMMARY

3 ]

4[

i n t r o d u c t i o n

The No Child Left Behind Act of 2001 (NCLB) promisesflexibility in exchange for greater accountability. NCLBaims to ensure that all students in America’s public schoolshave access to a high quality education and that schools areheld firmly accountable for student results. This includesasking states to define precisely what high quality meansin terms of critical indicators such as student performanceand teacher professionalism. Specifically, NCLB:

n Requires states to develop rigorous formulas for defin-ing adequate yearly progress, with the goal of studentproficiency on challenging subject matter in a specifiedtimeframe for all student subgroups.

n Ensures that states assess the performance of all studentsgrade-by-grade in reading and mathematics.

n Defines and requires “highly qualified” teaching and para-professional staff in order to ensure that educators havethe knowledge and skills necessary to succeed.

n Clarifies ways in which under-performing schools mustaddress their lack of progress.

n Improves upon the ways in which schools convey stu-dent performance and other data to parents and otherstakeholders.

n Introduces greater flexibility in allocation of funds.

An emerging criticism of the legislation is that it requiresstates to undertake expensive new initiatives without ade-quate funding, suggesting that NCLB is an “unfundedmandate.”The purpose of this study is to take a closer lookat the actual funding requirements of NCLB in order todetermine the extent to which that argument is valid.Research for this study:

n Reviewed the new requirements imposed by NCLB,including their quantifiable fiscal impact on states andlocal education agenciesa (LEAs), and associated newfederal funding.

n Quantified all significant “hard” costs that could be attrib-uted to compliance with specific mandates in NCLB,comparing them to new federal revenues above andbeyond what states had previously received from pro-grams in the federal Elementary and Secondary EducationAct and the Individuals with Disabilities Education Act(the relevant underlying federal statutes).

n Analyzed previous studies of NCLB revenues and costs,most of which purported to find insufficient federal dollars.

There is evidence to suggest that most of these hard costs– such as improved teacher content knowledge, increasedparental choice and stronger accountability systems – willlead to improved student achievement. After accountingfor such costs, there remain substantial new dollars to assiststates and LEAs with general school improvement. Wefind that studies claiming that NCLB contains insuffi-cient funding for general school improvement lack solidevidence regarding the cost of improvement. Further, manyof these studies make the questionable assumption thatbecause NCLB encourages states and LEAs to improveschools that are not adequately serving all students (by thestates’ own measures), the federal government now has theprimary responsibility for all necessary expenditures.Historically, the federal government has provided a frac-tion of elementary and secondary expenditures (approxi-mately 7% in recent years). States choosing to accept TitleI and other federal dollars should be assured that substantialfederal resources accompany new demands. There is, how-ever, no reason to assume that the fundamental federal role

has changed to the point that all new future K-12 needsare now the responsibility of the federal government.

Therefore, we conclude that the charge that NCLB is an“unfunded mandate” is false; additionally, we find that thelevel of federal funding provided to support implementa-tion of NCLB requirements has been – and is likely toremain – sufficient.

NCLB UNDER A MICROSCOP E > INTRODUCT ION

5 ]

NCLB aims to ensure that all studentsin America’s public schools have accessto a high quality education and thatschools are held firmly accountable forstudent results.

a We use the term “local education agencies” to include school districts and other regional administrative entities charged with delivery ofpublic education.

6[

m e t h o d o l o g y

In examining the financial impact on the U.S. public edu-cation system of the federal No Child Left Behind Act of2001 (NCLB), the authors of this report completed threecritical steps. The first step was to compare NCLB to itspredecessor, the Improving America’s Schools Act (IASA),and to other laws that NCLB affects.The second step wasto establish precise assumptions about the most effectiveand efficient options available to states and LEAs in com-plying with the law (i.e., not necessarily the options statesand LEAs actually choose). The last step was to determinethe cost of implementing those options.

There are many significant ways in which NCLB departsfrom IASA; however, both share common features. Bothacts require the establishment of clear academic standards,measures of student assessment, professional developmentfor staff, state planning, parental involvement, and correctiveaction when necessary (though NCLB increases demandsfor assessment and corrective action). In 1994, referringto IASA, the U.S. Department of Education wrote:

“The new Title I requires states to submit plans thatdemonstrate that they have challenging content stan-dards specifying what children are expected to knowand be able to do and performance standards in at leastthe curricular areas of mathematics and reading … A

state’s Title I plan must show how LEP students willbe served, and how Title I and other resources will beused for school wide reform strategies to meet the needsof all students.” 1

In examining both IASA and NCLB and reviewingreports and other materials published by public educa-tion agencies, the authors of this study identified fourareas in which NCLB introduces specific new require-ments that will have a necessary fiscal impact on statesand LEAs (“hard costs”). These areas – as discussed insection three of this report and summarized in Table 3 –are as follows:

1. Accountability (Adequate Yearly Progress and StudentAssessments)

2. Personnel (High Quality Teachers and Paraprofessionals)

3. Information Management (Database System for Dis-

In examining the financial impact on the U.S. public education system of thefederal No Child Left Behind Act of 2001(NCLB), the authors of this reportcompleted three critical steps.

aggregation of Student Data)

4. School Improvement (Corrective Action and StudentSupport)

First, we review hard costs in each of these four areas.Second, we discuss a number of considerations relevantto further expenditures on improving student achieve-ment across the board (which, in budgeting, we refer toas “general school improvement/surplus”). Third, wereview a number of previous studies on the costs ofNCLB, most of which claim to identify “unfunded man-dates.” Throughout this study, we also assume that statesfollow the legal mandate to use NCLB and federal spe-cial education dollars to add to – not replace – localexpenditures (“supplement, not supplant”); given the factthat the federal contribution to elementary and second-ary education is but a fraction of total spending, no imag-inable increase in federal funding would be sufficient tosupport reform if we assume otherwise. Further, impor-

tant federal regulations regarding accountability for stu-dents with disabilities were not yet implemented as of thepublication of this report; we assume that these regula-tions will be implemented in a flexible manner, permittingstates to design accountability policies that are appropriateto the special circumstances of these students. The impactof NCLB on special education costs will need to be mon-itored in coming years.

NCLB UNDER A MICROSCOP E > METHOD OLO GY

7 ]

8[

r e s u l t s

Federal Funding Calculations

There are numerous ways in which one might assess thefederal funding made available to assist states in comply-ing with NCLB. There are differences in the resourcesavailable for all public school students, for individualstates, and for individual LEAs and schools. Given thiscomplexity, this study considers the total federal elemen-tary and secondary funding increases available for NCLBpurposes and the total demand on states directly relatedto NCLB. For years extending beyond the 2003-04 schoolyear, we have used a projection based on historical trends.While the increase in federal funding for elementary andsecondary education has increased sharply in the first twoyears since the passage of NCLB (17.1 percent in 2002and 9.2 percent in 2003), we assume a more conservativeprojection of 7.2 percent (the average annual increasedating back to 1980) increases in later years. The studyassumes none of the expected funding increases availablefor Adult Education or Vocational Education. We includeone-half of the funding increases available for SpecialEducation under the Individuals with Disabilities Act(IDEA), given that a major purpose of IDEA (i.e., increas-ing the reading and math skills of special education stu-dents) is also a substantial purpose of NCLB. In Table 1,

the first row includes the total annual revenue for statesas a result of these NCLB-related federal programs. Thesecond row includes the total amount available in newfederal funds each year (i.e., the amount above and beyondthe funding level in the last year before NCLB). Thethird row indicates the annual additional costs to states ofcomplying with specific NCLB requirements and man-dates. The last row, or “bottom line,” reflects the amount offunds left over that states may use as they see fit for generalschool improvement and increasing student achievement.

Total Federal Funding Available Under NCLB is Sufficient for the Implementation ofNew Requirements

If funding continues to increase at least at the conservativerate we have assumed, cumulative increases through the2007-08 school year will total $46.4 billion over the 2001-02 baseline funding school year. The net result of our analysisshows a “surplus” for general school improvement in every fiscalyear through 2008 after states have covered the costs of all specificmandates (“hard costs”) of NCLB.

In addition to generating an overall projection through2007-08, we also studied in detail the first two years of

actual appropriations for NCLB by program. Again, ouranalysis indicates that it has been possible for states todevelop NCLB implementation plans and applicationsthat allow them to comply with the law given actual fed-eral appropriations. As in all of our calculations, we reliedonly on “new” resources, defined as those appropriated byCongress for elementary and secondary education above

and beyond appropriation levels prior to NCLB. Fundingincreases for elementary and secondary education since thepassage of NCLB (2002-03 and 2003-04 school years)have included:

n A cumulative increase of $6.4 billion – 28.5 percent intotal – for all elementary and secondary and special edu-cation (see Table 2).

n Average annual increases of 13.5 percenta versus anhistorical average of 7.2 percent.b

n A cumulative increase of $2.9 billion – or 33 percent –for Title I alone (see Table 2).

n An increase to $5.5 billion in total funds (2003-04) towhich Title VI flexibility provisions apply, which shouldpermit states to increase efficiency.c

1 Accountability (Adequate Yearly Progress and Student Assessments)

Under NCLB, states face more rigorous requirementsfor defining and achieving adequate yearly progress(AYP) and ensuring that all students are proficient inthe core competency areas of reading/English LanguageArts (ELA) and mathematics (and later, in science). Ina departure from IASA, NCLB requires that AYP goals,as well as testing results, be disaggregated for the follow-ing subgroups:

n Economically disadvantaged

n English language learners

n Major racial and ethnic groups

n Students with disabilitiesd

Adequate Yearly Progress (AYP) Plans – $0

NCLB requires each state to develop a plan with newtargets for AYP. Unlike IASA, NCLB requires that AYPtargets be applied to student subgroups. While the estab-lishment and revision of AYP targets for subgroupsrepresents a substantial shift in practice for many states,these are, nevertheless, activities typically associated witheducational policymaking and administration. States andLEAs should be able to utilize existing staff and capac-ity for these activities, which should not require theexpenditure of incremental new federal funds. Similarly,there is no basis for assuming that the costs of theannual determination of which schools and districts haveattained their targets for AYP have increased significant-ly (though, as indicated below, many states will need toinvest in enhancing state databases in order to permitsuch calculations).

NCLB UNDER A MICROSCOP E > RESULTS

9 ]

The net result of our analysis shows a “surplus” for general school improvementin every fiscal year through 2008 afterstates have covered the costs of all specificmandates (“hard costs”) of NCLB.

a In estimating available funds for later years, we assume the more conservative projection of 7.2 percent.b Dating back to 1980.c Title VI applies to Improving Teacher Quality State Grants, Safe and Drug Free Schools, 21st Century Learning Centers (at state level

only), Educational Technology and State Grants for Innovative Programs.d In addition to the four subgroups listed here, NCLB requires that states disaggregate test results by gender and migrant status, but states are

not required to establish AYP goals for these two additional subgroups.

10[


TA

BL

E1.

Tot

al P

roje

cted

NC

LB

Fun

ding

and

Gen

eral

Sch

ool I

mpr

ovem

ent/

Surp

lus b

y Sc

hool

Yea

ra

2002

-03

2003

-04

2004

-05

2005

-06

2006

-07

2007

-08

Act

ual

Act

ual

Proj

ecte

dPr

ojec

ted

Proj

ecte

dPr

ojec

ted

$26,

528,

912,

000

$28,

867,

895,

500

$30,

946,

383,

976

$33,

174,

523,

622

$35,

563,

089,

323

$38,

123,

631,

754

$3,5

21,9

49,0

86$5

,297

,261

,995

$6,7

98,2

69,9

50$8

,434

,780

,802

$10,

217,

222,

804

$12,

156,

791,

506

$1,9

64,1

69,3

38$3

,789

,493

,255

$6,0

13,5

03,5

34$6

,588

,704

,953

$6,8

21,8

81,0

17$7

,205

,013

,195

$1,5

57,7

79,7

48$1

,507

,768

,740

$784

,766

,416

$1,8

46,0

75,8

49$3

,395

,341

,787

$4,9

51,7

78,3

11

Tot

al E

lem

enta

ry a

ndSe

cond

ary

Fund

ing

Ava

ilabl

efo

r NC

LB

pur

pose

s(in

clud

ing

50%

of I

DE

A)

Tot

al I

ncre

ase

Ove

r L

ast Y

ear

of I

ASA

/ID

EA

(“

new

dol

lars

”)

Tot

al “

Har

d”C

osts

(“

new

cos

ts”)

Gen

eral

Sch

ool

Impr

ovem

ent/

Surp

lusb

aFu

nds

for

proj

ecte

d ye

ars

are

adju

sted

by

a fu

ture

infla

tion

fact

or o

f 2.4

5 pe

rcen

t (So

urce

:U.S

.Fed

eral

Res

erve

Ban

k of

Phi

lade

lphi

a).F

undi

ng a

vaila

ble

for

NC

LB

incl

udes

one

hal

f of

IDE

A S

peci

al E

duca

tion

fund

ing

and

all o

ther

ele

men

tary

and

sec

onda

ry e

duca

tion

fund

ing

excl

udin

g vo

catio

nal e

duca

tion

and

adul

t edu

catio

n fu

nds)

.b

Fund

s re

mai

ning

aft

er th

e “h

ard

cost

s”of

NC

LB

are

det

erm

ined

(e.

g.,t

estin

g,sc

hool

cho

ice

tran

spor

tatio

n,su

pple

men

tal s

ervi

ces,

etc.

) ar

e al

loca

ted

to th

e ca

tego

ry o

f “ge

nera

l sch

ool

impr

ovem

ent/

surp

lus.

”


11 ]

2001

-02

(pre

-NC

LB

)a20

02-0

320

03-0

4C

umul

ativ

e C

hang

e

$286

,000

$975

,000

$1,0

68,0

12$7

82,0

12

$8,7

62,7

21

$10,

350,

000

$11,

684,

311

$2,9

21,5

90

$993

,302

$1

,143

,500

$1

,188

,226

$1

94,9

24

$2,1

08,0

00

$2,8

50,0

00

$2,9

30,8

25

$822

,825

$439

,250

$4

72,0

17

$468

,949

$2

9,69

9

$845

,614

$1

,000

,000

$9

93,5

00

$147

,886

$713

,328

$7

63,0

00

$758

,041

$4

4,71

3

$385

,000

$3

85,0

00

$382

,498

($

2,50

2)

$387

,000

$3

84,4

84

$384

,484

$24,

838

$24,

838

$190

,000

$2

00,0

00

$198

,700

$8

,700

$115

,500

$1

20,3

68

$121

,573

$6

,073

$446

,000

$6

64,2

69

$685

,515

$2

39,5

15

$3,7

19,9

74

$4,3

36,4

02

$5,0

16,9

58

$1,2

96,9

84

$3,4

52,0

83

$2,8

82,3

56

$2,9

61,4

65

($49

0,61

8)

$22,

456,

772

$26,

528,

912

$28,

867,

895

$6,4

11,1

23

Rea

ding

Fir

st

ESE

A T

itle

I G

rant

s to

Loc

al E

duca

tiona

lA

genc

ies

Impa

ct A

id

Impr

ovin

g T

each

er Q

ualit

y St

ate

Gra

nts

Safe

and

Dru

g-Fr

ee S

choo

ls

Stat

e G

rant

s

21st

Cen

tury

Com

mun

ity

Lea

rnin

g C

ente

rs

Edu

catio

nal T

echn

olog

y

Stat

e G

rant

s fo

r In

nova

tive

Prog

ram

s

Stat

e A

sses

smen

ts

Cre

dit E

nhan

cem

ent f

or C

hart

er

Scho

ol F

acili

ties

Cha

rter

Sch

ools

Gra

nts

Indi

an E

duca

tion

Eng

lish

Lan

guag

e A

cqui

sitio

n

Spec

ial E

duca

tionb

Oth

er

Tot

al E

ach

Year

Net

Inc

reas

e to

Dat

e Si

nce

NC

LB

aSo

me

of th

e lin

e ite

ms s

how

n fo

r the

bas

elin

e sc

hool

yea

r 200

1-02

) ref

lect

fund

s tha

t wer

e av

aila

ble

for s

imila

r pur

pose

s,bu

t und

er d

iffer

ent p

rogr

amm

atic

line

item

s.T

hey

are

show

n he

re u

nder

the

rele

vant

NC

LB

line

item

in o

rder

to p

rovi

de a

fair

com

pari

son

betw

een

pre-

NL

CB

fund

ing

and

post

-NC

LB

fund

ing.

For

exam

ple,

the

Rea

ding

Firs

t lin

e ite

m in

200

1-02

con

tain

s fu

nds

appr

opri

ated

for

the

Rea

ding

Exc

elle

nce

Act

.Als

o,th

e Im

prov

ing

Tea

cher

Qua

lity

Stat

e G

rant

s lin

e in

200

1-02

con

tain

s fu

nds

appr

opri

ated

for

Cla

ss S

ize

Red

uctio

n an

d E

isen

how

er P

rofe

ssio

nal D

evel

opm

ent G

rant

s.b

IDE

A fu

ndin

g in

Tab

le 2

incl

udes

100

per

cent

of f

eder

al fu

ndin

g in

crea

ses.

TA

BL

E2.

aN

omin

al F

undi

ng I

ncre

ases

by

Pro

gram

Sin

ce th

e Pa

ssag

e of

NC

LB

by

Scho

ol Y

ear (

dolla

rs in

thou

sand

s)2

12[


2002

- 03

2003

-04

2004

-05

2005

-06

2006

-07

2007

-08

Cum

ulat

ive

Act

ual

Act

ual

Proj

ecte

dPr

ojec

ted

Proj

ecte

dPr

ojec

ted

Tot

al

$295

,800

,000

$295

,800

,000

$368

,830

,707

$738

,804

,066

$756

,904

,767

$914

,526

,030

$3,3

70,6

65,5

70

$1,1

52,7

44,3

38$2

,553

,619

,930

$4,1

70,2

33,0

46$4

,272

,403

,755

$4,3

77,0

77,6

47$4

,484

,316

,050

$21,

010,

394,

766

$127

,500

,000

$63,

750,

000

$65,

311,

875

$66,

912,

016

$68,

551,

360

$70,

230,

869

$462

,256

,120

$388

,125

,000

$876

,323

,325

$1,4

09,1

27,9

07$1

,510

,585

,116

$1,6

19,3

47,2

44$1

,735

,940

,246

$7,5

39,4

48,8

38

$1,9

64,1

69,3

38$3

,789

,493

,255

$6,0

13,5

03,5

35$6

,588

,704

,953

$6,8

21,8

81,0

18$7

,205

,013

,195

$32,

382,

765,

294

Acc

ount

abili

ty

(Ade

quat

e Ye

arly

Pro

gres

s an

d St

uden

t Ass

essm

ents

)

Pers

onne

l (H

igh

Qua

lity

Tea

cher

s)

Info

rmat

ion

Man

agem

ent

(Dat

abas

e Sy

stem

fo

r D

isag

greg

atio

n of

Stu

dent

Dat

a)

Scho

ol I

mpr

ovem

ent

(Cor

rect

ive

Act

ion

and

Stud

ent S

uppo

rt)

Tot

al E

ach

Year

TA

BL

E3.

“Har

d C

osts

”:N

ew N

CL

B re

quir

emen

ts a

nd th

eir f

isca

l im

pact

on

stat

es a

nd L

EA

s

Student Assessments Cumulative Total – $3.37 billionTitle I general education assessments – $2.51 billionSpecial populations assessments – $0.87 billion

NCLB clearly increases the financial burden on states inthe area of student assessments. NCLB requires the fol-lowing assessments over and above IASA requirements:a

n Four additional reading tests (in grade spans threethrough eight) administered annually;

n Four additional mathematics tests (in grade spans threethrough eight) administered annually;

n Three new science tests (in grade spans three throughfive, six through nine, and 10 through12) adminis-tered annually;

n Annual assessment of students with limited Englishproficiency (LEP); and

n Alternate assessments are permitted to be includedin calculations of AYP for some students with disabil-ities (alternate assessments are also required under pre-existing IDEA).

We have developed projections for each of these instru-ments based on independent research as well as a reviewof extant studies on the subject. While IDEA previous-ly required alternate assessments for students with dis-abilities, we conservatively budget additional expendi-tures in case compliance with NCLB requires additionalrefinement or new development in this area. Based oninformation obtained from state testing directors con-tacted for this study, we estimate that each new assess-ment will cost up to $1.2 million to develop and that thisdevelopment will take up to three years. We includedassumptions about assessment administration costs thatvary by instrument. We have also added an assumption,missing from some studies, on the development costs forongoing test maintenance (i.e., refreshment of forms fortest security purposes).

These assumptions reflect conservative estimates for amid-range test design option: they are based on each statedeveloping and administering custom tests that includemultiple-choice items, a significant number of shortanswer open-response items, and a limited number oflonger extended-response items on some tests/subjects.The lowest cost option would be tests relying entirely onmultiple-choice items while the highest cost option wouldbe all tests heavily dependent on essays or similar extend-ed performance tasks. Further, we assume that no stateshave in place the grade-by-grade testing required byNCLB (though some, such as Texas and North Carolina,already do), nor do we assume any savings from the shar-ing of test development costs through consortia of states(some of which are, in fact, occurring). Using NationalCenter for Education Statistics data on student popu-lations for calculating administration costs, we arrive attotal new student assessment costs of $3.37 billion forthe period of this analysis, as detailed in Table 4. Of thisamount, $2.51 billion are for general education assess-ments required under Title I, while $0.87 billion are forspecial population assessments that may be required underTitle III, Title I and overlapping requirements underIDEA. It is important to note that NCLB does not

require the inclusion of open-ended test questions, asassumed here, for the general education Title I assess-ments (such test items are assumed because most statesappear likely to include them and because it is likely thatsuch a mix reflects good policy); states wishing to save,or reallocate, testing dollars could choose to implementassessments limited solely to multiple-choice items, whichis permissible under NCLB.


13 ]

NCLB requires the following assessments over and above IASArequirements

a Biennial participation in the National Assessment of Education Progress (NAEP) is also required beginning in 2002-03. We have omittedthis from our cost analysis because the law specifies that funds are held at the federal level and administered directly. NAEP fundingincreased from $36 million in fiscal year 2001 to $107.5 million in fiscal year 2002 and $90.2 million in fiscal year 2003.

14[


A variety of federal revenue streams may be used to payfor these new testing costs.The most obvious are the newNCLB funds dedicated specifically to assessment andallocated to individual states based on a federal formula.Additional assessment funds appropriated by Congressabove this level have been distributed competitively, witha preference for consortia of states sharing the benefitsof the final product. Costs associated with the develop-ment of assessments for LEP students may also be paidfrom a portion of NCLB Title III revenues. Costs associ-ated with the development of alternate assessments forspecial education students may also be paid from a por-tion of IDEA revenues. All of these available revenuestreams must be taken into account.

Several studies have already been published concerningthe costs to states to comply with the assessment require-ments outlined in NCLB, with price tags ranging from$1.9 to $7 billion (some of these are reviewed at the endof this report).

Recently, the General Accounting Office (GAO) releaseda report on assessment development.3 The GAO con-cluded that the cost to states would be $1.9, $3.9 or $5.3billion through 2008 to develop and administer reading,mathematics and science assessments as required underTitle I. The three different GAO estimates vary prima-rily as a result of different assumptions about the typesof test items the states would employ and how the assess-ments would be scored.The GAO report has merit becauseit presents very useful data on overall Title I assessment

costs and recommends that states could benefit from thesharing of information on assessment development. TheGAO assumptions, however, are not appropriate fordetermining whether new NCLB requirements areadequately funded because they are not focused on theincrementally new obligations imposed by NCLB.a TheGAO estimates include the costs of complying with thepre-existing IASA requirements for those states that werenot already in compliance.

Our assumptions most closely match those of a previ-ous report developed by AccountabilityWorks for ELCin that, in order to isolate the impact of NCLB, we haveassumed that all states are compliant with IASA, nomore and no less.4 Thus, we have attributed additionalassessment costs to all states, including states that arealready compliant (or partially compliant) with NCLBand, conversely, we have made no special allowance forstates that did not meet the requirements set by IASA.Assuming that all states face additional assessment costsis clearly a conservative assumption given that some states,such as Texas and Florida, already have in place grade-by-grade assessment as required by NCLB.The assump-tion that all states are compliant with IASA is appropriatein that we are seeking to compare additional costs imposedby NCLB against additional dollars provided.

2 Personnel (High Quality Teachers and Paraprofessionals)Cumulative Total – $21 billion

High Quality Teachers: Evaluation of Current Teachers and Intensive Content-orientedRetraining for Those Who Require It Cumulative Total – $598 million

By the 2005-06 school year, all teachers of core academicsubjects must meet the “Highly Qualified” standards ofNCLB. By 2002-03, all new teachers hired using Title

The GAO assumptions, however,are not appropriate for determiningwhether new NCLB requirements are adequately funded because they arenot focused on the incrementally newobligations imposed by NCLB.

a In a response to the GAO report, the U.S. Department of Education questioned the methodology of the GAO, including: it did not collectsufficient kinds of data in order to inform its estimates of testing costs, as well as the variable characteristics and options available to individ-ual states in the construction, scoring and dissemination of test data; it incorrectly attributes the costs of developing new assessments to therequirements of NCLB; and it does not consider all possible sources of Federal funding. The U.S. Department of Education agrees with theoverall assertion that information sharing among states can reduce the financial burdens on individual states.


15 ]

aPr

ojec

tions

ass

ume

test

dev

elop

men

t will

cos

t $1.

2 m

illio

n pe

r tes

t for

fift

y st

ates

plu

s th

e D

istric

t of C

olum

bia

base

d on

dat

a ga

ther

ed fr

om s

tate

test

ing

dire

ctor

s co

ntac

ted

for t

his

stud

y (c

onta

cts

avai

labl

e up

on re

ques

t),a

nd a

ssum

e th

at te

sts

will

be

deve

lope

d ov

er a

thre

e-ye

ar p

erio

d.N

eith

er o

ur re

venu

e no

r our

exp

ense

pro

ject

ions

in th

is re

port

incl

ude

figur

es fo

r out

lyin

g ar

eas.

bPr

ojec

tions

ass

ume

a co

st o

f $10

per

stu

dent

bas

ed o

n da

ta g

athe

red

from

sta

te te

stin

g di

rect

ors

cont

acte

d fo

r thi

s st

udy.

The

add

ition

al g

rade

s to

be

adde

d w

ill v

ary

from

sta

te to

sta

te s

o w

e ha

ve u

sed

an a

vera

ge s

tude

nt e

nrol

lmen

t for

gra

des

3 to

8 o

f 3,6

98,1

05 a

nd m

ultip

lied

this

by fo

ur fo

r a to

tal a

nnua

l inc

reas

ed te

stin

g vo

lum

e of

14,

792,

418.

cPr

ojec

tions

ass

ume

annu

al m

aint

enan

ce c

osts

of $

500,

000

per t

est b

ased

on

data

gat

here

d fr

om s

tate

test

ing

dire

ctor

s co

ntac

ted

for t

his

stud

y.d

Sam

e as

with

read

ing,

but f

or th

ree

addi

tiona

l gra

des,

not f

our.

eA

ssum

es th

e de

velo

pmen

t of f

our n

ew te

sts

at a

cos

t of $

1.2

mill

ion

each

to b

e de

velo

ped

over

a tw

o-ye

ar p

erio

d.f

Ass

umes

the

test

s w

ill b

e ad

min

ister

ed to

all

K-1

2 E

nglis

h la

ngua

ge le

arne

rs.U

ses

NC

ES

Com

mon

Cor

e D

ata

for t

he 2

001-

02 s

choo

l yea

r for

all

K-1

2 st

uden

ts a

nd a

NC

ES

estim

ate

that

nat

ionw

ide

abou

t 5 p

erce

nt o

f stu

dent

s ha

ve li

mite

d E

nglis

h fo

r a to

tal t

estin

g po

pula

tion

of 2

,341

,045

.g

Ass

umes

ann

ual m

aint

enan

ce c

osts

of $

500,

000

per t

est b

ased

on

data

gat

here

d fr

om s

tate

test

ing

dire

ctor

s co

ntac

ted

for t

his

stud

y (c

onta

cts

avai

labl

e up

on re

ques

t).

hA

ssum

es th

at o

ne n

ew te

st is

add

ed a

nd d

evel

oped

ove

r a th

ree-

year

per

iod

and

that

hal

f of t

he c

ost i

s at

trib

uted

to N

CL

B (b

ecau

se o

nly

half

of th

e sp

ecia

l edu

catio

n fu

ndin

g in

crea

se is

att

ribut

ed to

NC

LB

in th

is an

alys

is).T

he D

epar

tmen

t of E

duca

tion

has

been

allo

win

g th

e st

ates

som

e le

eway

in th

e tim

ing

of th

e im

plem

enta

tion

of th

is re

quire

men

t as

long

as

the

stat

es a

re a

ctiv

ely

wor

king

tow

ards

a

solu

tion.

Ass

umes

that

the

test

is a

dmin

ister

ed to

1 p

erce

nt o

f the

K-1

2 20

01-0

2 po

pula

tion

from

NC

ES

(47,

687,

871

pre-

K-1

2 le

ss p

re-K

enr

ollm

ent o

f 866

,969

).i

Ass

umes

that

test

adm

inist

ratio

n w

ill c

ost $

20 p

er s

tude

nt.

jA

ssum

es a

nnua

l mai

nten

ance

cos

t of $

500,

000

per t

est.

TA

BL

E4.

Stud

ent A

sses

smen

ts fo

r Gen

eral

Edu

cati

on a

nd S

peci

al P

opul

atio

ns:D

evel

opm

ent,

Adm

inis

trat

ion

and

Mai

nten

ance

Cos

ts

by S

choo

l Yea

r

2002

-03

2003

-04

2004

-05

2005

-06

2006

-07

2007

-08

Cum

ulat

ive

Tot

al

$81,

600,

000

$81,

600,

000

$83,

599,

200

$246

,799

,200

$155

,261

,256

$159

,065

,157

$162

,962

,253

$477

,288

,666

$107

,059

,226

$1

09,6

82,1

77$1

12,3

69,3

90$3

29,1

10,7

93

$81,

600,

000

$81,

600,

000

$83,

599,

200

$246

,799

,200

$155

,261

,256

$159

,065

,157

$162

,962

,253

$477

,288

,666

$107

,059

,226

$109

,682

,177

$112

,369

,390

$329

,110

,793

$62,

699,

400

$64,

235,

535

$65,

809,

306

$192

,744

,241

$122

,221

,690

$122

,221

,690

$84,

277,

042

$84,

277,

042

$122

,400

,000

$122

,400

,000

$244

,800

,000

$23,

984,

007

$24,

571,

615

$25,

173,

620

$25,

790,

373

$99,

519,

615

$104

,499

,000

$107

,059

,226

$109

,682

,177

$112

,369

,390

$433

,609

,793

$10,

200,

000

$10,

200,

000

$10,

449,

900

$30,

849,

900

$4,9

14,3

23$5

,034

,724

$5

,158

,075

$15,

107,

122

$13,

382,

403

$13,

710,

272

$14,

046,

174

$41,

138,

849

$295

,800

,000

$2

95,8

00,0

00

$368

,830

,707

$7

38,8

04,0

66

$756

,904

,767

$9

14,5

26,0

30

$3,3

70,6

65,5

70

Rea

ding

/EL

AD

evel

opm

enta

Adm

inis

trat

ionb

Mai

nten

ance

c

Mat

hem

atic

sD

evel

opm

ent

Adm

inis

trat

ion

Mai

nten

ance

Scie

nce

Dev

elop

men

tA

dmin

istr

atio

nd

Mai

nten

ance

LE

PD

evel

opm

ente

Adm

inis

trat

ionf

Mai

nten

ance

g

Alte

rnat

e A

sses

smen

tD

evel

opm

enth

Adm

inis

trat

ioni

Mai

nten

ance

j

Year

ly T

otal

s

16[


I funding must meet these requirements. In general, ateacher, to be considered “Highly Qualified,” must havea bachelor’s degree, be fully certified, and have demon-strated knowledge of the core subjects taught. But eachstate decides on the specific criteria it uses to determinewhether an individual has met the federal mandate.Every state already requires a Bachelor’s degree and statesthemselves define what it means to be fully certified.a

Therefore, the key new federal requirement in NCLBis to ensure that teachers possess content knowledge inthe subjects that they teach.

With regard to current teachers, the most obvious con-cern would be teachers who are teaching “out of field”as a result of emergency certification or similar waiver,but there is no guarantee that even teachers who areteaching within their field possess all of the subjectknowledge they need. Under NCLB, teachers may dem-onstrate subject competency either through a test orthrough another objective evaluation system developed bythe state (such as, an evaluation tied to state academiccontent standards, conducted by an administrator).Whileit would be unreliable to estimate the cost of still devel-oping – and widely varying – state systems based onadministrative evaluation, one can develop a very rigor-ous set of assumptions for a test-based approach. Since,with respect to current teachers, most states appear to beusing administrative evaluations that rely on existingadministrator salaries rather than new “hard” costs, atest-based cost estimate is likely to be a high estimate.As the most fiscally conservative – though maximallyreform-oriented option – we assume that all currentteachers are assessed to ensure subject competence.

If we calculate that all teachers are assessed at a cost of$110 per teacher, the total cost for the current populationof teachers in the United States would be $330 million.b

Providing each teacher to be tested a study guide at a costof $25 each would cost an additional $82 million.

It is difficult to estimate the proportion of currentteachers likely to be deemed not “Highly Qualified” bythe states. In July 2003, a report by the U.S. Departmentof Education (USDoEd) estimated that more thanhalf of U.S. teachers would not have met the definitionif the statutory requirements were in place in 1999-2000.5 However, preliminary results reported by thestates are far more optimistic: while 11 states have notreported any data so far, “[t]hirty-three states reportthat at least four out of five classes in core subjects haveteachers who are ‘highly qualified.’”6 For the purpose ofthis analysis, we assume a somewhat higher proportionof current teachers not meeting subject matter compe-tence requirements than the initial state reports indicateis likely, though not as high a number as the USDoEdestimated. We assume that as many as one-quarter ofcurrent teachers would not meet the subject mattercompetence requirements and would need to be provid-ed intensive content-oriented professional developmentto address their deficiencies. We further include the costof re-evaluating the subject matter competence of thosewho fail using the same cost assumptions for the initialtesting. The content-based remedial training and the costof re-testing add an estimated $186 million in expenses.

As noted previously, a test-based option is only oneway by which states could choose to comply with therelevant requirements of NCLB. But we have selected itfor budgeting purposes because it is an efficient optionthat would provide reliable and consistent informationon the extent of teachers’ content knowledge and is likelyto lead to a conservative estimate of new costs. Thoughthis cost analysis is focused on new costs and new fed-eral revenues, it is worth noting that states and districtswould do well to review and, in some cases, re-directpre-existing federal and state funding for professional

a For example, whether “emergency certified” teachers must take education courses remains up to states. There is little research supporting theview that such courses typically raise student achievement or assist teachers in creating a disciplined learning environment.

b We have conservatively included all elementary and secondary teachers, although only teachers of core subjects must meet the federalrequirements.

development. As discussed later in this report, there isreason to believe that existing professional developmentexpenditures are often not the type of research-based,content-intensive experiences that are likely to have thestrongest effect.

High Quality Teachers: Recruitment and RetentionCumulative Total – $16.2 billion

The challenges of recruiting and retaining high qualityteachers in special education, mathematics and sciencein school districts throughout the country, but especiallyin those serving disadvantaged students, are well known.Individuals who possess degrees in the fields of mathe-matics and science can often be more selective in choos-ing a high-paying career than many of their fellow liberalarts graduates. The field of special education requiresadditional methodological skills and presents particulareveryday challenges in the classroom.

NCLB outlines a number of innovative ways in whichfederal teacher quality grants can be used to attract andretain high quality teachers, including:

n Professional development opportunities

n Signing bonuses

n Mentoring programs

n Performance bonuses

n Scholarships

n Differential pay

We expect that such reforms in the way existing fundsare being spent could go a long way toward addressingthe challenge of ensuring a “Highly Qualified” teacher in

every classroom.These include reforms in the way teach-ers are recruited, trained and compensated. Innovativeapproaches to attracting “Highly Qualified” teachersthrough alternate routes based on high standards – such as the American Board for Certification of TeacherExcellence – offer new avenues for states.Therefore, fewreasons exist to believe that after such reforms, non-inflationary, across-the-board compensation increaseswould be required in non-shortage areas in order to meetNCLB requirements for “Highly Qualified” teachers(however desirable these increases may be for other rea-sons, such as recognizing the social value of the teach-ing profession).

However, we do assume, conservatively, that additionalteacher compensation expenditures may be necessary toaddress shortage areas in science, mathematics and specialeducation. Therefore, we include the cost of closing thegap between current levels of teacher compensation inthese shortage areas and the level of compensation forpositions in the private sector requiring similar skills.a

We estimate that teaching positions are, on average, paidapproximately 14 percent below the market value forequivalent private sector positions (e.g., those employedin the Life Science and Engineering fields) that requirea Bachelor’s degree in math or science.b According to datafrom the National Education Association,7 the averagecompensation for classroom teachers is $45,822. A 14percent increase in compensation would cost $6,415per teacher. (Note: we are not advocating that the bestapproach is for every teacher in these shortage areas toreceive the same increase, but rather are using this fig-ure as an average for the purpose of calculation; thoughpolitically challenging, it is conceivable that a merit-based


17 ]

a Occupations used in calculating average compensation for those employed in the Engineering and Life Science sectors included: aerospaceengineers, agricultural engineers, biomedical engineers, chemical engineers, civil engineers, computer hardware engineers, electrical and electronicsengineers (except computer), environmental engineers, industrial engineers (including health and safety), materials engineers, mechanical engineers,mining and geological engineers (including mining safety engineers), nuclear engineers, petroleum engineers, agricultural scientists, food scientists,conservation scientists, and foresters. Occupations in the life, physical and social sciences for which middle or high school science teachers areunlikely be qualified (such as, those which typically require an advanced degree in medicine or a Ph.D.) were omitted.

b Drawing on Bureau of Labor Statistics sources, the mean compensation for elementary and secondary teachers in the most recent year available(2000) was $41,820 while the average compensation for professions with Bachelor’s degrees in the Life Science and Engineering fields was$60,368. We conservatively estimate that teachers are required to work 190 days each year, while individuals in the Life Science and Engineeringfields work approximately 240 days each year. When adjusted for the differences in days worked per year (by multiplying $60,368 by .79), weconclude that professions in the Life Science and Engineering sectors earned approximately $47,690, while teachers earned $41,820, for a similaramount of work (excluding supplemental teacher income through summer employment); thus, teachers earn 14 percent less.

18[


approach would award a greater proportion of this increaseto the best teachers in shortage areas).

According to the Digest of Education Statistics8 the totalnumber of elementary and secondary math and scienceteachers is 376,284.a Thus, a 14 percent average increasein compensation would cost approximately $2.4 billioneach year, thereafter, increasing with inflation. For budg-eting purposes, we assume this large increase is phasedin over two years (50 percent of the increase in 2003-04and the remainder in 2004-05).

Teaching positions in special education are also difficultto fill. While it is difficult to find a precise comparisongroup in the private sector for special education teachers,a conservative assumption would be to assume a similar14 percent average increase in compensation as well.According to the Digest of Education Statistics, the totalnumber of elementary and secondary special educationteachers is 309,440.Thus, a 14 percent increase in averagecompensation for special education teachers, assumingthe same level of compensation, would cost an additional$1 billion each year, increasing with inflation.b Weassume this large increase is phased in over two years – 50percent of the increase in 2003-04 and the remainderin 2004-05. (As noted above for math and scienceteachers, the 14 percent increase is a budgetary assump-tion that reflects an aggregate adjustment that is averagedacross all special education teachers – we are not neces-sarily advocating that every individual teacher, regard-less of merit, should receive the same size increase.)

No doubt, existing teacher contracts or, in some cases,state laws present challenges to implementation of thetype of reforms discussed above. While, in the longterm, it is essential that such structural obstacles to an eco-nomically rational and efficient teacher recruitment andcompensation system must be addressed, in the shortterm there are other ways to accomplish some of thesame ends through different means. For example, while

contractual or legal obstacles could initially preventdifferential (higher) compensation for teachers in short-age areas, it may be possible to offer signing bonuses or housing benefits for teachers in shortage areas (requiringsuch teachers to remain for a minimum number of years).Reform of recruitment and compensation policies is likelyto be a challenging undertaking, but it is an important oneif ambitious student achievement and teacher qualificationgoals are to be met at a reasonable cost to taxpayers.

High Quality Teachers: Evaluation of CurrentParaprofessionals, Content-oriented Retraining for Those Who Need It, and CompensationAdjustment for Recruitment and Retention of Individuals With Higher SkillsCumulative Total – $4.2 billion

There are 675,038 teachers’ assistants9 in the UnitedStates. In order to comply with NCLB, states musttest paraprofessionals for competency or hire para-professionals with a Bachelor’s degree or two years of post-secondary education. Two years of post-secondaryeducation would be an expensive proposition, so for budg-eting purposes we assume competency testing.

If we assume that state education agencies require allparaprofessionals to be tested over two years beginningin fiscal year 2002 using a nationally recognized test suchas ParaPro – at a cost of about $40 each or $27 million– then one can identify the need for professional devel-opment assistance.c If states paid for a study guide for

a This should be a conservative number as the Title II report includes full-time, part-time and long-term substitutes.b As mentioned earlier, we include only one-half of the Special Education funding increase as usable for (overlapping) NCLB purposes and

thus only half of the costs as well.c ParaPro is marketed and sold by Educational Testing Service, Inc

Reform of recruitment and compensationpolicies is likely to be a challengingundertaking, but it is an important oneif ambitious student achievement andteacher qualification goals are to be metat a reasonable cost to taxpayers.

each person taking the test (at a cost of $27.50), it wouldcost states another $19 million.

If we then make a very conservative estimate that one-third of all paraprofessionals fail, states would thus needto assist 225,013 paraprofessionals in meeting the require-ments of NCLB. Through a combination of remedial,content-oriented training and assumption of the costto re-take the exam by states, this would cost states anestimated $31 million.

In order to attract paraprofessionals who possess higherlevels of academic skills, it is reasonable to assume thatschools may need to offer higher hourly rates. Using theaverage annual compensation for a full-time equivalentparaprofessional of $19,430,10 a 5 percent average increasein compensation would cost the states $656 million peryear, increasing with inflation.

High Quality Teachers: Parental NotificationCumulative Total – $52 million

While reporting school performance data to parents andpublic agencies is not a new innovation of NCLB, ourresearch has indicated that some of the new reportingrequirements involve more than simple modification ofexisting communication forms. In particular, the require-ment to proactively report teacher quality data to parentswill require some additional resources.

LEAs currently keep data on teacher qualifications, butuntil now were not required to disseminate that informa-tion to parents. Thus, we have assumed that the costsof doing so will include data entry and report generation.We have further assumed that LEAs will contract withan outside vendor to develop teacher quality reports andwill assume the cost of data entry with existing resources.

Using an average school size of 600 students, it wouldcost each district about $455 per year to subscribe to aservice such as that provided by Transact.com that pro-duces and regularly updates communication forms. If

we conservatively assume that teacher reporting is 20percent of the total reporting burden, then the cost tostates would be about $8.2 million more per year to getaccess to necessary reports, increasing with inflation.

3 Information Management (Database System for Disaggregation of Student Data)Cumulative Total – $462 million

In order to assess options available for collecting andreporting disaggregated student data as required byNCLB, we contacted a number of state officials andcommercial vendors who have already undertaken orare planning to undertake database projects. For a statewith limited systems to develop an inexpensive yetNCLB-compliant database for the collection and dis-aggregation of student data, we estimate that the costto states would be approximately $2.5 million in initial

development and $1.25 million per year in ongoingmaintenance costs. (More sophisticated systems – suchas ones that would enable longitudinal tracking andreporting of student data and allow for periodic audit-ing – are substantially more expensive; while prefer-able, they are not required by NLCB).

For the purposes of this study, then, we calculate aninitial development cost to states of $127.5 million andongoing maintenance costs of $63.7 million, increasingwith inflation.a


19 ]

While reporting school performance datato parents and public agencies is not anew innovation of NCLB, our researchhas indicated that some of the newreporting requirements involve morethan simple modification of existingcommunication forms.

a These calculations assume a base cost of $2.5 million to develop the system and $1.25 million annually.

20[


4 School Improvement (Corrective Action and Student Support)Cumulative Total - $7.5 billion

School Improvement Plans and School Choice Options

Under NCLB an LEA must require any school that failsto make adequately yearly progress for two years to devel-op a school improvement plan, and to offer students schoolchoice transportation and supplemental services.

The U.S. Department of Education has estimated that7,000 out of about 85,000 schools are currently on schoolimprovement plans.11 NCLB is likely to increase thenumbers of LEAs reporting schools facing consequencesunder section 1116 of NCLB. Based on a compilation ofUSDoEd projections for school choice transportationand supplemental services by LEAs, see Table 5 for themaximum state costs for school choice transportation andsupplemental services (i.e., assuming, conservatively, thatall Title I schools fail to achieve AYP and are subject tothese provisions). We have then assumed that a full 75percent of districts will actually be designated in needof improvement.

Because not all students will exercise their right to schoolchoice or supplemental services and because the publicschool choice options could be constrained due to limita-tions of supply, we assume that the total average expenseon public school choice transportation and supplementalservices is 5 percent of Title I funds in 2002-03, 10 per-cent in 2003-04, and a maximum 15 percent in 2004-05and the out years (rather than the 20 percent maximumallowed by law). (The total cost to all states is conservative-ly estimated at about $388 million in 2002-03, $876 mil-lion in 2003-04, $1,409 million in 2004-05, and approxi-mately 7.2 percent more each year after that until 2007-08, depending on annual Congressional appropriations).Though the provision of supplemental services has beenhampered by initial state and district administrative start-

up challenges, we estimate that private sector supply willrespond quickly to demand once these are ironed out.a

General School Reform and Improvement

Under both IASA and NCLB, schools that fail to makeAYP for two consecutive years must be identified forimprovement. NCLB, then, sets aside substantially greaterfunding for school improvement than IASA.12 Furthermore,as illustrated above in Table 1, we find that the efficientstrategies for complying with the identifiable NCLB “hard”costs that we have already delineated would still leave asubstantial “surplus” of new dollars that would be availablefor further school improvement activities. More importantis how all funds, new and pre-existing, are expended; inother words, whether current dollars for initiatives thatare not the most effective and efficient approaches can bere-directed toward better uses. While there is extensiveevidence, reviewed below, suggesting that existing funds canbe re-channeled to substantially increase student achieve-

ment, we are not arguing that increased funding is nothelpful. Highly efficient school systems that are alreadymaking best use of current resources could improve fur-ther with additional funding (though the research belowsuggests that there are few such actual cases). Further,public school funding allocation decisions inevitably involvepolitical pressures and constituencies, so increases can behelpful in “greasing the skids” for change. Many of therequirements of NCLB (as discussed above) also do requireadditional funding, much of which is likely to spurimproved achievement.

Under both IASA and NCLB, schoolsthat fail to make AYP for two consecutiveyears must be identified for improvement.NCLB, then, sets aside substantiallygreater funding for school improvementthan IASA.

a For example, according to the New York Times, “Free Tutoring Reaches Only Fraction of Students”, August 29, 2003, obstacles exist for parents totake advantage of supplemental services for their child. The article suggests that at least one education service provider – Sylvan Education Solutions– has begun its own extensive outreach efforts into city neighborhoods. We anticipate that such aggressive forms of marketing will increase.


21 ]

Ala

bam

a

Ala

ska

Ari

zona

Ark

ansa

s

Cal

iforn

ia

Col

orad

o

Con

nect

icut

Del

awar

e

Dis

tric

t of C

olum

bia

Flo

rida

Geo

rgia

Haw

aii

Idah

o

Illin

ois

Indi

ana

aA

ctua

l am

ount

s rec

eive

d by

LE

As w

ill b

e sm

alle

r tha

n sh

own

here

due

to st

ate-

leve

l adj

ustm

ents

to F

eder

al T

itle

I al

loca

tions

.Sta

tes a

djus

t allo

catio

ns,f

or e

xam

ple,

to re

flect

LE

A b

ound

ary

chan

ges o

r the

crea

tion

of n

ew L

EA

s,in

clud

ing

char

ter s

choo

l LE

As,

that

are

not

acc

ount

ed fo

r in

the

USD

oEd

calc

ulat

ions

.Sta

tes a

lso a

re p

erm

itted

to re

serv

e up

to 1

per

cent

of a

lloca

tions

for a

dmin

istra

tion

and

mus

tre

serv

e 2

perc

ent o

f allo

catio

ns (r

ising

to 4

per

cent

in fi

scal

yea

r 200

4) fo

r sch

ool i

mpr

ovem

ent a

ctiv

ities

.The

se a

djus

tmen

ts w

ill re

duce

the

actu

al a

mou

nts a

vaila

ble.

bA

n L

EA

mus

t use

up

to a

n am

ount

equ

al to

20

perc

ent o

f its

Titl

e I,

Part

A a

lloca

tion

(the

“20

perc

ent r

eser

vatio

n”) r

ecei

ved

from

the

Stat

e to

cov

er c

hoic

e-re

late

d tr

ansp

orta

tion

cost

s for

stud

ents

who

ex

erci

se a

cho

ice

optio

n an

d to

pay

for s

uppl

emen

tal e

duca

tiona

l ser

vice

s for

stud

ents

who

se p

aren

ts re

ques

t suc

h se

rvic

es.T

he 2

0-pe

rcen

t res

erva

tion

may

incl

ude

Titl

e I,

Part

A fu

nds o

r fun

ding

from

oth

erFe

dera

l,St

ate,

loca

l,an

d pr

ivat

e so

urce

s.T

he a

mou

nt sh

own

in th

is co

lum

n is

the

USD

oEd’

s est

imat

e of

the

amou

nt th

at a

ffect

ed L

EA

s – th

ose

with

scho

ols i

dent

ified

for i

mpr

ovem

ent,

corr

ectiv

e ac

tion,

orre

stru

ctur

ing

– m

ay h

ave

to sp

end

to m

eet t

his r

equi

rem

ent.

Act

ual e

xpen

ditu

res w

ill d

epen

d on

such

fact

ors a

s the

num

ber o

f stu

dent

s exe

rcisi

ng e

ither

a c

hoic

e op

tion

or su

pple

men

tal e

duca

tiona

l ser

vice

san

d th

e co

sts o

f sat

isfyi

ng th

ese

requ

ests

.An

LE

A h

as d

iscre

tion

to d

eter

min

e th

e al

loca

tion

of th

ese

fund

s bet

wee

n ch

oice

-rel

ated

tran

spor

tatio

n an

d su

pple

men

tal e

duca

tiona

l ser

vice

s,ex

cept

that

it m

ust

spen

d at

leas

t one

-qua

rter

of t

he 2

0 pe

rcen

t res

erva

tion

– or

an

amou

nt e

qual

to 5

per

cent

of i

ts T

itle

I,Pa

rt A

allo

catio

n –

on e

ach

activ

ity if

ther

e is

dem

and

for b

oth

from

stud

ents

and

thei

r par

ents

.c

An

LE

A th

at m

ust a

rran

ge fo

r sup

plem

enta

l ser

vice

s is r

equi

red

to p

ay,f

or e

ach

child

rece

ivin

g se

rvic

es,t

he le

sser

of t

he a

ctua

l cos

t of t

he se

rvic

es o

r an

amou

nt e

qual

to th

e L

EA

’s T

itle

I,Pa

rt A

allo

catio

nre

ceiv

ed fr

om th

e St

ate

divi

ded

by th

e nu

mbe

r of p

oor s

tude

nts i

n th

e L

EA

,as d

eter

min

ed b

y es

timat

es p

rodu

ced

by th

e U

S B

urea

u of

the

Cen

sus.

Thu

s the

am

ount

show

n in

this

colu

mn

refle

cts t

he

stat

utor

y “c

ap”o

n pe

r-ch

ild e

xpen

ditu

res f

or su

pple

men

tal s

ervi

ces.

dT

his c

olum

n is

deriv

ed b

y di

vidi

ng th

e U

SDoE

d es

timat

ed tr

ansp

orta

tion

and

supp

lem

enta

l ser

vice

s tot

al a

llotm

ent b

y th

e U

SDoE

d es

timat

ed p

er c

hild

max

imum

.

Max

imum

Req

uire

d E

xpen

ditu

res

For

FY 2

002

Cho

ice-

Rel

ated

Tra

nspo

rtat

ion

And

Chi

ldre

n E

ligib

le fo

rPe

rcen

t of

Titl

e I

Allo

catio

naSu

pple

men

tal E

duca

tiona

l Ser

vice

sbSu

pple

men

tal S

ervi

cesc

Tot

al E

nrol

lmen

td

$ 15

4,93

8,44

1$

30,9

87,6

88

34,8

55

4.7%

$ 29

,751

,544

$ 5,

950,

309

4,35

4 3.

2%

$ 17

3,24

6,32

9$

34,6

49,2

66

39,5

83

4.3%

$ 97

,257

,824

$ 19

,451

,565

22

,113

4.

9%

$ 1,

448,

830,

457

$ 28

9,76

6,09

1 29

7,36

7 4.

8%

$ 96

,384

,634

$ 19

,276

,927

20

,128

2.

7%

$ 10

4,12

6,37

4$

20,8

25,2

75

16,4

40

2.9%

$ 27

,673

,803

$ 5,

534,

761

3,62

3 3.

1%

$ 34

,870

,179

$ 6,

974,

036

4,60

6 6.

1%

$ 47

6,51

8,78

9$

95,3

03,7

58

102,

089

4.1%

$ 31

3,33

0,50

4$

62,6

66,1

01

62,2

71

4.2%

$ 33

,671

,534

$ 6,

734,

307

6,17

3 3.

3%

$ 32

,795

,392

$ 6,

559,

078

7,58

8 3.

1%

$ 43

0,67

8,21

9$

86,1

35,6

44

73,0

91

3.5%

$ 15

2,66

9,18

0$

30,5

33,8

36

28,7

80

2.9%

TA

BL

E5.

Est

imat

ed M

axim

um C

ost o

f Fis

cal Y

ear 2

002

Cho

ice-

Rel

ated

Exp

ense

s13

22[


Iow

a

Kan

sas

Ken

tuck

y

Lou

isia

na

Mai

ne

Mar

ylan

d

Mas

sach

uset

ts

Mic

higa

n

Min

neso

ta

Mis

siss

ippi

Mis

sour

i

Mon

tana

Neb

rask

a

Nev

ada

New

Ham

pshi

re

New

Jer

sey

New

Mex

ico

New

Yor

k

Nor

th C

arol

ina

Nor

th D

akot

a

Ohi

o

TA

BL

E5.

(con

tinu

red)

Est

imat

ed M

axim

um C

ost o

f Fis

cal Y

ear 2

002

Cho

ice-

Rel

ated

Exp

ense

s

Max

imum

Req

uire

d E

xpen

ditu

res

For

FY 2

002

Cho

ice-

Rel

ated

Tra

nspo

rtat

ion

And

Chi

ldre

n E

ligib

le fo

rPe

rcen

t of

Titl

e I

Allo

catio

nSu

pple

men

tal E

duca

tiona

l Ser

vice

sSu

pple

men

tal S

ervi

ces

Tot

al E

nrol

lmen

t

$ 62

,955

,668

$ 12

,591

,134

12

,850

2.

6%

$ 73

,138

,888

$ 14

,627

,778

14

,304

3.

0%

$ 15

2,14

5,29

9$

30,4

29,0

60

30,7

39

4.7%

$ 21

3,13

3,49

9$

42,6

26,7

00

42,9

85

5.9%

$ 37

,942

,232

$ 7,

588,

446

5,80

7 2.

8%

$ 15

3,98

3,25

7$

30,7

96,6

51

25,3

78

2.9%

$ 22

0,64

5,90

7$

44,1

29,1

81

34,1

41

3.5%

$ 42

0,79

8,92

2$

84,1

59,7

84

63,6

45

3.7%

$ 11

2,96

4,54

9$

22,5

92,9

10

21,6

18

2.5%

$ 13

0,43

0,96

4$

26,0

86,1

93

24,8

42

5.0%

$ 16

3,74

3,25

8$

32,7

48,6

52

33,3

40

3.7%

$ 34

,294

,108

$ 6,

858,

822

6,47

9 4.

3%

$ 37

,640

,120

$ 7,

528,

024

6,58

9 2.

3%

$ 40

,690

,871

$ 8,

138,

174

9,02

9 2.

5%

$ 26

,874

,231

$ 5,

374,

846

3,62

3 1.

8%

$ 25

7,02

1,82

7$

51,4

04,3

65

41,0

69

3.1%

$ 82

,192

,818

$ 16

,438

,564

18

,769

5.

9%

$ 1,

027,

695,

871

$ 20

5,53

9,17

4 15

3,03

2 5.

3%

$ 21

4,42

2,37

0$

42,8

84,4

74

46,3

93

3.5%

$ 26

,529

,972

$ 5,

305,

994

3,66

7 3.

5%

$ 34

1,10

7,05

7$

68,2

21,4

11

59,9

76

3.3%


23 ]

Okl

ahom

a

Ore

gon

Penn

sylv

ania

Pue

rto

Ric

o

Rho

de I

slan

d

Sout

h C

arol

ina

Sout

h D

akot

a

Ten

ness

ee

Tex

as

Uta

h

Ver

mon

t

Vir

gini

a

Was

hing

ton

Wes

t Vir

gini

a

Wis

cons

in

Wyo

min

g

Gra

nd T

otal

s

TA

BL

E5.

(con

tinu

red)

Est

imat

ed M

axim

um C

ost o

f Fis

cal Y

ear 2

002

Cho

ice-

Rel

ated

Exp

ense

s

Max

imum

Req

uire

d E

xpen

ditu

res

For

FY 2

002

Cho

ice-

Rel

ated

Tra

nspo

rtat

ion

And

Chi

ldre

n E

ligib

le fo

rPe

rcen

t of

Titl

e I

Allo

catio

nSu

pple

men

tal E

duca

tiona

l Ser

vice

sSu

pple

men

tal S

ervi

ces

Tot

al E

nrol

lmen

t

$ 12

2,62

8,59

4$

24,5

25,7

19

28,0

39

4.5%

$ 94

,338

,817

$ 18

,867

,763

16

,527

3.

0%

$ 39

9,59

9,69

3$

79,9

19,9

39

62,7

70

3.4%

$ 33

3,29

4,83

5$

66,6

58,9

67

99,7

87

16.5

%

$ 34

,250

,183

$ 6,

850,

037

5,53

7 3.

5%

$ 14

2,36

3,25

3$

28,4

72,6

51

30,5

08

4.4%

$ 27

,405

,067

$ 5,

481,

013

4,91

9 3.

9%

$ 15

2,47

9,82

0$

30,4

95,9

64

32,1

50

3.5%

$ 86

2,75

6,27

5$

172,

551,

255

176,

700

4.2%

$ 43

,651

,344

$ 8,

730,

269

10,4

28

2.2%

$ 22

,381

,556

$ 4,

476,

311

2,27

5 2.

2%

$ 17

4,34

6,53

7$

34,8

69,3

07

37,8

40

3.3%

$ 14

2,69

8,86

2$

28,5

39,7

72

29,4

70

2.9%

$ 81

,032

,847

$ 16

,206

,569

13

,709

4.

8%

$ 14

9,74

6,33

6$

29,9

49,2

67

24,7

39

2.8%

$ 23

,956

,093

$ 4,

791,

219

2,43

4 2.

8%

$ 10

,244

,025

,000

$ 2,

048,

805,

000

1,95

9,16

64.

1%

24[


According to the text of NCLB, school plans for improve-ment must

“… Incorporate strategies based on scientifically basedresearch that will strengthen the core academic subjectsin the school and address the specific academic issues thatcaused the school to be identified for school improve-ment, and may include a strategy for the implementa-tion of a comprehensive school reform model … (and)adopt policies and practices concerning the school’score academic subjects that have the greatest likelihoodof ensuring that all groups of students … will meet theState's proficient level of achievement …”14

Fortunately, states and LEAs have a number of resourcesavailable that, should they have the political will to usethem, can raise student achievement in underperformingschools by redirecting pre-existing expenditures rather thanadding new costs.

A recent meta-analysis of the 29 most common compre-hensive school reform models15 sought to identify thosemodels proven to raise student achievement. Accordingto the authors,

“(Comprehensive school reform) focuses on reorganizingand revitalizing entire schools, rather than on implement-ing a number of specialized, and potentially uncoordi-nated, school improvement initiatives. In general, thefunding sources supporting the implementation of CSRhave been targeted toward the schools most in need ofreform and improvement: high-poverty schools with lowstudent test scores.”16

Those models identified as having the “strongest evidenceof effectiveness” and possessing “statistically significant andpositive achievement effects based on evidence from studiesusing comparison groups or from third-party comparisondesigns” include Direct Instruction and Success for All.17

Such models “have clearly established, across varyingcontexts and varying study designs, that their effects arerelatively robust and that the models, in general, can beexpected to improve students’ test scores.”18 While mostof such models are designed for the elementary grades,the Direct Instruction model has also been demonstrat-

ed to be highly effective at the high school level with stu-dents who are far behind academically in reading, mathand language skills.

A report published by the American Federation ofTeachers19 estimates the cost of implementing DirectInstruction as a comprehensive school reform program at$150-$200 per pupil, which includes materials and pro-fessional development. According to information postedon its Website, The Comer School Development Programcosts $5,000 for up to 5 schools within a district and$1,000 for each additional school. Leadership training atYale University costs $1,000 per participant and on-siteprofessional development costs are $1,200 per day for eachconsultant hired. The American Federation of Teachersreport estimates total costs for implementing the Successfor All program at “between $261,060 and $646,500 for aschool of about 500 students.”Thousands of schools serving

low socio-economic status populations have been able toredirect funds to implement these models at such costsprior to the passage of NCLB. The fact that many morethousands have chosen not to implement evidence-basedmodels suggests that there are other obstacles, beyond fund-ing, to fundamental reform.

A number of studies detailed later in this report20 drawconclusions about the appropriate level of “educationaladequacy” (i.e., funding) required to get students to per-form at a specified level of proficiency (e.g., under No ChildLeft Behind). In many of them, precise levels of fundingare appended to particular proficiency levels. Is the level ofaggregate funding correlated to student achievement? Severaldecades of research and hard data suggest that it is not.

According to Eric Hanushek, “Real spending per student– that is spending per student adjusted to remove generalinflation – has grown steadily and dramatically. Fromspending of $164/student in 1890, the average for the

Many of the requirements of NCLB (as discussed above) also do requireadditional funding, much of which islikely to spur improved achievement.

United States quintupled roughly every fifty years, reach-ing $4,622/student in 1990” (p. 72).21 As the fundinglevel increases, one would reasonably expect a corre-sponding rise in student achievement, or at least be ableto draw a causal link between levels of increased spend-ing and student achievement. However, this is not thecase. In his review of school-based studies, the authorconcludes that there is “no systematic relationship betweenresources and outcomes once one considers families andother factors that determine achievement.”22

Other researchers provide further commentary on the tenu-ous relationship between resources and student achievement:

“… The last several decades of research suggest that rela-tionships between resources and outcomes are not thatdirect. Researchers report that schools and teachers withthe same resources do different things, with differentresults for students’ learning. If so, resources are notself-enacting, and differences in their effects depend ondifferences in their use. That makes school improvementappear to be a much more complex enterprise, one thatdepends as much on what is done with resources aswith what resources are provided.”23

If, as research suggests, the way in which resources areemployed – including both human and material resources– is more important to the performance of a school thanoverall funding, then educators and policymakers mightwant to consider the research evidence on school improve-ment such as the following:

n Schools with similar demographic characteristics andfunding levels perform quite differently due to differ-ences in educational practices – the “opportunity gap”– and work performed by NCEA (formerly, Just forthe Kids) to identify and disseminate “best practices” fromhigh-performing schools. In the eight states in whichNCEA has implemented at scale, they have found thata majority of schools have a substantial opportunity gapbetween what they are currently achieving and whatthey could be achieving at present resource levels. Forexample, in fifth-grade mathematics in Texas, 2003 dataindicate that 77.9% of schools are performing substan-tially below the level attained by other schools with similar

students and resources.24

n Educators in U.S. schools do not maximize the use ofavailable instructional time, especially for diverse learn-ers25 and comparisons to other nations.26 One authorestimates that “American students are actively engaged inlearning for less than 40% of the time they are in school.”27

n Focusing on developing the professional skills of individ-ual teachers is more effective than class-size reductionstrategies (often cited as one of the most costly schoolimprovement strategies)28 and that some student groupingpatterns are more effective than others.29

n LEAs and schools do not accurately account for profes-sional development spending;30 new strategies by whicheducators can more precisely direct professional devel-opment funds to improve student performance;31 andstudies that suggest professional development that is notspecific to the needs of individual schools is unlikely torealize improvements in student achievement.32

n Teacher quality – and thereby, student achievement – maybe enhanced by financial incentives linked to developingthe professional knowledge of teachers rather than existingmethods of teacher compensation based on years of teach-ing, according to Odden.33

n Careful selection and implementation of evidence-basedinstructional interventions or whole-school reform modelscan raise student achievement.34

Finally, several elements of NCLB are designed to reducestate costs even as they improve results. For example, theReading First program to teach students to read usingresearch-based models (which are not widely implemented)could substantially reduce state and district special educationidentification and attendant costs.


25 ]

26[

b r i e f c o m m e n t a r i e s o n p r e v i o u s s t u d i e s o f n c l b f u n d i n g

The fiscal impact on states, LEAs or schools of complyingwith NCLB is determined by what actions are inspiredby the legislation and the actual costs of those initiatives.There will be variances among schools, and thus states, as aresult of geography, local economics, and student popula-tions. There will also be variances depending on the deci-sions of local educators given the options available.

In our review of existing fiscal impact studies of NCLB,we find a number of significant flaws or limitations that leadto questionable conclusions. Several studies failed to con-sider innovative reforms and approaches to compliance(many discussed earlier), instead relying on unnecessarilycostly assumptions. Other studies that fail to identify a causallink between funding levels, related expenses and increasedstudent achievement, nevertheless, go on to claim that largeincreases in expenditures are essential. Below are brief com-ments on a number of studies.

New Hampshire School AdministratorsAssociation Study

In a study of the impact of NCLB on the state of NewHampshire, for example, the authors concluded that anacross-the-board pay increase for teachers was necessary to

ensure teacher quality.35 The study did not contemplateany other options such as improved recruitment strategies,hiring bonuses, merit rewards, increases in compensation,or other incentives in shortage areas, or targeted profes-sional development. All of these other options are readilyavailable under NCLB. The study contains other erro-neous assumptions as well, including a substantial costfor LEAs to update their technology plans. Analysis byAccountabilityWorks actually indicates that no additionalcosts are required beyond what is already covered by priorfederal requirements and standard management practices.36

Further, the study suggests that NCLB will result in themisclassification of 4,400 students as requiring specialeducation but without any clear justification.

Democratic Congressional Study

A recent Democratic Congressional study37 asserts thatNCLB is under-funded in a number of areas, includingTitle I, IDEA, after-school opportunities, teacher quali-ty grants, and funding for children of military personnel.The report confusingly represents the legislated “author-ization levels” as the expenses needed to fulfill therequirements under NCLB without evidence to supportthe assertion that authorized levels of federal expenditure

are necessary. An important distinction needs to be madebetween authorized funding, which establishes the maxi-mum amount of funding possible under the law, andappropriations, which represents the funding actually pro-vided. While the authorized levels are the maximumamount envisioned by Congress, they do not, by them-selves, provide a justification for what is necessary. It iscommon for Congress to appropriate less than what wasoriginally authorized; our calculations support the actualappropriation levels passed by Congress as adequate.

General Accounting Office (GAO) Study 38

Another flaw common to a number of studies consists ofapplying costs necessary for fulfilling requirements of theold IASA to the new NCLB. For example, IASA requiredthat all states develop “high quality, yearly assessments” inmath and reading (or language arts) to be used at one gradeeach between three through five, six through nine, andten through twelve. Under NCLB, the testing require-ments in elementary and middle school math and readinghave been expanded to yearly testing in grades three througheight. States will thus have to administer six assessmentsat these levels, instead of two. The maximum added costto states, then, is the development, administration, scoringand maintenance of four additional tests.

In an April 2002 study (discussed previously), GAOreported:

“Many states may not be well-positioned to meet therequirements added in 2001. Only 17 states were incompliance with the assessment requirements of the 1994law in March of 2002; therefore, the majority of stateswill still be working on meeting the 1994 requirements asthey begin work toward meeting the new requirements.”39

It is not reasonable, however, to attribute increased costsassociated with the lack of compliance to IASA to NCLB,as this study has been widely interpreted.

Cost Studies of “Educational Adequacy”

In an evaluation of the impact of NCLB on the state ofVermont, the author cites a number of studies – includ-

ing several performed by Augenblick and Myers, Inc. –that use various methodologies to determine the levels offunding required for LEAs to achieve educational ade-quacy: the professional judgment method, the successfulschools (or school district) method, and the statisticalapproach.40 Regardless of the methodology, these studiesuniformly recommend across-the-board increases in perpupil spending in order to comply with the demands ofstandards-based reform that have been in place for nearlya decade. If one accepts the reasonable proposition thatstandards-based reform merely focuses educators’ attentionto the area of their work that matters most to policymakersand the public – high academic achievement of all students

– these cost studies do little to connect funding levels tostudent performance in meaningful ways. And whileresearch does not provide a strong causal link betweenfunding and student performance levels, the studies nev-ertheless assert that present funding levels are insufficient.The use of these studies to suggest that the funding forNCLB is insufficient to ensure success is therefore mis-guided. Due to the number and importance of these studies,what follows is an extended discussion of one example ofsuch an analysis.

A recent study of the cost of educational adequacy forColorado41 employs the same methodology as many ofthe cost studies cited by the author of the evaluation ofthe impact of NCLB on the state of Vermont. Because itcontains a detailed discussion of the assumptions used indetermining proposed funding levels, the Colorado studyis an ideal case study for evaluation of the assertion thatNCLB is an unfunded mandate.

The Colorado study and others – including ones per-formed for Indiana, Montana, Nebraska and New York42

– employ a methodology known as the “professionaljudgment approach” in which panels of state and local

NCLB UNDER A MICROSCOP E > BRIEF COMMENTARIES

27 ]

Another flaw common to a number of studies consists of applying costs necessary for fulfilling requirements of the old IASA to the new NCLB.

28[


educational experts determine the funding levels neces-sary for all students in a given state to receive an adequateeducation. However, studies that utilize this approach donot specify connections with the actual requirements ofNCLB (nor do the studies appear to take into account extantresources and spending for compliance with IASA):

“… It should be noted that while the NCLB was con-sidered by the professional judgment panels, it was notfully considered due to the lack of rules and regulationsin place at the time of this work. (We received the state’srequirements for the NCLB after the work of the pro-fessional judgment panels was completed.)”43

The panels of professional experts in the Colorado studywere unable to specify the linkage between increases inspending and higher levels of student performance.

“We asked panels to be as precise as they could, butprecision should not be over-interpreted; that is, panelmembers found it difficult to precisely link resources toperformance expectations … Finally, we treated eachgroup of students with special needs as if they wereindependent while, in reality, there may be cross-overamong groups that leads to some double counting ofresources (for example, some ELL students might alsobe eligible for free lunch).”44

The panels recommend funding for professional develop-ment activities that is unsupported by “best practice” researchin professional development – such as embedded, classroom-based professional development activities.45

The panels failed to take into account how existing resourcescould be reallocated to professional development and howthe professional development activities are tied to educa-tional research on effective teaching methods and practices;instead, they simply specified a blanket amount of funding:

“After reviewing the work of the other panels, the expertpanel agreed that school districts needed $1,000 perteacher for professional development each year, whichmight be paid as a stipend for time spent before thestart of or after the school year for students and/or topay for fees, materials, or travel associated with devel-opment activities.”46

The technology requirements specified by the panels in theColorado study do not appear to take into account existingfunding for technology staff or resources, nor is there aneffort to link recommendations for substantial technologyinvestment to the improvement of instruction or studentachievement; instead, panel members were provided witha menu from which to create a virtual wish list:

“In order to develop the technology needs, panels weregiven a standard list of equipment … In most cases,the panels wanted to see an extensive array of technol-ogy available in classrooms, in computer labs, in mediacenters, and for teachers and administrative staff.”47

Furthermore, there is little indication as to how the priceswere computers (e.g., why laptops must cost $2,309 orjustification for the distinction given between desktopcomputers estimated at $1,432 vs. “CD/DVD Towers”estimated at $2,355).48

Regarding calculations of per-pupil expenditures, theestimates provided in the Colorado study are purelyhypothetical:

“A note of caution is in order concerning these costs.They represent estimates based on the best judgmentsof many people, reviewed multiple times, and on esti-mated prices, based on statewide average figures. Wepresent them as precise figures reflecting the assumptionsthat were used to calculate them. But it is probably wiserto view them as indicative of an order of magnitude thatmight be slightly low or slightly high and that couldchange more substantially if other people, informed byexperience, research, and expertise, thought the objec-tives identified to the panels could be met even if somecomponents were modified or eliminated.”49

Regarding estimates of assessment costs, per pupil expen-ditures varied widely based on school size and grades servedwithout consideration to the economies of scale that largerschool systems might provide in mitigating assessmentcosts; rather, converse estimates are used in which mid-range school systems actually incur higher costs per pupilthan small ones.50

In addition to the “professional judgment approach,” theColorado study also makes use of the “successful schooldistrict approach”, a methodology that consists of deter-mining the funding levels for high-performing schooldistricts and tying the funding levels to other districtswith similar demographic characteristics.

No indication is provided as to how funding is currentlyspent in the “successful” school districts, (e.g., innovativeprograms, creative staffing arrangements or professionaldevelopment initiatives). Determinations for how fundsshould be spent avoid innovation and do not appear todiffer from approaches of the last several decades:

“The average basic expenditure number does not tell usanything about how the districts spend their money. Itonly tells us, on average, the amount of money districtsneed to provide programs and services for studentswith no special needs, in average districts, to be suc-cessful. Our belief is that districts can use this amountof money in the way they feel best meets the needs oftheir student population.”51

The two methods used (i.e., the professional judgmentand successful school district approaches) produced esti-mates that differed significantly. No indication or discus-sion is provided – backed by verifiable data – as to whichapproach is more appropriate for given states and LEAs.The studies do not adequately account for the fairly largedifference in the cost estimates resulting from the twoapproaches – on the order of 46% higher for the profes-sional judgment approach in Colorado and 25-27%higher for the professional judgment approach in Kansasand Maryland.52

In the Colorado study, the authors further note the following:

“The fact is that the base cost figure derived from theprofessional judgment approach is a goal in much thesame way that the expectation that all students willmeet state standards in 2013-14 is a goal. That is, thestate performance benchmark for 2002 represents par-tial achievement of an ultimate objective – the averageperformance benchmark for 2002 is about 70.4 percentof the expectation for 2013-14. Therefore, if $6,815

represents the full cost of meeting all standards, it seemsreasonable to assume that 70.4 percent of that amount,$4,798, would be the amount needed to reach the 2002performance benchmark. This amount compares favor-ably to the amount of basic revenue available in the 27successful districts in 2000-01 ($4,654 in 2000-01,which would be $4,794 if inflated by three percent to2001-02). Therefore, we suggest setting the base levelat $4,798 in 2001-02, increasing it by inflation everyyear until 2013-14 and raising it to a higher level inthose years in which the state expects performance toincrease (using the same proportional approach used tocalculate the 70.4 percent in 2002) until it reaches 100percent of $6,815 in 2013-14 adjusted by inflationevery year between 2001-02 and 2013-14.”53

The study thus calculates that per-pupil costs should riseas the percentage of students supposed to be at proficiencyincreases under NCLB without consideration of improve-ments in efficiency or increases in school and districtcapacity and competence. The study assumes that givenlevels of student performance are explicitly and directly tiedto corresponding levels of funding. There is little basis inresearch for this assumption.


29 ]

The studies do not adequately account for the fairly large difference in the cost estimates resulting from the two approaches

30[

e n d n o t e s

1 U.S. Department of Education, Office of Public Affairs, IASA 1994 Summary Sheets (Washington, D.C.: U.S. Government Printing Office, 1994).2 U.S. Department of Education, U.S. Department of Education Budget History Table,

available from http://www.ed.gov/about/overview/budget/history/index.html?exp=0; Internet; accessed 26 September 2003.3 United States General Accounting Office, Title I: Characteristics of Tests Will Influence Expenses; Information Sharing May Help States Realize

Efficiencies, available from http://www.gao.gov/highlights/d03389high.pdf; Internet, accessed 26 September 2003.4 AccountabilityWorks, Estimated Cost of the Testing Requirements in the No Child Left Behind Act,

available from http://www.schoolreport.com/AWNCLBTestingCostsStudy.pdf; Internet, accessed 26 September 2003.5 Meeting the Highly Qualified Teachers Challenge: The Secretary's Second Annual Report on Teacher Quality,

<http://www.ed.gov/about/reports/annual/teachprep/2003title-ii-report.pdf>6 Bess Keller, States Claim Teachers are ‘Qualified’. Education Week. 29 October 2003.7 National Education Association, Rankings & Estimates: Rankings of the States 2002 and Estimates of School Statistics 2003, available from

http://www.nea.org/edstats/images/03rankings.pdf; Internet, accessed 26 September 2003 (p. 65).8 National Center for Education Statistics, Digest of Education Statistics, 2002, available from http://nces.ed.gov/pubs2003/2003060b.pdf;

Internet, accessed 4 September 2003 (Chapter 2, p. 80).9 U.S. Department of Education, Common Core of Data 2001-02, available from http://nces.ed.gov/ccd/; Internet, accesses 26 September 2003.10 U.S. Bureau of Labor Statistics, 2001 National Occupational Employment and Wage Estimates.11 U.S. Department of Education, No Child Left Behind: A Parent’s Guide,

available from http://www.ed.gov/parents/academic/involve/nclbguide/parentsguide.pdf; Internet, accessed 26 September 2003.12 §1003 and §1116 of NCLB13 U.S. Department of Education, Fiscal Year 2002 Title I allotments by LEA (Washington, D.C.: U.S. Government Printing Office, 2002).14 §1116 of NCLB15 Geoffrey Borman, et al., Comprehensive School Reform and Student Achievement: A Meta-Analysis,

available from http://www.csos.jhu.edu/CRESPAR/techReports/Report59.pdf; Internet, accessed 3 September 2003.16 Geoffrey Borman, et al., p. 2. Comprehensive School Reform and Student Achievement: A Meta-Analysis,

available from http://www.csos.jhu.edu/CRESPAR/techReports/Report59.pdf; Internet, accessed 3 September 2003.17 Geoffrey Borman, et al., p. 29. Comprehensive School Reform and Student Achievement: A Meta-Analysis,

available from http://www.csos.jhu.edu/CRESPAR/techReports/Report59.pdf; Internet, accessed 3 September 2003.18 Geoffrey Borman, et al., p.37. Comprehensive School Reform and Student Achievement: A Meta-Analysis,

available from http://www.csos.jhu.edu/CRESPAR/techReports/Report59.pdf; Internet, accessed 3 September 2003.19 American Federation of Teachers, Building on the Best, Learning From What Works: Seven Promising Reading and English Language Arts Programs,

available from http://www.aft.org/edissues/downloads/seven.pdf; Internet, accessed 6 July 2003, 15.

20 William J. Mathis, No Child Left Behind: Costs and Benefits, available from http://www.pdkintl.org/kappan/k0305mat.htm#4a;Internet, accessed 3 September 2003.

21 Eric A. Hanushek, p.72. Spending on Schools, available from http://www-hoover.stanford.edu/publications/books/fulltext/primer/4.pdf;Internet, accessed 26 September 2003.

22 Eric A. Hanushek, p.81. Spending on Schools, available from http://www-hoover.stanford.edu/publications/books/fulltext/primer/4.pdf;Internet, accessed 26 September 2003.

23 David K. Cohen, et al., p. 5. Resources, Instruction, and Research: A CTP Working Paper,available from http://depts.washington.edu/ctpmail/PDFs/Resources-CRB-12-2000.pdf; Internet, accessed 26 September 2003.

24 Information provided by NCEA to AccountabilityWorks.25 National Education Commission on Time and Learning, Prisoners of Time,

available from http://www.ed.gov/pubs/PrisonersOfTime/index.html; Internet, accessed 26 September 2003.26 William Schmidt, et al., Facing the Consequences: Using TIMSS for a Closer Look at U.S. Mathematics and Science Education

(Dordrecht: Kluwer, 1999), 63-80.27 Gene V. Glass, Time for School, Its Duration and Allocation. ed. Alex Molnar, “School Reform Proposals: The Research Evidence”

(Greenwich, CT: Information Age Publishing, 2002), 80.28 Ludger Woessmann and Martin West, Class-Size Effects in School Systems Around the World: Evidence from Between-Grade Variation in TIMSS,

available from http://ideas.repec.org/p/wop/kieliw/1099.html; Internet, accessed 26 September 2003.29 James A. Kulik, An Analysis of the Research on Ability Grouping: Historical and Contemporary Perspectives (Storrs, CT: The National Research

Center on the Gifted and Talented, 1992).30 Karen Hawley Miles, et. al., “Understanding and Comparing District Investment in Professional Development: Methods and Lessons from

Four Districts (Working Paper SF-02-02),” Submitted to Educational Evaluation and Policy Analysis.31 Richard Elmore and Deana Burney, Investing in Teacher Learning: Staff Development and Instructional Improvement in

Community District #2, New York City. (New York: National Commission on Teaching and America’s Future and the Consortium for Policy Research in Education, 1997).

32 David K. Cohen and Heather H. Hill, Learning Policy: When State Education Reform Works. (New Haven, CT: Yale University Press, 2001).33 Allan Odden, et al., Enhancing Teacher Quality Through Knowledge-and Skills-Based Pay,

available from http://www.cpre.org/Publications/rb34.pdf; Internet, accessed 26 September 2003.34 Geoffrey Borman, et al., Comprehensive School Reform and Student Achievement: A Meta-Analysis,

available from http://www.csos.jhu.edu/CRESPAR/techReports/Report59.pdf; Internet, accessed 3 September 2003.35 New Hampshire School Administrator’s Association, Analysis of Cost Impact of ESEA – No Child Left Behind Act on New Hampshire,

available from http://www.nhsaa.org/pdffiles/ESEAcostimpactan.pdf; Internet, accessed 26 September 2003.36 Meave O’Marah and Theodor Rebarber, pp. 7-8. Financial Impact of the No Child Left Behind Act On the State of New Hampshire &

Review of the Cost Analysis of NHSAA (Concord, NH: Bartlett Center for Public Policy).37 Office of Democratic Leader Pelosi and Democratic Staff of the House Appropriations Committee,

GOP Funding Bill Shortchanges America’s Children by Underfunding Key Education Priorities,available from http://www.house.gov/appropriations_democrats/ChildrenShortchange.pdf; Internet, accessed 26 September 2003.

38 United States General Accounting Office, Title I: Characteristics of Tests Will Influence Expenses; Information Sharing May Help States RealizeEfficiencies, available from http://www.gao.gov/highlights/d03389high.pdf; Internet, accessed 26 September 2003.

39 United States General Accounting Office, pp. 18-19. Title I: Characteristics of Tests Will Influence Expenses; Information Sharing May HelpStates Realize Efficiencies, available from http://www.gao.gov/highlights/d03389high.pdf; Internet, accessed 26 September 2003.

40 William J. Mathis, No Child Left Behind: Costs and Benefits, available from http://www.pdkintl.org/kappan/k0305mat.htm#4a;Internet, accessed 3 September 2003.

41 Augenblick & Myers, Inc., Calculation of the Cost of an Adequate Education in Colorado Using the Professional Judgment and the Successful School District Approaches (Denver: Augenblick & Myers, Inc.).

42 Augenblick & Myers, Inc., Calculation of the Cost of an Adequate Education in Indiana in 2001-2002 Using the Professional Judgment Approach, (Denver: Augenblick & Myers, Inc., 2003); Calculation of the Cost of an Adequate Education in Montana in 2001-2002 Using the Professional Judgment Approach, (Denver: Augenblick & Myers, Inc., 2002.); Calculation of the Cost of an Adequate Education in Nebraska in 2002-03 Using the Professional Judgment Approach, (Denver: Augenblick & Myers, Inc., 2003);William Duncombe and Anna Lukemeyer, Estimating the Cost of Educational Adequacy: A Comparison of Approaches, available from http://www-cpr.maxwell.syr.edu/faculty/duncombe/special%20report/costofeducadeq.pdf; Internet, accessed 26 September 2003.

43 Augenblick & Myers, Inc., pp. III-2, III-3. Calculation of the Cost of an Adequate Education in Colorado Using the Professional Judgment and the Successful School District Approaches (Denver: Augenblick & Myers, Inc.).

44 Augenblick & Myers, Inc., p. IV-6. Calculation of the Cost of an Adequate Education in Colorado Using the Professional Judgment and theSuccessful School District Approaches (Denver: Augenblick & Myers, Inc.).

NCLB UNDER A MICROSCOP E > ENDNOTES

31 ]

32[

NCLB UNDER A MICROSCOP E > ENDNOTES

45 Thomas Corcoran and Margaret Goertz, “Instructional capacity and high performance schools,” Educational Researcher, 24 (9) (1995):27-31; Richard Elmore and Deana Burney, Investing in Teacher Learning: Staff Development and Instructional Improvement in CommunityDistrict #2, New York City. (New York: National Commission on Teaching and America’s Future and the Consortium for Policy Research in Education, 1997); Valerie E. Lee and Julia B. Smith, “Collective responsibility for learning and its effects on gains in achievement forearly secondary school students,” American Journal of Education, 104(2) (1996): 103-147; Thomas Corcoran, Helping teachers teach well:Transforming professional development. (Philadelphia, PA: Consortium for Policy Research in Education, 1995).


47 Augenblick & Myers, Inc., pp. IV-7. Calculation of the Cost of an Adequate Education in Colorado Using the Professional Judgment and theSuccessful School District Approaches (Denver: Augenblick & Myers, Inc.).



50 Augenblick & Myers, Inc., Tables IV-4A, IV-4B, IV-4C. Calculation of the Cost of an Adequate Education in Colorado Using the ProfessionalJudgment and the Successful School District Approaches (Denver: Augenblick & Myers, Inc.).

51 Augenblick & Myers, Inc., p. V-3. Calculation of the Cost of an Adequate Education in Colorado Using the Professional Judgment and theSuccessful School District Approaches (Denver: Augenblick & Myers, Inc.).

52 Augenblick & Myers, Inc., p. VI-1. Calculation of the Cost of an Adequate Education in Colorado Using the Professional Judgment and theSuccessful School District Approaches (Denver: Augenblick & Myers, Inc.).

53 Augenblick & Myers, Inc., p. VI-2. Calculation of the Cost of an Adequate Education in Colorado Using the Professional Judgment and theSuccessful School District Approaches (Denver: Augenblick & Myers,

nclb under a microscope - accountabilityworks.org · aw believes that the next few years represent...

Documents