Money for Nothing...2018/02/08 · Money for nothing? An analysis of the Oregon Quality Education...

Money fornothing?

An analysis of the OregonQuality Education Model

by Richard Vedder, Ph.D.

CASCADEPOLICYINSTITUTE

August 2000

www.cascadepolicy.org

��

Richard Vedder, an adjunct scholar to Cascade Policy Institute, is a distinguished profes-sor of economics at Ohio University and a consultant to the Joint Economic Committeeof Congress. An economic historian and labor economist, he has authored numerousbooks and monographs, as well as over 200 scholarly articles. He is the author of the 1993Cascade Policy Institute report, “The Economic Impact of an Oregon Sales Tax,” and hismost recent book is Can Teachers Own Their Own Schools? New Strategies for EducationalExcellence.

Professor Vedder received his Ph.D. in economics from the University of Illinois, and hasbeen a visiting professor at numerous colleges and universities, most recently serving asthe John M. Olin Visiting Professor of Labor Economics and Public Policy at the Centerfor the Study of American Business, Washington University in St. Louis.

Professor Vedder is a former elected member of the Athens (Ohio) City School Board, hiswife is a public high school guidance counselor, and his son and daughter teach in publicschools. For the past decade he has extensively analyzed the results from proficiency testsadministered in his state, in particular analyzing the relationship between resource use,socioeconomic factors, and student learning.

��

For the research of this report I benefited from the assistance of several individuals, nota-bly Joshua Hall, staff economist at the Joint Economic Committee of the Congress of theUnited States.

��

Founded in 1991, Cascade Policy Institute is Oregon’s premier policy research center.Cascade’s mission is to explore and promote public policy alternatives that foster indi-vidual liberty, personal responsibility and economic opportunity. To that end the Insti-tute publishes policy studies, provides public speakers, organizes community forums andsponsors educational programs. Focusing on state and local issues, Cascade offers prac-tical, innovative solutions for policy makers, the media and concerned citizens.

Cascade Policy Institute is a tax-exempt educational organization as defined under IRScode 501(c)(3). Cascade neither solicits nor accepts government funding, and is sup-ported by individual, foundation, and corporate contributions. Nothing appearing inthis document is to be construed as necessarily representing the views of Cascade, or asan attempt to aid or hinder the passage of any bill before any legislative body.

Copyright © 2000 by Cascade Policy Institute. All rights reserved.Photos used in this publication © 1999-2000 www.arttoday.com

ii

��

About the Author .................................................................................................................... ii

Acknowledgments ................................................................................................................... ii

About Cascade ........................................................................................................................ ii

Preface ..................................................................................................................................... v

Introduction............................................................................................................................ 1

QEM fact sheet ........................................................................................................................ 2

The Quality Education Model and contemporary educational research ............................. 3

Education reform and testing: Oregon and the nation......................................................... 4

Oregon public education spending in historical and geographic perspective ..................... 5

The Oregon QEM: Ignoring other approaches to reform .................................................... 8

The determinants of student achievement: What the literature tells us............................. 10

School expenditures and student achievement ................................................................... 11

Non-resource determinants of student achievement ......................................................... 16

Structural changes in education: What the QEM study ignored........................................ 17

Financial implications of the QEM and the impact on economic growth ........................ 19

Conclusions ........................................................................................................................... 20

Notes ...................................................................................................................................... 20

References .............................................................................................................................. 25

iii

Preface

Oregonians feel stronglyabout our public school system,but some things many believe

about it may not be true:“Things are different here,” “our

schools produce the beststudents in the nation,” and

“schools have been starved forfunds since the passage of

Measure 5 in 1990.”

Three reports releasedthis summer should cause

Oregonians to questions theseassumptions, and look for

answers outside the standardone-size-fits-all Kindergarten

through 12th grade schoolsystem.

First, an Oregonianinvestigative report found thatour state’s public school dropout rate is the highest in theNorthwest, and one of the

worst in the nation.

Second, the RandCorporation found that, afteradjusting for socioeconomic

factors favorable to Oregon, ourstudents perform nowhere nearthe best in the country, scoring

only near the average.

And third, this CascadePolicy Institute analysis of the

Oregon Quality EducationModel (QEM) finds that Oregonhas already put more moneyinto its public schools thanmany other states, yet ourresults are far from stellar.

Even worse, the newest modelfor improvement, the QEM,

relies on very costlyproscriptions that fail to pass

educational research and cost-benefit tests.

Richard Vedder,Distinguished Professor of

Economics at Ohio University,tried to find research-based

justifications for the elementsmaking up the Oregon QualityEducation Model. His report

deals extensively with one ofthe primary assumptions of the

QEM: that increased fundingwill positively affect student

achievement. Dr. Vedder showsthat this assumption is notsupported by the research.

Moreover, he demonstrates thatspending money in the wayscalled for by the QEM may

actually result in a decline instudent achievement.

Dr. Vedder looks atsignificant research ignored by

the Model and offers somevaluable insights as to what

does have a positive impact onlearning. Vedder identifies such

important factors as family

��

Oregonians feel strongly about our publicschool system, but some things many believeabout it may not be true: “Things are differ-ent here,” “our schools produce the best stu-dents in the nation,” and “schools have beenstarved for funds since the passage of Mea-sure 5 in 1990.”

Three reports released this summer shouldcause Oregonians to questions these assump-tions, and look for answers outside the stan-dard one-size-fits-all Kindergarten through12th grade school system.

First, an Oregonian investigative report foundthat our state’s public school drop out rate isthe highest in the Northwest, and one of theworst in the nation.

Second, the Rand Corporation found that,after adjusting for socioeconomic factors fa-vorable to Oregon, our students performnowhere near the best in the country, scor-ing only near the average.

And third, this Cascade Policy Institute analy-sis of the Oregon Quality Education Model(QEM) finds that Oregon has already in-vested more money in its public schools thanmany other states, yet our results are far fromstellar. Even worse, the newest model for im-provement, the QEM, relies on very costlyproscriptions that fail to pass educationalresearch and cost-benefit tests.

Richard Vedder, Distinguished Professor ofEconomics at Ohio University, tried to findresearch-based justifications for the elementsmaking up the Oregon QEM. His report dealsextensively with one of the primary assump-tions of the QEM: that increased funding willpositively affect student achievement. Dr.Vedder shows that this assumption is not

supported by the research. Moreover, he dem-onstrates that spending money in the wayscalled for by the QEM may actually be detri-mental to learning.

Dr. Vedder looks at significant research ig-nored by the Model and offers some valu-able insights as to what does have a positiveimpact on learning. Vedder identifies suchimportant factors as family environment andnotes that in general, more learning occursin schools where students, teachers and ad-ministrators are empowered to make theirown educational decisions.

Vedder also suggests fruitful avenues thatmight lead to better results—avenues that arebeing explored now in most states but thatthe QEM doesn’t even acknowledge. He dis-cusses market-based approaches to reform,which greatly increase the role of studentsand their families in allocating educationalresources. Among other findings, Vedder re-ports that the early research on vouchers islargely quite supportive, and that the highparental demand for charter schools is a mea-sure of their success.

Dr. Vedder’s report should be read by all Or-egonians interested in better understandingthe complex relationship between schoolfunding and educational results. Parents andteachers should be especially interested in thediscussion of alternative ways to provide bet-ter education at less cost.

While the education policy debate is far fromover, we believe this report will help moveOregon toward more effective ways of im-proving our education system.

Steve Buckstein, PresidentCascade Policy Institute

v

environment and notes that ingeneral, more learning occurs

in schools where students,teachers and administratorsare empowered to make theirown educational decisions.

Vedder also suggestsfruitful avenues that might leadto better results—avenues thatare being explored now in moststates but that the QEM doesn’t

even acknowledge. Hediscusses market-based

approaches to reform, whichgreatly increase the role of

students and their families inallocating educational

resources. Among otherfindings, Vedder reports that

the early research on vouchersis largely quite supportive, andthat the high parental demand

for charter schools is ameasure of their success.

Dr. Vedder’s report shouldbe read by all Oregonians

interested in betterunderstanding the complexrelationship between school

funding and educationalresults. Parents and teachersshould be especially interestedin the discussion of alternative

ways to provide bettereducation at less cost.

While the educationpolicy debate is far from over,

we believe this report will helpmove Oregon toward more

effective ways of improving oureducation system.

Steve Buckstein, President

Cascade Policy Institute

vi

cording to the chair of the Legislative Coun-cil, Mr. Lundquist, the QEM “is foremost aremarkable tool—unique in the nation—thatwill enable the lawmakers of this state andpotentially others to make reliable decisionsabout the funding of education. And for thefirst time, that funding can be linked to per-formance.”3

In this study, I perform an independent as-sessment of the QEM and Oregon’s effortsin public education. It is clear from the state-ments made by members of the LegislativeCommission and other state education offi-cials, that many leaders sincerely believe thatby following the guidelines outlined in theQEM, Oregon will improve student perfor-mance and the quality of life of Oregonians.The intentions of those responsible for theQEM are honorable and arguably even noble.Yet my own assessment is that an objectiveevaluation of the research evidence suggeststhat the QEM is highly flawed, and thatblindly following it will prove extremelycostly to Oregon taxpayers, without com-mensurate benefits. The QEM reports nooriginal research (e.g., relating studentachievement of Oregon children to such pro-posed mandates as greater instructional timeor smaller class size), is biased in its refer-ence to secondary research sources, is exces-sively reliant on the expertise of a singleeducator, and is narrowly limited in the op-tions it considers.

I further suggest that the report is flawed con-ceptually from the beginning by explicitlyminimizing cost considerations, one of thebasic ingredients in the cost-benefit analysisthat all good public policy analysis shouldincorporate.4 Examining research by re-spected scholars who take both costs andbenefits into account, I conclude that manyof the assertions of the QEM are highly ques-tionable from the standpoint of good public

��

The United States began a path towards re-forming its system of public educationshortly after the release in 1983 of A Nationat Risk.1 In the next decade, most states en-acted education reform legislation designedto improve student performance in theschools. In Oregon, the cornerstone piece oflegislation was the Oregon “Education Act forthe Twenty-First Century,” passed in 1991,which put in motion the standards and as-sessment system defining the Certificate ofInitial Mastery that was the primary account-ability tool of the school reform system. In1997, the then Speaker of the Oregon Houseof Representatives, Lynn Lundquist, formeda Legislative Council to develop the OregonQuality Education Model (hereafter, QEM).In the spring of 1999, the Model was releasedto the public in a report, that with appendi-ces, ran over 160 pages.2

The QEM attempts to create a goal and vi-sion of an effective, accountable school inwhich 90% of the students will meet state-wide achievement standards as measured bythe Certificate of Initial Mastery. Because theModel’s designers believe that students them-selves have too diverse circumstances to per-mit useful analysis, they created what they call“prototype” elementary, middle and highschools. The QEM unit of analysis is thus theschool building, not the individual student.

The QEM was largely developed under thedirection of an associate professor of educa-tion at the University of Oregon, DavidConley, although it was approved by thecouncil of 22 individuals, consisting mostlyof business leaders and current or retirededucators. In the QEM report, it is arguedthat by adopting the Commission’s recom-mendations Oregon public schools wouldproduce improved student performance. Ac-

The QEM reportsno original re-search, is biased inits reference to sec-ondary researchsources, is exces-sively reliant onthe expertise of asingle educator,and is narrowlylimited in the op-tions it considers.

1

��

The following fact sheet was distributed by the Legislative Council on the Oregon QualityEducation Model, on March 3, 1999.

��

� ��

�� !�� "��#��$��

��%��&

� �� '��#��'��

�� &

� �� ()�#��

�*�� &

� �� "��#��#�' ��

� �' "�� &

� ��'��#�� +��,��!��-��.��'��

/ ��#��0 ��#�� 1��#�� 2

��'�� #�� '�� %� �� #� �� '�� ##��

�� &

� ��3��'�� '�� 4��

��'�� '��'��#� �%� �� &

� ��'�� '��#��'�� #�'��. � � ��*��

5�6��.�� '��#�� #�� 7��#�'

�� &

� �� &�� '��

��6��8��#��'�� "&

� �� #�� '�� #��

��'��' ��'�&

2

policy and maximizing the use of resources.

The Model is seen by its authors as a way torelate costs to performance and not as a setof fixed proposals. Yet the initial QEM reportconcludes that “Costs for Full Implementa-tion of The Oregon Quality Education Modelas outlined in the Prototype Schools wouldbe $5.65 billion.”5 This is more than $1 bil-lion greater than Governor Kitzhaber’s ini-tial proposed 1999-2001 K-12 educationbudget of $4.5 billion. While actual approvedspending was $4.81 billion (an increase ofover 10 percent from the previous biennium),adopting any of the options being studiedwith the QEM would almost certainly in-crease spending substantially.

As indicated below, the assertion that theModel would lead to 90 percent of studentspassing the CIM test is not supported bymainline academic research.

��

��

��

The QEM purports to reach its conclusionsabout the educational needs of Oregoniansafter reviewing the corpus of research per-taining to the delivery of public educationalservices. In introducing the bibliography, theauthor of the QEM study notes that “the bib-liography references provide justification forboth tangible and intangible dimensions ofthe Model...”6 This appeal to scholarly author-ity, indeed, is appropriate, and is the hallmarkof scholarly work—reaching conclusions af-ter evaluating evidence that seeks the truth.Yet any study is only as good as theresearcher(s) doing the work, and the qual-ity of the research cited.

Major studies on which significant policyrecommendations are made are usually based

on the work of multiple scholars serving in aconsulting role. The use of several scholarsincreases the intellectual quality of the results,provides greater expertise on many topicswithin the general area of inquiry, and pro-vides a diversity of viewpoints. Thus the gov-ernment of Russia, when it recently wantedsome consultation on its plan for economicreform, invited nine scholars from five na-tions to assist it.7 When the U.S. governmentwanted a national evaluation of U.S. educa-tion in the early 1980s, it created a commis-sion with a staff including several scholarsand researchers, and then commissionedsome 40 scholarly studies from outside aca-demic experts.8

The Legislative Commission formed byformer speaker Lundquist, however, hired asingle outside consultant, David Conley ofthe University of Oregon, to provide someresearch direction to the project. Thus thegroup lacked the breadth of knowledge andthe diversity of views that are hallmarks ofmost major reassessments of public policy.

Moreover, the reference list accompanyingthe QEM report ignored a number of distin-guished scholars and writers on educationalreform. Based on my familiarity with the lit-erature of education reform, I selected thir-teen leading scholars whose work on variousaspects of education have commanded atten-tion in recent years. These included not onlyprofessors of education, but scholars in thefields of economics, English, history, politi-cal science, psychology, and sociology. Amajority of them of them are associated withdistinguished research institutions, such asHarvard University, Stanford University, theUniversity of Michigan, the University ofChicago, or the Brookings Institution. Onlyone of the thirteen names on this list is in-cluded among the nearly 200 people and or-ganizations referenced in the QEM report

The assertion thatthe Model wouldlead to 90 percentof students passingthe CIM test is notsupported bymainline academicresearch.

3

��

� ��!�"��!��##$%�##&

��

��

�� !��"

bibliography, and that individual was notcited for his major scholarly works. Theseomissions show that major themes coveredin much modern educational research wereignored, and that their findings were not con-veyed to the largely lay members of the Leg-islative Commission. This also suggests thatthe Oregon QEM might be based on a skewedor biased reading of modern scholarship.

��'

(��

Is the Oregon Quality Education Model“unique” as indicated by former speakerLundquist? After reading the Oregon QEMdocument and examining sample tests for theCertificate of Initial Mastery that is integralto the Oregon reform effort, I would concludethat the Oregon approach is not particularlyunique or different. A majority of Americanstates have instituted statewide testing of K-12 students to some degree. Many of themtie scholarly advancement in part to test re-sults, and some provide financial incentivesor sanctions for schools whose students showrelative high or low achievement levels withrespect to the tests.

To cite one example quite familiar to the au-thor, in Ohio statewide testing at several lev-els and in several subjects began in 1990, evenbefore the passage of the Oregon legislationmandating such testing. Students in Ohiowho fail any subject (mathematics, reading,writing, citizenship or sci-ence) on a ninth gradeproficiency test may notreceive a high school di-ploma (students with dis-abilities excepted).Promotion from thefourth to fifth grade iscontingent on passage ofproficiency tests adminis-

tered in fourth grade. Tests scores are an im-portant component in an assessment proce-dure that puts poor performing districts intoa category that can lead to sanctions up to thestate taking over management of the district.Districts showing good improvement on pro-ficiency test scores are given incentive grants.The notion of establishing standards and en-forcing accountability is thus not unique toOregon. Indeed, sanctions for poorly perform-ing schools are more explicit and substantialin many other states.

A related issue is whether the national moveto higher standards and accountability hasled to better academic performance. In ad-dition to greater testing and accountabilitystandards, the reform movement has involveddifferent forms of educational change in thevarious states. A few examples of popularchanges include: reducing class size; adopt-ing uniform curricular standards; adoptingnew approaches to instruction such as phon-ics, Core Knowledge, Direct Instruction,block scheduling, and so forth; easing teachercertification requirements; allowing greaterpublic school choice; and the opening ofcharter schools.

The evidence of positive results from the re-form movement in general is somewhat dis-appointing. While there are some examplesof clear improvement in student perfor-mance, the overall evidence is not unambigu-ously positive. Table 1 shows the average test

Major themes cov-ered in muchmodern educa-tional researchwere ignored in theQEM, and theirfindings were notconveyed to thelargely lay mem-bers of theLegislative Com-mission.

4

0

1000

2000

3000

4000

5000

6000

7000

8000

1966-67 1986-87 1997-98

Figure 1Growth in real Oregon spending per pupil (in 2000 $)

scores in reading at the fourth,eighth and twelfth grade levelson the National Assessment forEducational Progress adminis-tered in most states (but notOregon) over the period 1992to 1998 when the testing/stan-dards/accountability move-ment was in full swing.10

Aggregate average test scoresshowed no change at the fourthgrade level, some improvementfor eighth-graders, but a slightdecline at the twelfth gradelevel. To generalize, I concludethat the results were mixed,showing no clear overall trend.

The national historical experience to datesuggests that proponents of reform shouldbe cautious in claiming that their proposedchanges will bring about significant improve-ments in educational performance. This isnot meant to suggest that reform is “bad,” butrather that suggestions that a given plan ormodel will guarantee large gains in academicachievement should be received with greatskepticism.

(��

��

��

Proponents of the Oregon Quality EducationModel suggest that Oregon needs to investconsiderably more resources into public edu-cation in order to improve student perfor-mance, as measured by such instruments asthe Certificate of Initial Mastery. This criti-cal premise—that greater resource commit-ment will likely lead to noticeableimprovements in student performance—needs to be examined. First, it is worth ex-amining Oregon’s financial commitment to

its schools over time. Doing so, the follow-ing facts emerge:

• Oregon’s spending per pupil has more thandoubled in real terms in the lifetime ofmost adult Oregonians;

• The growth in spending adjusted for in-flation has continued to grow significantlyin the past decade;

• Oregon’s per pupil public educationspending exceeds the national average;

• The taxpayer burden of educating a pub-lic school child is well above the nationalaverage, in part because Oregon’s incomeper capita is below the U.S. average;

• The most significant cost item is person-nel; teacher salaries in Oregon are abovethe national average, even though Oregonworkers on average earn less than workersnationwide.

Elaborating, Figure 1 looks at current expen-ditures per K-12 student in Oregon for three

Oregon’s spendingper pupil has morethan doubled inreal terms in thelifetime of mostadult Oregonians.

5

��

��

��

��$

��)��!�(��

�*�+��,��!��#--%#&.

��#�� $�� %��

��

�$%&� ��%"�"�'�(�� %"�"

'�(��

�� !�� "#��"�� !�� !$

dates over the past one-third century, correct-ing for inflation using the Consumer PriceIndex for All-Urban Consumer (CPI-U). Theresults indicate that in real resource terms,spending per pupil more than doubled fromthe 1966-67 school year to the 1997-98 year.For technical reasons discussed below, thatis probably a very significant understatementof per pupil spending growth.

The reader might observe that the spendinggrowth in more recent years (e.g., the 11 yearsfrom the 1986-87 to the 1997-98 school years)has slowed down (arguably in part due to thepassage of a citizen’s initiative in 1990, Mea-sure 5, which placed a limit on local prop-erty tax rates for education). Using the officialCPI-U, the growth in per pupil spending isonly slightly more than 9.4 percent over thatperiod, reflecting a reduction (but not anending) of the rate of expenditure growthallowing for inflation.

However, it is widely acknowledged by ex-perts that the official CPI-U overstates therate of inflation. A congressional commissionof prominent economists led by MichaelBoskin concluded that the overstatement wasprobably about 1.1 percentage pointsa year.11 The Bureau of Labor Statis-tics has acknowledged some of theproblems, and has constructed anexperimental index that it reports inaddition to the official index. Usingan index that reduces reported an-nual inflation by 1.1 percentagepoints as the Boskin commissionsuggested, the growth in real spend-ing per pupil in Oregon from 1986-87 to 1997-98 rises to 22.1 percent;with the BLS experimental index, thereal growth is 13.5 percent. It is prob-ably safe to say that spending perpupil over that 11 year period rose13-22 percent in real terms.

Per pupil spending in Oregon has been con-sistently greater than the national average (seeTable 2). I selected eight dates at random overthe past one-third century, and in every singleyear current expenditures per pupil weregreater in Oregon than in the United Statesas a whole.

What is particularly remarkable is that percapita personal income in Oregon over thesame time has been below the national aver-age.12 This means the taxpayer burden in Or-egon, relative to the nation, is even greaterthan what the per pupil spending statisticsindicate. As Figure 2 shows, in 1997-98 it tookthe average Oregonian about eight more daysto earn the income to educate a K-12 publicschool student than was the average for thenation. Any suggestion that Oregon is rela-tively lax in its commitment to public edu-cation is clearly unfounded.

While it is beyond the scope of this study toreview this issue in great detail, I did exploreone important area of expenditure, teachersalaries. If teachers are poorly paid by na-tional norms, that might be a sign of a lackof commitment to instruction not revealed

Per pupil spendingin Oregon has beenc o n s i s t e n t l ygreater than thenational average.

6

in aggregate expenditure data. The evidencewas striking. Oregon teachers are paid morethan the average of teachers nationally. Ipicked nine years at random from 1940 tothe present and observed in seven of thoseyears higher salaries than the national norm.

For example, National Education Associationdata indicate that in 1998-99 the averageteacher salary in Oregon ($42,200) exceededthe national average by 7.1 percent. At thesame time, the average annual earnings of allOregon workers is estimated by the U.S. Bu-reau of Labor Statistics to be more than 7percent below the national average. Thus rela-tive to local labor market conditions, Oregonteachers receive about 14 percent more paythan is typical in the United States as a whole.

Moreover, that is probably an understatementof the pay advantage of Oregon teachers rela-tive to other workers. In the aggregate, ben-efits for all Oregon public school districtemployees total about 38 percent of salaries.13

By contrast, benefits for all full-time work-ers in the American economy average slightlyover 20 percent.14 Assuming a similar per-centage in Oregon, the state’s teachers are re-

ceiving substantially higherfringe benefits than the non-teaching work force.

Higher pay for Oregon teachersis not necessarily “bad.” Indeed,one could assume that relativelyhigh pay for Oregon teachers wasjustified as a way to obtain highquality instructors. At the sametime, it may be true for Oregonthat “the substantial increase inthe money spent on teacher sala-ries likely has not had as large animpact on educational attain-ment as would have occurred hadthe money been targeted to the

recruitment and retention of a more highlyqualified teaching force.”15 Certainly it isworth investigating whether there are betterways to improve teacher quality than increas-ing compensation.

In short, Oregonians have made a substan-tial financial commitment to public educa-tion. Real resources have increaseddramatically per pupil, and continue to rise.Spending on students and on teachers (asmeasured by average salaries) are high rela-tive to the national average for teachers, eventhough overall incomes in Oregon are belowaverage.

Some individuals cite statistics that seem toshow Oregonians fare relatively well on stan-dardized tests relative to students from otherstates. They might conclude that high perpupil spending in Oregon has paid off inhigher student performance. Reaching thisconclusion, however, is dangerous for severalreasons.

First, there is relatively little evidence on theperformance of Oregon students relative tothose in other states. Data are available for

0

20

40

60

80

100

120

Oregon United States

Figure 2Days it takes average person to educate pupil, 1997-98

In short, Orego-nians have made asubstantial finan-cial commitmentto public educa-tion. Real re-sources haveincreased dramati-cally per pupil, andcontinue to rise.

��

��

��

7

the Scholastic Aptitude Test used for collegeadmissions. They show Oregon students withcomposite scores that compare quite favor-ably with the rest of the nation, but only whenstates where a large proportion of studentsdo not take the SAT are eliminated. Partici-pation rates vary widely among states (par-ticularly since some states favor the use ofthe alternative ACT test for college admis-sions). The test also tells us very little abouteducational achievement at lower grade lev-els. Unfortunately for comparison purposes,Oregon has not been a continuing regularparticipant in the testing of the National As-sessment of Educational Progress (NAEP), analternative and more comprehensive measureof student academic achievement. Limiteddata suggest, however, Oregon mirrors thenational average in student achievement.16

Second, as discussed below, there is strongevidence that out-of-school factors such asfamily educational background, family cohe-siveness, welfare, and even religious commit-ment are important in explaining studentachievement. With respect to these factors,on balance Oregon compares favorably toother states. For example, research suggeststhat students with college educated parents,other things equal, do better in school. To il-lustrate, in 1998 on the National Assessmentof Educational Progress reading test admin-istered at the eighth grade level, the averagescore for students with parents with less thana high school education was 243 out of apossible 500, dramatically lower than the 274score for students whose parents had gradu-ated from college.17 In 1998, 27.7 percent ofadult Oregonians were college graduates, com-pared with 24.4 percent nationwide. Oregonranked 12th among the states in the propor-tion of college graduates.18 Therefore, even ifit were determined that Oregon students per-form above the national average, it would stillbe an open question whether this reflects high

resource involvement in Oregon or other fac-tors that research shows play a major role instudent achievement. Indeed, correcting fordemographic factors, a recent Rand Corpo-ration study suggests Oregon schools performat about the national average.19

Third, by some other measures of achieve-ment, Oregon does quite poorly. The drop-out rate has been rising dramatically, bothabsolutely and relative to national trends.One news account recently reported that “Or-egon schools post the worst dropout rate inthe Northwest and among the worst in thenation.”20

��(��'��

��

There are at least three major approaches toeducational reform followed in the UnitedStates:

1. Structural change: fundamentally alter themethod of delivery of public educational ser-vices, introducing student-centered financialincentives (e.g., vouchers), establishing char-ter schools, facilitating home schooling, andso forth. More conservatively, make majorchanges in rules by which schools operate,such as liberalizing (or ending) the licensingof teachers and administrators, implement-ing performance-based pay, or endingteacher tenure.

2. Curricular reform: maintain the currentmechanism of educational delivery, butchange ways of serving students, the natureof what is taught, or how students are taught.Establish minimum levels of knowledge re-quired for graduation or promotion to highergrades.

3. Financial reform: maintain the currentmechanism of educational delivery, but in-

Limited datasuggest Oregonmirrors the na-tional average instudent achieve-ment.

8

crease the resources provided to the schools.In some cases, this includes reducing the dis-parity between school districts in per studentexpenditures.

With all three approaches, increased testingof students is typically called for to evaluatethe success of reform efforts in improvingstudent achievement, although there is seri-ous debate about just what tests should beused and who should be responsible forthem.

The Oregon QEM largely ignores the firstapproach to educational reform, only par-tially deals with the second, and concentrateson the third. There is no discussion of vouch-ers, charter schools, or home-schooling. Evenmodest reforms like loosening licensing re-quirements to increase entry into teachingand administration are not discussed. Yetthese structural approaches are command-ing much attention in a majority of Ameri-can states. For example, the superintendentsor CEOs of the school districts in America’sthree largest cities (New York, Los Angeles,and Chicago) are now non-educators with-out training in school administration, andmany smaller cities (e.g., San Diego, Hous-ton, Seattle, Kansas City) have turned to non-traditionally trained leaders as well.21 YetOregon has resisted this innovation. Indeed,the president of the Portland Association ofTeachers, Richard Garrett, recently com-mented, “The practice of employing unli-censed teachers . . . has not yet found a placehere.”22 Yet it is working well elsewhere. 23

The QEM implicitly rejects the move to de-centralize educational decision-makingdown to the school or even family level, ahallmark of much of the educational reformmovement in the United States, in favor ofan educational model that says, citing justone example, that there should be one school

guidance counselor for every 250 students inevery public school in the state. It thus reaf-firms the highly centralized approach to edu-cational decision-making that is under attackin many states. It implicitly rejects the mar-ket approach that suggests that competitionfor public school dollars from students freeto choose their school might be an attractivealternative to the existing paradigm. Theimplicit assumption is that the current struc-ture of educational delivery works fine, andthat the introduction of competition or pa-rental choice into the educational process isnot needed or is undesirable.

The Certificates of Initial and AdvancedMastery are discussed extensively, and muchis made of the fact that the QEM will lead toenhanced student performance. There arespecific subject areas that are to be coveredin grades K-12, and examination at approxi-mately the 10th grade (CIM) and 12th grade(CAM) is designed to assure that standardsare met. Yet there is no real discussion of de-tailed curricular issues, such as the advan-tages/disadvantages of innovations like CoreKnowledge or the whole language approach,or the relative merit of using incrementalschool time to promote, say, science as op-posed to foreign language instruction. Whilesome discussion of the resources needed fortechnology enhancement is included, it doesnot outline specific plans for, say, distancelearning to enhance curricular offerings inrural high schools, or discuss the efficacy ofInternet-based instruction.

Yet the QEM is quite specific on the need forincreased resources. The proposal has spe-cific recommendations regarding class size,the numbers of certain types of support per-sonnel, the need to lengthen the school year,and so forth. These recommendations in gen-eral raise the per pupil cost of instruction,and implicitly assume that the provision of

The QEM reaf-firms the highlycentralized ap-proach to educa-tional decisionmaking that isunder attack inmany states.

9

these additional resources will enhance learn-ing. The major emphasis in the QEM studyis on the need for more resources to attainan ideal defined by the study.

In reaching its conclusions, the QEM studyignores historical evidence that suggests thatover a long time span, Oregon has reducedthe proportion of K-12 public spending thatactually goes for instruction. One option thatOregon has is to reallocate funds within ex-isting education budgets, for example, hir-ing more teachers to cut class sizes, butfinancing that by reducing the number of ad-ministrators, curriculum coordinators, trans-portation specialists, etc. That possibility isnot even discussed.

Figure 3 documents that the proportion ofspending on instruction in Oregon K-12budgets has declined over time. I looked atthe Digest of Education Statistics for three ran-dom years over a span of one-third century.24

The evidence is clear that a smaller propor-tion of spending directly supports efforts inthe classroom. More and more, resourcesgiven to the schools have been diverted frominstructional purposes. Whereas in 1961-62,Oregon devoted a far greaterproportion of educational bud-gets to instruction than was theaverage nationwide (59.28% vs.54.62%), that was not so in1995-96 (Oregon’s proportionof 53.89% barely exceeded thenational figure of 53.63%). Thusthe relatively significant declinein the proportion of budgets al-located to instruction was a phe-nomenon in Oregon, but notnationally, where the declineover time was modest.

To conclude, the QEM is not athorough investigation of alter-

native paths to educational reform, andlargely embraces the status quo, with the ex-ception of the call for greater financial re-sources. It ignores major reform efforts goingon across the country with respect to paren-tal choice, independent (charter) publicschools, alternative certification, etc. It makesno recommendations to implement or evenconsider any of these alternative ways to re-form, approaches that seem promising andrelatively cost-effective in many instances.The QEM ignores the rise in non-instruc-tional expenditures in Oregon, and fails toinvestigate its origins, its consequences, andpossible changes.

��

�� '�/��

��

The Oregon QEM study leaves the impres-sion that there is a strong academic researchjustification to conclude that the increaseduse of educational inputs leads to improve-ments in student performance. The bibliog-raphy cites a number of studies that supportthat perception, yet the study ignores a muchlarger body of literature that suggests that the

0

10

20

30

40

50

60

1961-62 1984-85 1995-96

Figure 3Percent of Oregon K-12 spending for instruction, 1961-96

Over a long timespan Oregon hasreduced the pro-portion of K-12public spendingthat actually goesfor instruction.One option thatOregon has is toreallocate fundswithin existingeducation budgets.

��

��

��

10

relationship between school resources andstudent achievement is, at best, extremelymodest. Even where learning does improvewith incremental resources, however, there isalso the issue of the relationship between thedegree of improvement to the magnitude ofthe cost of those resources. Resources are fi-nite, and money spent on educational spend-ing crowds out spending on other things suchas public health, infrastructure and privateconsumption. The Oregon QEM study com-pletely ignores that issue. Thus the QEM pro-poses increases in school resources on thebasis of biased and inadequate evidence andignores the fact that resources have a cost.

The literature on the resource/learning rela-tionship is immense, and attempting to sum-marize it all accurately in a few paragraphs isdifficult. Nonetheless, taken as a whole, theliterature seems to support the followingthree stylized facts:

1. Resources are of secondary importance. Themajority of the evidence suggests that thereis little relationship between expenditureson schools and the amount of learninggoing on in them. While in some situationsenhanced resources may make some dif-ference, the impact of increased spendingis typically small, and sometimes is evenperverse (more spending, less learning).While it may be a little strong to say that“money does not matter,” the literature onbalance suggests that is far closer to thetruth than the implicit assumptions of theQEM, which come closer to concludingthat “money alone matters.”

2. Out-of-school factors are important in ex-plaining student achievement. By contrast,most studies show that student perfor-mance is enormously impacted by familyfactors. For example, children from intactfamilies with two college-educated parents

are far more likely to succeed academicallythan those coming from broken familieswith relatively uneducated parents. Thepresence of public assistance seems to beinversely correlated with learning. Someresearch shows that children of churchgo-ers, other things equal, learn more thanchildren of those without a strong faithtradition. Parents who push their childrento excel tend to get results. Studies, for ex-ample, tend to show a strong correlationbetween school attendance and learning,and even between participating in extra-curricular activities and student perfor-mance.

3. The school environment can make a largedifference in student performance. Even af-ter controlling for differences in studentsocioeconomic status, spending per pupil,and so forth, on average students in Catho-lic and other private schools learn morethan those in public schools. School sys-tems where strong principals are able tomake decisions independent of the cen-tral office often succeed more than wheredecision-making is centralized. Teacherswith a strong knowledge base and a senseof mission can also make a difference. Ingeneral, more learning occurs where thereis a strong sense of school community,where students, teachers, and administra-tors feel empowered to make their owneducational decisions.

,��)��

��

The most authoritative and oft-cited work onthe spending/learning relationship has beenperformed by Professor Eric Hanushek of theUniversity of Rochester and the NationalBureau of Economic Research. WhileHanushek himself has performed a numberof studies explaining the relationship be-

In general, morelearning occurswhere students,teachers, and ad-ministrators feelempowered tomake their owneducational deci-sions.

11

tween resource usage and public school per-formance, his most important work has beensummarizing the results of hundreds of otherstudies on the topic. In his initial major pieceon the topic, published in 1986, Hanushekexamined some 147 studies.25 He found thatof the 65 studies explicitly examining expen-ditures per pupil, only 16 percent showed anystatistically significant relationship betweenspending per pupil and learning (and acouple of them showed a negative relation-ship). Similar findings were observed forother measures of resource use, such as theteacher-pupil ratio (a proxy measure for classsize), teacher educational training, teachersalaries, etc.

Hanushek has updated his results on severaloccasions. In the most recent update, pub-lished in 1997, Hanushek reports on nearly400 studies, concluding that “there is not astrong or consistent relationship betweenstudent performance and school resources,at least after variations in family inputs aretaken into account.” He also notes that thesefindings are compatible with other ap-proaches (e.g., meta-analysis) that look at theimpact of school expenditures on labor mar-ket outcomes.26

I myself, working with Joshua Hall (staffEconomist at the Joint Economic Commit-tee of the Congress), have run scores of re-gressions examining the spending/academicachievement relationship using a vast data-base for Ohio schools, looking at studentachievement at different grade levels, usingalternative measures of student performance,using different samples of schools, and soforth. Our findings remarkably parallelHanushek’s. Most of the time there is no ob-served statistically significant relationshipbetween spending and learning, and whenthere is a relationship, it is often very weak(it takes a massive amount of resource infu-

sion to even measurably impact student per-formance). And, as Hanushek reports, occa-sionally there is an observed negativerelationship between spending and studentperformance.27

Some authors have commented that while thegeneral findings of Hanushek are valid on thespending/learning relationship, some typesof spending are positively associated withlearning. For example, Heinbuch andSamuels note that direct instructional spend-ing is positively associated with learning, evenif other spending is not, suggesting that in-cremental funding targeted for instructiononly may have positive benefits.28 While thismay well be true (my own research withJoshua Hall is consistent with that finding),the reality is that a large percentage of incre-mental public funding tends to go for non-instructional activities, and that many of theelements of the Oregon QEM proposal in-volve non-instructional expenditures. More-over, the Oregon evidence (Figure 3 above)relating to instructional spending suggeststhat the decline over time in the proportionof Oregon educational budgets going for in-struction means that resources may well havebeen reallocated in a fashion detrimental tolearning.

The most important challenge to the studentexpenditure/performance studies (as re-ported by Hanushek) has come in studies bytwo respected economists, David Card andAlan B. Krueger. They argue that enhancedschool expenditures have positive social ef-fects in any case (notwithstanding theHanushek findings), based on their belief thatthere is a positive correlation between spend-ing (which Card and Krueger equate withschool quality), and later earnings in theworkplace. This implies that enhanced schoolspending raises labor productivity.29

Most of the timethere is no ob-served statisticallysignificant rela-tionship betweenspending andlearning, and whenthere is a relation-ship, it is often veryweak.

12

Julian Betts, reviewing a significant body ofliterature on the school spending/earningsissue, effectively argues against the Card andKrueger findings. After surveying the litera-ture, he concludes there is “very little” link-age between earnings and school resources.Moreover, the studies finding such a link arebased on workers over the age of 30 who typi-cally were educated before 1970, at a timewhen school resources per pupil were dra-matically lower than is typical today. Thestudies also likely suffer from a significantaggregation bias because they use state aver-ages of educational costs, rather than school-specific data. Those researchers usingschool-specific data show no significant cor-relation between school spending and earn-ings. Moreover, even if Card and Krueger andother writers were correct, they “suggest avery low or even negative rate of return toadditional spending per pupil at conven-tional rates of discount.”30 This finding ishighly consistent with Betts’ own researchusing data for white males from the NationalLongitudinal Survey of Youth.31

��

The Oregon QEM calls for limits on class sizeat all grades, implicitly assuming that reduc-tions in class size have significant positiveeffects on student achievement. The QEMwork group making recommendations onclass size stated, “Large class sizes can ham-per achievement of goals especially at K-3levels....Therefore, the Work Group recom-mends class sizes of 20 students or less forgrades K-3.”32

Before dealing with the basic assumption thatreduced class size has important positive ef-fects on learning, it is important to note thatclass size reduction is extremely expensive.Since public policy decisions need to com-pare costs with benefits, the fact that class size

reduction is extremely costly means that thepositive benefits need to be very large for thisreform to be cost effective, to offer as much“bang for the buck” as other options for ex-penditures, both within and outside of edu-cation.

An example will illustrate. Suppose a schoolhas an average class size of 25 in its primarygrades, and it wishes to reduce it to 20. Sup-pose there are 200 primary students and thatcurrently it takes eight teachers to servicethem. With class sizes of 20, the minimumnumber of required teachers would be 10. Ifthe two new teachers hired earn the sameaverage salary and benefits as the older ones(which they likely would in any case overtime), instructional costs will rise 25 percent.If those costs constitute 40 percent of totalexpenditures per pupil, then per pupil spend-ing will rise at least 10 percent. In fact, it islikely to rise far more, for two reasons. First,if 20 is an absolute maximum number of stu-dents allowed per class (as the Oregon QEMimplies), more teachers will likely be hiredthan the example above indicates since thenumber of students per grade is not typicallyneatly divisible by 20. For example, there maybe 50 first-graders in a school previously ser-viced by two teachers that now would requirethree—a 50 percent, not a 25 percent in-crease. Average class size would fall from 25to 16.7 in the first grade.

Second, the addition of teachers increasesother costs. There are additional neededclassrooms which imposes significant capi-tal costs, which are often at least partially dis-guised in the short run by the failure of schooldistricts to fully practice generally acceptedaccounting practices (e.g., depreciate build-ings). There may be a perceived need formore teacher aides to assist teachers, and evenmore custodial help to maintain greater in-structional space. In short, the total increase

The fact that classsize reduction isextremely costlymeans that thepositive benefitsneed to be verylarge for this re-form to be costeffective, to offeras much “bang forthe buck” as otheroptions for expen-ditures.

13

in costs very well may be closer to 20 percentthan the 10 percent rise associated withhigher direct instructional costs indicatedabove.

The extremely large costs associated withclass size reduction have been recognized bythe chief academic adviser on the OregonQEM, David Conley. Writing in 1993, Conleycited the late Albert Shanker saying that “a20 percent cut in class size would mean anequal increase in staff and education bud-get—a huge and unrealistic increase for oneitem.”33 Conley also makes another interest-ing point after a discussion of larger class sizesin Asian schools. “Teachers and parents tendto accept as dogma that smaller is better.Getting them to question this assumptionand to explore alternative organizational andgroup arrangements will be quite difficult inmany cases.”34 Small classes are politicallypopular, especially because the presumedbeneficiaries themselves pay no marginal costfor the smaller classes. Yet at the same time,there is abundant evidence that there are al-ternative ways of improving academic per-formance, most of them far cheaper andmore effective than class size reduction.

In the most far-reaching review of the litera-ture on class size, Eric Hanushek concludesthat “there appears to be little systematic gainfrom general reductions in class size.... Fromproduction function estimates, there is littlereason to believe that overall reductions inclass size will yield much in the way of posi-tive achievement gains.”35 In reviewing 277estimates of the class size/learning relation-ship, only 28 percent of the studies report astatistically significant relationship; of theseabout as many observe a positive relationship(bigger classes lead to greater learning) as anegative one.36 While other instructional con-siderations (the quality of the teacher in par-ticular) can make a difference, class size per

se is typically a relatively minor factor in ex-plaining variations in learning.37 This is notto say that class size is always fairly unim-portant, since in some situations the dynam-ics of the classroom setting might favor smallclasses. As Hanushek puts it, “The real diffi-culty is that we do not know how to describe,a priori, situations where reductions in classsize will be beneficial. Thus it is not possibleto legislate only good outcomes from the statecapitol....”38

While the preponderance of national stud-ies explaining variations in learning showlittle or no impact from reductions in classsize, they typically use “macro” (school orschool district) data, and do not focus spe-cifically on the class size issue. A major ex-ception is a study of the STAR(Student-Teacher Achievement Ratio)project, in which the state of Tennessee be-gan a significant class size reduction experi-ment in 1985. The results suggest that“smaller classes did produce substantial im-provement in early learning and cognitivestudies.”39 Student learning in classroomswith 13 to 17 students (average about 15) wascompared with that in classrooms of 22 to25 students (average about 24). Students inthe smaller classes showed initial learninggains that a second phase of the study (theLasting Benefits Study) suggested were long-lived.

While the STAR results are interesting, theirrelevance to the Oregon QEM is question-able on several grounds:

1. The results dealt with reductions in classsize from around 24 to 15. Oregon is notconsidering reducing class sizes to 15, pre-sumably on cost considerations. It mayvery well be that most of the observedlearning gains observed in Tennessee re-flected the reductions in class size below

While otherinstructional con-siderations canmake a difference,class size per se istypically a rela-tively minor factorin explainingvariations in learn-ing.

14

20 per class, a class size range not underconsideration in the Oregon QEM.

2. There were significant problems in the Ten-nessee experiment in measuring thelonger-term success of students owing tothe mobility of students into and out ofsmaller classes. “For example, less thanone-half of those in the original experi-ment in kindergarten remained in the ex-periment at the end of third grade.”40 Theattrition from the experiment was not ran-dom in nature.41 Indeed, the lower aca-demic achievement of students who leftthe experimental classes biases the find-ings towards the conclusion that class sizereduction had long run benefits.

3. To this day, the researchers in Tennesseehave refused to make the raw data fromthe experiment available to other scholarswishing to do their own independentanalysis of student performance, makingsome scholars skeptical of the conclusionsdrawn from the study.

4. The STAR study draws dubious policyconclusions based on the observation ofrelatively steady learning advantages tosmall classes observed in grades K-3. Ifsmall classes had persistent advantages, onewould expect the small class size learningdifferential to rise in, say, grades two andthree as a result of continuing educationalbenefits of smaller classes. In fact, the STARresults might support small class size inkindergarten, and arguably the first grade,but not grades two and three because thatexpected rising learning differential is notobserved.

To determine that massive additional re-sources should be devoted to class size reduc-tion based on a single study is irresponsible,particularly given the mass of alternative evi-

dence that shows modest or no benefits fromsuch reduction. Moreover, even if there arelearning advantages to small class size, theevidence from the STAR experiment andother sources suggest those advantages arerelatively modest relative to costs, which caneasily exceed one thousand dollars per pupilor more, the exact amount depending on theprecise extent of class size reduction, incre-mental capital costs, and the amount of sup-port personnel hired to assist new instructors.On this point, a new study by Alan Kruegerand Diane Whitmore does find long-termbenefits arising from small class size, but theyestimate the rate of return from investing insmaller classes is less than six percent, belowthe rate of return on many other forms ofinvestment.42 Even putting the evidence onclass size in the best possible light, the rate ofreturn on class size reduction is likely to bevery low.

��

The Oregon QEM study recommends addi-tional resources be expended in a number ofareas. A specific example is the call to haveone guidance counselor for every 250 stu-dents.43 Again, there is no published empiri-cal research, to my knowledge, in significantscholarly outlets that supports this conclu-sion. There is, however, some research thatindicates a negative relationship between ex-penditures on non-instructional resourcesand student achievement.44 To be sure, theremay be arguments for guidance counselors(and other resources) that are unrelated tothe instructional mission of schools, but theextension of the schools into non-instruc-tional areas is certainly going beyond the callfor providing greater learning opportunitiesfor children, the ostensible rationale for theOregon QEM.

The Oregon QEM calls for more spending

15

There is someresearch that indi-cates a negative re-lationship betweenexpenditures onnon-instructionalresources and stu-dent achievement.

on professional development for teachers andadministrators. Again, on the surface thissounds like a good idea. Yet whether this is infact justified on cost-benefit grounds cannotbe ascertained from the Model. Publishedstudies on the impact of teacher educationon learning adds to the scepticism. Fewerthan 10 percent of the results reported in thelatest Hanushek research review shows apositive statistically significant relationshipbetween teacher education and studentperformance.45 It may be that short workshoptraining exercises have a more positive im-pact than formal college courses, but againthere is little reason to be certain of that.

To be sure, I do not wish to leave the impres-sion that everything mentioned in the Or-egon QEM is without merit, or fails to pass arational cost-benefit test. The discussion of“prototype schools” speaks of the case forsmaller schools than conventionally exist. Asignificant body of literature argues thatlearning is more likely to occur in smallerschools. As one study put it: “the two primaryarguments for large schools, cost savings andcurriculum enhancement, pale in compari-son with the positive schooling outcomesapparently achieved by smaller schools.”46

"��%��

��

The emphasis in the Oregon QEM study ison increasing resources with a goal forachieving greater student achievement. Asindicated earlier, a substantial body of litera-ture stresses the importance of non-school(particularly family) factors, not to mentionthe nature of the school and community.While a major elaboration of these points isnot practical here, it is clear that many schol-ars believe the keys to promoting learningcome from outside the school. Moreover, re-search suggests that internal school policies

to promote learning are likely in many casesto require few or no incremental resources.

The role of the family and non-resources fac-tors in explaining learning differentials wassuggested as early as the 1960s by JamesColeman, the eminent sociologist, in the fa-mous Coleman Report.47 Expanding on thattheme, Coleman and associates in the early1980s argued forcefully that a strong senseof school community contributed to greatereducational achievement in Catholicschools.48 His work in some sense was sup-ported by a major study of American highschools published in 1990 by John Chubb andTerry Moe. Chubb and Moe found that re-sources were a secondary consideration inexplaining school success, but that schoolorganization, the power of teachers and prin-cipals and the role of parents played signifi-cant roles in explaining differentialachievement.49 Meanwhile, as the number ofstatistical studies of student achievementmultiplied, they increasingly found that fam-ily and other socioeconomic factors were im-portant determinants of studentachievement.

Other important writers involved in the de-bate over educational reform stressed theimportance of curricular content andmethod. Diane Ravitch and Chester Finn ar-gued that our children are woefully ignorantof major works that form part of our cul-tural heritage.50 This led to an attack on theanti-intellectualism of modern teacher edu-cation and a bemoaning of a lack of “culturalliteracy” by prominent William Blake scholarE.D. Hirsch.51 Prominent psychologistHarold Stevenson argued that learning inAmerican schools has been far behind thatof those in Japan and China.52 Several writ-ers argued that major structural changes inAmerican public education were necessaryto overcome the curricular and other prob-

It is clear thatmany scholars be-lieve the keys topromoting learn-ing come fromoutside the school.

16

lems associated with our schools.53 To thesewriters, special interest groups such as teacherunions and colleges of education were fight-ing piecemeal reform and that therefore moreradical changes were necessary, leading tocalls for vouchers, charter schools, privatiza-tion of education, home schooling, and soforth. This issue is discussed more extensivelybelow.

Another issue regarding school effectivenessis the source of funding. Many states have in-creased state funding, and reduced localfunding of public schools in the past decade.In Oregon, this shift was precipitated by thepassage of Measure 5 in 1990. In addition toplacing a limit on local property tax rates foreducation, the measure also required the stateto replace lost property tax revenues to schooldistricts for five years. This legal mandate hasexpired, nonetheless the Oregon QEM ac-cepts the current funding apportionment asa given. Yet, one might argue on theoreticalgrounds that accountability and performanceare likely to be greater when funding comesdirectly from the customer base utilizing theeducational services. Moreover, Harvard andNational Bureau of Economic Researchscholar Caroline Hoxby has found a nega-tive correlation between learning and theproportion of school funding coming fromnon-local sources.54 This suggests that learn-ing could be enhanced by shifting fundingsources partially back to local school districts(as well as by increasing the instructionalproportion of the budget).

The point being made here is that the Or-egon QEM is devoid of a discussion of thisferment in educational policy, implicitly as-suming that most or all of it is either irrel-evant or wrong. Thus the Oregon QEM studyhas no serious discussion of concepts likeCore Knowledge, no evaluation of the grow-ing research on alternative forms of educa-

tional delivery, no discussion of the implica-tions of centralized financing, and so forth.It is almost as if the last 15 years or so of de-bate on American public education did notexist.

,��

��'�/��

��

As indicated above, the Oregon QEM studyignored a national whirlwind of efforts toredefine the way educational services are de-livered in the United States. Some of theseefforts involve relatively modest changes inthe existing structure, such as introducingmerit or market-based compensation sys-tems, or easing restrictions on entry into theteaching profession. Other changes are mod-erately substantial, such as the introductionof charter schools largely free of the regula-tory strictures facing traditional publicschools. A growing number of experimentsinvolve substantial change, including market-based approaches that greatly increase therole of students and their families in allocat-ing educational resources.

Space does not permit an elaborate discus-sion of all potential forms of structuralchange. An abridged discussion of some ideasthat have received national attention willdemonstrate the weakness of the unstated butvery real assumption of the Oregon QEMthat significant structural reform is unwiseor not worth seriously exploring.

��

For several years, students in Milwaukee andCleveland have been able to attend privateschools at public expense, in some casesschools with a religious affiliation. Addition-ally, in recent years a number of privategroups have funded a significant number of

A growing numberof education re-form experimentsinvolve substantialchange, includingmarket-based ap-proaches thatgreatly increasethe role of studentsand their familiesin allocating edu-cational resources.

17

scholarships for low-income students in vari-ous cities.55 The theory is that by giving fundsto the consumers rather than producers ofeducation, consumer choice increases andeducational efficiency and outcomes are en-hanced. Instead of having public school mo-nopolies dictating to students where they willgo to school and what they will study, schol-arship (voucher) based public funding givesconsumers (students and their families) acommanding role in deciding where publiceducational funds will be spent. Assumingthese consumers are out to maximize theirchildren’s welfare, they presumably will avoidschools with a poor reputation and flock toschools with a record of accomplishment.This will stimulate poorly performing pub-lic schools to improve in order to survive ina market environment.

The early research on vouchers is largelyquite supportive of the concept. The mostextensive research effort has been under-taken by a team of scholars led by the dis-tinguished political scientist Paul E.Peterson, Henry Lee Shattuck Professor ofGovernment at Harvard University. Regard-ing Milwaukee, the results showed studentsreceiving vouchers did better on math andreading achievement than students attend-ing conventional public schools, controllingfor other factors that might explain perfor-mance.56 An independent assessment of thatprogram performed by Cecilia E. Rouse forthe National Bureau of Economic Researchconcluded that students selected for thevoucher experiment in Milwaukee did bet-ter in math achievement, although not read-ing.57

In their early assessment of the Clevelandvoucher program, the Peterson team con-cluded that for two schools for which gooddata were available, not only did studentswitness a rise in test scores in math and read-

ing, but parental satisfaction with the pro-gram was far greater than for students attend-ing conventional public schools.58 Similarresults are observed with respect to privately-funded vouchers experiments. For example,46 percent of private-school parents in Wash-ington, D.C. participating in a voucher pro-gram gave their school a grade of “A,”compared with 15 percent for non-voucherstudents attending public school.59

��

Coleman, Chubb and Moe and other schol-ars find that schools are more often success-ful when given considerable freedom tomanage their own affairs. These findings ar-gue for freeing public schools from most ofthe bureaucratic restrictions imposed bytheir own central administrations and stategovernments. This perception led to the be-ginning of charter schools in the early 1990s.In the 1999-2000 school year, over 1,600 char-ter schools were in operation in the UnitedStates, enrolling over 250,000 students.60 En-rollments have been rising 50 percent or morea year.

If parental demand is a measure of success,then the charter school movement has beenquite successful since a majority of suchschools have a waiting list for admission.61

Oregon is a latecomer to this movement, withS.B. 100, the charter school bill, being signedinto law only on May 29, 1999. Oregon’s char-ter law, like that of many states, does not granttrue complete autonomy to charter schools,for example requiring that at least one-halfof teachers have certification.62 Any seriousdiscussion of Oregon public educationshould evaluate at length the probable growthin charter schools, their implications for con-ventional schools, etc. The Oregon QEMstudy does not do this.

The early researchon vouchers islargely quite sup-portive of theconcept.

18

��

As indicated above, Oregon has been slow inoffering nontraditional routes to becominga teacher. While technically permissible, allpersons seeking alternative means of certifi-cation must receive approval of the TeacherStandards and Practices Commission. Manystates do have more liberalized means of en-try into the profession. The current restrictedteacher certification system is more explain-able in terms of erecting barriers of entry intothe profession to enhance income for exist-ing members rather than in terms of any tan-gible educational benefits. There is littledoubt that reduced barriers to entry increasesthe pool of available teachers which should,other things equal, increase the quality ofthose actually hired.63 Moreover, there aresome indications that alternative certificationis a means of promoting increased minorityparticipation in teaching.64 Again, the OregonQEM Model does not discuss this significanttopic.

��

The topics above do not exhaust the avenuesfor structural change. Home schooling hasgrown dramatically, and arguably publicpolicy should move to accommodate homeschoolers, perhaps, for example, by onlineprovision of educational materials, distancelearning, and so forth. Merit pay for teachersdeserves consideration, as does providingdifferential pay for high-demand positionssuch as math and science teachers. Perhapslaws should be eased permitting greater con-tracting out of educational services with pri-vate providers such as Edison Schools.Similarly, so-called “prevailing wage” lawscould be relaxed for school construction tolower the costs of offering quality physicalfacilities. This list is not exhaustive. The pointis that any true evaluation of Oregon’s edu-

cational system needs to discuss items suchas these, and probably policy changes are nec-essary to accommodate innovative ways ofdelivering educational services.

0��

��

��

While the Oregon QEM study appropriatelyis concerned about K-12 public education,the study has broader implications for theOregon economy. Proponents might arguethat the earnings differentials between skilledand unskilled workers have grown and thatthe QEM will increase skills, making Oregona more desirable locale in terms of the NewEconomy.65 Opponents would observe thatthere is little or no certainty that the OregonQEM will actually improve skills of youngOregonians much, and that the financing ofthe QEM will have profound negative eco-nomic effects on the Oregon economy.66

The QEM report concludes that the cost ofphased to full implementation this biennumwould have ranged between $457 million and$1.15 billion above the Governor’s proposedbudget, a 10% to 25% increase.67 Oregoniansalready have a high income tax burden, andthe voters have rejected general sales taxationon numerous occasions. Past historical ex-perience and economic theory suggest thatan increase in already very high income taxrates would have a significant adverse impacton economic growth—another “cost” of theQEM ignored by its authors. A detailed analy-sis of the financial and economic conse-quences of the QEM is beyond this study, butthe evidence on the adverse impact of taxa-tion on economic growth makes it likely thatwhatever benefits that might arise from in-creased skills (and, based on previous re-search, I am dubious they would be verylarge) would likely be more than offset by the

The current re-stricted teachercertification sys-tem is moreexplainable interms of erectingbarriers of entryinto the professionto enhance incomefor existing mem-bers rather than interms of any tan-gible educationalbenefits.

19

"��

1. National Commission on Excellence inEducation, A Nation at Risk: The Imperativefor Educational Reform: A Report to theNation and the Secretary of Education(Washington, D.C.: Government PrintingOffice, 1983). Web: http://www.ed.gov/pubs/NatAtRisk/index.html.

2. Oregon Legislative Assembly, TheOregon Quality Education Model (Salem,OR, June 1999). Web: http://dbi.ode.state.or.us/qualityed.

3. Ibid., p2.

4. On this point, see the Oregon QEM,page 39: “The model was not developedwith attention to cost; it was developedwith attention to quality.”

negative consequences of increased taxation.At a minimum, the means and magnitudesof financing the QEM should be identifiedand the implications of that explored for thebroader Oregon economy.

��

The Oregon Quality Education Model im-plicitly ignores an immense body of litera-ture in reaching conclusions about needededucational change in Oregon. There is nodirect evidence based on original Oregon-specific research indicating that the changessuggested, such as mandatory low limits onclass size in the primary grades, would haveany positive effect whatsoever. The QEMimplicitly assumes that resources are notscarce, thus giving no consideration to therelationship between incremental benefitsand increased costs that could have addedone billion dollars to this biennium’s bud-get. There is no consideration of importantalternative approaches to improving perfor-mance such as introducing greater schoolcompetition and promoting parental choice.

By stressing “prototype schools,” the OregonQEM implicitly assumes that there is someideal school model that applies in every com-munity and learning environment. In real-ity, learning communities are highly diversein terms of student and teaching abilities andstrengths, cultural and socioeconomic influ-ences, and so forth. The needs and wants ofour children vary from city to city, school toschool, classroom to classroom. The central-ist thinking pervading the Oregon QEM ig-nores that reality.

Because trying to improve the educationalperformance of young Oregonians is com-mendable, perhaps a relatively apolitical andobjective commission could reevaluate thedirection of educational reform. Such a

group ideally would have a sizable staff ofnon-Oregonians without any vested interestin the outcome. This staff could include se-rious scholars from outside the college ofeducation milieu coming from a variety ofdisciplines such as economics, political sci-ence, psychology, sociology and English. Thisgroup could examine many possibilities forchange, not merely regulatory mandates thatincrease costs. Arguably it would conductoriginal Oregon-specific research evaluatingissues such as the spending/learning relation-ship and the importance of class size in anOregon context. Above all, it would relatecosts, including the broader effects of financ-ing any incremental spending, to benefits.The Oregon QEM study fails to do this. With-out a meaningful and objective evaluation ofits basic premises, proceeding to implementthe QEM would likely lead to a significantwaste of taxpayer resources.

Without a mean-ingful and objec-tive evaluation ofits basic premises,proceeding toimplement theQEM would likelylead to a signifi-cant waste of tax-payer resources.

20

http://www.ed.gov/pubs/NatAtRisk/index.html


http://dbi.ode.state.or.us/qualityed


13. See the Oregon Department of Educa-tion web site: http://www.ode.state.or.us/stats/schoolFinance/BUDsum9900.xls.

14. See the 1999 Statistical Abstract of theUnited States (Washington, D.C.: Govern-ment Printing Office, 1999), p. 443; on theweb at http://www.census.gov/prod/www/statistical-abstract-us.html.

15. Hamilton Lankford and James Wyckoff,“The Changing Structure of TeacherCompensation, 1970-94,” Economics ofEducation Review. 16(4), 1997, pp. 371-384;the quote is from p. 382 and is not specifi-cally directed at Oregon teacher compensa-tion.

16. As one example, in fourth and eighthgrade mathematics in 1996, the sameproportion of Oregonians scored at theadvanced level as was the case nationally,while a slightly lower proportion scored atthe “below basic” level. For more data, seehttp://www.heritage.org/schools/oregon.html.

17. See the NAEP Internet site, op. cit., formore details.

18. U.S. Bureau of the Census, 1999 Statis-tical Abstract of the United States (Washing-ton, D.C., Government Printing Office,1999, p. 171). Web: http://www.census.gov/prod/www/statistical-abstract-us.html.

19. “Oregon’s just-average schools,” TheOregonian, July 27, 2000. For the originalstudy, see David W. Grissmer et al., Improv-ing Student Achievement: What NAEP StateTest Scores Tell Us (Santa Monica, CA: RandCorporation, 2000), available online athttp://www.rand.org/publications/MR/MR924/.

21

5. Ibid., p. 11. The QEM also offers aPhased Implementation plan which wouldraise spending to $4.95 billion.

6. Ibid., p. 95.

7. I was part of that group.

8. For details see the A Nation at Risk website, op. cit.

9. A search of educational literature usingthe Nexis-Lexis and Social Citation Indexdatabases confirm that the names that Ithought should be referenced were in factwidely cited in the literature, indeed farmore so than the most prolific writers citedin the QEM study. My list of thirteenscholars: Chester Finn, Diane Ravitch, E.D.Hirsch, Theodore Sizer, Eric Hanushek,Caroline Hoxby, John Chubb, Terry Moe,James Coleman, Harold Stevenson, HerbertWahlberg, Myron Lieberman, and PaulPeterson. The only one of these referencedin the QEM study was Wahlberg.

10. Information on the NAEP test results isonline at http://nces.ed.gov/nationsreportcard/site/home.asp.

11. Michael J. Boskin et al., “ConsumerPrices, the Consumer Price Index, and theCost of Living,” Journal of Economic Per-spectives.12(1) Winter 1998, pp. 2-26.

12. In 1967, per capita personal income inOregon was 15 percent below the nationalaverage; in 1999 (preliminary), it was fivepercent below that average. See U.S. Bureauof the Census, Historical Statistics of theUnited States, Colonial Times to 1970(Washington, D.C.: Government PrintingOffice, 1975), pp. 243-245 and the web sitefor the Bureau of Economic Analysis, athttp://www.bea.doc.gov/.

http://www.ode.state.or.us/stats/schoolFinance/BUDsum9900.xls


http://www.census.gov/prod/www/statistical-abstract-us.html


http://www.heritage.org/schools/oregon.html

http://www.heritage.org/schools/oregon.html



http://www.rand.org/publications/MR/MR924/

http://www.rand.org/publications/MR/MR924/

http://nces.ed.gov/nationsreportcard/site/home.asp


http://www.bea.doc.gov/

20. Bill Graves, “1 in 3 Misses Graduation,”The Oregonian, June 25, 2000. Web: http://www.oregonlive.com/news/oregonian/index.ssf?/news/oregonian/00/06/lc_12drop25x.frame.

21. Tamar Levin, “Educators Are BypassedAs School System Leaders,” New York Times,June 8, 2000, p. A1. Premium subscribers tothe Times’ web site may find the storyarchived via http://www.nyt.com.

22. Richard Garrett, “Look beyond class-rooms when examining ‘crisis,’” Oregonian,June 6, 2000. Web: http://www.oregonlive.com/oped/index.ssf?/oped/00/06/ed060602.frame.

23. A personal note: my son and daughterteach in Georgia public schools, neither ofwhom took a single education coursetowards meeting licensing requirementsbefore being hired, but both of whom hadstrong academic backgrounds in theirmajors. One of them has won two teachingawards. That probably could not happen inOregon.

24. See the National Center for EducationStatistics, Digest of Education Statistics,various years, published by the Govern-ment Printing Office. Available on the webvia http://www.nces.ed.gov.

25. Eric Hanushek, “The Economics ofSchooling,” Journal of Economic Literature.24(3), September 1986, pp. 1141-1177.

26. Eric A. Hanushek, “Assessing the Effectsof School Resources on Student Perfor-mance: An Update,” Educational Evaluationand Policy Analysis. 19(2), Summer 1997,pp. 141-164.

27. Our most recent paper using the Ohio

data is Richard Vedder and Joshua Hall,“Private School Competition and PublicSchool Salaries,” Journal of Labor Research.21(1), Winter 2000, pp. 161-168.

28. Susan E. Heinbuch and Judith A.Samuels, “Getting More Bang for theEducational Buck: A MicrofinancingApproach,” Public Productivity & Manage-ment Review. 18(3), Spring 1995, pp. 233-242.

29. See David Card and Alan B. Krueger,“Does School Quality Matter? Returns toEducation and the Characteristics of PublicSchools in the United States,” Journal ofPolitical Economy. 100(1), February 1992,pp. 1-40. See also their “School Quality andBlack-White Earnings: A Direct Assess-ment,” Quarterly Journal of Economics.107(1), February 1992, pp. 151-200.

30. Julian R. Betts, “Is There a Link betweenSchool Inputs and Earnings? Fresh Scrutinyof an Old Literature,” in Gary Burtless, ed.,Does Money Matter: The Effect of SchoolResources on Student Achievement and AdultSuccess (Washington. D.C., BrookingsInstitution, 1996). See page 184 for thequote.

31. Julian R. Betts, “Does School QualityMatter? Evidence From the NationalLongitudinal Survey of Youth,” Review ofEconomics and Statistics. 77(2), May 1995,pp. 231-247.

32. Oregon QEM, p. 116.

33. David T. Conley, Roadmap to Restruc-turing: Policies, Practices and the EmergingVisions of Schooling (Eugene, OR: ERICClearinghouse on Educational Manage-ment, 1993), p. 258. This is written as partof a discussion about allowing more non-

22

http://www.oregonlive.com/news/oregonian/index.ssf?/news/oregonian/00/06/lc_12drop25x.frame



http://www.nyt.com

http://www.oregonlive.com/oped/index.ssf?/oped/00/06/ed060602.frame



http://www.nces.ed.gov

certified personnel into the classroom.

34. Ibid., p. 382.

35. Eric A. Hanushek, “The Evidence onClass Size,” in Susan E. Mayer and Paul E.Peterson, eds., Earning and Learning: HowSchools Matter (Washington, D.C.:Brookings Institution Press, 1999), pp. 131-168. The quote is from p. 161.

36. Ibid., p. 149.

37. On the importance of the quality ofteachers, see Eric A. Hanushek, “The Trade-Off Between Child Quantity and Quality,”Journal of Political Economy. 100(1), Janu-ary-February 1992, pp. 84-117.

38. “The Evidence on Class Size,” op cit., p.162. The Rand study, op cit., argues thattargeting class size reduction for lowerincome or disadvantaged children mightmake sense.

39. Frederick Mosteller, “The TennesseeStudy of Class Size in the Early SchoolGrades,” The Future of Children. 5(2),Summer/Fall 1995, pp. 113-127. The quoteis from p. 113.

40. Eric Hanushek, “Some Findings froman Independent Investigation of the Ten-nessee STAR Experiment and from OtherInvestigations of Class Size Effects,” Educa-tional Evaluation and Policy Analysis. 21(2),Summer 1999, pp. 143-163.

41. See Alan B. Krueger, “ExperimentalEstimates of Education Production Func-tion,” National Bureau of Economic Re-search Working Paper 6051 (Cambridge,MA: NBER, June 1997), and HarveyGoldstein and Peter Blatchford, “Class Sizeand Educational Achievement: A Review of

Methodology with Particular Reference toStudy Design,” British Educational ResearchJournal. 24(3), 1998, pp. 255-268.

42. Alan Krueger and Diane Whitmore,“The Effect of Attending a Small Class inthe Early Grades on College-Test Takingand Middle School Test Results: Evidencefrom Project Star,” National Bureau ofEconomic Research Working Paper#W7656 (Cambridge, MA: NBER, June2000).

43. In the interest of full disclosure, I ammarried to a public high school guidancecounselor.

44. See, for example, Gary M. Anderson,William F. Shughart and Robert D. Tollison,“Educational Achievement and the Cost ofBureaucracy,” Journal of Economic Behaviorand Organization. 15(1), January 1991, pp.29-45.

45. Hanushek, “Assessing the Effects,” op.cit., p. 144.

46. William F. Fowler, Jr., and Herbert J.Walberg, “School Size, Characteristics, andOutcomes,” Educational Evaluation andPolicy Analysis. 13(2), Summer 1991, pp.189-202; the quote is from p. 200.

47. James S. Coleman et al., Equality ofEducational Opportunity (Washington,D.C.: Government Printing Office, 1966).

48. James S. Coleman, Thomas Hoffer andSally Kilgore, High School Achievement:Public, Private and Catholic Schools Com-pared (New York: Basic Books, 1982).

49. John E. Chubb and Terry M. Moe,Politics, Markets and America’s Schools(Washington D.C.: Brookings Institution,

23

1990).

50. Diane Ravitch and Chester E. Finn, Jr.,What Do Our 17-Year Olds Know? A Reporton the First National Assessment of Historyand Literature (New York: Harper & Row,1987).

51. E.D. Hirsch, Jr., Cultural Literacy: WhatEvery American Needs to Know (Boston:Houghton Mifflin, 1987), and his TheSchools We Need and Why We Don’t HaveThem (New York: Doubleday, 1996).

52. Harold W. Stevenson, The LearningGap: Why Our Schools and Failing andWhat We Can Learn from Japanese andChinese Education (New York: SummitBooks, 1992).

53. Representative studies include ChesterFinn, We Must Take Charge: Our Schoolsand Our Future (New York: Free Press,1991), and Myron Lieberman, PublicEducation: An Autopsy (Cambridge, MA:Harvard University Press, 1993).

54. Caroline M. Hoxby, “Does CompetitionAmong Public Schools Benefit Studentsand Taxpayers?” National Bureau of Eco-nomic Research Working Paper #4979,December 1994. All the recent NBERworking papers on education are obtain-able via http://www.nber.org.

55. For example, Cascade Policy Institutefacilitates the Children’s ScholarshipFund— Portland, part of a national privatescholarship program for low-income K–8students.

56. Jay P. Greene, Paul E. Peterson andJiangtao Du, “Effectiveness of SchoolChoice: The Milwuakee Experiment,”Program on Education Policy and Gover-

nance, Kennedy School of Government,Harvard University, Occasional Paper 97,March 1997, on the web via http://data.fas.harvard.edu/pepg/mil.htm.

57. Cecilia E. Rouse, “Private SchoolVouchers and Student Achievement: AnEvaluation of the Milwaukee ParentalChoice Program,” Quarterly Journal ofEconomics. 113(2), May 1998, pp. 553-602;see also the National Bureau of EconomicResearch Web site at http://www.nber.org/papers/W5964.

58. Paul E. Peterson, William G. Howelland Jay P. Greene, “An Evaluation of theCleveland Voucher Program After TwoYears,” Harvard University, June 1999,available on the Web at http://www-vdc.fas.harvard.edu/PEPG/clev2rpt.pdf.

59. Patrick J. Wolf, William G. Howell, andPaul E. Peterson, “School Choice in Wash-ington, D.C.: An Evaluation After OneYear,” Harvard University, February 2000;Executive Summary available on the web athttp://www-vdc.fas.harvard.edu/PEPG/dc00x.pdf.

60. See U.S. Department of Education,Office of Educational Research and Im-provement, The State of Charter Schools2000 (Washington, D.C.: U.S. Departmentof Education, 2000). For a useful Web sitecontaining this and other information, goto http://www.uscharterschools.org.

61. For a representative account from aparent on the need for and success ofcharter schools, see Chiara R. Nappi, WhyCharter Schools: The Princeton Story (Wash-ington, D.C.: Thomas B. Fordham Founda-tion, October 1999). Web: http://www.edexcellence.net/library/wcs/wcs/html.

24

http://www.nber.org

http://data.fas.harvard.edu/pepg/mil.htm


http://www.nber.org/papers/W5964

http://www.nber.org/papers/W5964

http://www-vdc.fas.harvard.edu/PEPG/clev2rpt.pdf


http://www-vdc.fas.harvard.edu/PEPG/dc00x.pdf


http://www.uscharterschools.org

http://www.edexcellence.net/library/wcs/wcs/html



62. For a general description of schoolchoice efforts in Oregon and other states,see Nina Shokraii Rees, School Choice 2000(Washington, D.C.: Heritage Foundation,2000), available on their Web site at http://www.heritage.org.

63. On this point, see Carla J. Stevens andMicah Dial, “A Qualitative Study of Alter-natively Certified Teachers,” Education andUrban Society.26(1), November 1993, pp.63-77, or “Dena E. Franke, “The AlternateRoute: Testimonial from a Texas Teacher,”Educational Leadership. 49(3), November1991, pp. 34-35.

64. See Sharon Brennan and Traci Bliss,“Increasing Minority Representation in theTeaching Profession through AlternativeCertification: A Case Study,” Teacher Educa-tor 34(1), Summer 1998, pp. 1-11.

65. On this point, see R.J. Murnane, J.B.Willet, and F. Levy, “The Growing Impor-tance of Cognitive Skills in Wage Determi-nation,” Review of Economics and Statistics.77(2), 1995, pp.251-266, or John Bishop,“Achievement, Test Scores, and RelativeWages,” in Marvin H. Kosters, ed., Workersand Their Wages (Washington, D.C.: AEIPress, 1991), pp. 146-86.

66. The literature on this point is volumi-nous. For a summary, along with somestatistical findings relating to state andlocal taxation as it relates to economicgrowth, see Richard Vedder, State and LocalTaxation and Economic Growth: Lesson forFederal Tax Reform (Washington, D.C.:Joint Economic Committee of CongressStaff Report), December 1995. Web: http://www.senate.gov/~jec/sta&loc.html.

67. Oregon QEM, pp 48-49.

1��

Anderson, Gary M., William F. Shughart andRobert D. Tollison. 1991. “EducationalAchievement and the Cost of Bureaucracy.”Journal of Economic Behavior and Organiza-tion. 15(1), January, pp. 29-45.

Betts, Julian R. 1995. “Does School QualityMatter? Evidence From the National Longi-tudinal Survey of Youth.” Review of Econom-ics and Statistics. 77(2), May, pp. 231-247.

__________ . 1996. “Is There a Link Be-tween School Inputs and Earnings? FreshScrutiny of an Old Literature.” In GaryBurtless, ed., Does Money Matter: The Effectsof School Resources on Student Achievementand Adult Success. Washington, D.C.:Brookings Institution Press.

Bishop, John. 1991. “Achievement, TestScores, and Relative Wages.” In Marvin H.Kosters, ed., Workers and Their Wages. Wash-ington, D.C.: AEI Press.

Boskin, Michael J. et al. 1998. “ConsumerPrices, the Consumer Price Index, and theCost of Living.” Journal of Economic Perspec-tives. 12(1), Winter, pp. 2-26.

Card, David and Alan B. Krueger. 1992. “DoesSchool Quality Matter? Returns to Educationand the Characteristics of Public Schools inthe United States.” Journal of PoliticalEconomy. 100(1), February, pp. 1-40.

__________ . 1992. “School Quality andBlack-White Earnings: A Direct Assessment.”Quarterly Journal of Economics. 107(1), Feb-ruary, pp. 151-200.

Chubb, John E. and Terry M. Moe. 1990. Poli-tics, Markets and America’s Schools. Washing-ton, D.C. : Brookings Institution.

25

http://www.heritage.org


http://www.senate.gov/~jec/sta&loc.html


Coleman, James S. et al. 1966. Equality ofEducational Opportunity. Washington, D.C.:Government Printing Office.

__________ , Thomas Hoffer and SallyKilgore. 1982. High School Achievement: Pub-lic, Private and Catholic Schools Compared.New York: Basic Books.

Conley, David T. 1993. Roadmap to Restruc-turing: Policies, Practices and the EmergingVisions of Schooling. Eugene, OR: ERIC Clear-inghouse on Educational Management.

Finn, Chester E., Jr. 1991. We Must TakeCharge: Our Schools and Our Future. NewYork: Free Press.

Fowler, William J., Jr., and Herbert J. Walberg.1991. “School Size, Characteristics, and Out-comes.” Educational Evaluation and PolicyAnalysis. 13(2), Summer, pp. 189-202.

Franke, Dena E. 1991. “The Alternate Route:Testimonial From a Texas Teacher.” Educa-tional Leadership. 49(3), November. pp. 34-35.

Garrett, Richard. 2000. “Look Beyond Class-rooms When Examining ‘Crisis.’” TheOregonian, June 6. Online at: http://www.oregonlive.com/oped/index.ssf?/oped/00/06/ed060602.frame.

Goldstein, Harvey and Peter Blatchford. 1998.“Class Size and Educational Achievement: AReview of Methodology with Particular Ref-erence to Study Design.” British EducationalResearch Journal. 24(3), pp. 255-268.

Graves, Bill. 2000. “1 in 3 Misses Graduation.”The Oregonian, June 25. Web: http://www.oregonlive.com/news/oregonian/i n d e x . s s f ? / n e w s / o r e g o n i a n / 0 0 / 0 6 /lc_12drop25x.frame.

Greene, Jay P., Paul E. Peterson and JiangtaoDu. 1997. “Effectiveness of School Choice:The Milwaukee Experiment.” Program onEducation Policy and Governance, KennedySchool of Government, Harvard University.Occasional Paper 97, March. Web: http://data.fas.harvard.edu/pepg/mil.htm.

Grissmer, David W. et al. 2000. Improving Stu-dent Achievement: What NAEP State TestScores Tell Us. Santa Monica, CA.: Rand Cor-poration.

Hanushek, Eric. 1986. “The Economics ofSchooling.” Journal of Economic Literature.24(3), September, pp. 1141-1177.

__________ . 1992. “The Trade-Off Be-tween Child Quantity and Quality.” Journalof Political Economy. 100(1), January/Febru-ary, pp. 84-117.

__________ . 1997. “Assessing the Effectsof School Resources on Student Performance:An Update.” Educational Evaluation andPolicy Analysis. 19(2), Summer, pp. 141-164.

__________ . 1999. “The Evidence onClass Size.” In Susan E. Mayer and Paul E.Peterson, eds., Earnings and Learning: HowSchools Matter. Washington, D.C.: BrookingsInstitution Press.

__________ . 1999. “Some Findings froman Independent Investigation of the Tennes-see STAR Experiment and from Other Inves-tigations of Class Size Effects.” EducationalEvaluation and Policy Analysis. 21(2), Sum-mer, pp. 143-163.

Heinbuch, Susan E. and Judith A. Samuels.1995. “Getting More Bang for the EducationalBuck: A Microfinancing Approach.” PublicProductivity & Management Review. Spring,18(3), Spring, pp. 233-242.

26









Hirsch, E.D., Jr. 1987. Cultural Literacy: WhatEvery American Needs to Know. Boston:Houghton Mifflin.

__________ . 1996. The Schools We Needand Why We Don’t Have Them. New York:Doubleday.

Hoxby, Caroline. M. 1994. “Does Competi-tion Among Public Schools Benefit Studentsand Taxpayers?” National Bureau of Eco-nomic Research Working Paper 4979. Cam-bridge, MA: NBER, December. Available onthe Web via http://www.nber.org.

Krueger, Alan B. 1998. “Experimental Esti-mates of Education Production Function.”National Bureau of Economic ResearchWorking Paper 6051. Cambridge, MA: NBER,June. Web: http://www.nber.org.

__________ and Diane Whitmore. 2000.“The Effect of Attending a Small Class in theEarly Grades on College-Test Taking andMiddle School Test Results: Evidence fromProject STAR.” National Bureau of EconomicResearch. Working Paper 7656. Cambridge,MA: NBER, April. Web: http://www.nber.org.

Lankford, Hamilton and James Wyckoff.1997. “The Changing Structure of TeacherCompensation, 1970-94.” Economics of Edu-cation Review. 16(4), pp. 371-384.

Lieberman, Myron. 1993. Public Education:An Autopsy. Cambridge, MA: Harvard Uni-versity Press.

Levin, Tamar. 2000. “Educators Are Bypassedas School System Leaders.” New York Times,June 8, p. A1. Archived for premium subscrib-ers via http://www.nyt.com.

Mosteller, Frederick. 1995. “The TennesseeStudy of Class Size in the Early School

Grades.” The Future of Children. 5(2), Sum-mer/Fall, pp. 113-127.

Murnane, Richard J., J.B. Willet, and FrankLevy. 1995. “The Growing Importance ofCognitive Skills in Wage Determination.” Re-view of Economics and Statistics. 77(2), pp.252-266.

Nappi, Chiara R. 1999. Why Charter Schools:The Princeton Story. Washington, D.C.: Tho-mas B. Fordham Foundation, October. Website: http://www.edexcellence.net/library/wcs/wcs/html.

National Assessment of Educational Progress.Various dates. Home page: http://nces .ed.gov/nat ionsrepor tcard/s i te/home.asp.

National Center for Education Statistics.Various dates. Digest of Education Statistics.Washington, D.C.: Government Printing Of-fice. Web: http://www.nces.ed.gov.

National Commission on Excellence in Edu-cation. 1983. A Nation at Risk: The Impera-tive for Educational Reform: A Report to theNation and the Secretary of Education. Wash-ington, D.C.: Government Printing Office.Web: http://www.ed.gov/pubs/NatAtRisk/index.html.

Oregon Department of Education. 2000. Fi-nancial statistics. Online at: h t t p : / /www.ode.state.or.us/stats/schoolFinance/BUDsum9900.xls.

Oregon Legislative Assembly. 1999. The Or-egon Quality Education Model. Salem, OR,June. Web: http://dbi.ode.state.or.us/qualityed/.

“Oregon’s just-average Schools.” 2000. TheOregonian, July 27.

27

http://www.nber.org

http://www.nber.org

http://www.nber.org

http://www.nyt.com






http://www.nces.ed.gov








Peterson, Paul E., William G. Howell, and JayP. Greene. 1999. “An Evaluation of the Cleve-land Voucher Program After Two Years.”Harvard University, June. Web: http://www-vdc.fas.harvard.edu/PEPG/clev2rpt.pdf.

Ravitch, Diane and Chester E. Finn, Jr. 1987.What Do Our 17-Year Olds Know? A Reporton the First National Assessment of History andLiterature. New York: Harper & Row.

Rees, Nina Shokraii. 2000. School Choice 2000.Washington, D.C. Heritage Foundation, June.Available on the Web at http://www.heritage.org.

Rouse, Cecilia E. 1998. “Private SchoolVouchers and Student Achievement: AnEvaluation of the Milwaukee Parental ChoiceProgram.” Quarterly Journal of Economics.113(2), May, pp. 553-602. Web: http://www.nber.org.

Stevens, Carla J. and Micah Dial. 1993. “AQualitative Study of Alternatively CertifiedTeachers.” Education and Urban Society. 26(1),November, pp. 63-77.

Stevenson, Harold W. 1992. The LearningGap: Why Our Schools Are Failing and WhatWe Can Learn from Japanese and ChineseEducation. New York: Summit Books.

U.S. Bureau of Economic Analysis. 2000. “Re-gional Income Statistics.” Web: http://www.bea.doc.gov.

U. S. Bureau of the Census. 1975. HistoricalStatistics of the United States, Colonial Timesto 1970. Washington, D.C.: GovernmentPrinting Office.

__________ . Various years. Statistical Ab-stract of the United States. Washington, D.C.:Government Printing Office. Web: http://

www.census.gov/www/statistical-abstract-us.html.

U.S. Department of Education, Office ofEducational Research and Improvement.2000. The State of Charter Schools 2000. Wash-ington, D.C.: U.S. Department of Education.Web site: http://www.uscharterschools.org.

Vedder, Richard. 1995. State and Local Taxa-tion and Economic Growth: Lessons for Fed-eral Tax Reform. Washington, D.C.: Joint Eco-nomic Committee of Congress. Web: http://www.senate.gov/~jec/sta&loc.html.

__________ and Joshua Hall. 2000. “Pri-vate School Competition and Public SchoolSalaries.” Journal of Labor Research. 21(1),Winter, pp. 161-168.

Wolf, Patrick J., William G. Howell, and PaulE. Peterson. 2000. “School Choice in Wash-ington, D.C.: An Evaluation After One Year.”Harvard University, February. Web: http://www-vdc.fas.harvard.edu/PEPG/dc00x.pdf.

28





http://www.nber.org

http://www.nber.org






http://www.uscharterschools.org





��

��

��

��

�� !"�#$��$%&��

'''&"�#$��$%&��

�(��

www.cascadepolicy.org

mailto:[email protected]

Money for Nothing...2018/02/08 · Money for nothing? An analysis of the Oregon Quality Education...

Documents

Transcript of Money for Nothing...2018/02/08 · Money for nothing? An analysis of the Oregon Quality Education...