Orwell Award Announcement SusanOhanian.Org Home

No Child Left Behind: Where Does the Money Go? Part 1

Susan Notes: Unfortunately, I can't fit all this blockbuster piece in one piece so I'm doing it in two parts. The tables and Notes read much better at the EPRU website, but I want to be part of spreading the word, and so I post it here.

Contact: Gerald Bracey (703) 317-1716 (email) gbracey1@verizon.net or Alex Molnar (480) 965-1886 (email) epsl@asu.edu

This report examines how No Child Left Behind (NCLB) dollars flow from the federal government through states and districts and into private coffers. It does not attempt to make many connections between NCLB and efforts to control curriculum and instruction for political or ideological purposes. That has been done, and done well, by others. This report also does not attempt to determine if the law's goals are attainable. That, too, has been done elsewhere. This report makes the case that the law enriches many private companies, especially those historically close to the Bushes and their friends.

No Child Left Behind: Description of Program Essentials
Although dwarfed by programs such as Medicare and Social Security, NCLB--Title I of the federal government's Elementary and Secondary Education Act of 2001--represents the largest single federal involvement ever in education. For the 2004-2005 school year, Congress authorized $18.5 billion, and $12.342 billion, a sum quite close to the President's budget request, was actually appropriated. For school year 2005-2006, Congress authorized $20.5 billion; the president has requested $13.342 billion.

NCLB's advocates tout it as a major increase in aid to the states although several analyses have concluded that the law will cost the states more money than they will receive. Questions have been raised about how the money is to be distributed, and how much of the money of Title I flows through states, districts and schools, and into the treasuries of private companies. The answer to these questions vary with the number of years that a school or district has been labeled "failing."

To comply with NCLB, each state submitted to the U.S. Department of Education a plan indicating a baseline of achievement for 2002-2003 and how the state will move from that baseline to attain the final goal: 100 percent of students at "proficient" or better by the year 2013-2014.
Proficient is uniquely defined by each state. The plan also indicates how much schools must raise test scores each year in order to make Adequate Yearly Progress (AYP) on tests administered by the state. In grades three through eight and one grade in high school, all students must be tested annually in reading and math, with science to be added in some grades in 2007. President Bush has proposed to extend the program to test more grades in high school, but the fate of this proposal is uncertain.

The amount of money a school is eligible for under Title I depends on its level of poverty. Technically, schools that do not receive Title I funds do not have to test and do not fall prey to the sanctions imposed by failing to make AYP. In reality, most districts make use of an "embarrassment factor" (publicly displaying their test scores) and test all students and rate schools not receiving Title I funds as well. To date, only a few school districts have refused to take part in NCLB, forfeiting whatever Title I money they would have received otherwise.

At the school level, AYP is required for all subgroups of students; the subgroups are formed by subject tested, grade, gender, ethnicity, socio-economic status, special education status, migrant status, English Language Learner status, and percent of each group who took the tests (95 percent of the students in each group must take the test to meet the law's requirements). Not all subgroups are applicable to all schools. States choose, with U.S. Department of Education approval, the minimum number of students in a subgroup for that group to be reported for NCLB purposes. This can yield puzzling results. For example, in Maryland one school failed to make AYP because its 12 special education students failed to make AYP. Next door in Virginia, another school made AYP even though its 24 special education students failed to make AYP. At the time, the minimum number for reporting in any subgroup was five in Maryland, but 50 in Virginia.

If any one subgroup fails to make AYP, the entire school is labeled to be "in need of improvement" the media typically use "failed" and whatever sanctions apply for that year apply to the whole school, not just the subgroup(s) that did not make AYP.

Schools that do not meet their growth targets are subject to sanctions which increase in severity with each successive year of failure. Each sanction has different implications regarding how Title I money will be distributed. It is important, therefore, to examine the sanctions that apply with each successive year. A summary of the consequences of failing to make AYP for different numbers of consecutive years follows. The sanctions are cumulative. For example, if a school fails to make AYP three years in a row, the school bears the sanctions from both years two and three (there are no sanctions for a single year of failure).

Essentials of NCLB Sanctions
This section begins with a description of the sanctions incurred by schools after each consecutive year of failure. The descriptions are followed by a discussion of what the sanctions cost the districts and schools, and how they benefit private for-profit corporations. The discussion is organized by what happens at the end of each consecutive year a school fails to make AYP. Some costs, however, were not anticipated by NCLB and are independent of year-contingent sanctions (i.e. states hiring firms with expertise in detecting cheating on tests, and states developing databases that permit the state to track individual students as long as they remain in the state's public schools).

End of Year 0

There are cost considerations from the outset because most states did not operate a testing program of the magnitude required by the law. Some states did not test in all required grades and some states did not test annually. To date, the U.S. Department of Education has adamantly insisted that all grades covered by the law be tested each year. Thus, many states have experienced large increases in the costs of their testing programs and the reporting systems needed to provide the results to parents, as also required by the law. CTB/McGraw-Hill, Harcourt Assessment, NCS Pearson, Riverside (Houghton Mifflin), and Educational Testing Service are companies that dominate the K-12 achievement test market. Few states develop tests, score them, and report the results using state civil servants.

In addition to testing, the law requires that curricula adopted for beginning reading instruction, known as Reading First, be backed by "scientifically based research." The phrase "scientifically based research" occurs 111 times in the 1,100 pages of the law. This has required a number of states to adopt new curricula that meet criteria established through the U. S. Department of Education, or risk having applications for Reading First funds rejected. Many of the same firms that supply tests also supply textbooks and other curricular materials.

End of Year 1
The school is placed on a "watch list" and must develop a plan to improve.

End of Year 2
The school must offer all students the option to transfer to a different, "successful" school within the district. The sending school must pay for any transportation costs. If the failing school later makes AYP, the students who left are not required to return, but the sending school no longer has to pay for transportation.
Initially, the U. S. Department of Education held that crowding at a receiving school could not be used to refuse students from sending schools. The receiving schools needed to build capacity and add classrooms, mobile units, extra teachers, etc. The logistical realities of the choice option, however, have resulted in little enforcement of this provision. In cities where schools are already crowded, only a tiny percentage of eligible students have received the choice option. In some rural areas, choosing a different school can require rides of several hours, and in parts of Alaska and Hawaii, a plane ride. For many small districts, no option exists.

In addition, negative reactions from receiving schools have constricted some choice programs. For instance, in 2003-2004, some 8,000 New York City students availed themselves of the choice option. Reactions from principals at the receiving schools led the city to limit the number of transfers to 1,000 students in 2004-2005.

Where the choice option cannot be utilized by all students in a failing school, it is supposed to be offered to the "neediest" first--those with the lowest test scores. It is not always the lowest scoring students who ask to change schools. In Fairfax County, Virginia, a district of some 160,000 students, two Title I schools failed to make AYP. Although neither of the schools would be labeled "low performing" in a normative sense, some subgroups did not make AYP. The students who opted to go to other schools were largely high-scoring students.

End of Year 3
The school must offer "supplemental educational services" (SES). These services can vary but typically consist of: (1) after-hours programs of small group work, (2) individual tutoring at the school, or (3) online instruction at the school or at the student's home. Each state develops a list of qualified providers. Some districts have used district teachers to provide such tutoring. Chicago Public School provided the SES for a time, but was later declared ineligible by the U.S. Department of Education because Chicago is itself a "failing" district. The U.S. Department of Education�s action in this case implies that competent teachers do not exist in the Chicago system and that the dynamics of one-on-one tutoring or small groups are no quite different from the 30-to-1 pupil-to-teacher ratio often found in urban classrooms. The U.S. Department of Education has contended that if the schools cannot improve achievement during the regular school day, they cannot do so in the hours after school, either, nor do they deserve the chance to try.

End of Year 4
The school must opt from a variety of corrective actions:

  • Replace school staff relevant to the failure (staff refers to teachers).

  • Put in place a new curriculum.

  • Decrease management authority at the school.

  • Appoint outside experts to advise the school.

  • Extend the school year or the school day.

  • Restructure the internal organization of the school.

  • End of Year 5
    The school must opt from a variety of corrective actions:
  • Reopen as a charter school.

  • Replace all or most of the staff.

  • Contract with an outside entity to operate the school.

  • Institute other significant governance and staffing changes likely to improve the school.

  • Annual Reports from All Schools, Districts and States
    In addition to the annual assessments in grades three through eight, states must produce and disseminate annual report cards that provide information on the results of such assessments. The state report cards must show the trends for the state in the subcategories listed earlier, and a comparison of the state's performance to the state's goal. The state must also provide data on district progress listing the schools in each district that failed to make AYP. The report cards must also show the high school graduation rate, the professional qualifications of the teachers in the state, the percentage of teachers with emergency or provisional credentials, and the percentage of classes not taught by highly-qualified teachers as defined by NCLB. Descriptions must compare these indicators separately for high- and low-poverty schools.

    School districts must prepare similar report cards showing district-level data and data for each school.

    NCLB Dollar Consideration Year By Year

    Year Zero: Testing

    At the time the No Child Left Behind Act (NCLB) was passed, very few states maintained sufficiently large testing programs to satisfy the law's requirements. Table 1 shows the number of additional tests needed to comply with the law varied from state to state as reported by the Government Accounting Office (GAO) report.

    Table 1: Number of Tests States Needed to Add to State Testing Programs to Comply with Title I
    Number of Tests Needed Number of States
    None 5
    1-3 4
    4-6 6
    7-9 17
    10-12 10
    13 or more 10
    Source: Government Accounting Office. (2003, May). Characteristics of tests will influence expenses; information sharing may help states realize efficiencies (GAO-03-389). Washington, DC: Author Note: Presence of the District of Columbia and Puerto Rico brings total to 52
    The GAO estimated that if states used only multiple-choice questions, they would spend $1.9 billion over a six-year period (from 2002 to 2008) to develop, score, and report results. If they used all open-ended questions, they would spend $5.3 billion. Testing with the then-current mix of multiple-choice and open-ended questions would cost $3.9 billion. NCLB would provide sufficient funds to cover the costs of the multiple-choice-only option, but not the other two. One study from Accountability Works argued that funds for testing would be sufficient, but many dismissed the study's assumptions as unreasonable. Note that the projections are for meeting the Title I requirements. The number of Title I schools varies a great deal from district to district. If districts test all schools and not just Title I schools, the cost of testing would rise substantially, even using only multiple-choice tests.

    No doubt the projections from both the GAO and Accountability Works greatly underestimated how much money states would actually spend over the period. They estimated only compliance costs, not performance costs. That is, they did not take into account what expenditures would be required to bring all children to a level near 100 percent proficiency. The National Council of State Legislatures concluded that at best, the federal funding would come close to covering compliance costs, but not begin to touch the far more expensive performance costs.

    The projections did not take into account the extensive purchase of "test prep" materials. In addition, at the time the reports were published (early 2003 for the GAO study, early 2002 for that from Accountability Works), the use of "formative evaluation" had not taken hold the way it has now. Formative evaluation uses tests to diagnose particular problems a student is having, or to measure progress on the objectives that will be covered by the "big" test used for NCLB. It is not clear that the tests now being offered as "formative" differ in any meaningful way from the "summative" tests that are used to determine if the school has made AYP except that it is often easier to get individual student data from the formative assessments. In any case, ETS' John Oswald said, "The formative-assessment market is a big one, and it's growing like crazy, and it's accompanied by a lot of professional development to help teachers use those."

    In late May 2005, two of the largest players on the testing scene, Harcourt Assessment and McGraw-Hill, rolled out on-line testing systems that make extensive use of so-called formative testing. The Harcourt system has two components, "periodic assessments called Class Links that are guided by each state's academic content standards, and annual tests called Class Views that are guided by the blueprints for state accountability tests." The McGraw-Hill system contains "timed assessment of weekly reading passages to help diagnose students' overall reading ability, fluency, decoding skills and comprehension skills." McGraw Hill's system also contains a weekly assessment for language arts.

    Eduventures, a Boston-based education industry, marketing, and research firm, projected that annual industry revenues from state assessments would grow from $572 million in 2003 to $810 million in 2006. The projections for this period are linear. If one assumes the same growth rate for some years not in the Eduventures projection, then revenues for state testing programs come to about $5.4 billion over the same period as the GAO spending projections (2002 to 2008). In 2003, Eduventures put the total spending on K-12 testing, including test prep in 2003, at $1.81 billion with a projected rise to $2.29 billion by 2006. Of the $1.81 billion, Eduventures estimated that $334 million would go to the larger companies' subcontractors.

    Eduventures contends that nine companies currently account for approximately 87 percent of the market:
    Major Publishers
  • CTB-McGraw-Hill

  • Harcourt Assessment

  • Pearson Assessment

  • Riverside Publishing (Houghton Mifflin)
  • Education Testing Service

  • Smaller Publishers

  • Touchstone Applied Science Associates

  • Measurement, Inc.

  • Nonprofits
  • Measure Profits

  • Northwest Evaluation Association

  • Harcourt Assessment is linked to the various Harcourt publishing houses and all are part of the Reed Elsevier empire that earned $5.6 billion in sales in 2001. CTB/McGraw-Hill is linked to the various other McGraw-Hill publishing endeavors. For the nine months ending September 30, 2004, McGraw-Hill enjoyed revenues of $3.84 billion and net profits of $566 million. Pearson has seen rapid growth including the acquisition of the test-scoring firm, NCS Corporation. It recently passed Riverside to become the third largest company in the field.

    Riverside's best known product, no doubt, is the Iowa Tests of Basic Skills, but it recently entered the online arena with the purchase of Edusoft which permits teachers to make "mini-tests" that will estimate how close children are to passing the state test. According to Edusoft's website, its data warehouse contains 100,000,000 student scores from 500,000 assessments. The website also states that 300 districts signed up "last year."

    A few other players have substantial contracts. Educational Testing Service returned aggressively to the K-12 market. Its state testing contracts garner about $90 million a year with $275 million arriving from California over a five-year period. ETS made its move shortly after abandoning its tradition of having educators run the company and installing a marketing executive as president. Data recognition has been growing and now has about $100 million in annual revenues (not all from state testing), and the American Institutes for Research has entered the field with yearly revenues of about $40 million.

    Observers have often noted that the testing industry is virtually unregulated. Said Walt Haney of Boston College, "There is more public oversight of the pet industry and the food we feed our dogs than there is for the quality of tests we make our kids take." The industry is not accountable for the tests that are used to hold schools accountable.

    It is only when someone accidentally discovers an error that quality control procedures and accountability enter the picture. Martin Swaden, a Minnesota lawyer and father of a public school student, threatened to sue the state of Minnesota for refusing to let him look at his daughter's test and answer sheet at the same time. The state yielded. The daughter had failed the state's then-high school exit test and the father reasoned that the best way to help her pass was to see what items had caused her difficulties. What he found, though, was that the answer sheet had been mis-keyed. Some 47,000 Minnesota students received lower scores than they deserved, 8,000 failed when they should have passed--including Swaden's daughter--and 525 seniors had been wrongfully denied diplomas. NCS Corporation, later purchased by Pearson for $2.5 billion, eventually settled for $8 million. This is one example of several instances where a testing company error has punished students.

    The probability of mistakes has likely increased since the arrival of NCLB. Test use was already growing and the various testing companies were raiding each other for talented employees. In the opinion of some, the increase in testing has pushed the companies beyond capacity. An April examination of the "employment opportunities" window of the ETS website found 98 professional openings, 65 of which were still available in May. States now insist on obtaining results from scoring-reporting services in time frames "weeks" that were heretofore considered impossible and, indeed, might have been before recent technological advances. States have begun to build substantial penalty clauses into their contracts covering not only errors but specific dates by which the companies must have satisfactorily completed specific services.

    — Gerald W. Bracey, George Mason University
    Education Policy STudies Laboratory, Education Policy Research Unit, Arizona State University



    This site contains copyrighted material the use of which has not always been specifically authorized by the copyright owner. We are making such material available in our efforts to advance understanding of education issues vital to a democracy. We believe this constitutes a 'fair use' of any such copyrighted material as provided for in section 107 of the US Copyright Law. In accordance with Title 17 U.S.C. Section 107, the material on this site is distributed without profit to those who have expressed a prior interest in receiving the included information for research and educational purposes. For more information click here. If you wish to use copyrighted material from this site for purposes of your own that go beyond 'fair use', you must obtain permission from the copyright owner.