Orwell Award Announcement SusanOhanian.Org Home


The Mismanagement of Reading First: Summary of Evidence, Part 1

Susan Notes: We should all be grateful that Success for All personnel are speaking out about Reading First--and documenting the federal bias so meticulously. Too many academics remain silent on this issue. Does Success for All have a dog in the show? Of course they do. So should we all. There is a fundamental issue here, one that goes far beyond what reading material a given teacher uses.

People who teach mathematics should pay special attention to this report. Your curriculum is now being chosen by the Other.

Here is an article about the Randy Best enterprise mentioned in this research, Entrepreneur pursues dream of educational empire, by Joshua Benton, Dallas Morning News.


This is Part 1 of two parts.

Executive Summary
In 2002, the U.S. Congress appropriated $1 billion per year for Reading First, an ambitious program intended to place "proven methods of early reading instruction" in low-performing schools. Yet in practice, this intention was ignored by the U.S. Department of Education administrators who instead promoted the use of commercial textbook programs lacking any scientific evidence of effectiveness. Many of the key consultants entrusted with program management have serious conflicts of interest involving the very textbooks and training programs that have benefited from Reading First funding.

This document summarizes evidence assembled by the Success for All Foundation, a nonprofit organization that disseminates one of the research-proven reading programs that was largely excluded by Reading First. The summary focuses on six key questions:

1. Did the U.S. Department of Education promote use of certain reading programs?
Overwhelming documentary evidence shows that the Department of Education promoted use of five traditional, commercial basal textbooks, in violation of federal local control laws. In particular, state Reading First applications proposing anything other than these textbooks were rejected – and state funding was denied - until the states proposed to emphasize them or use them exclusively.

2. Did the U.S. Department of Education promote the use of DIBELS?
Again, overwhelming evidence documents the promotion by the federal government of a single reading progress assessment, called DIBELS, in preference to other assessments. Largely unknown before Reading First, DIBELS is effectively mandatory in Reading First-funded schools.

3. Did the U.S. Department of Education promote specific experts to provide professional development?

In the process of reviewing state applications, the U.S. Department of Education forced most states to use professional development services from a small group of selected individuals connected to Reading First leaders.

4. Did the U.S. Department of Education promote the use of the three-tier model of instruction?

The three-tier model is a reading instruction plan promoted by the three technical assistance centers funded by the U.S. Department of Education. Although it lacks any evidence of effectiveness and was mentioned in only three state's proposals, it has been aggressively promoted among Reading First schools nationally, pushing out alternative models with far better evidence of effectiveness.

5. Has the U.S. Department of Education fulfilled the Reading First Act's emphasis on scientifically-based reading research?
Although Reading First administrators and consultants speak about the importance of "scientifically-based reading research" (SBRR), research has in fact played almost no role in programs or practices promoted by Reading First. To the contrary, Reading First has promoted the use of commercial basal textbooks, supplementary texts, assessments, and professional development with little or no evidence of effectiveness in preference to well-researched alternatives. Important references to research have been interpreted to mean only the report of the National Reading Panel, from which five "key elements" of reading instruction were derived.
These five elements (phonemic awareness, phonics, fluency, vocabulary and comprehension) are present in virtually any reading program.

6. Have there been conflicts of interest among Reading First leaders and consultants?
The implementation of Reading First has been substantially influenced by a small group of consultants, many of whom earn substantial income from the publishers of programs promoted by Reading First. Reid Lyon, a key architect of Reading First, recently left government to join a company that made enormous profits from Reading First. Former Secretary of Education Rod Paige, who ran the Department of Education while the Reading First program was developed, has joined the same firm.

Congress created Reading First to direct significant resources to serve at-risk children with scientifically validated programs. Instead, these funds have been substantially diverted to forcing states and districts to purchase the products of large publishing companies that lack any evidence of effectiveness. Congress and the Department of Education must take immediate action to reform Reading First to enable it to fulfill what Congress intended the program to accomplish.


Note: This version extends and fully replaces earlier drafts of this paper.


Overview

In 2001, the U.S. Congress passed President George W. Bush's signature education initiative, No Child Left Behind (NCLB). The legislation called for the use of science to inform practice, recommending "scientifically-based research" as a basis for practice and policy more than 100 times.
One centerpiece of NCLB is Reading First, a $1 billion per year program that is providing grants to thousands of schools and districts to implement "scientifically-based reading practices" in grades K-3 in mostly high-poverty schools. Congress was very clear about what Reading First was supposed to do: "(4) To provide assistance to State educational agencies and local educational agencies in selecting … programs, learning systems, and strategies to implement methods that have been proven to prevent or remediate reading failure within a State."

The Reading First Guidance (U.S. Department of Education, 2002) likewise states:


"Quite simply, Reading First focuses on what works, and will support proven methods of early reading instruction in classrooms. The program provides assistance to States and districts in selecting or developing effective instructional materials, programs, learning systems, and strategies to implement methods that have been proven to teach reading."

Further, the Guidance defines in detail what is meant by "scientifically-based reading research. The definition requires "systematic, empirical methods that draw on observation or experiment" and acceptance by a "peer-reviewed journal or panel of independent experts." The guidance details how states and districts should review research on programs to ensure that it is scientific.

Given the clear focus of the legislation, it was widely expected that Reading First schools would adopt programs and practices with strong evidence of effectiveness. And the US Department of Education has continued to talk about science. For example, in a February 2006 Report to Congress, the US Department of Education wrote, "One of the most notable aspects of the No Child Left Behind Act of 2001 (NCLB) is its emphasis on the use of scientifically based research to ensure that Federal funds are targeted to educational programs and practices that have evidence of their effectiveness. NCLB aims to end the use of unproven practices and methods that may actually be harmful to students and detrimental to student achievement. Prior to NCLB, educational fads were often the driving forces behind the selection of programs and practices. The law's focus on scientifically based research is helping to prevent the use of untested practices in our Nation's classrooms…." Yet the reality is exactly the opposite.

In practice, the U.S. Department of Education and its contractors administering Reading First have not only ignored programs with strong evidence of effectiveness, they have actively worked to exclude the few reading programs that do have strong evidence. In particular, two programs have suffered under Reading First: Our own non-profit Success for All program, and Direct Instruction. (Direct Instruction (DI) uses a textbook, Reading Mastery, that is published by McGraw-Hill, and supplements it with extensive professional development, usually from the nonprofit National Institute for Direct Instruction (NIFDI). Reading Mastery, the book, has been allowed under Reading First, but the full DI program has not.)

Success for All and Direct Instruction are by far the most extensively and successfully evaluated of all reading programs. The Comprehensive School Reform Quality Center at the American Institutes for Research (CSRQ, 2005) recently gave Success for All and Direct Instruction its highest ratings for evidence of effectiveness among 22 comprehensive school reform models, rating 31 studies of SFA and 10 studies of DI as "conclusive." A 2003 article in the Review of Educational Research (RER) by Borman, Hewes, Overman, & Brown identified 46 rigorous experimental-control comparisons evaluating Success for All and an essentially identical program called Roots & Wings, of which 29 were third-party evaluations. The RER review found 40 experimental-control studies of Direct Instruction, of which 38 were third party (also see Adams & Engelmann, 1996). Most recently, a national randomized evaluation once again found substantial positive results of Success for All on reading outcomes (Borman et al., 2006). Many of these studies of both programs were published in the most selective journals in education. Published reviews by Herman (1999), Traub (1999), and others have also concluded that Success for All and Direct Instruction have solid, replicated evidence of effectiveness for the specific populations targeted by Reading First. In fact, the Florida State Technical Assistance Center, which reviewed evidence on programs other than the traditional basals, concluded that there is substantial evidence supporting the effectiveness of SFA

Yet these programs have been substantially shut out of Reading First. Instead, Reading First funds have overwhelmingly gone to support traditional basal textbooks-Scott Foresman, Macmillan/McGraw-Hill, Harcourt, Houghton-Mifflin, and Open Court. Schools are also adopting commercial supplemental materials, including Voyager Passport. Even though Success for All and Direct Instruction are theoretically eligible for Reading First funding in many states (they appear on many state lists of "approved" reading programs), only two schools (out of almost 4800 schools funded) have used Reading First funds to adopt Success for All. About 3% of Reading First grants have gone to schools already using Success for All. Even this modest proportion is being eroded as Reading First personnel pressure schools to drop SFA in order to keep their Reading First funding (see below).

There is a very limited research base for the achievement effects of traditional basals. Of the five top-selling basal textbook programs, only Open Court has ever been evaluated in a published study. Yet when the U.S. Department of Education selected a state to feature in a White House celebration of Reading First, it selected Michigan – a state that listed these top five commercial basal series as its core reading programs for Reading First, without even pretending to review their scientific basis. Michigan's application was among the first to be accepted by the Department of Education, sending a clear signal to all other states: Reading First is about commercial basals, not science.

Why did the Department of Education violate the clear intent of Congress to have Reading First grantees use programs with strong evidence of effectiveness? One clue appears in a recent interview with Reid Lyon, a principal designer of the program (Salvato, 2006):

"What we originally wanted in Reading First was that if you want to buy a program with federal money, it should have gone through clinical trials to be sure it was effective. But there weren't enough programs that went through that level of rigor…only a limited number of programs would be available. The Department of Education made the decision to make the criteria more general."

The programs Lyon mentions that had "been through clinical trials" could only have been Success for All and Direct Instruction. What he was saying is that because too few programs had been rigorously evaluated, Reading First had to greatly loosen the criteria. What is a mystery, however, is why the Department then failed to highlight the few programs that did meet their original standard of evidence, and in fact appears to have instead directed grantees toward the unresearched basal series produced by the large commercial publishers.

One reason may be that the Reading First program has been compromised by its reliance on several key leaders with significant connections to textbook publishers. Among three technical assistance centers charged with management of the program, two were led by individuals who were both authors of the 2007 Scott Foresman basal and members of the design team for Voyager Passport. The company that produced Voyager, estimated to be worth $5 million before Reading First, was recently sold for $380 million. Reid Lyon himself, who was reported to have personally forced New York City to adopt Voyager or risk losing its Reading First funding, later left government to work for the entrepreneur who founded Voyager. Secretary of Education Rod Paige, in office during the implementation of Reading First, also has now joined this same company. Reading First virtually mandated use of a reading assessment called DIBELS created by a leader of one of the technical assistance centers. Individuals with conflicts of interest were sent as technical advisors to states revising their Reading First proposals. Publishing companies with ties to these individuals were given special opportunities to ensure that their materials qualified for Reading First funding. These and many other conflicts of interest are far beyond what is normal in government, and are deeply troubling.

After the state awards were made, things went from bad to worse. Reading First national staff and technical assistance centers strongly promoted specific instructional practices to be used at the classroom level. In the jargon of Reading First, this is called a "three tier" model. The Department of Education continues to promote this model nationwide, strongly suggesting that schools receiving Reading First funding must implement it, even though nothing in the authorizing legislation said anything about it. The model involves teaching children using basal texts (tier 1), assessing their progress, giving additional instruction if needed, and then providing a 30-minute small group remedial instruction program for children who do not meet standards (tier 2). Those who still do not succeed are given more extensive intervention and ultimately referred to special education (tier 3).

While there is nothing wrong with the idea that struggling students need additional instruction, the 3-tier model promoted to Reading First is very specific about grouping strategies, use of group rather than one to one remediation, and many other particulars. This 3-tier model provides one way of structuring instruction, but there is no evidence that it enhances children's achievement. The outcomes of the three-tier model have never been evaluated in comparison to control groups. Before even a single experimental study has been published, Reading First technical assistance contractors have been promoting the three-tier model nationally, making Reading First, which was supposed to focus on proven programs, instead a $6-billion-dollar pilot test. Because the Success for All program uses a somewhat different structure – notably, it combines "tier 1" and "tier 2" so that children who struggle or fall behind receive help immediately - in state after state, Reading First schools using the Success for All program are being pressured to drop or eviscerate it in favor of a basal program, even in circumstances when the schools have made substantial gains on their state assessments and on DIBELS, the assessment favored by Reading First.

It is difficult to imagine that reading outcomes for at-risk children will magically improve because schools use traditional basal texts – in most cases, simply more recent versions of the ones they've always used. In the long run, reading outcomes will only improve when teachers receive high-quality professional development on programs that are known from rigorous research to improve student achievement. Reading First is a giant step backward. It is being used not to achieve the noble goals of the NCLB legislation but to instead substitute unresearched products of commercial publishers for programs that are truly scientifically validated.
This paper updates the evidence collected by the Success for All Foundation to document the mismanagement of Reading First.

Sources of Information

Information on the administration of Reading First comes from many sources. The most useful are as follows.

State Reading First Proposals
We have obtained the final Reading First proposals from most states, as well as initial drafts in selected states.

Reviews of Successive Drafts of State Reading First Proposals
We obtained Department of Education reviews of successive drafts of almost all state Reading First applications. Most states had to submit their RF applications many times; Rhode Island submitted six versions over an 18-month period. We obtained each of the successive drafts of Reading First proposals from several states, and matched them to their reviews.

Reading First Monitoring Reports
Under a Department of Education contract, the American Institutes for Research carries out annual reviews of implementations of Reading First, including subgrants to individual districts and schools as well as programs implemented at the school level. Reading First has rescinded funding to districts and schools based in large part on these reports, so they are taken very seriously.

RMC's Funded Proposal to Create the Technical Assistance Centers
RMC was the only bidder who responded to an RFP to set up the RF technical assistance centers at the University of Oregon, University of Texas, and Florida State. The proposal approved by the Department of Education provides a clear blueprint for the later stages of program implementation.

Articles in General and Education Publications
Investigative articles on Reading First have appeared in USA Today (Toppo, 2005 a, b), Education Week (Manzo, 2005 a, b, c), Title I Monitor (Brownstein & Hicks, a, b) and Education News (Salvato, 2006). These include interviews with Department of Education offices, ED contractors, state RF leaders, and many others, as well as reviews of many documents.

PowerPoint Slides, Binders, and Handouts from Reading First Academies and Other RF Events
Materials distributed by the U. S. Department of Education cover topics such as selecting instructional materials, defining scientifically based reading research, and the three-tier model.

Financial and Outside Employment Disclosure Forms for RF Consultants
Dr. Edward Kame'enui, now Assistant Commissioner for Special Education Research in the U.S. Department of Education, was required to file a financial disclosure form. Dr. Sharon Vaughn, a professor at the University of Texas at Austin, is required to file annual notices of external employment. We were informed that Reid Lyon was exempt from this policy. The University of Oregon, which employed Drs. Simmons and Kame'enui, refused to provide any financial disclosure documentation.

Information from Web Sites for Technical Assistance Centers, Publishers, and Other Organizations
These include materials ratings, information on the three-tier model, and membership of Reading First consultants on design teams for commercial products as well as links to those commercial products.

Emails To and From State RF Officials
Under the Freedom of Information statutes of each state, we obtained emails relating to Reading First from state RF leaders and others.

This document draws on these and other sources to answer a set of questions we, and others, have raised about the Reading First program. In addition, the Success for All Foundation continues to press the US Department of Education for additional documents relevant to the operations of the Reading First program.


Did the Department of Education Promote Use
of Certain Reading Programs?


The No Child Left Behind Act of 2001 prohibits the Department of Education from promoting particular curricula or textbooks:

"(b) LOCAL CONTROL.-Nothing in this section shall be construed to-'(1) authorize an officer or employee of the Federal Government to mandate, direct, review, or control a State, local educational agency, or school's instructional content, curriculum, and related activities;" PL 107-110, sec. 9526, General Prohibitions

From the beginning, Department of Education officials have claimed that there is no sponsored or approved list of reading programs for use in Reading First, and they have maintained this position over time. In 2005, in a letter to Senator Richard Lugar (R-Indiana), Assistant Secretary Ray Simon stated, "It is clear that Congress did not intend for the federal government to decide which scientifically based reading programs would be used with Reading First, and we have strongly supported the core principle that States, districts, and schools make these decisions" and "We have repeatedly stated that the Department does not approve reading programs…." The ED Web site repeats this assertion: "Just like every other aspect of No Child Left Behind, states and local communities maintain control.

• States and local schools have the flexibility to determine how reading programs are selected, as long as the selected program has been scientifically proven to work.

• There is no federally prescribed reading program.

• States are responsible for the quality of the local programs they fund, and for ensuring that these programs rely on scientifically based reading research." http://www.ed.gov/programs/readingfirst/nclb-reading-first.html

Yet the evidence is overwhelming that there is indeed a list of favored reading texts. It consists of the six top-selling commercial basals, as follows:
Basal: Publisher
Scott Foresman: Pearson
Harcourt Trophies: Harcourt
Macmillan: McGraw Hill
Houghton-Mifflin: Houghton-Mifflin
Open Court: McGraw Hill
Reading Mastery: McGraw Hill

Other basal texts and reading programs are used in various states, but the numbers are comparatively small. For example, approximately 3% of schools received RF funding to continue implementation of Success for All, and a few other programs are also used in isolated schools.

How Did the Department Promote Its Favored List of Textbooks?
From comparing the state proposals to the federal reviews, reading interviews reported in the press with state officials involved with Reading First, and speaking with individuals who served as federal reviewers, the process that led states to end up with this list of basal series has become clear. The federal reviewers made it very difficult for state proposals to be funded. They made detailed critiques in each of many areas, including materials, assessments, and professional development, and states had to pass in each area separately to receive funding. Most, and perhaps all states had to revise and resubmit their proposals at least once, and some had to do so up to six times. Each resubmission meant delay in funding and the possibility that states would never be funded, so the pressure on state proposal writers was intense. Some state officials were fired or transferred when their proposals were not accepted. At one point, then-Undersecretary Eugene Hickok threatened states that if they did not quickly submit acceptable proposals, their funding would be redistributed to other states whose proposals had been accepted.

After each unsuccessful evaluation, top state officials had a telephone conference with Chris Doherty, the federal Reading First director. It was in these conversations, we believe, that states were pushed toward the favored basals and assessments. There is no indication that Doherty specified textbooks by name, but we have been told that he did recommend that state officials "look at the Oregon list" (i.e., the six textbooks) or "look at the Michigan list" (five of the six). For example, a North Dakota State Department official was quoted by Brownstein & Hicks (2005a) as saying, "Even though there was no approved list of assessments or core programs, you don't get approved unless you have certain assessments or core programs. There must have been a list somewhere." Although the reviewers never suggested specific basal textbooks, as soon as a state limited itself to a set of basals from the above list, and excluded all others, reviewer criticism in the "materials" category ceased. Even after their proposals were approved, several state departments, including Massachusetts, Illinois, Maine, and Kentucky, were contacted by Doherty and told to reduce their list of recommended basals to remove programs other than the favored six.

Again, although rumors abound, we have no independent reason to believe that the Department has ever published a list of the basal programs favored by Reading First. However, it has engaged in a consistent set of practices that have had the same effect. These are as follows.

1. In the initial presentations where states learned about the requirements of Reading First, known as the Secretary's Reading Leadership Academies, presentations on selecting instructional materials showcased basal textbooks (specifically, Harcourt Trophies, Houghton-Mifflin, and Open Court) and state adoption lists (California and Texas) as examples of acceptable materials. These examples were given in a binder distributed to all participants in the first Reading Leadership Academy. In 2002, the Association of American Publishers questioned the U.S. Department of Education on the provision of these "examples." Although ED, approximately 4 months later, published a letter on its web site clarifying that programs cited in meetings were intended only as "examples," the agency never took serious action to counter the widely held public perception that the basal programs mentioned, or standard commercial basals in general, were intended to be used in Reading First schools. Some states referred to "the USED approved list of materials" in their Reading First proposals. States that chose the any of the six favored basals were never criticized for doing so. Although ED has stated many times that there was no "approved list," the similarity of programs both approved and in use across states seems to indicate that states and districts understood that they should propose to use basal textbooks listed by the Department itself as exemplars (see Manzo, 2005 a, b).

2. Shortly after the Reading First legislation was approved, the Florida Center for Reading Research (FCRR), which later became one of the three technical assistance centers, produced a review of the research based behind various reading programs. It gave high ratings to Success for All and Direct Instruction. Soon thereafter, it removed these ranks from its web site. Instead, its web site provided narrative reviews of various programs, including information on the research behind each. However, it stated that six basals, the same ones favored on the Oregon List and favored by reading First in all of its practices, did not need to be reviewed. These are: Harcourt, Houghton-Mifflin, MacMillan, Scott Foresman, Open Court, and Reading Mastery. Although the narrative reviews gave high marks to Success for All and Direct Instruction for their research support, the statement that the six favored basals did not need to be reviewed for their research base clearly indicates that these are "safe" choices, while others require a level of evidence that the favored basals do not need (and, not incidentally, do not possess).
Significantly, one of the basals, Scott Foresman, was included on the "safe" list only if it included a supplement designed to fill in apparent gaps. No other program was permitted to submit a supplement to improve its rating, and the exact Florida list, with its requirement for the supplement, appears verbatim and in the same order on several approved state applications.

3. Although the Department of Education never told any state in writing that it must use a basal, according to media reports, several state directors were told this verbally by Chris Doherty, the director of Reading First. In general, ED referred to the above list by referring states to the "Oregon review," or the "Oregon website." In practice, the Oregon review produced a list of the six basals listed above. Originally, there were seven programs ranked highly by the Oregon review; Success for All was listed fifth. Yet states that adopted or referred to the Oregon list invariably listed the six textbook programs, omitting SFA. In 2004, without any additional review, the Oregon list was "updated" to rank SFA seventh, but this was after all Reading First grants had been made to the states.

The Oregon list, which generated complaints from the Association of American Publishers, among others, for its lack of a transparent review processes (Brownstein & Hicks, 2005b), was created by a state panel dominated by researchers from the University of Oregon, who later became the leadership of the University of Oregon's RF technical assistance center. The University of Oregon was the home of one of the programs, Reading Mastery. Three of the University of Oregon researchers were authors of the Scott Foresman 2007 basal and two of them were authors of an earlier Scott Foresman remedial program. Three were members of a four-member external design team for Voyager (the fourth member was Sharon Vaughn, from the University of Texas, who also is extremely influential within Reading First – see below).
Why was the Oregon list, and no other, supported by Reading First? There were several other states that carried out equally extensive analyses of reading programs, including a coalition of the States of Washington, Idaho, Montana and Alabama that considered research on program outcomes; this review was even available earlier than the Oregon list. As noted earlier, in 2002-03, the Florida Center for Reading Research (FCRR) at Florida State University carried a detailed review of reading programs that did consider evidence of effectiveness for programs other that six favored basals. Both the Washington coalition and the FCRR gave very positive ratings to Success for All and Direct Instruction, because of the extensive research behind them. The Oregon review did not consider research on program effectiveness. Yet to our knowledge, ED never directed states to look at the Washington coalition or Florida research reviews, nor at any other state reviews. Absent other evidence, it appears that Oregon is emphasized because unlike the Washington Coalition or other states its list corresponds exactly to the list of the six top-selling commercial basals, which the Department has promoted under Reading First.

4. The Michigan proposal was among the first state RF proposals to be funded, and it was the first state to distribute funds to districts and schools. Michigan did not carry out its own review of the degree to which various textbooks reflected principles of scientifically-based reading research, but instead simply listed the five top basal series. The quick approval of the Michigan proposal sent a powerful message to other states, that Reading First is intended for use with commercial basal series, regardless of evidence. Later, in a celebration of Reading First at the White House and in an address with Reid Lyon at NIH, President Bush highlighted the Michigan Reading First program as an exemplar.

5. The research base for the basal programs proposed for use in each state was of no consequence in ED reviews of state RF applications. States such as Michigan, Wisconsin, and Rhode Island simply designated the top five basals, without any pretense of reviewing research on these programs or on the principles they incorporate. In fact, in describing its process of reading program selection, Michigan wrote that its procedure was simply "to place phone calls to all major publishers requesting examination copies for review." In the case of Rhode Island, discussed below, reviewers did not ask for any justification for the basals on the Michigan list that Rhode Island adopted verbatim under pressure, although they repeatedly asked for detailed and rigorous reviews of evidence on any basal not on the Michigan list. California specified just the two basals on its state adoption list, and many other states also simply designated their state adoption lists, drawn up according to completely different standards that had nothing to do with SBRR, as their approved lists for Reading First. These state adoption lists primarily consist of the top five basals, produced by large companies with the resources to go through the state adoption processes.

6. Oklahoma proposed to fund RF schools to adopt programs that had at least three years of longitudinal data, a procedure entirely consistent with the definition of "scientifically based research" in the Reading First legislation. Through two drafts with versions of this provision, Oklahoma was criticized on the basis that this restriction would limit schools to a small number of programs. When Oklahoma dropped this language, but instead limited schools to traditional basals from its state adoption list, the criticism on this topic ceased, and the fourth draft was approved for funding.

7. Rhode Island had to submit six revisions of its Reading First application. We obtained all six versions and the corresponding reviews. They make it clear how the Department forced states to use the favored commercial basals, and nothing else.

In its first two proposals, Rhode Island would have required that LEAs
purchase "high-quality reading programs that meet the test of having a scientific research base and are "comprehensive" and "systematically and explicitly address all of Reading First's five…essential components of reading…."

Federal reviewers rejected this proposal, on the basis that it "does not include the rigorous and clearly defined standards the State will use to evaluate the research base of instructional programs and strategies."

In its third draft, Rhode Island inserted the Michigan list of commercial basals: Houghton-Mifflin, Harcourt, Open Court, Macmillan, and Scott Foresman. The Rhode Island list was obviously copied from Michigan; it is in the same font, and includes a requirement that Scott Foresman be supplemented with additional material on fluency, as was the case in the successful Michigan proposal (and first suggested on the FCRR web site). However, Rhode Island's third draft also allowed districts to propose other programs if they justified them based on SBRR.

The federal reviewers again rejected this formulation. In its fourth draft, the "Instructional Strategies and Programs" section was virtually identical to that in Draft 3, except that it deleted one sentence:
"LEAs that use other high-quality programs…"
This section was finally accepted. As soon as Rhode Island limited its schools to the five basals from the Michigan proposal (which Michigan itself accepted with no scientific review whatever), the reviewers had no further concerns about their reading programs.

8. Wisconsin had to submit four revisions of its Reading First proposal before it was accepted. We obtained all four drafts. The progression is a mirror image of Rhode Island's experience. In its first two submissions, Wisconsin proposed a process of allowing districts to justify the choice of any core reading program according to SBRR. Reviewers criticized this as not specific enough. Finally, the state imported verbatim into its proposal the Michigan list. As in Rhode Island, the list is in the same font as the Michigan proposal and, like Michigan, requires Scott Foresman to provide a fluency supplement.

Wisconsin schools could theoretically apply to implement programs not on the list, but if they did so, they had to submit their own Consumer's Guide ratings and take a substantial chance that their proposals would be rejected if they made the wrong choice. Clearly, the safe option for schools was to propose to use any of the five basals.

9. In its approved Reading First application, Maine did not specify specific basals. However, after the proposal was approved, Chris Doherty discovered that the state was restricting schools to two basals, Scott Foresman and Rigby. He wrote to the Maine Commissioner to tell her that Rigby "does not appear to be aligned with scientifically based reading research." The state substituted Houghton-Mifflin for Rigby, and this was accepted with no review of Houghton-Mifflin (or of Scott Foresman).

10. Maryland simply adopted the Oregon list, in this case including Success for All. It had a panel do "Maryland annotations" on the Oregon list. The "Maryland annotations" criticized Success for All because the grades 2-3 materials had been revised since the Oregon review. Maryland's review team was so constrained by the Oregon review that it would only approve the older edition, solely on the basis that this is what Oregon reviewed.

11. Vermont's approved application included a suggestion that districts "give careful consideration" to programs recommended by other states. It gave the Michigan, California, and Washington lists as examples. As noted earlier, neither Michigan nor California even pretended to do scientific reviews of their recommendations. Michigan's list was simply the top five basals, and California's was Open Court and Houghton-Mifflin. This aspect of the Vermont proposal was accepted without comment by federal reviewers.

12. Among the states from which we received all drafts of state proposals, several did not specify a list of permitted or recommended textbooks. Yet the ultimate effect of limiting schools to textbooks on the favored list was the same. Georgia Reading First school applicants were required to review the language arts textbooks they had just adopted from the Georgia state adoption list according to their fit with SBRR. That list included all of the Michigan list basals (Harcourt, Houghton Mifflin, MacMillan, Scott Foresman, Open Court). As happened in many states, the Georgia state adoption list became the de facto Reading First list, even though state adoption processes never consider evidence of effectiveness. A virtually identical process took place in Indiana, where applicants could keep the textbooks they had just adopted from the state's adoption list, as long as they submitted a review of their already-adopted basal using the Consumer's Guide.

13. The states described above are only unique in that these are the states from which we had access to all proposal drafts. In other states, however, we have the reviews of each successive proposal and the final proposal, and the pattern is equally clear. The federal reviewers kept complaining until the state settled on any subset of the top commercial basals, or settled on a process that would have the same result.

14. In multiple federal reviews of state proposals we obtained for almost every state, there is not a single criticism of any state for restricting RF grants to schools using any of the favored basals. In contrast, criticism for states suggesting other programs is constant. Similarly, annual monitoring reports criticize schools that chose programs other than these basals, and never criticize schools for choosing any of the six basals.

15. The State of Illinois initially approved a long list of reading materials for use under Reading First funding. According to correspondence between ED and Illinois, and e-mails internal to the Illinois State Board of Education (ISBE), ED officials pressured Illinois to significantly reduce its list of approved texts. After working with the University of Oregon, Illinois agreed to limit schools to the same five basals adopted by Michigan. Districts that had not initially adopted one of these programs were informed, in writing, that their grants would terminate unless they switched programs. Districts were notified that they could pay for reviews of programs not on the list of five, and such requests were made by districts wanting to implement Success for All. However, ISBE never honored –or even acknowledged- any of these requests.

16. States whose initial Reading First applications were not approved (in practice, nearly all states) were assigned "technical assistance consultants," many of whom were affiliated with basal textbooks or related programs. For example, the Kentucky Commissioner complained to ED about a consultant who is a certified DIBELS trainer (Brownstein & Hicks, 2005a). The Illinois consultant was a trainer for the Sopris West publishing company. Because these consultants were assigned to states after their initial applications had been rejected, the consultants had great influence on state officials anxious to receive their Reading First funding.

17. According to the funded technical assistance proposal submitted by the RMC Research Corporation, ED was complicit in the pressure that states were under to use basal textbooks. In its application to establish the Reading First technical assistance centers, RMC proposed to develop professional development strategies to support the products of "major publishers." This was apparently accepted without comment by the Department of Education.

18. Since 2003, Reading First has been aggressively promoting a "three-tier model." The three-tier model involves teaching using a basal reader, then providing a supplemental program (tier 2) for students who do not succeed in core "tier 1" instruction. In many states, Success for All and other programs are being forced out of Reading First schools because of a perceived "lack of fit" with the three-tier model. About one third of all Success for All schools that once received Reading First funding have been forced to drop Success for All or risk losing their RF funding. The Reading First emphasis on the three-tier model has had the effect of promoting the use of commercial basal series, and of commercial supplemental texts, such as Voyager Passport and the Scott Foresman Early Reading Intervention.

The Department of Education denies that the three-tier model is mandatory under Reading First, claiming that states choose their own models of instruction. Overwhelming evidence shows that it is heavily promoted, however. The three-tier model has been extensively presented at all of the national Reading First conferences, as well as to national gatherings of the state Reading First directors. It is described in detail on the websites of the University of Texas and University of Oregon technical assistance centers. Schools and districts that deviate in any way from the elements of the three-tier model are criticized in the annual monitoring reports, and can lose their Reading First funding on this basis. New Mexico, after receiving two consecutive negative monitoring reports, developed – in close consultation with ED – a list of "non-negotiables" under Reading First. This list of non-negotiables, which every RF site had to sign an assurance that it would follow, requires the three-tier model.

The RMC proposal to establish the RF technical assistance centers stated that it would introduce and disseminate the three-tier model as the core of its professional development plan for Reading First. The three-tier model, which has never been evaluated in even a single published study in comparison to a control group, has become the de facto instructional program for thousands of schools, yet it was unheard-of before Reading First and would likely, absent this level of promotion, never have been adopted by these schools.
Because the three-tier model promotes the use of standard commercial basal and supplementary textbooks and opposes the use of programs such as Success for All and Direct Instruction that use different instructional strategies, it is one more means by which Reading First supports use of unresearched commercial programs in preference to scientifically proven programs.

19. In a 2003 speech in New Jersey, Reid Lyon, the NICHD official who was an architect of Reading First, stated in response to a question that states and schools would be wise to use any of the favored commercial basals in their Reading First applications, and gave several examples. Presumably, he presented similar opinions in many speeches elsewhere, as various state documents indicate his frequent speaking engagements regarding Reading First.

20. A 2005 article in an Ohio newspaper noted how at McGraw-Hill, publisher of three of the six "Oregon list" basal texts, "business was booming" due to No Child Left Behind, and they added 500 employees in Ohio alone. As noted earlier, Voyager, whose design team was primarily composed of individuals who were later involved central in Reading first technical assistance centers, increased in value from about $5 miillion before Reading First to $380 million in 2005. A the same time, since Reading First began, the staff of the nonprofit Success for All Foundation has been cut by 60%. Whether by intention or not, Reading First has had an unequivocal effect in increasing the use of commercial programs lacking evidence of effectiveness and reducing the use of non-traditional programs that do have strong evidence of effectiveness – exactly the opposite of what the law requires.

— The Success for All Foundation
Revision


INDEX OF RESEARCH THAT COUNTS


FAIR USE NOTICE
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.