States Pressed to Refashion Reading First Grant Designs

Evidence is mounting that federal employees and their agents may have directed or even pressured states to choose specific assessments, consultants, and the criteria for evaluating core reading programs as conditions for getting funding under the Reading First initiative, possibly in violation of federal law.

Education Week found such a pattern of behavior in an examination of thousands of pages of correspondence and official documentation obtained through open-records requests, as well as interviews with education officials across the country.

The close oversight of the $1 billion-a-year program has allowed a handful of commercial reading programs, assessments, and consultants to reap much of that money, while others have been shut out of the competition, according to documents and confirmation by several state officials.

Among the complaints that federal representatives have overstepped their authority:

• In January 2003, Kentucky officials complained to the U.S. Department of Education after Christopher J. Doherty, the director of the federal Reading First program, told them they would have to use what is known as the DIBELS assessment to get their grant approved. Kentucky officials also complained that a federal consultant had suggested that they hire specific experts to train teachers.

• Also that year, Georgia Reading First officials complained that a federal consultant had suggested that if the state adopted a list of core programs, it would improve the state’s chances of getting a grant.

• Illinois was told in 2003 by the Education Department to drop the use of its state literacy assessment for Reading First schools and instead use the Dynamic Indicators of Basic Early Literacy Skills, or DIBELS, even though the state’s grant proposal had already been approved and districts were beginning to implement the program.

• In 2002, Oklahoma was repeatedly advised by a federal review panel to change its selection criteria for reading texts.

All but a few states, in fact, changed their initial plans for Reading First after federal reviewers rejected or sent back their grant proposals for specific revisions. A study by the Washington-based Center on Education Policy found that it was common practice for states to adopt the DIBELS assessment and the “Consumers Guide for Evaluating a Core Reading Program” after their initial proposals were rejected. Those instruments had little to prove their worth.

Both the assessment and the guide were written by researchers at the University of Oregon, which later was chosen as one of three regional technical-assistance centers for Reading First.

As more information about the grant-approval process unfolds, former federal officials, as well as publishers and others are calling for greater transparency in the administration of the grant program.

“The federal government should have put in place, or should put in place right now if they hadn’t before, a conflict-of-interest agreement that clearly makes everybody’s advice above board,” argued Susan B. Neuman. She helped launch Reading First as the Education Department’s assistant secretary for elementary and secondary education from July 2001 to January 2003.

Federal officials, however, maintain that they pressed state and local officials only to meet the law’s demand for research-based materials, assessments, and practices and provided counsel on how they could do that.

“In fact, what we’ve said about Reading First is that there is no approved list of programs or assessment, truthfully,” Mr. Doherty said last week. “Over a couple-of-years the period, some states may say it seems as though we were pressed into this and pressed into that, but what was going on is this: This is not an ‘anything goes’ reading program.”

In correspondence to states and districts, in fact, federal officials repeatedly included reminders that they cannot help them select specific programs and products that meet the law’s requirements.

But some documents and interviews indicate that such officials, including Mr. Doherty, gave directives in phone calls and closed-door meetings.

Purposely Prescriptive

Reading First is restrictive by design. Many federal education initiatives over the past decades had high ideals but imposed limited accountability. Too often, however, the recipients did what they pleased and ended up implementing ineffective and faddish instructional approaches.

The initiative is intended to help the nation’s low-performing and disadvantaged schools provide professional development for teachers and purchase high-quality, research-based materials and assessments for grades K-3.

While the No Child Left Behind Act—particularly the Reading First initiative, which is part of the nearly 4-year-old law—is considered to be the most prescriptive of any federal education law to date, it forbids federal employees “to mandate, direct or control a state, local educational agency, or school’s specific instructional content, academic achievement standards and assessments, curriculum, or program of instruction.”

Such intrusion has long been a concern, said Christopher T. Cross, who helped draft that prohibition in the Elementary and Secondary Education Act in the 1970s and later served as an assistant secretary of education in President George H.W. Bush’s administration. The No Child Left Behind law is the latest renewal of the 40-year-old ESEA.

Mr. Cross, now a policy consultant with Cross & Joftus LLC in Danville, Calif., said the alleged interference regarding Reading First may violate that provision.

“What the feds have said and not said in regard to Reading First gets pretty close to the edge of what the law allows, or crosses the line in my view,” he said.

Education officials, publishers, researchers, and critics of the law have alleged as much since early in its implementation.

Now, documentation has turned up to support those claims. Information gathered by Education Week, including hundreds of e-mail messages between officials in several states and federal representatives for Reading First, reveal some of the negotiations. In at least some cases, states yielded—often reluctantly—to the recommendations of federal consultants, and adopted materials and tests that those states had not included in their original proposals.

Several officials said the demands made by federal officials and consultants felt intrusive and violated the rights of the state or local authority to make such decisions.

Kentucky Issues

Kentucky officials, for example, complained 2½ years ago to then-U.S. Secretary of Education Rod Paige that advice they had received from the federal review team and Mr. Doherty was inappropriate. After the state’s Reading First plan was rejected three times, Mr. Doherty told them, in a conference call, that they would have to change their choice of assessment, according to Starr Lewis, the associate commissioner for teaching and learning in Kentucky.

“During the conference call, Chris told us that we would not get funded using [the assessment we chose],” she said. “The bottom line was everybody sitting around the table, and there were six or so of us, understood very clearly that we were … to use DIBELS.”

At that point, according to Kentucky education department spokeswoman Lisa Y. Gross, “staff felt so frustrated with the whole process. … Finally, they were just throwing up their hands and said, ‘We’ll use DIBELS.’ ”

Moreover, state schools Superintendent Gene Wilhoit wrote in a Feb. 25, 2003, letter to Mr. Paige that the fact that a member of the federal review team was a DIBELS trainer raised “serious issues concerning conflict of interest,” and that the committee formed by RMC Research Corp. to evaluate which assessments met Reading First requirements included Roland H. Good III, a University of Oregon researcher who designed DIBELS.

In fact, that assessment committee included several of Mr. Good’s colleagues at the University of Oregon, in Eugene, including Edward J. Kame’enui, now the commissioner of the National Center for Special Education Research at the federal department, and Deborah Simmons. Both are authors for publishers that have competed for Reading First money. Mr. Kame’enui, who was the director of the Reading First technical-assistance center at the University of Oregon until his appointment as commissioner, and Ms. Simmons, now a professor of education at Texas A&M University, are the authors of the “Consumer’s Guide.”

In response to the Kentucky complaint, Eugene W. Hickok, the U.S. undersecretary of education at the time, wrote that the federal panel had been “appropriately meticulous in holding state plans to the highest standards.” After reviewing the state’s claims, Reading First staff members at the department did not find anything inappropriate, he said in a letter.

Mr. Hickok also wrote that reviewers had been “screened for real and perceived conflicts of interest.”

Later that year, in June 2003, Georgia officials complained to the Education Department about advice they had received from a federal consultant assigned to help them revise the state Reading First application. The consultant suggested the officials have “the courage” and “step up to the plate” and put together a state list of approved reading programs, a move that is prohibited by state law.

Mr. Doherty and Everett Barnes, the RMC president, apologized to state officials and confirmed that the law does not require states to list approved programs.

The Education Department hired the Portsmouth, N.H.-based RMC Corp. to help states revise grant proposals that were deemed inadequate. The company later won a $36 million federal contract to oversee technical-assistance centers for Reading First, located at Florida State University, the University of Oregon, and the University of Texas at Austin.

“The suggestion [by the consultant] was entirely inappropriate and shows a lack of understanding of the role and authority an agency such as yours has in implementing state law, regulation, and policy,” Mr. Barnes wrote in a June 18, 2003, e-mail message to the Georgia education department.

Mr. Barnes maintains that the Georgia incident was a misunderstanding. The language, he said, was part of the banter sent by e-mail between consultants—and inadvertently attached to a message to Georgia officials—but was not intended as official advice.

Mr. Doherty said he responded quickly to any such complaints.

Many state officials have praised Mr. Doherty and RMC for their advice and support.

“Truthfully, the technical assistance we got from the U.S. Department of Education and technical-assistance centers was mostly excellent,” said Ms. Lewis.

‘Require DIBELS’

In Illinois, the directive from the Education Department to use DIBELS came after the state’s $32.8 million grant had already been approved, in September 2002, and the state had issued grants to districts. While the proposal endorsed by federal reviewers included plans to use a state test to assess students’ reading skills, by spring 2003, representatives of the federal Education Department’s Reading First staff told Illinois officials a change was required.

“The U.S. Department of Education told us to look again at [the state test] to see if it matched their criteria. In any case, [they said], ‘We want you to require DIBELS,’ ” said Gail Lieberman, who was the manager of curriculum and instruction for the state education department at the time and worked closely on the Reading First application and initial implementation. Ms. Lieberman, who is now retired, works as a consultant to the state education department on the No Child Left Behind Act.

“Even though [our grant proposal] had been accepted,” she recounted, “and we have a letter stating that our grant was approved, [the Reading First staff at the Education Department said], ‘You will, like everyone else, use DIBELS.’ ”

Oklahoma’s application was sent back several times after reviewers questioned the plan’s choice of assessments and its mandate that districts choose only those reading programs with three years of longitudinal research demonstrating their effectiveness. Such a standard would exclude all but a “small pool” of texts, the report by the federal review panel said.

A subsequent review, which also deemed the state’s revised application “unready for funding,” again questioned the state’s restrictive approach. Oklahoma eventually included DIBELS in its proposal. The state also agreed to use the less-rigorous “Consumer’s Guide” to evaluate core reading programs, on which only a few brand-name products tend to score well, according to some observers.

Some 40 states agreed to use DIBELS as a core assessment for Reading First schools. Several have adopted it for all schools and have bought hand-held computers and database services to help teachers track students’ progress on the test. Although the assessments are free, the data-monitoring service costs $1 per pupil. The hand-held computers and other packaged sets of assessments are sold separately. Nearly all states included the “Consumer’s Guide” in their final applications as a required tool for evaluating texts.

‘Overly Intrusive’

As states were submitting their proposals in 2002 and 2003—most of which were sent back for revisions before winning federal approval—state officials were reluctant to publicly express their concerns for fear of having the federal money withheld or delayed, according to Charlotte Postelwaite, who conducted a 2003 survey of state Reading First directors as the chief education policy analyst for the Council of State Governments. For fear that there would be repercussions for states that criticized the process, the Lexington, Ky.-based council did not release the results of the survey.

But Ms. Postelwaite, who is now teaching at a Kentucky middle school, said many state officials she spoke with were upset by the demands made by the Education Department and its representatives.

“You cannot allow a vendor to dictate [grant implementation], yet with so many people [representing] the DIBELS sitting on the assessment committee and serving as consultants to Reading First,” she said, “the vendors were telling [states] what they were going to put in their grants.”

Officials in several states, she said, had to go an extra step and get approval from their legislatures to change assessments.

“We understood pretty early on that we had to be pretty prescriptive, and we had to tell districts if you want this money, this is what you have to do,” said Mike Fry, who has retired as the North Carolina education department’s chief consultant for language arts.

North Carolina officials had to ask the legislature for a waiver from the state’s assessment law—which allows only oral testing of K-2 children—in order to meet Reading First’s testing mandates.

“It felt overly intrusive,” Mr. Fry said, “but once we talked to other states, we understood that this is not just somebody picking on North Carolina.”