An analysis of 48 states’ accountability plans under the “No Child Left Behind” Act of 2001, which were obtained directly from the states and other sources, suggests that federal officials will have their hands full in deciding which proposals meet the letter or intent of the law.
The plans detailing how states propose to comply with the law, a reauthorization of the Elementary and Secondary Education Act, raise a host of thorny questions and put forth many innovative solutions that federal officials will need to address in the coming months.
Education Week filed a Freedom of Information Act request with the U.S. Department of Education on Jan. 31 that sought access to the states’ accountability workbooks. Although the department had yet to provide the newspaper with access to any of the documents as of last week, spokeswoman Melinda Malico said it would do so this week. Meanwhile, Education Week itself collected the data from all but Oklahoma, Pennsylvania, the District of Columbia, and Puerto Rico.
The plans, which all 50 states, the District of Columbia, and Puerto Rico had to submit by the end of January, spell out how and under what timelines the jurisdictions will ensure that all students perform at the “proficient” level on state tests by 2013-14.
Under the law, schools that fail to meet annual performance targets for two years in a row, known as “adequate yearly progress,” are identified as needing improvement. Those that receive federal Title I money are subject to increasingly severe penalties if they continue to miss their targets.
Most of the plans attempt to increase the certainty that schools identified for improvement deserve that designation, and that the classification is not just a result of sampling or measurement error. (“Small Schools Pose Big Challenges,” this issue.)
Many plans also propose to raise the targets more rapidly in future years than at the outset of the law’s implementation, in an attempt to reduce the number of schools that could fail to meet their goals in the short run.
The federal Education Department must decide whether such proposals are reasonable, given the law’s demanding accountability provisions, or are attempts to circumvent the intent.
That task is unlikely to be easy for the department, given the wide variations across the states, many of which already have state accountability systems in place; the detailed, complex, and technical nature of some of the proposals; and the need to be consistent in making such judgments.
While the details can seem overwhelming, much is at stake, including continued state cooperation in carrying out the law, decisions on which students actually get included in state accountability systems, and determinations of how many schools are identified as underperforming.
‘Rich With Ideas’
All the plans are only drafts at this point. During the coming months, the department will conduct “peer reviews” with every state to determine whether the proposals comply with the federal legislation and, if not, where they must be revised. States must submit their final plans to the federal government by May 1.
“I think that the plans are rich with ideas for states to use as they go through the peer-review process, because it truly is a negotiation,” said Patricia F. Sullivan, a deputy executive director of the Council of Chief State School Officers.
“In terms of how the department will roll this out,” she added, “I think it will identify areas where we need to look harder at the regulations and, frankly, look harder at the statute.”
To date, the Education Department has approved the plans of Colorado, Indiana, Massachusetts, New York, and Ohio. (“Ed. Dept. OKs First Accountability Plans,” Jan. 15, 2003.) Letters to those states clarifying what remains to be done are expected to go out shortly. In addition, the department is planning to release a set of proposed rules soon to clarify some remaining issues.
By late last week, the agency had conducted peer reviews with another eight states and had at least a dozen more scheduled. Department officials said they could not comment on the content of the plans before the completion of the reviews.
For their part, state officials stressed that changes are likely not only because of the peer-review process, but also because the plans were due at a time when many states were changing governors, schools chiefs, and legislators following the November elections. Many of the ideas states are considering require approval by state school boards or changes in state law.
Michigan’s plan, for example, points out that the former governor instituted most of the state’s reward-and-recognition programs for schools. “With a new governor and the turnover in two-thirds of the Michigan legislature because of term limits,” it says, “it is not certain that these programs will be continued, especially given Michigan’s current budget crisis.”
State funding to provide technical assistance to Michigan’s underperforming schools has been reduced from $10 million to $2 million this fiscal year.
Moreover, while all states technically met the Jan. 31 deadline, some plans contain gaping holes, such as timelines and criteria for making adequate progress.
The law requires that states set initial targets, or “starting points,” for the percent of students who must perform at the proficient level on state reading and math tests, using data from the 2001-02 school year. Those proposed starting points vary widely by state—from 77.5 percent of elementary students proficient in reading in Colorado, to 13.6 percent of students proficient in English/language arts in grades 2-8 in California—based in part on how rigorously states define proficiency.
States must then set intermediate goals that raise those targets, like stair steps, leading up to the final goal of having every single student reach the proficient level by 2013-14.
A school must meet the targets both for its student population as a whole and for specific subgroups: racial and ethnic minorities, students from low-income families, students with disabilities, and students with limited fluency in English.
States also must identify at least one other academic indicator—graduation rates for high schools and another of their choice for elementary and middle schools—that they will use to judge whether schools have made adequate progress. Schools also must test at least 95 percent of students in each subgroup.
‘Go Slow to Go Fast’
One of the most vexing questions now facing the federal Education Department is how states set those goals.
At least a dozen states, including two whose plans have been approved by the department—Indiana and Ohio—are considering some degree of back-ended lift. They are proposing what one document describes as a “go slow to go fast” approach.
It’s based on a provision in the law that requires states to raise their goals in equal increments, with the first jump in 2004-05 and additional increases occurring within three-year spans. Essentially, the increment remains the same each time, but the states propose to raise the bar once every three years initially, and then every year closer to the 2013-14 deadline.
That means, in some cases, states are promising to improve as much in a single year toward the end of the 12-year period as they would over three years at the outset. In some states, half the total gains would be made in the last three years, from 2011 to 2014.
Reed Hastings, the president of the California board of education, reflected a rationale shared by officials in many states. He argued that it’s reasonable to expect the state to make “faster progress” once it has put in place changes required by the law, including new rules on teacher qualifications and supplemental services for students in lagging schools.
“We believe that we’ll have significant improvement between now and 2007,” he said, “but even more steep after that, once we have all the infrastructure developed.”
But in an electronic newsletter, Chester E. Finn Jr., the president of the Washington-based Thomas B. Fordham Foundation, compared the strategy to “balloon mortgages,” and suggested that states were gambling that “somebody else will be on duty” by the time the bigger payments hit.
“To believe that this approach is plausible, you have to believe that academic gains will be made in U.S. schools at an accelerating pace, indeed that as the going gets hardest—moving those last, toughest kids over the hump to proficiency—the rate of improvement will speed up,” wrote Mr. Finn, who was an assistant secretary of education during the Reagan administration. “Does that sound right to you?”
Mitchell D. Chester, the assistant superintendent for policy development in Ohio, called Mr. Finn’s criticisms “way off base.” Even the state’s initial targets, he said, require “sustained and substantial improvements,” especially for low-achieving subgroups.
For example, of 227 Ohio schools with at least 30 African-American students, more than 90 percent would have failed to make adequate progress in 2001-02 based on the performance of that subgroup.
“If you look at performance in the aggregate, you think about these targets differently than if you look at these targets for each subgroup in the state,” Mr. Chester said. “I think that’s the key point.”
Virtually all states are grappling with the fact that their initial proficiency targets are far higher than the current performance of their lowest-performing subgroups, such as students with disabilities. That reality has led at least one state, Oregon, to propose setting different initial targets for each subgroup based on current test scores, with those groups that are furthest behind having to gain faster to catch up by 2013-14. The law requires the same starting point for each subgroup.
Bill Auty, the associate superintendent for assessment and evaluation for the Oregon education department, argued that the state’s strategy “would do the most to close the achievement gap” over the long run.
“What we’re proposing is a system that we think works,” he said. “We’re going to look and see what finally gets approved, and how that matches with the law, and then determine our course.”
Defining Failure
Under the ESEA, schools are identified for improvement if they don’t make adequate progress two years in a row. Another question facing federal officials is what constitutes two years of failure.
More than a half-dozen states are proposing that schools be identified for improvement only if the same subgroup fails to meet the target in the same subject for two consecutive years.
For example, if the percentage of Hispanic 4th graders who were proficient in reading fell below the target two years in a row, the school would be identified as needing improvement. But that would not be the case if Hispanic students missed the reading target one year and black students missed it the next; or if Hispanic students missed the reading target one year, but the math target the next.
State officials argue that such an approach would target schools with a pattern of failure and reduce the potential for misidentifying schools in need of improvement based on temporary blips in performance.
But the department may disagree. Mississippi’s plan, dated Jan. 13, includes details of the revisions it made based on a peer review conducted Jan. 10. Notes the memo: “The ‘same subgroup-same subject’ method for determining failure to meet AYP over successive years has been deleted.”
In their revisions, state officials proposed that failure to make adequate progress over consecutive years be defined as missing the target in the same content area for two years running. At least 18 states are proposing that approach, including some whose plans have been approved.
An even more troublesome issue is how to define which students belong in certain subgroups, most notably children with disabilities or with limited English proficiency, and how to include them in state testing and accountability systems.
Some states are considering keeping students in the limited-English-proficient category until they score at a proficient level on tests of English fluency for two consecutive years. Other states have proposed that once children are identified as students with disabilities or limited fluency in English, their test results continue indefinitely to be calculated for AYP purposes based on their original subgroup designation.
Another option would be to identify the students that way at least as long as they remained in the same school, even after the youths had been redesignated as fluent in English or were no longer receiving special education or English-language services.
Otherwise, the argument goes, those subgroups are unlikely to meet the proficiency targets because as soon as students’ performance improves enough, they are no longer counted in those categories.
“The irony is that the more successful you are in teaching those students English, the higher your probability that the subgroup would be identified for not making AYP,” said Mark A. Leal, the director of assessment and accountability for the Alaska education department. “I’m sure that’s something that we’ll have to negotiate.”
Delaware has asked to weight the test scores of students who fit in multiple categories across those subgroups, so that their results do not count disproportionately. “That’s one we’re really pushing for,” said Valerie A. Woodruff, the state secretary of education.
The law requires that states include all students in their testing systems. Those with limited fluency in English or with disabilities must be given appropriate accommodations—such as more time—to measure their performance. States also must provide alternative assessments for students whose disabilities are so severe they cannot take the regular test, even with accommodations.
To the extent practical, states also must test English-language learners in their native tongues, although they must take reading tests in English once they have resided in the United States for three years.
Michigan has proposed allowing districts to exempt LEP students from state tests—if they’ve been enrolled in U.S. schools for less than three years—until the state devises an alternative assessment.
The proposal, approved by the state board of education Jan. 23, argues that “it is educationally unsound, unfair, and demoralizing for limited-English-proficient students to take the [state] tests they cannot read or comprehend.”
Michigan asserts it is impractical to administer the state test in the approximately 125 languages spoken in the homes of English-language learners and unfair to give it in some of those languages but not others.
Federal officials have threatened to reduce funding if the policy is not revised. “This is the administration that talks local control and flexibility, and that’s what we getneither,"said board President Kathleen N. Straus.
States also are all over the map in how they plan to incorporate into their calculations of adequate progress the test scores of students with disabilities who take alternative assessments. Most states indicate they cannot make a decision until the federal government provides more guidance, as it will in forthcoming regulations.
Perhaps the biggest challenge for federal officials is how to judge those states whose plans depart most from the letter of the law.
To retain the core features of its existing accountability system, for example, Kentucky wants to rate schools every other year, based on an index that combines test results across seven content areas, all levels of student achievement, and such indicators as attendance and graduation rates. Each school is judged based on how it performs relative to its own trajectory for getting all students to proficiency by 2013-14.
State officials argue that the system “requires schools to focus reform on the breadth of Kentucky’s curriculum,” and that the combination of individual measures is more statistically stable.
“We think that what we’re doing is complementary to what the federal law requires,” said Lisa Gross, a spokeswoman for the Kentucky education department.
Finally, federal officials must decide when a state has truly integrated the concept of adequate yearly progress into a unified accountability system for all schools. State proposals range from basing the core of their own accountability systems on the AYP judgments to simply reporting whether a school has made adequate progress on school report cards, alongside existing accountability ratings.
