Targeting Subgroups

How students with disabilities and limited English are unfairly measured under NCLB by Terri Duggan Schwartzbeck
By its very title, the No Child Left Behind Act makes a promise to consider the needs of individual students in raising student achievement. But as superintendents across the country have found, this presents a major challenge.

As states have released their lists of schools that failed to meet this year’s targets, one issue in particular emerged. Many schools are meeting the goals in all but one or two subgroups: limited English proficient students and students with disabilities. By requiring these groups of students meet the same targets at the same times as all other students, the law imposes a one-size-fits-all approach that ignores the individual child.

Several key issues in assessing LEP students and students with disabilities under NCLB are also emerging. The first issue is that, by definition, LEP students are not proficient in English, and by definition, students with disabilities have special needs that caused them to be labeled as such in the first place.

Second, within the accountability structures of NCLB, how are those students assessed and counted? There are significant variations from state to state in the types of modifications available and how groups of students are counted toward adequate yearly progress.

Finally, in the massive communications challenge presented by NCLB and AYP, this is perhaps the most challenging. How can and should school leaders face the issues these groups of students present, while taking care not to blame the students and still holding themselves accountable for continual improvements in performance?

Disabilities Assessed
NCLB is not the first law to require that students with disabilities be tested. The Individuals with Disabilities Education Act long has required these students be tested and that those test scores count toward something. What changed under NCLB is that now schools are accountable for having all special education students—minus one percent—meet the same standards and test score goals as all other students in the school—and at the same time.

In the past, school systems dealt with the needs of assessing special education students in a variety of ways. First, the individual education plan agreed to by the child’s parents and teachers always has taken priority. The IEP spells out goals and objectives for the child and how the student will be assessed.

One way this was achieved was with out-of-level testing. If a 5th-grade student was reading at the 3rd-grade level, then he or she took the 3rd-grade reading test. A student also might have an option to take an alternative assessment (if he or she has a low cognitive function).

Students with disabilities and those lacking English proficiency also can receive accommodations for taking an assessment, such as extra time or having questions read aloud.

NCLB puts limits on many of these methods. Now out-of-level testing is prohibited (although that prohibition was waived for this year). And according to regulations proposed in March 2003, just 1 percent of students—those with the most severe cognitive disabilities—may take an alternate assessment that holds them to standards more appropriate to their abilities, rather than the 3 or 4 percent that had been taking it previously. In some school districts, such as those located near major medical centers or that had particularly strong special education programs, that number might be even higher. Further, in special services districts—those that serve only students with special needs—and in some educational service agencies, all students being served have special needs.

States do have considerable flexibility in some aspects of how schools and districts will be held accountable for the achievement of special-needs students. Some offer different modifications for assessing students, while others have detailed performance portfolios linked to state standards that are used to assess students with special needs. But these policies vary widely from state to state. Further, each state had the option of choosing a specific minimum subgroup size, or “N” value. If a state chose an N of 30, for example, a school with 50 students in a tested grade but only 15 special education students would not be held accountable for those 15 because the subgroup size is too small to be statistically reliable.

Individual Focus
As a result of these policies, many school leaders are in a difficult position.High numbers of schools have been identified as failing to make AYP based solely on the performance of students with disabilities. In Byng, Okla., Superintendent Steve Crawford has one school with 38 students in 5th grade, 17 of whom are students with disabilities, and 14 of those special education students did not pass the test. This put the whole class at a pass rate below the target set by the state.

“I’m more worried about what we’re doing with these kids than I am about the school failing [to make AYP],” Crawford says. “We’ve got these 14 kids who will feel like it’s their fault the school failed. We’re telling kids they’re failures.”

While his state took the innovative approach of choosing a higher subgroup size for its special education students (52 for special education, 30 for all other groups), that didn’t help in this case. Instead, while his special education students were not held accountable as a subgroup, their performance still affected the whole school because the class was so small.

Crawford is concerned that those three students with disabilities who did pass the test might have been misplaced in special education. Charlie Kyte, executive director of the Minnesota Association of School Administrators, had the same reaction when the first run of test scores in Minnesota identified almost all of the alternative high schools in the state as failing to make AYP. For schools that did meet the target, an independent audit might be needed, he indicated.

These issues lead many educators to wonder why the focus does not remain on the IEP and the goals and objectives it sets for individual students. The IEP should remain at the forefront, says Crawford. “Let’s make sure the child grows annually. Let’s set targets and try to reach them. I want those children to have targets they can reach.”

Limited English
As with special education students, NCLB does not represent the first time LEP students have been assessed either. The Elementary and Secondary Education Act of 1994 required states to “adopt a standards-based system where all students, including limited English proficient students, also known as English language learners, are expected to reach the highest standards.” However, unlike NCLB, it did not specify that the standards would be the same level and at the same time as all other students. And, like special education students but unlike all other subgroups, the very definition of LEP students as not proficient in English creates a conundrum.

LEP students are unique among all of the NCLB-required subgroups (which also include race and ethnicity, poverty and disability) in that students tend to exit the subgroup over time upon attaining proficiency in the English language while new ones continue to enter. This leads to another key policy area that states had to make decisions on in the accountability plans they negotiated with the U.S. Department of Education throughout spring 2003.

Some states, including California, Ohio and South Carolina, stop counting a student’s scores as part of the LEP subgroup as soon as he or she demonstrates language fluency. Others require a passing score on the fluency assessment for two years in a row. Still others states continue to provide monitoring and follow-up services to the student and thus continue to count them as LEP for up to three years in some states.

This gets at the heart of one of the most complex elements of NCLB: rather than basing annual progress on a cohort of students, for example, looking at third grade performance one year, then that same group of students as fourth graders the next year, NCLB takes a snapshot of each grade each year. This year’s third grade student performance is compared to next year’s third grade students. In the case of LEP students, by continuing to take snapshots of the subgroup, the effect of students gaining fluency in English and exiting the group while new students continue to move in, the gains made by those students are often lost.

Non-English Speakers
Like special education students, LEP students also may have access, depending on the state’s policy, to other modifications and accommodations in taking assessments, such as simplified English versions or tests written in the student’s native language. However, tests may not be available in all languages spoken by students as this could be cost-prohibitive.

In addition, state policies on testing students with little or no proficiency vary. Some states, such as Virginia, found their existing testing policies for LEP students landed them in hot water. That state allows students who have just moved to the United States to be exempted from assessments for the first year, a policy that was not in compliance with NCLB, which requires that 95 percent of all students be tested. The waivers put the numbers in many schools well below 95 percent and flagged the schools as not making AYP.

Virginia strongly objected to this requirement. "To expect a newly arrived child to speak English and pass is ridiculous," Fairfax County Superintendent Daniel A. Domenech told The Washington Post. In a letter accompanying Virginia’s final accountability plan (one of the last to be approved by the U.S. Department of Education), Virginia Board of Education President Mark C. Christie stated that amendments in the plan complying with this requirement were only there because “USED has mandated them, and we agree only under strong protest. We do not believe these amendments represent sound or rational policies, especially the intention of USED to apply future testing policies, to which we have already agreed, to this past academic year on a retroactive basis for the purpose of determining AYP for Virginia schools.”

Disqualifying Scores
Just across the Potomac River in Montgomery County, Md., many schools found themselves in a conundrum over policies relating to both students with disabilities and students with limited English proficiency. At the heart of the debate was the question, “what is an accurate test score for LEP and special education students who receive accommodations while taking an assessment?”

On this year’s 3rd-grade statewide reading test, several questions were read aloud to some students who were not able to physically read the test. In some cases, students were visually impaired or LEP students who could understand spoken English still had difficulty reading from a page. This was done in accordance with existing policies requiring such accommodations be provided and with the individual education plans of some of the special education students.

However, somewhere in between the end of the test and the score reports received by the school district, the state education agency invalidated that section of the test for those students, and the entire test was given the lowest possible score. That figure was used in calculating AYP, resulting in far fewer schools in Montgomery County making AYP than would have been the case without the substituted scores.

That part of the test was invalidated was not a surprise to Superintendent Jerry Weast. In the past, that student’s score simply would not count, while the student and his or her family still would receive a score report providing feedback. Under NCLB, however, all students must be tested and reported in AYP (schools have a leeway of just 5 percent) so those scores needed to count.

The state education agency’s recalculations affected more than 3,000 students in 30 elementary schools in Maryland. Sixteen of those schools are in Montgomery County. Not only have these schools been falsely identified in many cases as not making adequate yearly progress, but those individual students have been denied the opportunity to show what they know on the state assessment—and contribute an accurate score to their school, says Weast.

“We don’t think it’s a fair shake for the kids. It hurts the students,” the superintendent adds, “because it doesn’t give a score. It hurts the schools because it gives a label. And it hurts the district” because of the cost of providing choice, supplemental services and other services to comply with the federally required sanctions that come into play.

Weast says Montgomery County has been disaggregating test data for five years. “We are for accountability in our district and for assessment ... [but] you want it to be a fair shake.”

Action Options
What is a school leader to do when the issue of limited English proficient students or students with disabilities suddenly surges to the forefront in your district or school? Perhaps a school previously considered high-achieving will be labeled because of one group’s performance. Perhaps, as in Byng, Okla., a school will be a victim of the numbers game.

In Montgomery County, score reports for affected students were delayed while the district worked with the state to provide a clear explanation. “Some of the parents don’t speak English themselves and are confused,” Weast says. “And the parents of special education students were outraged. They are used to the IEP ruling and want to see what academic progress their child is making.”

It’s important to remember that parents and the public show an understanding of the complexity of these issues. The 2003 Phi Delta Kappa poll of public attitudes reported that when asked if students with disabilities should be held to the same standards as other students, 67 percent of the public (66 percent of nonparents and 68 percent of parents) said they should not.

First and foremost, the priority of a school leader faced with this issue should be to make sure the lines of communication with parents and the public are open. Parents of students in schools that were labeled as failing to make progress should receive special attention. By how much did the students not make the target? Did they still improve from last year? Did they achieve other separate goals?

Further, all parents and the community should have the big picture of the data explained to them. How did each group perform, and are there explanations they should know about? One thing Crawford found was that while his one school with the 17 5th graders in special education did not make AYP, the students still performed better than the special education students at a neighboring school that had a smaller percentage of special education students and that did make AYP.

That leads to the most important conclusion about AYP. It is never the whole picture of a school’s performance. Parents and the public need to understand that. AYP is based on the scores from two tests on mathematics and reading/language arts. And parents and the public continually dismiss the single test as an indicator of how a school is doing—72 percent of the public thought it was not possible to accurately judge a student’s proficiency in English and mathematics on the basis of a single test, according to the Phi Delta Kappa poll. They know they and a student’s teachers and principal know more about an individual student’s performance than any federally mandated rating. So school leaders should be sure to portray AYP, particularly as it relates to students who need extra help, as what it is—just one measure of how a school is doing. In the case of LEP students and students with disabilities, more than any other group, there’s more to the picture.

NCLB places a much-needed focus on closing the achievement gap and improving the educational opportunities for all students. But in the case of these two special groups of students, the focus should be not on blaming them or labeling them, but on continually working to improve outcomes for them as individuals.

Terri Duggan Schwartzbeck is a policy analyst at AASA. E-mail: tschwartzbeck@aasa.org