Just weeks before states release their lists of schools that have not met 鈥渁dequate yearly progress鈥 targets under the main federal K-12 law, many states are still negotiating with federal officials over changes to their accountability plans designed to reduce those numbers.
But whether the proposed changes are common-sense measures that would better and more reliably identify needy schools or attempts to duck accountability is largely in the eye of the beholder. Moreover, the sheer volume of last-minute revisions could make it harder to tell if schools have really improved, or if the rules of the game have simply changed.
鈥淚t is going to be important to watch this process very carefully,鈥 said Ross Wiener, the policy director for the Education Trust, a nonprofit group in Washington that works to raise student achievement. 鈥淚t would be a concern if it appears that there has been improvement, when in fact student achievement has not improved.鈥
|
Raymond Simon, the assistant secretary for elementary and secondary education in the U.S. Department of Education, said in an interview last week: 鈥淚 have to believe that any state鈥檚 requests for amendments are done for the purpose of making a plan better and more defensible. I don鈥檛 see any state trying to avoid accountability.鈥
He said the department had denied some requests either because they could not be accommodated under the law or because states had not provided the data to support their proposals.
The federal No Child Left Behind Act, a 2陆-year-old revision of the Elementary and Secondary Education Act, requires states to determine whether schools and districts have made adequate yearly progress. AYP is based on whether they have met annual performance targets and tested at least 95 percent of both their overall student enrollments and individual groups of children who are poor, speak limited English, come from racial- or ethnic-minority backgrounds, or have disabilities.
Education Department officials promised that if states submitted requests to change their accountability plans by April 1, the agency would get back to them within 30 days, in time to make those changes based on 2003-04 test results. But while department officials have been in constant contact with states, by late last week they had given final, written approval to only 20 of the approximately 45 that submitted requests by April.
鈥淲e鈥檙e certainly aware that they need to make those decisions pretty quick, and that鈥檚 why we鈥檙e struggling very hard to get some final answers out,鈥 said Mr. Simon. 鈥淲e want to be sure, before we release any information for any state, 鈥 that we have everybody on board and feel good about it.鈥
The most common amendments sought by states, and approved by federal officials, take advantage of expanded flexibility announced earlier this year on calculating test- participation rates and including students with disabilities or limited English proficiency in AYP decisions. (The proposed rule changes for LEP students were published in the June 24 Federal Register.)
In addition, many states requested changes in the statistics they use to calculate adequate progress. More than a half-dozen states, for example, have asked to raise the minimum number of special education or LEP students who must be tested before those groups count in determining whether a school has met its performance targets.
Such changes were denounced last week by advocates for special education and LEP students, who fear that many schools will escape accountability for teaching those youngsters.
鈥淚 think it sends a big, wrong message,鈥 said Martha L. Thurlow, the director of the National Center on Educational Outcomes, located at the University of Minnesota-Twin Cities. 鈥淚t seems to me, at this time, we should be holding the numbers steady and really looking at our data to figure out what we need to be doing to provide the instructional supports that those students need.鈥
Last year, for example, 65 schools in Alaska failed to make adequate progress solely because they missed the targets for their students with disabilities. This year, the state will publicly report results when schools test as few as five students with disabilities or limited English proficiency. But schools will no longer have to meet annual performance targets for those groups if they include fewer than 40 children.
Les Morse, the director of assessment and accountability for the Alaska education department, said that change excludes about 85 percent of the state鈥檚 schools from having to meet AYP targets for their special education students, up from about 65 percent last year.
Mr. Morse said the measure would give schools time to put more programs in place for those children. 鈥淲e don鈥檛 want to let those schools off the hook,鈥 he said, noting that the state will closely examine the data and perhaps revisit its decision in 2007.
Federal officials, meanwhile, rejected a request by Tennessee to include its gifted and talented students in its special education subgroup.
鈥淚 don鈥檛 think we actually expected that one to be approved,鈥 said Kim Karesh, the communications director for the Tennessee education department. State law, she said, places gifted students under the special education umbrella, 鈥渟o we were trying to match our state definitions with the federal requirements.鈥
Tennessee also sought unsuccessfully to increase the minimum subgroup size for special education and LEP students from 45 to 55, said Julie P. McCargar, the director of federal programs for the state education department. 鈥淲e had to provide data, and we couldn鈥檛 support it,鈥 she said. No Tennessee school or district with at least 45 special education students made adequate yearly progress last year, she said.
Sizable Shift?
69传媒 and districts that do not meet their annual performance targets can still make AYP under a provision known as 鈥渟afe harbor.鈥 But to do so, the percent of students who score below the 鈥減roficient鈥 level on state tests must drop by 10 percent from the previous year. That鈥檚 often a substantial difference for schools or subgroups that start well below state targets.
At least five states have received approval to give schools more leeway in meeting the safe-harbor requirements by adding a 75 percent 鈥渃onfidence interval鈥 within which their performance could fall and still meet the safe-harbor provisions.
Sheila Barron, an assistant professor of education at the University of Iowa, warned that adding a confidence interval 鈥渞epresents a sizable shift in the accountability requirements of No Child Left Behind.鈥
That鈥檚 because the confidence band can be so wide, particularly for small schools or subgroups, that schools could meet the safe-harbor requirements even if they showed little, if any, improvement.
While such a gradual rate of improvement may be a more reasonable expectation, Ms. Barron said, it does not coincide with the law鈥檚 goal that all students will be proficient by 2013-14. Federal officials said they plan to review state data annually to see what effect the confidence interval is having on safe-harbor decisions.
Observers also note that most of the proposed changes mean that larger, more diverse schools and districts鈥攕uch as those in urban areas鈥攚ill continue to be subject to the law鈥檚 more stringent accountability provisions, while smaller, more homogeneous ones will not.
鈥淚f that鈥檚 what people want, they ought to say it,鈥 said Brian Gong, the associate director of the Center for Assessment, a nonprofit research and consulting group based in Dover, N.H. 鈥淏ut it doesn鈥檛 fit very well for probably 60 percent of the schools in this nation, which are in rural or suburban or small districts.鈥
District Flexibility
A number of states also are asking for more flexibility in determining whether districts make adequate progress. Without such leeway, they say, all or almost all their districts will end up identified as needing improvement.
鈥淭he problem is we鈥檝e got 24 districts and they鈥檙e all huge,鈥 said Ronald A. Peiffer, the deputy state superintendent for academic policy in Maryland. When the state applied the same minimum group size to both schools and districts last year, not a single district made adequate progress. 鈥淚t鈥檚 like having a smoke detector that goes off when you burn toast,鈥 Mr. Peiffer said. 鈥淚t鈥檚 senseless.鈥
Maryland requested that, at the district level, a subgroup would have to represent 15 percent of the total student population before it counted for making AYP determinations. At least five other states have sought increases in the subgroup size at the district level, according to an analysis conducted for the Washington-based Council of Chief State School Officers. So far, federal officials have rejected that approach.
In Tennessee, where 92 percent of districts did not make adequate progress last year, 鈥渨e felt that it was not a good representation of what was really happening in our school system,鈥 said Ms. McCargar. The state unsuccessfully asked to increase the minimum group size to 200 at the district level.
But federal officials did approve another change: Tennessee will identify districts as needing improvement only if they miss their AYP targets for both elementary/middle and high schools for two years in a row. That would have dropped the proportion of districts identified last year to 65 percent. Other states are waiting to hear whether the additional flexibility will apply to all states.
More than a dozen states also want to make it easier for schools to meet the 鈥渙ther academic indicator鈥 required to make adequate progress鈥攇raduation rates for high schools鈥攅ither by using a confidence interval or by giving schools credit for making progress from the previous year, even if they have not met an absolute target.
Ellen Forte-Fast, a Washington-based consultant who is analyzing the accountability plans for the CCSSO, said states want to cut the number of schools and districts identified for improvement because of legitimate concerns that they won鈥檛 be able to help them all. But, she said, it鈥檚 not clear whether the proposed changes will improve states鈥 ability to identify the schools and districts most in need of help.
States, agreed Mr. Gong of the Center for Assessment, 鈥渁re afraid of overidentifying schools that shouldn鈥檛 be identified. But many of the things they鈥檙e doing are taking schools off the radar screen; they鈥檙e not paying as much attention to that.鈥
Still others question whether federal and state efforts to soften the edges of various requirements of the law address the more fundamental issue: whether the law鈥檚 targets are realistic.
|
鈥淧eople are becoming increasingly desperate when faced with untenable demands for improvements in performance,鈥 Daniel M. Koretz, a professor of education at Harvard University, said at the CCSSO鈥檚 annual conference on large-scale assessment, held in Boston last month. 鈥淭he fundamental issue is not how states should respond to this. It鈥檚 whether we should be doing this.鈥
