TOPICAL REVIEW SEVEN Accountability for Students with Disabilities Who Receive Special Education: Characteristics of the Subgroup of Students with Disabilities A Summary of Quantitative Findings from the Educational Policy Reform Research Institute (EPRRI) September 2006 The Institute for the Study of Exceptional Children and Youth University of Maryland 1308 Benjamin Building College Park, Maryland 20742-1161 301.405.6509 • 301.314.9158 The Educational Policy Reform Research Institute Accountability for Students with Disabilities Who Receive Special Education: Characteristics of the Subgroup of Students with Disabilities A Summary of Quantitative Findings from the Educational Policy Reform Research Institute (EPRRI) Margaret J. McLaughlin University of Maryland Kimber Malmgren University of Wisconsin, Madison Victor Nolet Western Washington University September 2006 The Institute for the Study of Exceptional Children and Youth University of Maryland 1308 Benjamin Building College Park, Maryland 20742-1161 301.405.6509 • 301.314.9158 visit our web site at www.eprri.org Funding for this research work was provided by the U.S. Department of Education, Office of Special Education Programs (Grant # H324P000004). Opinions expressed in this paper are those of the authors, and do not necessarily reflect the views of the U.S. Department of Education or the Office of Special Education Programs. Any or all portions of this document may be reproduced and distributed without prior permission, provided the source is cited as: Educational Policy Research Reform Institute (EPRRI) (date). Title of publication. College Park, MD: University of Maryland, College Park Educational Policy Research Reform Institute. Retrieved [today's date], from the World Wide Web: www.eprri.org The Educational Policy Reform Research Institute 1 eprri The Educational Policy Reform Research Institute EPRRI, funded by the U. S. Department of Education's Office of Special Education Programs, investigates the impact of new educational accountability systems on students with disabilities and on special education. EPRRI addresses the research needs of policymakers and other key stakeholders by identifying critical gaps in current knowledge, seeking promising strategies, and publishing Topical Reviews, Policy Updates, and Issue Briefs. The Institute is a joint venture of the Institute for the Study of Exceptional Children and Youth at the University of Maryland, the National Center on Educational Outcomes at the University of Minnesota, and the Urban Special Education Leadership Collaborative. 2 Accountability for Students with Disabilities in Special Schools and Settings Table of Contents Topical Review Highlights 1 Introduction 3 Underlying Assumptions of NCLBA and Accountability Policy 5 What is EPRRI? 7 Overview of Research Design and Methods 9 Assessment Programs of the Four Study States 11 California 11 Maryland 18 New York 21 Texas 26 Summary 32 State and District Data Sources and Findings 35 Data Sources 35 Findings 36 Summary of Findings 51 Topical Review Summary of Findings 53 The Educational Policy Reform Research Institute 3 Topical Review Highlights Accountability for students with disabilities who receive special education services is now a result of policy requirements in the Individuals with Disabilities Education Improvement Act of 2004 and the No Child Left Behind Act of 2001. Together these pieces of federal legislation require that students participate in statewide assessments, that their participation and results be publicly reported, and that the results of the disability subgroup factor into measures of Adequate Yearly Progress (AYP). While the basic tenets of the laws have been adjusted to add some flexibility, such as the provision that an alternate assessment based on alternate achievement standards used to measure students with the most significant cognitive disabilities, can be used to count up to 1.0 percent of the total population of students as proficient, in general the subgroup of students with disabilities is now treated similarly to other student groups in school accountability. The purpose of this Topical Review is to provide a picture of what the implementation of IDEA and NCLB has produced in terms of participation and performance in statewide accountability measures. This is accomplished by focusing on EPRRI’s four case study states (California, Maryland, New York, and Texas), and partner districts within each state. and the different implementation avenues that the districts have pursued. In addition to variable findings is clear evidence that considerable change happens over time in states to complicate the implementation of educational reform initiatives such as those of standards-based reform and accountability for all students. The data from the four case study states reveal some unintended outcomes of reform – such as the finding that as participation increased, performance decreased. Clearly the intent of the reform was to increase participation first, and then to increase performance as well. The possibility that schools are seeking loopholes to avoid responsibility for the performance of the disability subgroup is difficult to identify in a study such as this one. It is more likely that these types of unintended outcomes will be identified in case study research. This type of research is another part of the EPRRI project, and is also available at the EPRRI Web site www.eprri.org. It is not possible to understand the state assessment participation and performance results without having a sense of the context of the states and the districts – the student population, the history and nature of the assessment and accountability systems, and the policies that surround the assessment system, such as those related to the use of accommodations. These are investigated within this Topical Review, along with the participation and performance results of the states and districts. The diversity in results is striking, and reflective of the different policy directions that the states have taken The Educational Policy Reform Research Institute 1 eprri 2 Accountability for Students with Disabilities Who Receive Special Education Introduction Current policy requirements for accountability relating to students with disabilities are found in two federal laws: the Individuals with Disabilities Education Improvement Act of 2004 (IDEA) and the No Child Left Behind Act of 2001 (NCLBA). Both pieces of legislation require states and local school districts to be held accountable for the performance of students with disabilities on state assessments. Students with disabilities are considered a subgroup under NCLBA and all requirements pertaining to general education students also pertain to this subgroup. Special education policy as defined by IDEA has increasingly been aligned with the accountability requirements of Title I of NCLBA. Policymakers concerned about the low expectations and poor post-school outcomes of students with disabilities have placed a strong emphasis on linking the educational programs of these students to the general curriculum and to state standards. The IDEA requires states and districts to include students with disabilities in local and statewide assessments with accommodations where appropriate and to document this on individual student IEPs. If a student with a disability will not participate in the regular state assessment, the IEP must include a statement of why that assessment is not appropriate for the child and how the child will be assessed. For those students with disabilities who are not able to participate in state assessments even with accommodations, states were required to develop an alternate means of assessment. States and districts are also required to report the performance of students with disabilities on state and district assessments with the same frequency and in the same detail that they use to report the performance of nondisabled students [§612(a)(17)(B)(iii)]. In addition, states must re- port the performance of students with disabilities on any alternate assessment [§612(a)(17)(B)(iii)]. Other accountability provisions in special education include the requirement that states establish performance goals and indicators for students with disabilities and report progress toward these goals to the U.S. Secretary of Education and the public every 2 years. While these IDEA requirements promote accountability, the Title I requirements of NCLBA are now the primary tool in U.S. schools for accountability for students with disabilities. Under NCLBA, states must assess at least 95 percent of all students and students in each of five target groups, including students with disabilities. In addition, states must publicly report disaggregated subgroup performance as long as student confidentiality is maintained. However, schools are only accountable for groups that are large enough to allow statistically valid and reliable conclusions to be made regarding their performance on state assessments. This minimum number for subgroup accountability is determined by each state. States must set separate annual measurable objectives defined as annual yearly progress (AYP) targets in mathematics and reading/language arts ensuring that all groups of students remain on a trajectory toward proficiency by 2013–14. These targets increase over time and must be the same for all schools serving the same grades and for all groups of students within schools.1 1 The NCLB includes a “safe harbor” clause that can be applied to any subgroup or subgroups of students that fail to meet the statewide goal. In this situation, the school can still make AYP if the percentage of students below proficient falls by 10 percent, and the subgroup or subgroups in question meet the 95 percent participation requirement and makes progress on the other required additional indicator. The Educational Policy Reform Research Institute 3 eprri Recognizing that grade-level assessments would not be appropriate for some students with disabilities, the NCLBA regulations give states and school districts the flexibility to measure the achievement of students with the most significant cognitive disabilities against alternate achievement standards and to count the “proficient” scores of these students as proficient in the calculation of AYP [200.13(c)(1)(i)]. However, the number of proficient scores on alternate achievement standards at the local education agency (LEA) and state levels must not exceed 1.0 percent of all students in reading/language arts and in mathematics (Federal Register, Vol. 68, No 236, Tuesday, December 9, 2003, Rules and Regulations, pg. 68703). An alternate achievement standard is an expectation of performance that differs in complexity from a grade-level achievement standard. Under NCLBA, individual states are allowed to define alternate achievement standards as long as they are aligned with the state’s academic content standards, promote access to the general curriculum, and reflect professional judgment of the highest achievement standards possible [34 C.F.R. § 200.1(d)]. Thus, with some modification, the subgroups of students with disabilities are now treated similarly to other student groups. That means that similar assumptions that underlie NCLBA apply to students with disabilities.  Accountability for Students with Disabilities Who Receive Special Education Underlying Assumptions of NCLBA and Accountability Policy The 2001 NCLBA is based on a set of interrelated assumptions about educational accountability that have been evolving for some time at the federal level as well in state policies. Current accountability schemes are grounded in certain assumptions including: inferences about the performance of a subgroup based on performance trends over time. The idea that the school should be the unit of accountability (e.g., consequences and improvement) is challenged if school test results are inexact and vary unsystematically. A. Common content and achievement standards are essential for achieving educational equality This report endeavors to help policymakers, administrators, and practitioners better understand the issues facing policymakers as they implement the assumptions of standards-based accountability with the subgroup of students who receive special education. The report is one of two that synthesize a set of analyses and studies that have been conducted under the auspices of the Educational Policy Reform Research Institute (EPRRI). This report presents the quantitative analyses, and its purpose is to examine specific features of the subgroup of students with disabilities as they relate to the specific accountability requirements. Separate reports present a summary of the qualitative research involving state, local, and school-level administrators and practitioners (see www.eprri.org). B. “Closing the achievement gap” between specific student subgroups is a central goal of educational reform C. Achievement can be reliably measured D. School is the unit of accountability and improvement E. Consequences (e.g., rewards and sanctions) are necessary to prompt schools to act on performance data Under NCLBA, these assumptions are applied to all students and the specific student subgroups defined in the law. Yet, students with disabilities, specifically those who receive special education services under Part B of IDEA challenge some of these foundational assumptions. For example, the concept of universal standards is challenged by the heterogeneity of the population and validity and reliability of the assessment results has also been problematic given issues surrounding accommodations. Furthermore, the notion of “closing the achievement gap” assumes that subgroup performance is solely the result of educational opportunity as opposed to intra-child factors. Finally, consequences are dependent on the ability to accurately measure across subgroup performance and policymakers must be able to make defensible The Educational Policy Reform Research Institute 5 eprri 6 Accountability for Students with Disabilities Who Receive Special Education What is EPRRI? The Educational Policy Reform Research Institute (EPRRI), funded by the U.S. Department of Education’s Office of Special Education Programs, is a project of the Institute for the Study of Exceptional Children and Youth at the University of Maryland, in collaboration with the National Center on Educational Outcomes and the Urban Special Education Leadership Collaborative. EPRRI investigates the impact of new educational accountability systems on students with disabilities and special education programs, develops Topical Reviews, Policy Updates, and Issue Briefs that review current accountability policies and practices, conducts Policy Symposia to identify and analyze current policy issues related to accountability reforms, and conducts research in collaboration with four core states: California, Maryland, New York, and Texas. The Educational Policy Reform Research Institute 7 eprri 8 Accountability for Students with Disabilities Who Receive Special Education Overview of Research Design and Methods This report provides a descriptive profile of selected characteristics of the subgroup of students with disabilities who were receiving special education within public schools in a total of eight school districts within four states: California, Maryland, New York, and Texas. The report is based on quantitative data collected at state and district levels during three school years: 1999-00; 2000-01; and 2001-02. This report begins by describing the assessment programs of each state, then follows with the quantitative participation and performance data for all students and for students with disabilities. The Educational Policy Reform Research Institute 9 eprri 10 Accountability for Students with Disabilities Who Receive Special Education Assessment Programs of the Four Study States Each of the four states whose data are included in this report has a unique assessment system used to determine the academic achievement of students in the state. Information about each state’s assessment system was retrieved from a number of documents and other sources, primarily the Internet. Following is a description of the assessment and accountability system that was in place in each of the four states between the school years 1999-00 and 2000-01. Also provided are details regarding future changes. It is important to note that three of the four states made substantive changes to their accountability systems during the 2002-03 school year. California California’s assessment program, known as the Standardized Testing and Reporting Program (STAR), was authorized in October 1997. In November 1997, the Stanford Achievement Test Series, Ninth Edition, Form T (SAT 9) was chosen as the STAR test for a period of 5 years. The SAT 9 is a multiple-choice test that allowed comparisons to be made to a national sample of students. School districts in California were required to test all students in grades 2 through 11. Students in grades 2 through 8 were tested in reading, mathematics, written expression, and spelling. Students in grades 9 through 11 were tested in reading, writing, mathematics, science, and history/social science. The only exemptions allowed were for special education students if their Individual Education Plans (IEPs) explicitly exempted them from such testing and for any students whose parent or guardian submitted a written request for exemption. Currently, the state office of education states that students receiving special education services who have IEPs specifying exemptions from STAR must participate in an alternate assessment (i.e., the California Alternate Performance Assessment [CAPA]). In 1999 two additions were made to the STAR program. Test items were added in language arts and mathematics in order to create two additional tests, the STAR augmentation of the SAT 9 (now called the California Standards Tests – CST) and the SABE/2. The latter was designated by the State Board of Education for use with Spanish-speaking limited English proficient students. Following the reauthorization of the STAR program in 2001, several additional changes were made to the STAR program. Three more CSTs were included: Grade 4 and 7 writing tests, Grade 9-11 end-of-course science tests, and Grade 9-11 history-social science tests. Table 1 shows the CST tests and the grade levels that were tested in 2004. In addition, the California State Board of Education designated the California Achievement Tests, Sixth Edition Survey (CAT/6) to replace the SAT 9, beginning in the 2002-2003 school year. The CAT/6 now provides the normative component of California’s assessment system. The assessment reflects national standards and facilitates comparisons of California students to students from around the nation. The normative group was obtained in the fall of 1999 and the spring of 2000. California students in grades 2–11 are tested in reading, mathematics, and language. Additionally, students in grades 2–8 are tested in spelling and students in grades 9-11 are tested in science. The National Percentile Rank (NPR) and mean scale score are reported for each grade level along with the percent of students performing above the 75th NPR, at or above the 50th NPR, and above the 25th NPR. The California High School Exit Examination (CAHSEE) is administered in February and March and The Educational Policy Reform Research Institute 11 eprri Table 1. California Standard Tests (CTS) Subtests and Grade Levels Tested. CST Subtest English Language Arts Mathematics General Mathematics (Grade 6 & 7 Standards) Algebra I Integrated Math 1 Geometry Integrated Math 2 Algebra II Integrated Math 3 Summative High School Mathematics History – Social Science (Grade 6, 7, & 8 Standards) World History U.S. History Science (Grade 4 & 5 Standards) Biology/Life Sciences Chemistry Earth Science Physics Integrated/Coordinated Science I Integrated/Coordinated Science II Integrated/Coordinated Science III Integrated/Coordinated Science IV covers the content areas of English Language Arts and Mathematics. Assessment Accommodations Revisions were made to testing accommodations for students with disabilities after the passage of the NCLBA. Currently, students receiving special education services in California, through either an IEP or a Section 504 Plan, are eligible for testing accommodations if the IEP team deems them unable to complete state assessments without their use. These accommodations and modifications may be used on all the statewide assessments, which are: California Achievement Test (CAT/6), California Standards Test (CST), Spanish Assessment of Basic Education (SABE2), California High School Exit Examination (CAHSEE), and the California English Language Development Test (CELDT). For further Grade Level Tested 2 – 11 2–9 8 – 11 8 – 11 8 – 11 8 – 11 8 – 11 8 – 11 8 – 11 9 – 11 8 10 11 5 9 – 11 9 – 11 9 – 11 9 – 11 9 – 11 9 – 11 9 – 11 9 – 11 information regarding the CAT/6, CST, and SABE2, go to http://www.cde.ca.gov/ta/tg/sr or http://www. startest.org. Information on the CAHSEE can be found at: http://www.cde.ca.gov/ta/tg/hs/index.asp. While some test variations, such as clarification of test directions and extra time (on the CST, CELDT, and CAHSEE only) may be provided to all students, assessment accommodations generally fall into the following three categories: • Category 1 accommodations are referred to as test variations and may be used by any student who regularly uses them in the classroom. Examples of Category I accommodations are: Using a marker or mask to maintain place, special lighting, and individual or small group test administration. 12 Accountability for Students with Disabilities Who Receive Special Education • Category 2 accommodations may be used by a student with an IEP or Section 504 plan. The accommodations must be indicated in the eligible student’s IEP or Section 504 Plan. Examples of Category 2 accommodations are: Completing a test across multiple days, using extra time on a test, being tested in the student’s home or in the hospital, being tested at a time of day beneficial to the student; marking answer in the test booklet for an adult to transcribe, using a scribe; and, using large print or Braille tests. • Category 3 accommodations are referred to as modifications in California and are considered to be nonstandard accommodations that may have scoring implications. Examples of Category 3 accommodations are: Using a dictionary, having a reading passage read aloud, using a calculator on a mathematics or science test, and dictating a response to a scribe. Table 2 indicates which of the selected accommodations were permitted in California in 2001, 2003, and 2005. A few of the accommodations (e.g., large print, magnification equipment, amplification equipment, calculator, individual administration) have remained the same over this time period, but most accommodations have changed. For example, in 2001, read aloud questions was not permitted for reading passages on the Stanford 9 Reading Comprehension Test and the California Language Arts Standards Test, and in 2003 this accommodation was not allowed on any non-reading/ELA tests and allowed with implications for scoring on the CAT6 and the CST reading tests. Currently, this accommodation is allowed for math, science, and history-social science tests and allowed with implications for scoring on reading, language, and spelling tests. Regarding the “with breaks” accommodation, in 2001, a student’s score was not aggregated if there were unspecified breaks taken within a test or subtest, however in 2003 and 2005, this accommodation was allowed. Table 2. California: Selected Accommodations, 2001, 2003, and 2005. Accommodation Large Print Braille Read Aloud Questions Magnification Equipment Amplification Equipment Spell Checker Calculator Scribe Write in Test Booklets Extended Time With Breaks Individual Administration Administration in Student’s Home 2001 2003 2005 A A A AI A A AC AC/AI AC/AI A A A A A A AI A AI AI AI AI AC/AI A A A AC AI A AC AI A A A A A A A Note: A = Allowed; AI = Allowed with implications for scoring and/or aggregation; AC = Allowed in certain circumstances; P = prohibited The Educational Policy Reform Research Institute 13 eprri Information on California’s accommodation policies does not necessarily reflect the controversy that has surrounded accommodations use in the state. This is true particularly in regard to the use of accommodations during the CAHSEE exit exam, which resulted in a lawsuit in May 2001 (Juleus Chapman et al. v. California Department of Education). The state initially indicated the students with disabilities would have to request a waiver to use accommodations during the exit exam. This was quickly stopped by the judge who heard the case; an immediate solution was imposed by the judge who indicated that all students with disabilities who needed accommodations to participate in the exit exam could receive whatever accommodations they needed. The state established (in response to legislative and state board actions, SB 964) an advisory panel to consider options and provide input to identify recommendations for alternatives to the high school exit exam for students with disabilities. These alternatives were to provide a different way for students to demonstrate that they have met the graduation standards when they need accommodations that produce invalid scores for the exit exam. The final recommendation was to provide an exemption from the requirement for students with disabilities for at least one year. Despite resistance from the State Department of Education, the one-year hiatus of the graduation requirement for students with disabilities was enacted due more likely to lawsuit threats rather than the panel recommendation. In May 2006, a ruling was made on a second lawsuit involving the CAHSEE. In Valenzuela v. California Department of Education, an Alameda Superior Court Judge ruled that the state of California could not withhold a high school diploma from any student in the Class of 2006 who had not passed the high school exit exam, but had met all other graduation requirements. Upon hearing this decision, Jack O’Connell, California’s State Superintendent of Public Instruction, filed an appeal with the California Supreme Court to overturn the ruling. A few weeks later, the Supreme Court reversed the Alameda County Superior Court decision and reinstated the high school exit exam as a requirement for graduation for the Class of 2006. The Supreme Court also referred the case to the 1st District Court of Appeals, and oral arguments in the case are scheduled to be heard on July 25, 2006. Alternate Assessments For students with the most complex and severe cognitive disabilities, accommodations and modifications may not be sufficient to enable their participation in state- and district-wide assessments. These students may either participate in the same assessments at a grade-level different from the one in which they are presently enrolled (out-of-level testing) or participate in the California Alternate Performance Assessment (CAPA), intended to individually assess the state’s Academic Content Standards. Out-of-level testing is not permissible for students in grades 2, 3, or 4. Students with IEPs or Section 504 Plans in grades 5-11 may be tested no more than two grade levels below their enrollment grade. Any student taking an out-of-level test must participate in all the assessments given at that grade level. The California Alternate Performance Assessment (CAPA) was designed by Educational Testing Services and was administered for the first time in Spring 2003. Results from the CAPA were included in the 2004 Base API, which is described later in this section. The CAPA is designed to be an alternate to California’s Standardized Testing 1 Accountability for Students with Disabilities Who Receive Special Education and Reporting (STAR) Program for students with more significant disabilities in grades 2-11 who are unable to take a large-scale assessment, even with accommodations. The CAPA is an on-demand performance event assessment aligned to a subset of the California content standards, specifically the ELA and mathematics standards. For those students with severe disabilities who are not assigned to a grade level, subtracting 5 from the student’s chronological age determines the student’s grade for testing purposes. For example for accountability purposes, a 12-year-old would be enrolled in grade 7 in a graded program. The CAPA has five levels: Level I (grades 211); Level II (grades 2-3); Level III (grades 4-5); Level IV (grades 6-8); and Level V (grades 9-11). Most students eligible for the CAPA will take the level corresponding to their grade. However, some students with complex, profound disabilities may be eligible only for Level I. This level represents an opportunity for students with the most significant disabilities to demonstrate their skills and independence. As with the STAR assessments, a student may be exempted from CAPA by parental waiver. Accountability Accountability in California’s educational system is defined by the Public Schools Accountability Act (PSAA) of 1999, which authorized the creation of a new educational accountability system for California public schools. The Public School Performance Accountability Program consisted of the following three component parts: (a) the state Academic Performance Index, known as the API; (b) the Immediate Intervention/Underperforming Schools Program and (c) the Governor’s High Achieving/Improving Schools Program. California reports its API as: (1) a base component and (2) the growth component. The base report is published each January or February to inform schools on the amount of progress they need to make on that year’s Spring assessments to meet their individual API target. In the fall, the growth reports are published to show if schools have reached their growth targets based on performance on the previous spring’s assessments. The State Board of Education has approved five performance standards for the state assessments: advanced, proficient, basic, below basic, and far below basic. Students performing at the proficient level, the state’s desired performance level for all students, score at or above the 50th percentile on the CAT/6. The California Standards Test (CST) requires the individual to reach a score of 350 or higher, while the CAPA requires a score of 35 or higher to be at the proficient level. This definition of proficient is used to calculate Adequate Yearly Progress (AYP) under the NCLBA. The cornerstone of California’s Public Schools Accountability Act of 1999 is the Academic Performance Index [API]. This figure is used to measure the growth and academic performance of schools by assigning a numeric value, ranging from 200 to 1000, to their academic improvements; the target score for all schools is 800. The weight of each test in the school’s API varies by grade level as shown in the Tables 3 and 4. To calculate a school’s API score, each student’s national percentile rank on the norm-referenced test, standards-based performance level on the CST, and performance level on the CAHSEE (if applicable), are weighted and combined to produce a summary result for each area. Those summary results are then also weighted and combined to yield a number The Educational Policy Reform Research Institute 15 eprri Table 3. California: Academic Performance Index (API), Elementary and Middle Schools (Grades Two through Eight). 2000-01 API Cycle 2000 API Base and 2001 API Growth NRT Content Area English Language Arts (ELA) NRT (Reading) (Language) (Spelling) CST Mathematics NRT CST Total 30% 15% 15% 2001-02 API Cycle 2001 API Base and 2002 API Growth NRT CST 2002-03 API Cycle 2002 API Base and 2003 API Growth NRT CST 2003-04 API Cycle 2003 API Base and 2004 API Growth NRT CST & CAPA 24% (12%) (6%) (6%) 12% (6%) (3%) (3%) 12% (6%) (3%) (3%) 40% 100% 40% 48% 48% 36% 8% 64% 36% 20% 8% 32% 80% 20% 32% 80% Note: NRT = Norm-referenced test (Stanford 9 through 2002; CAT/6 Survey beginning in 2003) CST = California Standards Test CAPA = California Alternate Performance Assessment Source: California Department of Education. 2003-04 Performance Index Growth Report: Information Guide. October 2004, p. 22. Table 4. California: Academic Performance Index (API), High Schools (Grades Nine through Eleven). Content Area ELA NRT (Reading) (Language) CST CAHSEE Mathematics NRT CST CAHSEE Science NRT CST Social Science NRT CST Total 2000-01 API Cycle 2000 API Base and 2001 API Growth NRT 20% 20% 2001-02 API Cycle 2001 API Base and 2002 API Growth NRT CST 16% (8%) (8%) 2002-03 API Cycle 2003-04 API Cycle 2002 API Base and 2003 API Growth 2003 API Base and 2004 API Growth NRT CST CAHSEE 6% (3%) (3%) 24% NRT CST & CAPA 6% (3%) (3%) 32% 35% 10% 10% 20% 20% 3% 3% 16% 18% 5% 5% 20% 20% CAHSEE 3% 3% 5% 20% 20% 100% 76% 24% 12% 20% 73% 15% 12% 73% 15% Note: NRT = Norm-referenced test (Stanford 9 through 2002; CAT/6 Survey beginning in 2003) CST = California Standards Test CAPA = California Alternate Performance Assessment CAHSEE = California High School Exit Examination Source: California Department of Education. 2003-04 Performance Index Growth Report: Information Guide. October 2004, p. 22. 16 Accountability for Students with Disabilities Who Receive Special Education between 200 and 1000, which is the school’s API. Prior to July 2003, the minimum annual growth target for each school with an API lower than 800 was 5% or a minimum of one point. Schools with an API above 800 must maintain that API. According to the Public Schools Accountability Act of 1999, if a school meets their API growth criterion, the school is eligible for the Governor’s Performance Award Program, which includes monetary rewards, special commendations and school honor roll, or the Certified Staff Performance Incentive Award. Schools ranked in the lower half of the state and not meeting their growth targets will be identified for the Immediate Intervention/Underperforming Schools Program (II/USP). If a school fails to meet its API growth target, the school is recommended for the II/USP and it could be selected to receive improvement funding. If the school fails to meet its API growth target in the following year, local interventions will be put into place such as removing the barriers toward improved student achievement or requesting the High Priority Schools Grant from the state amounting to $200 per child and not less that $50,000 per school. If local interventions do not help the school to meet its target in the third consecutive reporting cycle, the school is deemed low-performing and the “Superintendent of Public Instruction shall assume all the legal rights, duties, and powers of the governing board with respect to that school” (Summary of SB 1552, 52055.5c, 1999). The State Board of Education then has the option of at least one of the following: • Revising the attendance policy for the students, allowing them to attend any public school with space; • Allowing parents to form a charter school; • Assigning the management of the school to a university or other high education professional; • Reassigning employees of the school; • Renegotiating the teachers’ contract at the end of the contracted year; • Reorganizing the school; or • Closing the school. The API aligns with the AYP requirement of the NCLBA, and the state continues to report both results under the general heading of Accountability Progress Reporting (APR). Progress of one point or more on the API will translate to the same amount of gain in the AYP calculator. Schools with fewer than 100 valid scores may be paired with a school in the same LEA with a grade outside the testing program and the scores of both schools will be aggregated across no more than three years, or a statistical test will be applied to achieve a 95% confidence interval in AYP determination (California Department of Education, 2004). At the elementary level, scores from the CAPA have been aligned with the performance levels from the CST; the CAPA scores can be integrated with those of the students assessed with the CST to produce one measure of proficient for calculation of a school’s AYP. At the high school level, the CAHSEE is scored as “pass/ no pass.” Participation rates in all tests have been calculated by requiring that all students complete the demographical portion of the Student Answer Document whether or not they are participating in the standard assessments. The Educational Policy Reform Research Institute 17 eprri All schools in California have been advised that they must improve by annual increments of 10.8 in their English and Language Arts programs before 2014 and that they must improve by increments of 10.5 in their Math programs by the year 2014. However, there have been no changes to the consequences that a school will face for failing to meet its AYP goal since the adoption of the NCLBA. According to the most recent information available, California is continuing to discuss how best to align its present state interventions for low performing schools with the specific requirements of the NCLBA. This will require extensive legislative and regulatory changes. improve instruction in schools. The assessments were used to measure the performance of Maryland schools, not of individual students. Thus, individual student scores were not reported. The MSPAP assessments measured how well students solved problems cooperatively and individually, and how well students applied what they learned to real world problems, and how well students could relate and use knowledge from different subject areas. The MSPAP assessments were criterion-referenced performance tasks linked to student content standards. Performance tasks were administered in reading, mathematics, writing, language usage, science, and social studies. Before NCLBA, California required that schools be accountable for a child’s performance once the child had been in that school district for a year. This has been changed so a student must be in a school for a year’s time before the district is accountable for the student’s performance. At this time, California does not have a way to longitudinally track high school students from entry to graduation, as required by NCLBA. Since the MSPAP was designed for school accountability purposes and not to measure individual student achievement, only a portion of the entire assessment was administered to any given student. In each content area, MSPAP results were reported through five proficiency levels with Level 1 being the lowest level of proficiency and Level 5 the highest. Performance standards were established for both schools and school systems. In order to meet the Satisfactory level of performance, 70% of a school’s students had to score at proficiency Level 3 or above on the assessments. To meet the Excellent standard, 70% of students in a school had to score at level 3 and above, with at least 25% of the students scoring at Level 2 or higher. Maryland In May 1990, the Maryland State Board of Education approved the Maryland Learning Outcomes to be used beginning in 2000. An accountability system was developed to assess schools’ progress toward achievement of these learning outcomes. This system became known as the Maryland School Performance Assessment Program (MSPAP). The MSPAP assessments were administered each May to Maryland’s 3rd, 5th, and 8th graders to test their mastery of the basics and how well they applied knowledge in authentic problem-solving situations. The primary purpose of the MSPAP assessments was to provide information that could be used to Maryland stopped using the MSPAP after the 2001-02 school year and implemented the Maryland School Assessment (MSA) during the 200203 school year. Maryland also discontinued the use of a set of high school exit exams called the Maryland Functional Tests in 2004 and phased in the High School Assessments (HSA). These endof-course assessments measure students’ knowledge 18 Accountability for Students with Disabilities Who Receive Special Education in the core subject areas of English, Algebra/Data Analysis, Government, and Biology. The geometry HSA fulfills the NCLB requirement for a grade 10-12 mathematics assessment. Beginning with the graduating class of 2009 (students entering grade nine in fall of 2005), students will be required to earn a satisfactory score on the HSA to earn a Maryland High School Diploma. No students are exempt from participation in the current Maryland Assessment Program (MAP), and all students must participate in either MSA or the Alternate Maryland School Assessment (ALT-MSA). Assessment Accommodations Students with disabilities currently have access to certain accommodations when taking tests within the Maryland Assessment Program (MAP), provided that the accommodations are aligned with and are a part of daily instruction, and are justified and documented in the student’s IEP, Section 504 Plan, or ELL Plan. There have been some changes in which accommodations were permitted in Maryland since the implementation of NCLBA. Table 5 indicates how Maryland policies have changed regarding selected accommodations since 2001. Roughly half of the selected accommodations were the same for 2001, 2003, and 2005; however, there have been some notable changes across the years. For example, students using a Braille version of the CTBS/5 test in 2001 did not have their scores aggregated, but in 2003 and 2005 this accommodation may be used without implications for scoring. Also, while scores for students using a spell checker in 2001 were invalidated for the language use portions of the CTBS/5, MSPAP, and HSA, this accommodation was not mentioned in 2003 materials and is currently allowed for students. As indicated in Table 5, questions could be read to students who needed that accommodation when the MSPAP was used for accountability purposes; however, students who used the read aloud accommodation received the lowest possible score on the test. In 2002—when test scores were first used for NCLBA accountability purposes—there was much turmoil because most schools in the state were Table 5. Maryland, Selected Accommodations, 2001, 2003, and 2005. Accommodation Large Print Braille Read Aloud Questions Magnification Equipment Amplification Equipment Spell Checker Calculator Scribe Write in Test Booklets Extended Time With Breaks Individual Administration Administration in Student’s Home 2001 A AI AI 2003 A A AC/AI A AI AC A A AI AC A A A A A A A A A A 2005 A A AC/AI A A A A A A A A Note: A = Allowed; AI = Allowed with implications for scoring and/or aggregation; AC = Allowed in certain circumstances; P = prohibited The Educational Policy Reform Research Institute 19 eprri placed into the “needs improvement” category as a result of the widespread use of the read aloud accommodation. In 2003—after overwhelming school district and public demand—the Maryland accommodations policy was revised. Questions can no longer be read aloud on the grade 3 and 4 general reading processes part of the MSA. If the read aloud accommodation is used at other grade levels, the test administration is considered to be nonstandard; however, the score is invalidated only for certain portions of the MSA directly related to general reading processes. Maryland is one of the few states that use a compensatory strategy to aggregate the scores of each student. Scores for students who use a nonstandard accommodation or who are unable to take a portion of a test are based on the remaining sections of the test. In Maryland the aggregate score is the total score across all items and dimensions being assessed. According to Ryan (2002), this approach “allows high scores on some measures to compensate for low scores on others” (p. 298). Alternate Assessment Prior to 2003 students with severe cognitive disabilities who were not able to participate in MSPAP, even with accommodations, were permitted to take the state’s alternate assessment, the Independence Mastery Assessment Program (IMAP). The IMAP assessment tested students in their functional life skills in the areas of academics, communication/ decision making, career/vocational, community, recreation/leisure, and personal management. The IMAP assessments consisted of an on-demand performance assessment accompanied by a portfolio. Student IMAP scores were not counted in the accountability calculations prior to the NCLBA. Between three and five percent of Maryland’s spe- cial education population were eligible to take the alternate assessment. Students participated in IMAP in the same grades as they participate in MSPAP (grades 3, 5, and 8) and in grade 11. A student for whom the IMAP would be inappropriate because of severe medical complications may be excused from the IMAP administration, as determined by a student’s IEP team and documented in the IEP. When the MSPAP program ended, the IMAP alternate assessment was also phased out. As previously mentioned, since 2003 the alternate assessment in Maryland has been called the Alternate Maryland School Assessment (ALT-MSA) which is based on the Maryland standards. Student with severe and complex cognitive disabilities that are unable to participate in the MSA take the ALTMSA. A portfolio is assembled for the ALT-MSA that contains artifacts of student work that document student mastery of standards. The results are reported at three levels of proficiency: Basic, Proficient, and Advanced. Accountability As part of its efforts at school reform, Maryland also produced School Performance Reports, or report cards, which served as the primary accountability tool for MSPAP at the school, system, and state level. These included the School Performance Indicator (SPI), which was a mathematical calculation based on schools’ attendance rates, MSPAP scores, Maryland Functional Test scores, and (for high schools) dropout rates. Individual student scores were not reported for the MSPAP since the test was to be used for school-level accountability and not as a measure of individual student achievement. An SPI index score of 100 indicated that on average a school was meeting the State’s performance standards. A School Improvement 20 Accountability for Students with Disabilities Who Receive Special Education Team, which is present in every public school in the state, was charged with analyzing school and district MSPAP data and using it to adjust curriculum and instruction to meet the Maryland Learning Outcomes (MLOs). Sanctions and awards were determined based on these report cards. For schools that experienced significant progress, monetary rewards and recognition were given. The Maryland State Department of Education conceded in 2001 that the state standard of 70% students meeting proficiency was challenging. At that time, the state had not yet met any of the MSPAP standards. In 1999 only 8 of 24 school systems had more than 50% of their students scoring at Level 3 proficiency or higher. For this reason, focus was placed on schools’ improvement of scores, rather than their actual percentage or comparison to other schools. Thus, only those schools with large portions of students scoring below Level 3, and were declining or not making substantial and sustained progress were identified by the state as reconstitution-eligible schools. Such schools were obliged to submit proposals for how to resolve the problem and, after approval, received supplemental funding, technical assistance, and monitoring. They were expected to make major changes in staffing and school programs. Reconstitution-eligible schools were removed from the list when they showed improvement for three consecutive years and met the state average SPI. In certain cases, when schools still did not make sufficient progress, the state intervened with state reconstitution. In 2000, the first three schools in Maryland were reconstituted, meaning that oversight was shifted from local school system to a third party. As discussed above, in September 2002 the Maryland Department of Education announced that the Maryland School Assessment (MSA) would replace MSPAP beginning in the spring of the 2002-03 school year. Currently, the MSA measures student achievement in grades 3-8 reading and math and grade 10 reading, and is based on the Maryland Content Standards. The MSA is also given in geometry after students complete a geometry course, regardless of the students’ grade level. The Maryland Content Standards specify what students in grades K-12 should know and be able to do in four core content area: Mathematics, English/Language Arts, Science, and Social Studies. The MSA meets the federal requirements under NCLBA that state assessments be administered annually and yield individual student scores. To build the MSA, testing experts began with commercial, norm-referenced tests to produce a norm-referenced score. To create a criterion-referenced component, test items that matched the Maryland Content standards were identified and new items were written to ensure coverage of the content standards. The format of MSA is a mixture of multiple choice questions and short-essay questions. The norm-referenced score will be used to compare students’ performance to same grade peers nationally, and the criterion-referenced score will demonstrate how well the students mastered the Maryland Content Standards. The MSA has three levels of proficiency: basic, proficient, and advanced. The state will use only the criterionreferenced score to make accountability decisions for AYP. In addition, the High School Assessments (HSAs) were launched in 2001-02. New York Before the passage of NCLBA, New York State did not have a uniform accountability system, so districts varied on how they tracked school perThe Educational Policy Reform Research Institute 21 eprri formance. However, during the 1996-97 school year the New York State Education Department published a Resource Guide that put forth learning standards for seven areas: English language arts; The Arts; Social Studies; Math, Science, and Technology; Languages other than English; Health, Physical Education, and Family Consumer Sciences; and Career Development and Occupational Studies. These learning standards formed the basis for the education of all children in New York (www. nysatl.nysed.gov/standards.html) and contributed to New York State schools’ Adequate Yearly Progress standards. In 1999-00, a System of Accountability for Student Success (SASS) was established. This system became New York State’s single accountability system designed to provide information about school effectiveness in preparing students to meet New York State’s learning standards, as indicated above. These standards are aligned with curriculum and the New York State Assessment Program (NYSAP) across elementary, middle, and high school levels. Assessments are administered at 4th grade and 8th grade in English language arts and mathematics. The basic structure of this test has remained the same since 1999. However, in 2003, state administrators began to embed field test questions into the test. Students are also tested on social studies in 5th grade and 8th grade, technology education in 8th grade, and science in 4th grade and 8th grade. At the high school level, the Regents Comprehensive Examination is used as the indicators for NCLBA, along with the annual high school dropout rate. The Regents Comprehensive Examination tests English Language Arts, Social Studies, Science, Math, and Foreign Language, and students are assigned a performance score of 1, 2, 3 or 4, with 4 being the highest attainable level and 2 being the minimum level considered passing. The Regents Competency Test is given to high school students with learning disabilities and the test assesses reading, writing, and mathematics. The Regents Competency Test is graded on a pass/fail basis. Beginning in 2002, the NYSAP began to be tabulated using number-correct scoring that was manipulated into a scale score. New York has established four levels of performance on the NYSAP to determine a school’s Performance Index, which reflects achievement toward state standards. Students receiving a score of Level 1 are identified as having serious academic deficiencies and have shown little or no proficiency in the New York state content standards for their grade level. Level 2 represents students who have shown some knowledge and skill in each of the required state standards for elementary or middle school level students but still need extra assistance to achieve all of the standards required to pass the assessments. A Level 3 means that a student has met standards, while a Level 4 means that a student has exceeded standards. Assessment Accommodations In New York students with disabilities are permitted to use accommodations to meaningfully access the standard assessments. The accommodations that are permitted on the statewide examinations have changed due to the NCLBA. The Board of Regents found it necessary to make changes in accommodations permitted for each of the tests. Table 6 displays the accommodations provided in the NYSAP before and after the NCLBA became effective. Table 7 indicates how New York policies have changed regarding selected accommodations since 2001. Only one of the selected accommodations, 22 Accountability for Students with Disabilities Who Receive Special Education Table 6. Accommodations Permitted in New York: Before and After NCLBA. BEFORE NCLBA AFTER NCLBA Grades 4 and 8 ELA Reading - Oral reading or signing of reading passages (not listening), multiple choice, and/or extended response items Grades 4 and 8 ELA Reading - All directions/items/questions within the grade 4 ELA Session 2, Part1 and Grade 8 ELA Session 1, Part 2: Listening part of test may be read aloud. Grades 4 and 8 ELA Writing All directions, instructions/items/questions may be read aloud to student Students may NOT use spell-checking and/or grammar-checking devices on any parts of ELA test Students may NOT have requirements for spelling, paragraphing, and/or punctuation waived for extended writing times found within the Grade 4 ELA Session: Part 1; Session 2: Part2; Session 3; or in the Grade 8 ELA Session1: Part2; Session 2:Part2 Grade 4 Math Grades 4 and 8 ELA Writing Use of spelling/grammar checking device on tests measuring spelling and grammar permitted Deletion of spelling, paragraphing, and/or punctuation requirements on tests assessing spelling and/or grammar allowed Elementary, Middle, and High School Math - Use of calculator or abacus on math tests measuring calculation skills such as addition, subtraction, multiplication, division is permitted - No calculators allowed for all students Grade 8 Math - Grade 4 Science; Grade 5 Social Studies; Grade Science and Social Studies; Intermediate-Level Technology Educational Tests - Assessments did not exist before NCLB Book 1= No calculators allowed (test is based on basic knowledge of addition, subtraction, multiplication, and division) Book 2 = calculators allowed for students with disabilities if indicated in IEP or Section 504 Plan. Grade 4 Science; Grade 5 Social Studies; Grade Science and Social Studies; Intermediate-Level Technology Educational Tests - Grade 4 Science; Grade 5 Social Studies; Grade 8 Science and Social Studies; Intermediate-Level Technology Educational Tests - Assessments did not exist before NCLB Tests must be read aloud to students with disabilities whose IEP or Section 504 Plan require this accommodation Students with disabilities can use calculators if indicated by IEP or 504 Plan Grade 4 Science; Grade 5 Social Studies; Grade 8 Science and Social Studies; Intermediate-Level Technology Educational Tests - - Tests must be read aloud to students with disabilities whose IEP or Section 504 Plan require this accommodation Students with disabilities can use calculators if indicated by IEP or 504 Plan The Educational Policy Reform Research Institute 23 eprri read aloud questions, has changed since the implantation of NCLBA. In 2001, read aloud was allowed; however, in 2003, this accommodation was allowed only in certain circumstances. For example, the listening part of the test could be read aloud to students on the Grade 4 ELA Session 2, Part 1 and Grade 8 ELA Session 1 Part 2, but no other parts of the student test books could be read aloud. Alternate Assessment Students with severe disabilities must be declared exempt from the general assessment by the Superintendent and Committee of Special Education (CSE) based on the following criteria: (1) severe cognitive disability, significant deficit in communication/language, or significant deficits in adaptive behavior, (2) requires a highly specialized educational program that facilitates the acquisition, application, and transfer of skills across natural environments, and (3) requires educational support systems, such as assistive technology, personal care services, health/medical services, or behavioral intervention. These decisions are made on a case by case basis. Although students may be exempt from testing in the standard ELA and mathematics assessments due to disabilities and limited English proficiency, they must then take an alternate assessment that models the standard test. In 1999, the State Education Department began development of an alternate assessment for students with severe disabilities. A statewide task force of educators, researchers, parents, and advocates, along with the State’s alternate assessment testing contractor developed guiding principles, process and participation criteria, guidelines for creating assessment tasks to measure progress on the alternate performance indicators, and a scoring rubric. The New York State Alternate Assessment (NYSAA) was created for students with severe disabilities who are at the ages of 10-11, 14-15, or 17-18. The NYSAA is a data folio assessment in which students demonstrate their performance toward meeting the alternate performance indicator level of the New York State Learning Standards. In addition to this assessment, an alternate assessment Table 7. New York: Selected Accommodations, 2001, 2003, and 2005. Accommodation Large Print Braille Read Aloud Questions Magnification Equipment Amplification Equipment Spell Checker Calculator Scribe Write in Test Booklets Extended Time With Breaks Individual Administration Administration in Student’s Home 2001 A A A A A A A A A A A A 2003 A A AC A A A A A A A A A 2005 A A AC/AI A A A A A A A A A Note: A = Allowed; AI = Allowed with implications for scoring and/or aggregation; AC = Allowed in certain circumstances; P = prohibited 2 Accountability for Students with Disabilities Who Receive Special Education has been developed for students who have limited English proficiency in correspondence to the ELA assessment. This assessment is called the New York State English as a Second Language Achievement Test (NYSESLAT). For the assessments of Grade 4-Elementary Level Science, Grade 8-Intermediate Level Science and Grade 8-Intermediate Level Social Studies, written translation is provided in Chinese, Korean, Russian, Haitian Creole, and Spanish. In addition, these assessments may be translated orally for those students whose first language is not available in the written translation from the Department (www.emsc.nysed.gov/ciai/testing/ eliinfogen/gr48sciss03.htm). The scores of students who participate in the NYSAA are included with a maximum of 1% of the scores used in calculating the Performance Index. In March 2002 New York State Department of Education modified the NYSAP system for certain students with disabilities by adding a provision that permitted locally selected assessments to be used in lieu of the state assessments (4th and 8th grade and commencement assessments). Students who may qualify for locally selected assessments are those who, due to a performance gap resulting from the students’ unique disability needs, are unable to meet the grade/age level expectations of the regular state assessments and do not meet the eligibility criteria for the NYSAA. The locally selected assessments must be standardized, measure a student’s achievement of the New York State learning standards, and be of sufficient technical quality to measure progress toward the student’s performance indicator level. For purposes of accountability, students who participate in locally selected assessments are counted as though they performed at level one on the NYSAP elementary and middle school ELA and mathematics assessments. Locally developed assessments were first available in the spring of 2002 and are to be in place for two years while the Department of Education conducts pilot elementary and intermediate assessments of students who have been recommended by the CSEs. In April 2003 the provision was extended for the 2003-04 school year. Accountability The performance of all students, regardless of whether they take the standard or alternate assessment, must be reported through the Local Educational Agency Program (LEAP). LEAP is an electronic software device that collects and reports on State assessments administered in the elementary and middle school levels. In using this device, it will determine the school’s Performance Index, which is the percentage of full-year students tested who scored at Level 2 and above, and the percentage who scored at Level 3 and above on each assessment. This will determine if the school has achieved the State standard and its AYP target. Another data collecting and reporting device is called the System for Tracking Educational Performance (STEP). STEP is software designed to track all students’ performances in grades 9-12. This tracking device also determines if schools have met the State standard of the Performance Index and AYP. Prior to NCLBA, AYP in New York State was determined by using the performance of Title 1 schools, rather than measuring the performance of all students. A district was considered to have made AYP criterion if at least 50% of its schools met the State standard or reached AYP on an assessment. Since the No Child Left Behind Act was established, the following changes have been made to STEP: the NYSESLAT, NYSAA, and the The Educational Policy Reform Research Institute 25 eprri graduation/dropout rate are now included in the system’s accountability and the performance index increased. Now the district is held accountable for all student subgroups, including individuals who have been placed in programs outside of the district (i.e., approved private placements). Also prior to NCLBA there was a single measure for elementaryand intermediate-level ELA and a single measure for elementary- and intermediate-level math that combined results of Grade 4 and 8 assessments. Under NCLBA, these measures are now separated by grade level. LEAP and STEP determine if a school has met each State standard and appropriately labels each school as “Meeting Standard,” “Below Standard,” or “Farthest from Standard.” Schools that are farthest below State standards are designated as Schools Under Registration Review (SURR). SURR, which was developed before the NCLBA, is a program created by the Board of Regents in attempt to close the gap of student performance. A SURR school is labeled “School in Need of Improvement” if the school performs below State standards and fails to make AYP for 2 consecutive years. A school may also be labeled a “Corrective Action School” if the school, which was previously under “Need of Improvement,” fails to make AYP in 2 out of the next 3 years. Even though SURR was created prior to the NCLBA, the program has changed since the implementation of NCLBA. Schools used to be expected to demonstrate that 90% of their students were achieving at the minimally acceptable performance level (Level 2). If a school did not demonstrate this, the school was placed on registration review and given 3 years to improve performance. If no improvement was evident, the schools had to close or reorganize. However, these schools were not provided with substantial guidance on how to reorganize. Now, a Local Assistance Plan (LAP) and AYP Target must be established for the school to raise its State standards. The district in which the school is located develops the LAP. Since NCLBA, there is a need for additional assistance to ensure that a sufficient percentage of the subgroups within a SURR school will achieve proficiency in ELA and mathematics assessments. In addition, the Commissioner established the AYP Target for each school below State standards. For elementary and middle schools that are below the targeted performance level, the targets are set in 3-year increments and the school is expected to close the gap by 15% each year. For high schools, the goal is to demonstrate improvement over the previous year’s performance. Texas From 1980-1984 Texas assessed minimum basic skills in reading, writing, and mathematics with the Texas Assessment of Basic Skills (TABS) which became the Texas Assessment of Minimum Skills (TEAMS) from 1985-1989. Changes in state law resulted in a new assessment, the Texas Assessment of Academic Skills (TAAS), which was first administered in fall, 1990. The TAAS was a criterion-referenced test designed to measure whether students had met the content standards covered in the statewide curriculum, known as the Texas Essential Knowledge and Skills (TEKS), in reading, mathematics, and writing. The TEKS, which became effective September 1998, consisted of two sets of subject areas: “foundations” and “enrichment.” The “foundation” subject areas included English language arts and reading, mathematics, science, and social 26 Accountability for Students with Disabilities Who Receive Special Education studies. The “enrichment” subject areas consisted of languages other than English, fine arts, health, physical education, technology applications, and career and technology education. All of the subjects in the “foundation” curriculum are mandatory for state graduation credits. Under the “enrichment” curriculum, health, technology applications, and physical education are required for state graduation credits and the remaining subjects are recommended. In the “foundation” curriculum, districts were required to provide instruction in the essential knowledge and skills of the appropriate grade levels, whereas in the “enrichment” curriculum, districts were expected to use essential knowledge and skills as guidelines for instruction. The Texas Assessment of Academic Skills (TAAS) was used to assess the school districts and schools on the TEKS. Student scores were reported on school report cards as the percentage of students who “passed” the TAAS. Before the 1996-97 school year, Texas statutes permitted for exemption of students receiving special education from taking the Texas Assessment of Academic Skills (TAAS). However, starting in 1996-97, students with disabilities who had been exempt from the TAAS were required to take an alternate assessment. Students were now assessed at their appropriate instructional levels, as determined by their admission, review, and dismissal (ARD) committees (similar to IEP teams in other states), rather than at their assigned grade level. The student’s ARD team then determined whether a student would take one, more than one, or none of the TAAS subject tests. This decision was to be documented in the student’s IEP. The student’s IEP was also supposed to document any accommodations and/or modifications that the student would use on either the TAAS or the alternative assessment. In 1999 the Texas Legislature mandated that the TAAS be replaced by a new assessment system known as the Texas Assessment of Knowledge and Skills (TAKS). The TAKS was administered beginning in the 2002-03 school year. The TAKS measures the statewide curriculum in reading at Grades 3-9; in writing at Grades 4 and 7; in English Language Arts at Grades 10 and 11; in mathematics at Grades 3-11; in science at Grades 5,10, and 11; and social studies at Grades 8, 10, and 11. The Spanish TAKS is administered at Grades 3 through 6. The TAKS requires students to do more analysis and use more higher-order thinking skills than had been required on the TAAS. There are three performance categories for students on the TAKS: “Commended Performance,” which recognizes high level performance; “Met the Standard,” which means the student passed the test; and “Did Not Meet the Standard,” which means the student failed the test. Satisfactory performance on the TAKS at Grade 11 is a prerequisite to a high school diploma. Assessment Accommodations Testing accommodations for students with disabilities were based on the accommodations that the student routinely received in the classroom, the needs of the student, and the accommodations allowed for the test. The only accommodations that were not allowed on the TAAS were ones that would invalidate the test results. Table 8 shows the accommodations permitted and not permitted on the TAAS during the 2001-02 school year. Districts were also allowed to request permission from the Texas Education Agency to use other accommodations. Specific accommodations must be documented on the student’s IEP to in order to be used as an acThe Educational Policy Reform Research Institute 27 eprri Table 8. Texas: Permitted and Prohibited Accommodations, TAAS, 2001-02. Permitted Accommodations - Braille and large-print versions - Magnifying glass, color transparency, or place marker - Variety of methods for responses: * Handwriting * Typewriting * Computer keyboard entry * Verbal responses - Marking responses in the test booklet rather than the answer booklet - Individual administration of the test - Test administrator reading aloud the mathematics, social studies, and science test questions. - Instructions before and after test may be signed or translated into Native Language. - Writing prompt may be signed to student with hearing impairment - On composition section, may dictate composition or tape record essay or type Prohibited Accommodations - Reading assistance or writing and reading tests - Use of English-Languages or foreign language reference materials. - Translating during the test - Test questions, answer choices, passages, and writing prompts may not be rephrased or clarified - Use of calculator, slide ruler, or math chart - Other accommodations that would make the test invalid commodation on the TAKS. If a student with a disability needs accommodations that are not allowed on the TAKS, the student must take an alternative assessment (e.g., the SDAA II). There are five accommodations not allowed on the TAKS: • Student may not receive reading assistance (except on the Grade 3 mathematics test); • Student may not use foreign-language reference materials; • Student may not use a calculator on Grades 3-8 mathematics tests or the Grade 5 science test; • Test items may not be translated; and • Test questions, answer choices, selections, and writing prompts may not be rephrased or clarified. Table 9 indicates how Texas policies have changed regarding selected accommodations since 2001. Many of the accommodations were the same for all three years; however, there were some notable changes during the years reviewed in this study. For example, magnification equipment was not specifically listed as an accommodation in 2001 or 2003, but it was included in the accommodations policy for the 2004-05 school year. In 2001, students could only write their responses in the test booklet if they had a disabling condition that interfered with their ability to record machine-scorable responses—but that stipulation has been removed in more recent years. The read aloud accommodation was not permitted on the reading and writing tests during any 28 Accountability for Students with Disabilities Who Receive Special Education Table 2001, 2003, and 2005. Table9.9.Texas: Texas:Selected SelectedAccommodations, Accommodations, 2001, 2003, and 2005 Accommodation Large Print Braille Read Aloud Questions Magnification Equipment Amplification Equipment Spell Checker Calculator Scribe Write in Test Booklets Extended Time With Breaks Individual Administration Administration in Student’s Home 2001 A AC AC 2003 A A AC AC P AC AC P P A A A A A A 2005 A A AC A AC A A Note: A = Allowed; AI = Allowed with implications for scoring and/or aggregation; AC = Allowed in certain circumstances; P = prohibited of the years reviewed in this study, but was allowed on mathematics, social studies, and science tests if the student was identified as having dyslexia or a related disorder. Alternate Assessment Special education students in Grades 3-8 who are receiving instruction in the Texas Essential Knowledge and Skills (TEKS) but for whom TAKS is an inappropriate measure of their academic progress, participate in the State-Developed Alternative Assessment (SDAA II). In 2005, the SDAA for grade 9 and 10 will be available. The SDAA II assesses the areas of reading, writing, and mathematics. It is administered on the same schedule as TAKS and is designed to measure annual growth based on appropriate expectations for each student as decided by the student’s ARD committee. Students who participated in the SDAA II in 2003 were included as non-participants for calculating the AYP assessment measures. A proposal to evaluate the SDAA II performance results for AYP was developed for the 2004 school year. A small number of students with disabilities, usually those with the most severe cognitive disabilities, may be exempt from both the TAKS and the SDAA II if the ARD committee determines that neither are appropriate methods of assessing the student’s progress. These students participate in a Locally Designed Alternate Assessment (LDAA). The ARD committee determines the areas in which a student is not exempt from state testing and if the student meets the performance criteria established in the IEP. Beginning with the 2000-2001 school year, the local school district must report the results of students grade 3 through 8 assessed on an LDAA to the Texas Education Agency. Students tested on an LDAA in 2003 were not included in the AYP calculations but the state began to collect and evaluate the LDAA performance results for AYP for the 2004 school year. Accountability Texas has had an accountability system in place for more years than most other states. State statutes in place during the 1990s required annual district The Educational Policy Reform Research Institute 29 eprri and campus (e.g., school-level) performance ratings of “Exemplary,” “Recognized,” “Academically Acceptable,” and “Academically Unacceptable.” Ratings were based on TAAS scores of disaggregated subgroups (e.g., African American, Hispanic, White, and economically disadvantaged) and drop out rates. Over the period of time that TAAS was used, a number of changes were made to the rating criteria. For example, in 1999 a rating of “Academically Acceptable” meant that at least 45% of “all students” and each student subgroup must have passed each section of the TAAS. In 2000 the “Academically Acceptable” rating was increased to require that 50% or more students passed the TAAS. In 2000, for the first time, TAAS results in reading, mathematics, and writing for special edu- cation students in grades 3-8 and 10 were included in the calculations of the campus ratings. Additional acknowledgement indicators, such as attendance rate, campus comparable improvement in reading and/or math, college admissions test results, or recommended high school participation rate, were used to give supplementary recognition to districts and campuses. Table 10 shows the standards that districts and campuses were expected to reach in 2002. Prior to NCLBA, if a school district or campus received the lowest accountability rating, an “Academically Acceptable” peer review team was sent to visit the site(s) to determine any deficiencies. A preliminary report of the peer review team’s findings Table 10. Accountability Rating Standards for 2002 (Base Indicator Standards). Exemplary Recognized Academically Acceptable/ Acceptable At least 80% passing each subject area At least 55% passing each subject area Academically Unacceptable/ LowPerforming Below 55% passing each subject area (“all students” and each student group) At least 80% passing each subject area (“all students” and each student group) At least 50% passing each subject area (“all students” and each student group) Below 50% passing each subject area (“all students” (“all students” only) only) (“all students” only) (“all students” only) 1% or less 5% or less Above 5% Spring 2002 TAAS At least 90% o Reading passing each o Writing subject area o Mathematics (“all students” and each student group) o Social At least 90% Studies passing each subject area 2000-01 Dropout Rate 2.5% or less (“all students” (“all students” (“all students” and each and each and each student group) student group) student group) 30 Accountability for Students with Disabilities Who Receive Special Education (“all students” and each student group) was written and sent to the district from the Texas Education Agency (TEA). The district and campus would develop and implement an improvement plan to address the area(s) of poor performance. The district and/or campus were required to provide written evidence of the improvements made in the areas of deficiencies to the TEA. The TEA would then determine if more corrective measures needed to be conducted or if the file should be closed. If a school district or campus received the lowest accountability ratings for two or more consecutive years, the level of state intervention would increase. In 1995, the Public Education Grant (PEG) was created to allow parents with students attending poor performing schools to transfer their students to schools in other school districts that had higher performance results. In response to the NCLBA requirements, Texas created an Adequate Yearly Progress (AYP) Plan that would annually evaluate all public school districts, campuses, and the state as a whole. The Texas Adequate Yearly Progress Plan was approved by the United States Department of Education (USDE) in June 2003 as meeting the NCLBA requirements. The 2003 AYP status of Texas and its specific school districts and campuses were based on the results of Texas Assessments Knowledge and Skills (TAKS). Each public school district, campus, and the state as a whole are given an AYP Status Label. These AYP Status Labels are based on the districts, campuses, and state meeting criteria on three measures. NCBLA requires states to include at least three measurements in their accountability system: Reading/Language Arts, Mathematics, and one other measure (graduation rate for schools and districts offering grade 12 or attendance rate for schools and districts that do not have grade 12). NCLBA requires that the performance standard of “meet AYP” needs to be met for: the “all student” group as well as each “student subgroup” (i.e., African Americans, Hispanic, White, economically disadvantaged, special education and limited English proficiency students) which meets minimum size criteria. In Texas, if a “student group” within districts and campuses consists of 50 or more students (summed across grades 3-8 and 10) and make up at least 10% of all test takers in the subject or consists of 200 or more students, even if it does not make up 10% of test takers in the subject, the “student group” meets minimum size criteria and is evaluated against performance standards. If the performance standard is not met by the “all student” group or a separate “student subgroup,” it must be shown that there was a 10% decrease in the percent of students not passing the “Met Standard” performance standards on TAKS and any improvement on the other measure. AYP standards were set to increase every year to reach 100% proficiency by 2014. The 2003-04 AYP criteria that each school district, campus, and the state needs to meet is displayed in the Table 11. In addition, if school districts, campuses, or the state does not meet the Standards for Reading/Language Arts and Mathematics assessments presented in Table 14, they may prove to meet the performance gains criteria that are displayed below as well. If a school district or campus does not meet its AYP standards for Reading/ Language Arts, Mathematics, and one of the other measures, the state will impose sanctions on the school district or campus. If the school district or campus receives funding under Title 1, Part A and fails to make the AYP standards for two consecutive years, the The Educational Policy Reform Research Institute 31 eprri Table 11. Texas AYP Criteria and Standards. Reading/Language Arts 2003-04 tests (TAKS, SDAA, LDAA, and RPTE in Grades 38 & 10) All students and each student group that meets minimum size requirements: African American Hispanic Economically Disadvantaged Special Education Limited English Proficient Mathematics 2003-04 tests (TAKS, SDAA, and LDAA in Grades 3-8 & 10) All students and each student group that meets minimum size requirements (see above) Other Measures** All students Graduation Rate Class of 2003 Attendance Rate 2002-2003 Performance Standard: 47% % counted as proficient on test* for students enrolled the full academic year subject to the Federal 1% cap Participation Standard: 95% Participation in the assessment program for students enrolled on the date of testing (no more than 5% of students absent) Performance Standard: 33% % counted as proficient on test* for students enrolled the full academic year subject to the Federal 1% cap Participation Standard: 95% Participation in the assessment program for students enrolled on the date of testing (no more than 5% of students absent) Graduation Rate Standard 70.0% or any improvement Graduation Rate for high schools, combined elementary/secondary schools offering Grade 12, & districts offering Grade 12 OR Performance Improvement: 10% decrease in percent not proficient on test* and any improvement on the other measure (Graduation Rate or Attendance Rate) OR OR Average Participation Rate: 95% participation based on combined 2002-03 and 2003-04 assessment data Performance Improvement: 10% decrease in percent not proficient on test* and any improvement on the other measure (Graduation Rate or Attendance Rate) OR Average Participation Rate: 95% participation based on combined 2002-03 and 2003-04 assessment data Attendance Rate Standard 90.0% or any improvement Attendance Rate for elementary schools, middle/junior high schools, combined elementary/secondary schools not offering Grade 12, & districts not offering Grade 12 Key: TAKS = Texas Assessment Knowledge and Skills SDAA = State-Developed Alternative Assessment LDAA = Locally-Determined Alternate Assessment RPTE = Reading Proficiency Tests in English school district or campus is identified for School Improvement as required in Section 1116 of the NCLBA. The parent(s)/guardian(s) of a student in a Title 1, Part A school that is identified as “Year 1 of School Improvement” will have the option of transferring the student to another school in the district not in school improvement status. Students in Title 1, Part A schools identified as “Year 2 of School Improvement” are eligible for supplementary services, such as tutoring, remediation, and/or academic intervention outside of the regular school day. If a school district or campus does not meet the AYP standards for Reading/Language Arts, Mathematics, and one of the other measures for 2 consecutive years and is not receiving funds under Title 1, the school district or campus will be required to amend their school improvement plan to address the deficit areas. These sanctions will be periodically reevaluated as decisions are made related to the state accountability system. Summary Evident from the changes that have occurred in EPRRI’s four study states is how difficult it is to 32 Accountability for Students with Disabilities Who Receive Special Education establish a performance profile in a state. Nonetheless, we attempted to document, both quantitatively and qualitatively, aspects of state and district performance trends for students with disabilities. We have also attempted to better understand factors that may influence those trends as well as identify schools that are getting better results for students with disabilities. Following is a compilation of our quantitative investigation. Separate EPRRI reports document qualitative findings. The Educational Policy Reform Research Institute 33 eprri 3 Accountability for Students with Disabilities Who Receive Special Education State and District Data Sources and Findings As noted, we conducted our research in the following states: California, Maryland, New York, and Texas and in two local school districts from each of these states: Long Beach Unified (CA), New Haven Unified (CA), Montgomery County Public Schools (MD), Carroll County Public Schools (MD), Cypress Fairbanks Independent School District (TX), Garland Independent School District (TX), Rochester City Schools (NY), and North Colonie Central School District (NY). To select the districts, EPRRI staff and state representatives identified important accountability features across which the school districts were selected to vary. These key accountability features included high stakes versus low stakes accountability consequences, recentness of reforms, stability versus instability of reform efforts, participation of students with disabilities in all accountability reports, the use of alternate assessments, and the availability of school and district-level data on the performance of students with disabilities. Representatives from each state assisted in securing the participation of the districts. However, during the course of the study and as a result of the passage of NCLBA, all of the districts moved into highstakes accountability and the availability of publicly reported data increased. Further, as noted earlier in this report, each state initiated major changes in its state assessment in order to comply with the NCLBA requirements. dents receiving special education services, percent of students identified as English Language Learners (ELL), race and ethnicity, assessment participation rates for students with disabilities, performance of students receiving special education services, and performance of students not receiving special education services. Performance data were retrieved for elementary and middle school levels in the core subject areas of reading/ELA and mathematics. We did not include high school data because it was inconsistently reported among the four states. The data regarding assessment participation rates and the performance of students were those available for each tested grade in a state. For example, the New York assessments were administered at the 4th and 8th grades whereas California assessed students at all grade levels between 2nd and 8th. Students with disabilities attending special schools, centers, or nonpublic settings were not included in these analyses unless the student’s participation and performance were reported at the home school and were included in that school’s accountability. The data were collected primarily through Web site searches, school and district report cards, policy documents, technical reports, and memoranda on state, district, and school Web sites. Table 12 provides the primary Web sites for the performance and demographic data. Data Sources We examined various reported data obtained from the Web sites of each core state and district for the school years 1999-00, 2000-01, and 200102. Data included: general education and special education enrollment figures, percent of students receiving free and reduced lunch, percent of stuThe Educational Policy Reform Research Institute 35 eprri Table 12. State Assessments, Data Sources and Web Sites. State California Maryland New York Texas State Assessment Standardized Testing and Reporting Program (STAR) Maryland School Performance Assessment Program (MSPAP) Accountability for Student Success (SASS)/New York State Assessment Program (NYSAP) Texas Assessment of Academic Skills (TAAS) Source for Data Used in Report Star Reports Dataquest Maryland School Performance Report School District Report Cards "655 Report" AEIS Reports District Snapshot Report To verify the data collected, we contacted relevant special education state and district personnel and district assessment and accountability personnel by phone and/or e-mail. These persons provided valuable additional information, assisted in interpretation of the statistics, and helped to resolve inconsistencies in the data. Additionally, during on-site interviews, data were reviewed by key personnel to check for accuracy; they answered additional questions about the variables and interpretations of the data. Tables 13 and 14 present the state and district key demographics. The following section presents key findings related to assessment participation and performance for the 2000-01 school year. We opted to present only the last year of data after analyses revealed little change occurred in the demographics of the four states and eight districts. On average, general education and special education student enrollment remained relatively stable over the 3 school years. Web Site http://star.cde.ca.gov/star200 1/default.htm http://www.cde.ca.gov/demo graphics/ http://msp.msde.state.md.us/i ntroduction/index.asp http://www.emsc.nysed.gov/ http://www.emsc.nysed.gov/i rts/ch655_2002/home.html http://www.tea.state.tx.us/per freport/aeis/index.html http://www.tea.state.tx.us/per freport/snapshot/index.html Findings The percentages of students with disabilities reported as participating in the state large-scale assessments in each state and district are in Tables 15 through 25. As previously indicated in the Data Sources Section, only elementary and middle school level grade data were collected. We present performance as the percent of students who scored at or above the state-defined acceptable achievement standard. We refer to this as “proficient” although all states did not use this term during the years that we collected data. Percentages were calculated using the total number of students within a group who took the assessment as the denominator. The final column of each table shows the difference between the percent of general education students who met proficiency and the percent of students with disabilities who met the proficiency standard. Further, we report data only for 2 years of our data collection. Assessment participation rates and performance trends are presented by state. 36 Accountability for Students with Disabilities Who Receive Special Education Table 13. Selected State and School District Demographic Data: 2000-2001 School Year. State/School District Enrollment California Long Beach Unified New Haven Unified Maryland Carroll Co. Montgomery Co. New York North Colonie Central Rochester City Texas Cypress-Fairbanks Garland 6,050,895 93,694 13,775 852,929 27,528 134,180 2,823,602 5,500 37,885 4,021,641 63,497 50,312 % Economically Disadvantaged % Students Receiving Special Education Services (k-12) 46.8 68.3 26.7 30.0 8.0 22.7 51.0 6.3 67.2 49.3 20.0 33.0 10.4 7.4 8.9 13.1 12.2 14.0 11.8 9.1 15.1 12.0 11.0 13.0 Ethnic Group % Students Receiving Bilingual/ESOL Services 25.0 33.8 25.0 2.8 0.3 7.6 8.4 2.5 6.5 13.0 10.0 13.0 % African American % Hispanic % White % Other 8.4 19.7 9.8 37.1 2.3 21.2 20.1 3.2 62.8 14.0 10.0 17.0 43.2 45.4 29.8 4.8 0.8 16.2 18.4 1.4 18.9 41.0 23.0 29.0 35.9 17.8 19.8 53.4 95.6 49.0 55.1 88.5 16.1 42.0 59.0 47.0 12.5 17.1 40.6 4.7 1.2 13.6 6.4 6.9 2.2 3.0 8.0 7.0 2000-2001 Public School Enrollment *(The percentage of students receiving special education services for all states and districts was calculated by dividing the total number of students in the given state or district by the total number of students receiving special education services in grades kindergarten through 12 and multiplying by 100). Table 14. Selected State and School District Staff and Financial Statistics, 2000-2001 School Year. State/School District California Long Beach Unified New Haven Unified Maryland Carroll Co. Montgomery Co. New York North Colonie Central Rochester City Texas Cypress-Fairbanks Garland Average Number of Students/Teacher 20.7 21.0 20.0 15.3 17.5 15.7 12.9 14.3 11.7 14.7 15.5 16.3 Total Expenditure per Pupil for General Education $6,360 6,516 6,583 7,622 6,582 10,200 6,150 5,425 5,547 5,923 5,674 5,006 1 2002 data. Note: California defines the total expenditure per pupil as the "current expense of education per unit of average daily attendance"; Maryland defines total expenditure per pupil as the "local operating budget from federal, state, and local sources." The Educational Policy Reform Research Institute 37 eprri California California has the largest enrollment of the four study states, with over six million students. Nearly half (46.8%) of the students in the state are economically disadvantaged. About 10% of California’s students receive special education services and 25% of the students receive services for English for speakers of other languages [ESOL]. The largest ethnic group represented by California students is Hispanic (43.2%) and the smallest is African American (8.5%). On average, class size in California is about 20 students per teacher, which is the largest class size of the four study states. In terms of average per pupil expenditures for the 2000-01-2001-02 school year, California ranked second behind the state of Maryland. Beginning with the 2000-01 school year it was possible to calculate a proxy for assessment participation in California by dividing the number of students reported as receiving any special education service (taken from STAR performance http://star. cde.ca.gov/ ) by the number of special education students enrolled in December (taken from http:// data1.cde.ca.gov/dataquest). Caution must be taken when interpreting these participation rates due to student mobility from December to the time that state assessments are administered in the spring. Tables 15 to 17 present the California participation data. The participation rate for students with disabilities taking the SAT 9 Reading increased slightly in grades 3 and 8 between the years 2000 and 2002. Similarly, New Haven Unified School District showed a dramatic increase in participation at all grade levels assessed. However, data for Long Beach Unified revealed an increase in participation only at grade 3. State average participation of students with disabilities in the SAT 9 Mathematics increased at all grades except grade 4 where a 0.2% decrease occurred. The assessment participation rate for students with disabilities in Long Beach Unified was below the state average and decreased at grades 4, 5, and 8. The New Haven Unified School District reported increased participation rates at all grade levels from 2000-2002. Based on the California performance data (Table 18) there was an increase in performance of students with disabilities on the 3rd and 8th grade CST Reading assessment. However, in grades 4 and 5, participation levels decreased and the performance levels increased. Long Beach’s participation rates decreased in grades 3, 5, and 8, and increased 2% at grade 4 between 2000-01 and 2001-02 school years. Grades 3 and 5 performance levels increased for students with disabilities. The New Haven Unified School district showed a 20% increase in participation at grade 3 with only a slight increase at grades 4 and 5. The performance of students with disabilities declined at grades 3 and 4. In grade 5, both performance and participation increased and both decreased at grade 8. The participation rate of students with disabilities on the CST Mathematics was only available at grades 3, 4, and 5. The state indicated there was only a slight increase in participation in 3rd grade, and a decrease of participation in grades 4 and 5 between the two years. While student performance increased at each of these grades, the Long Beach Unified School District data document a decrease in participation at grades 3 (-2.7%) and 5 (-1.8%), and an increase at grade 4 (3.2%). Conversely, performance levels increased at grades 3 and 5. In New Haven Unified School District the participa- 38 Accountability for Students with Disabilities Who Receive Special Education Table 15. California State and District-Level Participation Rates for Students with Disabilities on SAT 9 Reading Assessment: 2000-2002. State/School District California Long Beach Unified New Haven Unified 3rd Grade % 2000-01 2001-02 55.0 56.8 58.5 60.8 67.1 84.8 4th Grade % 2000-01 2001-02 50.9 50.8 52.6 51.8 63.4 69.4 5th Grade % 2000-01 2001-02 50.1 49.4 46.5 45.6 68.3 71.8 8th Grade % 2000-01 2001-02 50.9 51.3 52.0 51.8 60.6 67.7 (The percentages were calculated by dividing the total number of students in each grade who receive special education services by the number of students who participated, in the SAT9, regardless of accommodations, and multiplying by 100). Table 16: Numbers and Percentage Change of Special Education Students Enrolled in December and Tested on SAT 9 2000-01 and 2001-02 School Years. Grade Level Tested 3 4 5 8 % Change Long Beach USD Enrolled Tested 00-01 605 484 546 600 01-02 640 597 509 570 3.6 00-01 354 255 254 312 01-02 389 309 232 295 4.3 New Haven USD Enrolled Tested 00-01 82 93 102 96 01-02 79 111 85 102 1.1 00-01 55 59 70 58 California Enrolled 01-02 67 77 61 69 13 00 - 01 50,172 55,698 57,724 50,342 01-02 50,342 54,238 56,761 52,286 -.1 Tested 00-01 27,597 28,354 28,907 25,620 01-02 28,906 27,840 28,282 26,975 1.3 Table 17. California State and District-Level Participation Rates for Students with Disabilities on ELA California Standards Tests: 2000-01 and 2001-02. State/School District California Long Beach Unified New Haven Unified 3rd Grade ELA 00-01 01-02 64.8 66.6 77 76.7 58.5 78.5 4th Grade ELA 00-01 01-02 64.3 58.6 72.3 74.4 64.5 63.1 5th Grade ELA 00-01 01-02 65.4 63.2 78 75.6 70.6 69.4 8th Grade ELA 00-01 01-02 67.3 68.5 82.8 75 72.9 72.5 (The percentages were calculated in the same manner as those previously mentioned). tion rate for students with special needs increased dramatically at grade 3 (17.6%) and showed a slight increase for grade 4 as well. Grade 5 participation decreased by 2.2%. The fluctuations in the relationships of performance and participation across years and grade levels make it difficult to determine if the achievement gap between students with disabilities and their general education peers has begun to close in reading or mathematics. However, the state-level performance for students with disabilities increased at all grades while assessment participation rates varied. The achievement gap between general education students and students with disabilities in California decreased over the 1998-99 school year to 2000-01 school years. For grades 3 and 5, the difference between general education students and special education students who reached proficiency on the CST Reading decreased by 2.0% and in grade 4 the gap closed by 1.0%. However, the two school districts varied. The gap in Long Beach widened The Educational Policy Reform Research Institute 39 eprri Table 18. California Reading and Mathematics: Percent of State Assessment Participants Meeting or Exceeding Proficiency Standards on the SAT9. Reading Grade Level Grade 3 Grade 4 Grade 5 Grade 8 State/School District (A) California Long Beach Unified New Haven Unified California Long Beach Unified New Haven Unified California Long Beach Unified New Haven Unified California Long Beach Unified New Haven Unified Mathematics Grade Level State/School District (A) Grade 3 California Long Beach Unified New Haven Unified California Long Beach Unified New Haven Unified California Long Beach Unified New Haven Unified California Long Beach Unified New Haven Unified Grade 4 Grade 5 Grade 8 General Education (B) 98-99 99-00 00-01 42.0 45.0 47.0 32.0 37.0 39.0 49.0 52.0 51.0 43.0 46.0 48.0 33.0 32.0 41.0 46.0 55.0 54.0 44.0 45.0 47.0 33.0 33.0 36.0 44.0 47.0 49.0 50.0 51.0 52.0 39.0 39.0 42.0 51.0 56.0 53.0 General Education (B) 98-99 99-00 00-01 49.0 57.0 61.0 46.0 56.0 59.0 59.0 69.0 68.0 45.0 52.0 56.0 39.0 45.0 52.0 50.0 64.0 66.0 47.0 52.0 56.0 39.0 42.0 47.0 50.0 60.0 64.0 47.0 50.0 51.0 37.0 40.0 45.0 50.0 52.0 55.0 Students with Disabilities (C) 98-99 99-00 00-01 23.0 28.0 30.0 26.0 29.0 27.0 31.0 12.0 24.0 21.0 25.0 27.0 25.0 24.0 29.0 14.0 24.0 10.0 17.0 21.0 22.0 18.0 19.0 25.0 17.0 9.0 20.0 13.0 15.0 15.0 11.0 14.0 9.0 18.0 17.0 7.0 Students with Disabilities (C) 98-99 99-00 00-01 29.0 38.0 41.0 36.0 41.0 43.0 45.0 43.0 41.0 22.0 29.0 32.0 31.0 29.0 36.0 17.0 30.0 27.0 19.0 24.0 28.0 23.0 26.0 31.0 22.0 14.0 16.0 13.0 15.0 15.0 8.0 15.0 8.0 18.0 11.0 4.0 0 Accountability for Students with Disabilities Who Receive Special Education % Difference Between General Education Participants and Those with Disabilities (B - C) 98-99 99-00 00-01 19.0 17.0 17.0 6.0 8.0 12.0 18.0 40.0 27.0 22.0 21.0 21.0 8.0 8.0 12.0 32.0 31.0 44.0 27.0 24.0 25.0 15.0 14.0 11.0 27.0 38.0 29.0 37.0 36.0 37.0 28.0 27.0 33.0 33.0 39.0 46.0 % Difference Between General Education Participants and Those with Disabilities (B - C) 98-99 99-00 00-01 20.0 19.0 20.0 10.0 15.0 16.0 14.0 26.0 27.0 32.0 23.0 24.0 8.0 16.0 16.0 24.0 34.0 39.0 28.0 28.0 28.0 16.0 16.0 16.0 28.0 46.0 48.0 34.0 35.0 36.0 29.0 25.0 37.0 32.0 41.0 51.0 by 6% for students in grade 3; 4% for students in grade 4; and 5% for students in grade 8. Grade 5 was the only grade to decrease the difference between students in general education and students with disabilities who reached proficiency over the course of the three years. New Haven Unified School District’s achievement gap between general education students and special education students increased from a low of 2% in grade 5, to as high as 13% in grade 8. This finding may have resulted from increased participation of students with disabilities in the CST Reading in this school district. Data retrieved for the 1998-99 through 200001 school years on the CST Mathematics show an upward trend at all grade levels assessed for both the general education students and the students with disabilities. Similarly, Long Beach and New Haven reported increased general education student performance at all grade levels. Thus, the achievement gap between general education students and students with disabilities on the CST Mathematics was not reduced over the 3 school years. In grades 3 and 5, there is no change in the gap, while at grade 8 the gap increased by 2%. At grade 4 there was an 8% decrease. Data from Long Beach show the achievement gap increasing at grades 3, 4, and 8 and stable at 16% at grade 5. In New Haven, the gap increased by 13% to 20% across the grade levels. Maryland Maryland is the smallest of the four study states, with 852,929 students. Maryland reported the lowest percentage of economically disadvantaged students (30%). Approximately 13% of Maryland’s students received special education services and just under 3% of the students in the state participated in ESOL programs and services. More than one third of the students were African American (37.1%), just over half of the total student population was White (53.4%), and Hispanic students represented under 5% of the total student population. On average, Maryland’s class size was about 15 students per teacher and reported total expenditure per pupil of about $7,600. During the data collection period, Maryland assessed students in five content areas at grades 3, 5, and 8 using the MSPAP. However, we report assessment results only for reading and mathematics. Maryland reported participation and exemption rates for students with disabilities on the state assessments at the state, district, and school level by content area and by grade level. However, the state reported participation as the number of students with disabilities out of the total number of students enrolled rather than the total number of students with disabilities enrolled. We chose to report the participation rates presented in Tables 19 and 20 as the percentages of students with disabilities who participated in state assessments out of the total number of students in special education per grade. We did this calculation in an attempt to provide comparable data across the four states. Table 21 and Figure 1 present the percent of students with disabilities who met the state standard of “proficient” on the state MSPAP Reading and Mathematics tests over the three school years. The data on the MSPAP Reading show an inverse relationship between the participation rate and the proficiency levels of students with disabilities. For example, between 1998-2001 the percentage of students with disabilities in grade 3 who reached proficiency decreased by 3.8%, while the number of students who took the MSPAP increased by 2.4%. These trends were mirrored in the two school disThe Educational Policy Reform Research Institute 41 eprri Table 19: Maryland State and District-Level Participation Rates for Students with Disabilities: MSPAP Reading: 1998-99, 1999-00, and 2000-01. 3rd Grade State/School District Maryland Carroll County Montgomery County 5th Grade 8th Grade 98-99 52.5 99-00 55.8 00-01 54.9 98-99 52.7 99-00 54.6 00-01 52.3 98-99 70.9 99-00 72.9 00-01 70.5 49.5 55.7 60.5 52.5 65.8 55.3 78.2 88.1 89.1 48.9 60.0 53.6 53.5 59.7 54.4 71.1 72.1 68.3 Table 20: Maryland State and District-Level Participation Rates for Students with Disabilities: MSPAP Mathematics 1998-99, 1999-00, and 2000-01. 3rd Grade State/School District Maryland Carroll County Montgomery County 5th Grade 8th Grade 98-99 78.3 99-00 79.9 00-01 95.8 98-99 96.5 99-00 96.5 00-01 96.2 98-99 95.2 99-00 94.7 00-01 95.3 95.0 92.0 96.6 94.6 97.1 96.3 93.1 94.6 96.7 62.5 76.9 96.3 98.1 97.3 97.1 95.7 94.7 95.1 tricts. For example in Carroll County, 11% more of the special education population took the MSPAP in grade 3, but there was a 3.6% decrease in the number of those students who met the proficiency standards. Likewise, in Montgomery County, the participation of students with disabilities increased in grades 3 and 5, but the level of proficiency in these grades decreased; students with disabilities in grade 8 showed a decrease in the participation level over the 3 years, while the percentage of students meeting proficiency increased. For general education students, state-level performance on the MSPAP Reading increased at grades 5 and 8, but decreased at third grade. Both Carroll County and Montgomery County showed a decline in the performance of general education students at those same grade levels. The performance gap between general education students and those with disabilities fluctuated over the three years at the state-level. In grades 3 and 5, the gap decreased between the years 1998-99 to 1999-00, but then increased the following year. 2 Accountability for Students with Disabilities Who Receive Special Education Table 21. Maryland Reading and Mathematics: Percent of State Assessment Participants Meeting or Exceeding Proficiency. Reading Grade Level State/School District (A) General Education (B) 98-99 41.2 49.4 48.9 41.4 54.2 50.0 25.3 33.2 34.2 Grade 3 Maryland Carroll Co. Montgomery Co. Grade 5 Maryland Carroll Co. Montgomery Co. Grade 8 Maryland Carroll Co. Montgomery Co. Mathematics State/School Grade Level District (A) Grade 3 Grade 5 Grade 8 99-00 39.2 46.3 44.2 44.6 52.4 52.3 26.8 29.6 35.6 00-01 36.5 39.6 37.6 44.6 48.0 49.4 26.6 29.7 33.4 General Education (B) Maryland Carroll Co. Montgomery Co. Maryland Carroll Co. Montgomery Co. Maryland Carroll Co. Montgomery Co. 98-99 38.9 47.6 52.1 46.2 56.9 61.2 49.0 68.8 66.1 99-00 40.1 48.0 49.2 46.7 55.5 60.4 50.4 68.0 66.0 00-01 37.8 43.6 43.4 42.6 47.6 52.2 47.0 61.5 63.0 Students with Disabilities (C) 98-99 29.4 32.7 42.1 24.3 33.6 35.1 5.3 8.1 9.3 99-00 29.7 30.9 30.7 27.8 35.0 33.8 6.5 7.4 12.7 00-01 25.6 29.1 21.7 25.4 21.8 26.9 7.2 6.8 11.0 Students with Disabilities (C) 98-99 31.9 41.2 50.8 27.0 33.9 44.3 18.4 31.8 32.2 99-00 33.2 42.6 37.6 27.6 31.1 40.1 19.8 27.9 32.3 00-01 26.2 29.5 21.7 20.7 21.6 22.5 16.0 24.1 24.8 Difference Between General Education Participants and Those with Disabilities (B - C) 98-99 99-00 00-01 11.8 9.5 10.9 16.7 15.4 10.5 6.8 13.5 15.9 17.1 16.8 19.2 20.6 17.4 26.2 14.9 18.5 22.5 20.0 20.3 19.4 25.1 22.2 22.9 24.9 22.9 22.4 Difference Between General Education Participants and Those with Disabilities (B - C) 98-99 99-00 00-01 7.0 6.9 11.6 6.4 5.4 14.1 1.3 11.6 21.7 19.2 19.1 21.9 23.0 24.4 26.0 16.9 20.3 29.7 30.6 30.6 31.0 37.0 40.1 37.4 33.9 33.7 38.2 Percentage Figure 1. Maryland 8th Grade Reading Test, Percentage of All Students and Percentage of Students Receiving Special Education Services Who Met or Exceeded Standards on Statewide Tests, 1999-2001. Maryland--Tot. Pop. Maryland--Sp. Ed. 40 35 30 25 20 15 10 5 0 1999 2000 Year 2001 Carroll County--Tot. Pop. Carroll County--Sp. Ed. Montgomery County--Tot. Pop. Montgomery County--Sp. Ed. The Educational Policy Reform Research Institute 43 eprri Notably, the participation rates of students with disabilities increased in the 1999-00 school year, the same year that the gap between the general education students and students with disabilities decreased. In Carroll County, the gap decreased at grade 3. For example, in grade 3 the 1998-99 school year the performance gap was 16.7%, in 1999-00 the difference between these groups decreased to 15.4%, and in 2000-01 the difference between general and special education groups decreased to 10.5%. At grade 8, the gap fluctuated but showed an overall decrease between general education and special education students. In Montgomery County, there was an increasing performance gap at grades 3 and 5. However, at 8th grade the gap narrowed by 2.5% over the three years. The statewide participation rate for students with disabilities on the MSPAP Mathematics between the school years of 1998-99 and 2000-01 steadily increased in 3rd grade while the percentage of students with disabilities achieving proficiency decreased. In contrast, the participation rate for students with disabilities in grade 5 slightly declined by 0.3%. In grade 8 the participation rate fluctuated. There was an inverse relationship between the participation rate and the percent of 8th grade students with disabilities reaching proficiency. The proficiency percentages increased by 1.4% from 1998-99 to 1999-00 and decreased by 3.8% from 1999-00 to 2000-01. Within the districts, Carroll County also exhibited an inverse relationship between the participation rates and the proportion of students with disabilities who attained proficiency. For example, in grade 8 there was a steady increase of 3.6% in the participation rate for special education students, but the percentage of students with disabilities who reach the proficient level decreased by almost 8 points. In Montgomery County at grades 5 and 8, there was a decrease in the reported participation of students with disabilities and performance levels of special education students also decreased. At grade 3 the participation rates of students with disabilities increased by 33.8% over the 3 years but performance decreased by 29.1%. Based on state-level data collected for the MSPAP Mathematics, general education students’ performance fluctuated over the course of the 3 years. However, in 2000-01, the performance on the MSPAP Mathematics was the lowest it had been over the 3 years. There were similar trends across the two districts. For example, Carroll County evidenced a steady decline in grade 3 performance by 0.4% from 1998-99 to 1999-00 and 4.4% from 1999-00 to 2000-01. In Montgomery County, grade 3 performance went from 52.1% to 49.2% to 43.4% for general education students. According to state-level findings, the achievement gap between general and special education students who reached the proficient level on the MSPAP Mathematics widened at all grade levels assessed. For example, the state level gap at grade 3 increased by a total of 4.6%. Both individual counties in the state reported increased achievement gaps from 1998-99 to 2000-01. In grade 3 in Carroll County, the gap widened by a total of 7.7% between 1998-99 to 2000-01 and, in Montgomery County, the grade 3 gap widened by a total of 20.4%. New York New York’s public school enrollment (2,823,602) was slightly more than half of the student enrollment in California. Similar to California and Texas, about half of the students in New York were economically disadvantaged. About 8% of New York’s students received English for Speakers of Other Languages (ESOL) and the percent of students who  Accountability for Students with Disabilities Who Receive Special Education received special education services for grades K-12 in New York was 11.9% which was approximately 2% above California (10.1%) and slightly less than Maryland (12.5%) and Texas (12.0%). About half of New York’s students were White; African American and Hispanic students each represented about one fifth of the state’s students. New York had the smallest average class size in each of the four study states with approximately 13 students per teacher and the highest per pupil expenditure ($11,040). New York assessed all students statewide in 4 and 8th grades in English language arts and mathematics. New York did not provide special education participation data at the state level on its Web site or in its Annual Report to the Governor and the Legislature on the Educational Status of the State’s Schools. This report, known as Chapter 655, required the Board of Regents and the State Education Department to submit an annual report to the Governor and the Legislature. Participation data on students with disabilities was gathered from the report entitled “Updated Special Education Data” submitted to the Board of Regents on April 16, 2002. th We were able to gather 1999-2000 data from the “Performance Report of Educational and Vocational Services and Results for Individuals with Disabilities, 1999-00 – Volume 2.” To determine participation we calculated the total number of special education students reported as “tested” in the state to those identified as “exempt” and “absent” in each school district. The total number tested (aggregate of all public school districts) was then divided by this aggregate denominator. The formula is indicated below: total # of students with disabilities tested total # students with disabilities + students “exempt/absent” We compared the participation on the 8th grade reading assessment using the report submitted to the Board of Regents which stated that 81.9% of the students with disabilities participated in the assessment whereas district-by-district calculations yielded a 84.4% participation rate. Table 22 presents the participation data that we calculated. Assessment participation rates were generally high, with a minimum of about 62%. However, participation on 4th and 8th grade reading assessments decreased in North Colonie by 24.2% across the three years. Rochester City also reported declines in participation in 4th grade reading and mathematics. State participation rates remained relatively stable during this time. Students with disabilities in New York consistently performed below general education students (See Table 23 and Figure 2). At the state level, the achievement gap in 2000-01 ranged from 29.6 percentage points in 8th grade mathematics to 37.1 percentage points in 8th grade reading. The 8th grade achievement gap in North Colonie was higher than the state’s gap, while Rochester City’s achievement gap was smaller. The 4th grade achievement gap in each district was smaller that the state’s achievement gap. Over the three-year period, this achievement gap widened in some areas and decreased in others. The discrepancy between the percentage of students with disabilities and general education students scoring at or above proficient widened for 4th grade reading at the state level, as well as in 4th reading The Educational Policy Reform Research Institute 45 eprri Table 22. New York ELA and Mathematics: Percent of Students with Disabilities Participating in Assessments and Change in Participation. English Language Arts Assessment Grade Level Grade 4 Grade 8 Mathematics Assessment Grade Level Grade 4 Grade 8 State/School District New York North Colonie Rochester City New York North Colonie Rochester City % Participation 98 - 99 99 - 00 00 – 01 81.5 100.0 93.7 ND 91.9 78.9 88.3 93.3 88.3 84.4 83.0 72.2 84.0 75.8 80.4 80.6 62.2 67.5 State/School District New York North Colonie Rochester City New York North Colonie Rochester City % Participation 98 - 99 89.9 82.8 96.1 84.7 100.0 73.3 99 - 00 89.3 96.6 95.5 85.0 81.5 79.2 00 – 01 86.3 92.3 81.7 81.9 100.0 65.9 and 8th grade mathematics in Rochester City. While the gap decreased in all other areas, the degree of improvement varied. At the state level, the decreases were almost invisible. For example, the difference in 4th grade mathematics was only 0.9 percentage points, from 31% to 30.1%. This was also the case for Rochester City, where there was a decrease from 25.9% to 23.2% in 8th grade reading. The North Colonie Central School District demonstrated the most dramatic decreases in the achievement gap. For example, in mathematics, the gap narrowed from 31.4 %to 10.4% in 4th grade, (a decrease of 23.7 percentage points), and 65.0% to 42.8% in 8th grade, (a decrease of 22.2 percentage points). proficient were 20 to 30 points higher than the state in both subject areas, while Rochester City’s proficiency percentages were 40 to 50 points lower. Furthermore, both the state average and the Rochester City 8th graders had proficiency percentages about 30 percentage points lower than their 4th graders in math. These differences were less pronounced among students with disabilities, with the greatest differences seen between 4th and 8th grade in mathematics. In Rochester City, percentages of both groups of students who scored proficient in 8th grade mathematics were extremely low, ranging in the three years from 11.5 to 13.8 for all students and from 2.1 to 3.4 for students with disabilities. In 1998-99 the proportion of 8th grade general education students in North Colonie who scored In North Colonie, there was a 24-point decrease in 4 grade reading proficiency among special eduth 6 Accountability for Students with Disabilities Who Receive Special Education Table 23. New York English Language Arts and Mathematics: Percent of State Assessment Participants Meeting or Exceeding Proficiency. Reading Grade Level State/School District (A) New York1,2 North Colonie Central1,2 Rochester City1,2 Grade 8 New York1,2 North Colonie Central1,2 Rochester City1,2 Mathematics State/School District Grade Level (A) Grade 4 Grade 4 Grade 8 New York1 North Colonie Central Rochester City New York1 North Colonie Central1 Rochester City1 General Education Students (B) Students with Disabilities (C) 98-99 99-00 00-01 49.0 59.0 60.0 73.5 87.3 86.2 27.9 42.3 45.8 49.0 45.0 45.0 83.3 82.7 79.9 27.5 31.8 29.2 General Education Students (B) 98-99 99-00 00-01 19.1 25.6 25.6 27.3 42.9 64.0 7.5 12.9 21.1 9.3 8.4 7.9 23.5 20.0 25.0 1.6 2.2 6.0 Students with Disabilities (C) 98-99 67.0 88.3 44.0 38.0 82.6 11.5 98-99 36.0 54.2 21.0 7.8 17.6 3.1 99-00 65.0 91.5 42.0 41.0 79.3 13.8 00-01 69.0 89.6 51.2 39.0 82.1 12.1 99-00 35.0 57.1 17.7 9.7 25.0 2.1 00-01 38.9 79.2 28.6 9.4 39.3 3.4 Difference Between General Education Students and Those with Disabilities (B - C) 98-99 99-00 00-01 29.9 33.4 34.4 46.2 44.4 22.2 20.4 29.4 24.7 39.7 36.6 37.1 59.8 62.7 54.9 25.9 29.6 23.2 Difference Between General Education Students and Those with Disabilities (B - C) 98-99 99-00 00-01 31.0 30.0 30.1 34.1 34.4 10.4 23.0 24.3 22.6 30.2 31.3 29.6 65.0 54.3 42.8 8.4 11.7 8.7 1 In New York the test is called English Language Arts. New York state-wide estimates computed from the "655 Report” and the "Achievement and Placement of Students with Disabilities" news article released on April 23, 2002. The district figures are from the School District Report Cards. 2 Percentage Figure 2. New York 8th Grade English Language Arts Test, Percentage of All Students and Percentage of Students Receiving Special Education Services Who Met or Exceeded Standards on Statewide Tests, 1999-2001. 90 80 70 60 50 40 30 20 10 0 New York--Tot. Pop. New York--Sp. Ed. 1999 2000 2001 North Colonie--Tot. Pop. North Colonie--Sp. Ed. Rochester--Tot. Pop. Rochester--Sp. Ed. Year The Educational Policy Reform Research Institute 47 eprri cation students over the three years, from 46.2 to 22.2. However, there was a 36.7-percentage point increase in proficiency for students with disabilities, from 27.3 to 64.0. Similarly, there were 25-point and 21.2 point increases in the percent scoring proficient for 4th and 8th grade mathematics (respectively). In the case of 4th graders, this could partially account for the 36.7 point increase in proficiency percentages and corresponding 24 point decrease in the achievement gap. Texas Texas enrolled 4,059,619 students in early childhood through grade 12 and identified about half as economically disadvantaged in the 200001 school year. The enrollment for Kindergarten through grade 12 was 3,913,848. About 11.9% of Texas’ students received special education services and about 12.6% received ESOL services. About 40.6% of Texas’ students were Hispanic, 42% were white, and the remaining 20% were Black, Asian, and American Indian. The student to teacher ratio was 14.8 and the expenditure per pupil was $6,638, which was the second highest per pupil expenditure of all four study states included in these analyses. Every student enrolled in a Texas public school in grades 3, 4, 5, 6, 7, 8, and 10 was given the opportunity to participate in state assessments in English Language Arts, mathematics, science and social science. However, there are circumstances under which some students were not tested, for example students with disabilities could receive an ARD (Admission, Review, and Dismissal) exemption for every test. An important difference between Texas and the other three states is that students served in special education in Texas could take the SDAA (State-Developed Alternative Assessment) in grades 3-8. This is an off-level assessment that is designed for students with disabilities who were receiving instruction in Texas state standards but for whom the regular state assessment (TAAS) was considered inappropriate by the students’ IEP due to the student’s instructional team level. The SDAA was administered at the same time as the state TAAS and assess reading, writing, and mathematics. This test was designed to measure annual growth and was administered for the first time in 2001. Participation rates for both assessments are presented in Table 24. Between the 1998-1999 and 2000-2001 school years, there was a marked decrease in the achievement gap between all students and special education students on the TAAS. Table 25 and Figure 3 reveal that the percentages of both all test takers and special education students meeting and exceeding proficiency on the TAAS were generally high at both the state and district levels during this time. In 2000-01, 90.2% of all fifth-grade test takers and 81.1% of special education test takers met or exceeded proficiency in reading. Likewise, in the mathematics portion of the grade 5 TAAS, 94.6% of all test takers demonstrated proficiency, while 87.4% of special education students did the same. Our two study districts, Cypress-Fairbanks and Garland, also had high proficiency rates. For example, in 2000-01, 94.5% of all fifth graders in Cypress-Fairbanks met or exceeded proficiency on the TAAS reading, and 98.3 % did so in mathematics. Fifth grade special education test takers had proficiency rates of 90.7% in reading and 97.7% in mathematics. In Garland School District, the grade 5 2000-01 percentages were 94.6% for all students and 87.3% for special education students in reading, and 97.0% and 94.4% in math, respectively. 8 Accountability for Students with Disabilities Who Receive Special Education Table 24. Participation Rates for Students with Disabilities in TAAS and SDAA. State/ School Aggregate Participation Rates ELA and Mathematics Assessments District Texas Tested SDAA Only Not Tested ARD Exempt3 Absent Other CypressTested Fairbanks SDAA Only Not Tested Ard Exempt Absent Other Garland Tested SDAA Only Not Tested ARD Exempt Absent Other State and local data from these years illustrates a relatively small achievement gap between general and special education students on the TAAS. For example, at the state level in 2000-01, the difference between general education participants and those with disabilities was less than 10 percentage points in grades 3, 4, and 5 in both reading and mathematics. The differences were greatest in grade 8, with percentages of 14.6 in math and 15.7 in reading proficiency. In the two study districts, the gap was even smaller, with the difference in percent proficiency ranging between 0.6, in Cypress-Fairbanks for grade 5 mathematics, to 10.7, in GSD for grade 8 reading. The achievement gap between general education students and those receiving special education 1999-2000 2000-2001 2001-02 46.7 NA 53.3 50.5 .7 2.0 52.9 NA 47.1 45.5 .5 1.1 50.7 NA 49.3 46.6 .5 2.1 89.4 45.3 10.6 7.8 .9 1.8 88.5 35.8 11.5 9.8 9.6 1.2 92.9 45.8 7.1 5.7 .7 .6 89.4 48.5 10.6 8.1 1 1.4 88.4 38 11.6 9.5 1.4 .6 91.6 47.8 8.4 6.8 1 .5 consistently decreased in reading and mathematics at all grades over the three years in both reading and mathematics. At the state level, the achievement gap decreased each year at all four grade levels for both content areas. The same trend was observed for the two districts, but with slightly more variance. For example, Cypress-Fairbanks showed decreases each year in grades 4 and 8 in reading, and grades 4, 5, and 8 in mathematics. The differences for grade 3 reading proficiency percentages changed from 2.5 to 2.0 to 2.4 between 1998-99 and 2000-01. Similarly, the difference in grade 5 reading percentages shifted from 6.3 to 5.6 to 5.9 during these years. The achievement gaps in Garland decreased each year in grades 3 and 5 in reading, and grades 5 and 8 in mathematics. In the grades 4 and 8 reading The Educational Policy Reform Research Institute 49 eprri Table 25. Texas Reading and Mathematics: Percent of State Assessment Participants Meeting or Exceeding Proficiency. Reading Grade Level (English Version) State/School District (A) Texas Grade 3 Cypress-Fairbanks Garland Texas Grade 4 Cypress-Fairbanks Garland Texas Grade 5 Cypress-Fairbanks Garland Texas Grade 8 Cypress-Fairbanks Garland Mathematics State/School District Grade Level (A) (English Version) Grade 3 Grade 4 Grade 5 Grade 8 Texas Cypress-Fairbanks Garland Texas Cypress-Fairbanks Garland Texas Cypress-Fairbanks Garland Texas Cypress-Fairbanks Garland All Students (B) Students with Disabilities (C) 98-99 99-00 00-01 88.0 87.9 86.8 95.0 94.4 91.9 93.9 92.6 93.2 88.8 89.9 90.8 96.1 94.8 94.7 95.7 95.0 95.1 86.4 87.8 90.2 95.7 94.1 94.5 93.3 92.6 94.6 88.2 89.6 91.9 93.4 95.6 96.4 91.5 91.9 93.4 All Students (B) 98-99 99-00 00-01 78.5 80.1 80.5 92.5 92.4 86.9 88.4 83.5 87.2 75.2 81.6 85.0 89.6 91.2 92.4 91.4 88.5 89.0 69.2 74.6 81.1 89.4 88.5 90.7 80.1 83.6 87.3 63.7 68.9 76.2 79.4 86.0 92.6 81.1 83.2 82.7 Students with Disabilities (C) 98-99 83.1 93.9 91.2 87.6 95.5 94.9 90.1 97.1 95.0 86.3 92.5 87.4 98-99 71.3 86.0 84.0 72.7 87.7 90.4 74.6 92.7 87.6 58.8 76.2 65.6 99-00 80.6 91.2 87.4 87.1 94.7 92.7 92.1 97.9 96.1 90.2 95.3 90.5 00-01 83.1 91.8 89.0 91.3 96.7 94.4 94.6 98.3 97.0 92.4 97.3 93.3 50 Accountability for Students with Disabilities Who Receive Special Education 99-00 69.9 89.0 79.7 77.0 88.7 86.3 81.6 95.1 92.6 70.7 90.0 79.6 00-01 74.6 89.6 83.2 85.2 95.4 90.2 87.9 97.7 94.4 77.8 96.0 86.3 Difference Between all Participants and Those with Disabilities (B - C) 98-99 99-00 00-01 9.5 7.8 6.3 2.5 2.0 2.4 5.5 9.1 6.0 13.6 8.3 5.8 6.5 3.6 2.3 4.3 6.5 6.1 17.2 13.2 9.1 6.3 5.6 5.9 13.2 9.0 7.3 24.5 20.7 15.7 14.0 9.6 3.8 10.4 8.7 10.7 Difference Between all Participants and Those with Disabilities (B - C) 98-99 99-00 00-01 11.8 10.7 8.5 7.9 2.2 2.2 7.2 7.7 5.8 14.9 9.4 6.1 8.1 6.0 1.3 4.5 6.4 4.2 15.5 10.5 7.2 4.4 2.8 0.6 7.4 3.5 2.6 27.5 19.5 14.6 16.3 5.3 1.3 21.8 10.9 7.0 Figure 3. Texas 8th Grade Reading Test, Percentage of All Students and Percentage of Students Receiving Special Education Services Who Met or Exceeded Standards on Statewide Tests, 1999-2001. Texas--Tot. Pop. 120 Percentage 100 Texas--Sp. Ed. 80 60 40 20 0 1999 2000 2001 Year Cypress Fairbanks--Tot. Pop. Cypress Fairbanks--Sp. Ed. Garland--Tot. Pop. Garland--Sp. Ed. assessments, the achievement gap did widen, but only slightly (from 4.3 to 6.5 to 6.1 in grade 4, and 10.4 to 8.7 to 10.7 in grade 8). While impressive, these increases must be considered in context with participation rates on TAAS and SDAA for students with disabilities. As discussed previously, although the rates of students with disabilities participating in the state alternate assessments are high, the percentages of students taking the “regular” TAAS are much lower. In 200102, 48.5% of students with disabilities in Texas took only the SDAA, up from 45.3% in 2000-01. Only 40.9% took the TAAS. Similar participation rates were reported for the Cypress-Fairbanks and Garland districts. Summary of Findings Data regarding the participation rates and performance levels of all students and students with disabilities were collected in four states: California, Maryland, New York, and Texas. Table 26 shows a basic and general summary of the results. At the state level, it appears that California’s participation rates stayed stable or slightly increased, the number of students meeting proficiency generally increased, and the achievement gap between all students and students with disabilities remained stable or decreased slightly. In Maryland, participation rates generally increased with a few exceptions depending upon grade level and subject area, generally decreased in the percentage of students meeting proficiency levels (again with a few exceptions), and the achievement gap lessened in reading but widened in math. New York data reveal that participation rates generally decreased with one exception, proficiency levels of students with disabilities generally increased, and the achievement gap was variable depending on the grade level and subject area. Finally, Texas showed the clearest results with increased levels participation and proficiency and decreased achievement gaps, however the increases in participation in Texas were in assessments that had a different proficiency criterion. The Educational Policy Reform Research Institute 51 eprri There are some notable limitations of the results. Each of the states has a different way of administering tests and reporting results; therefore, it may not be appropriate to make direct comparisons across the states. Although we attempted to calculate and report the data in a fair and consistent manner, we realize that there are differences across states that may be better accounted for by differences in their assessment systems than by differences in the participation and proficiency rates of their students. For example, as previously explained, many students with disabilities in Texas take the alternate assessment which can alter the state’s reported results. Also, while Table 26 provides a succinct and very general overview of the findings, it only reports state-level data and oversimplifies the complexities inherent in these data. Much volatility exists at the district level and most definitely at the school level. Problems can also arise with missing data and a lack of continuity in reported data across the years. Table 26. State-Level Summary of the Trends for Students with Disabilities over the 3 Years of Data Collection. State CA MD NY TX Participation of Students with Disabilities Generally stable or slightly increased Generally increased with a few exceptions Generally decreased with one exception Increased, but primarily in alternates Percentage of Students with Disabilities Meeting Proficiency Generally increased Generally decreased with a few exceptions Achievement Gap between All Students and Students with Disabilities Generally stable or slightly decreased Decreased in reading and increased in math Generally increased Variable Increased Decreased 52 Accountability for Students with Disabilities Who Receive Special Education Topical Review Summary of Findings There are many reasons for the lack of clearcut findings on the participation and performance of the subgroup of students with disabilities in accountability. These are evident in the contextual and quantitative analyses included in this Topical Review which illustrate both the instability and complexity of the evolving system of accountability. Starting from the description of the context of the four core states as background, it is quite evident that assessment systems were shifting even before the No Child Left Behind Act was enacted. Variability that directly affected students with disabilities emerged in how their participation was defined in state guidance, the nature of assessment systems that were designed for them, and the accommodation policies that were developed to promote or discourage their inclusion in assessments and reporting of their results. With NCLBA, each state initiated major changes in its assessment and accountability system to comply with the law. As these findings demonstrate, the K-12 educational policy environment has been quite fluid for much of the past decade and likely will be for the foreseeable future. The changes in each of the four states, initiated partly in response to changes in federal law, interact with the overall complexity of implementing a standards-driven system with the students with disabilities. The NCLBA accelerated the changes and interrupted some initiatives that were just beginning to be implemented (e.g., alternate assessments). The overall work of EPRRI has been guided by five research questions: 1. How are students with disabilities affected by educational accountability reforms? 2. What are the criteria to which special education has historically been held accountable? 3. What impact have educational accountability mechanisms had on students with disabilities at the system levels? 4. How do broad educational policies that incorporate high-stakes accountability include students with disabilities? 5. What changes could be made to better align special education policy with accountability reform? Findings from our four states that are presented in this report shed some light on how students with disabilities have been included in accountability reforms, both pre and post NCLBA. The findings suggest that full participation of these students in the accountability system did not come without the federal mandate. However, the findings are less clear regarding the impacts on the performance of students with disabilities. Data for our analyses were drawn from state and district Web sites for the years 1998-1999 through 2001-2002 for the grades in which the states assessed English language arts and mathematics. We examined data for the four core states (California, Maryland, New York, and Texas) and targeted districts in those states (California – Long Beach and New Haven; Maryland – Carroll County and Montgomery County; New York – North Colonie Central and Rochester City; Texas – Cypress Fairbanks and Garland). The states originally varied in terms of the stakes of their accountability systems, recentness of reforms, stability of reform efforts, participation of students with disabilities in all accountability reports, the use of alternate assessments, and the availability of school and districts-level data on the The Educational Policy Reform Research Institute 53 eprri performance of students with disabilities. These variances changed with the implementation of NCLBA, when all states and districts moved into a high-stakes accountability situation and the availability of publicly reported data increased for all. The difficulty in reaching conclusions from our analyses also arose because of data challenges. There were often missing data, or data elements that changed from one year to the next. In addition, as the numbers of students within the disability subgroup decreased, findings were less reliable, making results more volatile from year to year. This made it easier to examine state level data than district level data. Conclusions The 2004 Individuals with Disabilities Education Improvement Act and the 2001 No Child Left Behind Act both require that students with disabilities participate in statewide assessments, and that their participation and performance results be publicly reported and included in measures of AYP. The picture of implementation painted in this Topical Review for EPRRI’s four case study states (California, Maryland, New York, and Texas) and two districts within each of these states for the years 1998-99 through 2000-01, with specific years varying by state, is one of variability. Yet there are some conclusions to be drawn from the analyses. It is clear that as more students with disabilities participate, there is a tendency for performance to reduce, at least initially – unless there is something about the assessment system (such as in Texas) that obviates that. Yet, over time, with access to the curriculum and improved instructional interventions, it is hoped that with stable participation there will be evident increases in performance. Second, the data lead one to wonder whether, under current circumstances, it will be possible to know there has been a change in the achievement of students with disabilities, or we are simply seeing the effects of policy changes. The nature of assessments changed in nearly every one of our case study states. Similarly, the accommodation policies changed as well. To what extent these effected the performance of students with disabilities is unknown. Finally, it is not evident here whether there has been a closing of the gap in achievement. It may be that it is too early – that by 2000-01 is really still a baseline period. Contextual data show that states and districts are paying attention to the disability subgroup (they are disaggregating their data, they are examining and adjusting their accommodation policies), as do data from qualitative case studies (see www.eprri.org). First, the disaggregation of data for the subgroup of students with disabilities has opened up information that has not been available before. This has revealed that the performance of most students with disabilities – but not all – is low. It has also shown that the performance of students with disabilities can increase. Exactly what the increases in performance are tied to is the critical question. 5 Accountability for Students with Disabilities Who Receive Special Education