NCEO Synthesis Report 29

State Assessment Policies on Participation and Accommodations for Students with Disabilities:  1997 Update

by Martha L. Thurlow, Allison L. Seyfarth, Dorene L. Scott, and James E. Ysseldyke

Published by the National Center on Educational Outcomes

September 1997

Any or all portions of this document may be reproduced and distributed without prior permission, provided the source is cited as:

Thurlow, M.L., Seyfarth, A., Scott, D.L., & Ysseldyke, J.E. (1997). State Assessment policies on participation and accomodations for students with disabilities: 1997 update (Synthesis Report No. 29). Minneapolis, MN: University of Minnesota, National Center on Educational Outcomes. Retrieved [today's date], from the World Wide Web: http://education.umn.edu/NCEO/OnlinePubs/Synthesis29.html

Executive Summary

States' policies on the participation of students with disabilities in district or state assessments, and the accommodations available in those assessments, continue to change rapidly. Legislation (IDEA, Title I), research funding, and states' ongoing work on these issues are promoting the participation of students with disabilities in the assessments. In this report, we summarize states' current policies on the participation of students with disabilities in large-scale assessment, and the accommodations available for those students. Among the generalizations from these summaries are that:

• State participation and accommodation policies change frequently.

• For participation decisions, state policies often rely on the IEP team and the involvement of parents.

• Many policies indicate that states have begun to offer partial participation in testing or alternate assessments for students with disabilities.

• Testing accommodations have become very common, with nearly every state with a policy offering some accommodations.

• The most commonly offered accommodations include Braille or large-print editions of tests, the use of a proctor or scribe, extended time, and allowing for individual or small group administration of assessments. The accommodations that are most controversial (i.e., offered by some states and prohibited by others) include reading a test aloud and use of calculators.

States that offer both norm-referenced tests and criterion-referenced tests will generally offer more accommodations in their criterion-referenced tests than in their norm-referenced tests. The accommodations available for these two types of tests are most similar for setting accommodations (e.g., administering an assessment individually, in small groups, or in an alternate location).

.

Some Information on Statewide Assessments

Statewide assessments are common in the United States. Most often, they are tests or other performance measures that are intended to document the educational achievement of the students in a state. In 1996, 48 states reported that they either had a state assessment in place or were developing one. Iowa and Wyoming were the only states without a statewide assessment in development or in place (see Figure 1) (Bond, Braskamp, & Roeber, 1996).

Despite the prevalence of statewide tests, there remains a lack of state level information on the performance of students with disabilities (see Thurlow, Langenfeld, Nelson, Shin, & Coleman, 1997). There probably are many reasons for this (see Elliott, Thurlow, & Ysseldyke, 1996), but primary among them are state policies. State policies, in turn, must be viewed within the context of the variations in state assessments.

Figure 1. Status of Statewide Assessments

(from Bond et al., 1996)

Statewide assessments vary widely across the states that use them. One of the variants is the number of tests that make up the assessment program in a state. Some states, such as North Dakota, administer one test to their students, while other states, such as Maryland, administer four or more tests to students throughout their schooling. Another related issue is the content areas tested. Different content areas may be presented as separate tests or as subtests of a larger test. Nearly all states include assessments of English or language arts, mathematics, and writing. Fewer states assess science and social studies. Some states assess other areas, such as citizenship and geography. The types of assessments used are another source of variability among states. Some states use traditional multiple-choice tests, while others use constructed response, portfolios, performance events, writing samples, or other forms of student assessment. Many states use a combination of various types of assessments.

Another way in which assessments differ is in the unit of comparison. Norm-referenced tests are tests in which students' results are compared to those of a normative group. Examples of these assessments include the Iowa Tests of Basic Skills and the California Achievement Test. Criterion-referenced tests, in contrast, involve comparing students' results to previously set standards. Criterion-referenced tests tend to be tests that are state designed, such as Hawaii's Test of Essential Competencies. Some states use norm-referenced tests exclusively, others use criterion-referenced tests exclusively, and a third group uses some combination of both measures.

As statewide assessments become more common, their complexities become more obvious. Some of the complexity involves what cognitive skill will be tested, even when the content area is the same. For example, for a reading test, assessors must decide whether ability to decode or to comprehend text is being measured. Another complexity is how to test, which involves some of the issues addressed above, as well as how to make modifications or accommodations to tests or the testing environment to enable students with disabilities or students with limited English proficiency to participate in testing. A third complexity is who to test, whether to include students with disabilities, and what such decisions mean for those students and their families.

What We Know About Who Gets Tested and How

For a number of years, the National Center on Educational Outcomes (NCEO) has been examining who gets tested, and how tests are given to students with disabilities. This work revealed some important trends over time. In the early 1990s, McGrew, Thurlow, Shriner, and Spiegel (1992) looked at the participation of students with disabilities in both national and statewide assessment, and found that 34 out of the 49 reporting states had formal or written decision rules on the participation of students with disabilities in statewide assessments. In 1993, Thurlow, Ysseldyke, and Silverstein examined the literature on testing accommodations for students with disabilities, and updated and added to the earlier work on states' policies. Common testing accommodations were discussed in the report, as well as policy, legal, and psychometric considerations when using accommodations. In addition, a classification system was used for the different types of accommodations commonly offered. The four classes of accommodations included: presentation format, which were changes in how tests were presented and involved accommodations like providing Braille versions of the tests or orally reading the directions to students; response format, which were changes in the manner in which students gave their responses and included accommodations such as having a student point to a response or use a computer for responding; setting of the test, which could be at home, or in small groups; and finally, timing of the test, which could include extending the time allowed, or providing more breaks during testing.

Thurlow et al. (1993) provided information from a sample of states on their participation policies (who should participate in their statewide tests) and their accommodations policies (which accommodations could be used during testing). In 1993, there were 28 states with written policies on the participation of students with disabilities in their tests. At this time, there was a great deal of variability in the types of decision rules states had for the participation of students with disabilities. Some of the factors commonly considered at that time included the type of disability the student had, the degree of the student's impairment, and the percentage of time the student was mainstreamed or receiving special services. Rules sometimes called for looking at only one of these variables, but more commonly at a combination of the variables.

At the time the 1993 report was published, there were 21 states with written policies on accommodations. Again, there was a great deal of variability across states in the accommodations that were allowed. The types of accommodations that were most frequently allowed, and prohibited, were changes to the presentation format. Presentation format changes most frequently allowed included offering Braille or large-print versions of the tests. Those most frequently prohibited included oral reading, video, or signed presentations of the tests.

In 1995, NCEO updated the 1993 information in two separate reports. The report on participation reproduced and summarized information from the states' written guidelines on the participation of students with disabilities in their state tests (Thurlow, Scott, & Ysseldyke, 1995b). The number of states with written guidelines increased from the 28 in 1993 to 43 in 1995. Noteworthy variables mentioned most by states in the 1995 report were the involvement of the Individualized Educational Program (IEP) team in making decisions about the participation of students with disabilities in testing, the role of parents, issues related to partial testing, the placement or category of disability of the student, and the reporting of the students' results.

Policies for accommodations were also re-examined in 1995, with a total of 38 written guidelines provided by states, up from the 1993 total of 21 guidelines (Thurlow, Scott, & Ysseldyke, 1995a). Again, a number of accommodations proved controversial. Use of a scribe, in which a student can give answers to a person (scribe) who will write them down, was explicitly prohibited in one state, and allowed in 15 other states. The use of a calculator during testing was prohibited by five states and allowed by four states. Finally, reading a test aloud was prohibited by nine states and allowed by two states (this often depended on whether it was the reading test or other content area). Overall, while most states offered accommodations, there was little consistency in the apparent acceptability of various accommodations. Almost every state had revised its guidelines between the publication of the 1993 and 1995 reports.

The Need to Update What We Know

Since 1995, interest in state assessments, participation of students with disabilities in them, and use of accommodations has increased exponentially. New special interest groups have been set up on this topic, such as one of the Council of Chief State School Officers' (CCSSO) Special Education State Collaboratives on Assessment and Student Standards (SCASS), which focuses on students with disabilities and assessment. This has allowed a number of states and policy organizations to come together to wrestle with challenges, bring the latest information to the table, and produce helpful products to address such issues.

In part, increased interest in state assessment is due to the reauthorization of laws (e.g., Title I, IDEA), research, and states' applied experience. Both Title I and IDEA now require the participation of students with disabilities in state and district assessments, with accommodations when needed and appropriate. In addition, federal education agencies are providing research funds to explore ways to increase participation and examine the effects of accommodations. Both the Office of Special Education Programs (OSEP) and the Office of Educational Research and Improvement (OERI) have conducted two rounds of funding for projects to address these issues. There is a new realization that there are negative consequences for having accountability systems that do not include all students. Among the most commonly recognized is the increasing rate of referral to special education that occurs when students with disabilities can be exempted from tests that are seen as high stakes for schools or districts (Allington & McGill-Franzen, 1992; Zlatos, 1994).

Updating Procedures

To update the NCEO files on state participation and accommodations policies, we first made a decision about the conditions under which we would call for more recent information. If we had a policy document with a date more recent than 1995, or if the state did not have a statewide assessment, we did not attempt to update our files. Forty states in all were contacted. Upon completion of our analysis, all 50 states were provided with copies of summary tables for feedback (including previous information for states that we did not update). We accepted new documents through mid-March, 1997. In total, we updated 34 policies on participation and 32 policies on accommodations. A complete list of the policy documents is in the appendix.

In 1995, the text of all relevant state policies was included in the updates, with highlights presented prior to the actual policies. In this report, we have created tables that summarize the policies. In Table 1, we provide the definitions that we used when deciding whether a state's policy included language referring to a specific category in the tables.

Presenting policy information in tables makes the information easier to use, but sometimes obscures the complexity that underlies the policies. For example, the length and detail of the original source documents, which ranged from one sentence to 40 pages, is not apparent. Another difference is the specificity of the documents to the tests given. Some states, such as Maryland, specify accommodations for each test individually, while other states, such as Georgia, provide accommodations guidelines, but do not make them specific to the different tests given in the state.

After summarizing in table format the documents received from states, we sent a summary table for states to approve. States could indicate that there were no changes needed, ask for more information in order to decide whether the tables were accurate, or change the tables. If states indicated the need for a change after reviewing the summary table, we requested written documentation before making changes.

The information collected is summarized in two sets of tables in this reportÑone set for participation policies and a second for accommodation policies. Reporting policies, which were addressed previously by Thurlow et al. (1995b) are not included here. Instead, a separate analysis of state accountability reports has been conducted by NCEO (Thurlow, Langenfeld, Nelson, Shin, & Coleman, 1997).

Participation Policies

Currently, 40 of the 50 states have active policies on the participation of students with disabilities in statewide testing (see Figure 2). There are a number of reasons why states may not have active policies on participation. They might not have statewide assessments, or they could have assessments without having any policies in place. Other states have had their assessments suspended, and are in the development phase of new assessments and new guidelines. Still other states have had guidelines, but are currently in the process of revising them. For this document, we included only policies that are currently in use. Thus, if a state has a policy that is currently undergoing revision we did not include it in this report.

Figure 2. Status of Statewide Assessment and Participation Guidelines

 

Table 2 summarizes the variables included in the participation policies in each state. Note that we have used an asterisk (*) in this table to indicate that the specific criterion applies to only part of the assessment system in the state (e.g., it may apply to one test but not another or only under certain circumstances). It is evident in this table that nearly all states with assessment policies in place use the IEP team's decision as one of the primary criteria to determine whether a student participates in the statewide assessment. Of the specific criteria listed in the table, course content or curricular validity is the next most frequent criterion. Relatively uncommon (less than one-fourth of the states with criteria) are criteria referring to the technical characteristics of measures (validity or reliability) or the special education services received. Just slightly over one-fourth of the states with policies specifically referred to the role of the parent/guardian in the decision-making process, with one state specifically prohibiting their involvement.

Note: Ten states are not included in this table. IA, NE, and WY had no state assessment; AZ, CA, CO and MA had the state assessment suspended; VT, VA and WV were drafting guidelines.

† Kentucky does not allow any exclusion. Guidelines determine placement in the regular assessment or the alternate assessment.

* = true only for certain tests within the state's assessment system. See Table 3 for specification of tests.

Next to the IEP, "other" criteria were the most frequently found in our analysis of state guidelines. This reflects the tremendous variability in specific criteria included within state policies. We have summarized the "other" criteria used by states in Table 3, along with specifications noted for the Table 2 criteria (i.e., explanation of asterisks).

Perhaps most obvious in Table 3 is the diversity of other criteria that states use, from requiring certification of a medical condition to examining the motivation of a student to be like her or his peers. The most frequently mentioned "other" criterion refers to the meaningfulness of testing for the studentsÑseven states have criteria that allow for exclusion of a student if the results are anticipated to reflect the disability rather than the student's ability. Other frequently mentioned criteria involve (a) the exclusion of a student with disabilities based on a specific disability (some allowing that as a reason to exclude, others disallowing it), (b) concerns about whether testing might adversely affect a student, and (c) issues of whether appropriate accommodations are available.

In Table 4, we sum up additional testing options that some states make available: out-of-level testing, partial participation in testing, and alternate assessment. Partial participation appears to be the most popular of the three options, with about 40% of states with policies providing this option for students with disabilities. Out-of-level testing and alternate assessments are significantly less popular. A total of five states (over 10% of the 40 states with policies) allow out-of-level testing while another five disallow the practice. According to the state policy documents gathered for this report, alternate assessment is currently available or in development in eight of the 40 states that have participation policies (20% of states with policies). One example of an alternate assessment is Kentucky's Alternate Portfolio Assessment program (Ysseldyke et al., 1996). This is a program designed for students with moderate to severe cognitive disabilities that prevent them from completing a regular course of study even with modifications. These students are assessed using a portfolio composed of their best classroom work. This is intended to document their progress toward Kentucky's academic expectations for students in the alternate assessment system.

Accommodations Policies

Currently, 39 of the 50 states have active policies on accommodations (see Figure 3). There are a number of reasons why states might not have active policies on accommodations. Some states have assessment systems, but are currently developing or revising their accommodations guidelines. Other states do not have assessment systems in place, sometimes due to suspension of the system. For the purposes of this document we included only policies that are currently in use. States with policies currently under revision (or that were not approved by March, 1997) were not included.

Table 5 summarizes information on accommodations policies in four areas (presentation, response, setting, scheduling), plus two other factors commonly considered in making accommodations decisions (used for instruction, IEP determined). In this table the information is presented at the broadest level (i.e., are there any accommodations allowed in each of the major types?). When viewed this way, nearly every state allows some accommodations of nearly every type (note that those cells with XO indicate that an accommodation is both allowed and prohibited). Of the 39 states with specific accommodations policies (South Dakota allows no accommodations), 31 offer some accommodations of nearly every type, nearly 80% of states with policies. The rare exceptions are states like Kentucky, Missouri, New Mexico, and Vermont, which do not list accommodations but instead indicate that the decision is IEP-determined and/or the accommodation is one used for instruction. In a couple of states (Indiana, Oklahoma) accommodations are allowed in all areas except one.

Figure 3. Status of Statewide Assessment and Accommodation Guidelines

 

 

 

XO = Accommodation allowed in some situations, prohibited in others

Presentation Accommodations

Table 6 includes a more detailed listing of presentation accommodations. These alter the presentation of the test or test directions. Examples of these accommodations might be providing a large-print version of the test, or reading the test aloud to a student. Evident in this table is that presentation accommodations are widely allowed by states, with Braille or large-print editions of the tests most commonly offered (31 and 32 states offering them, respectively). Reading the test aloud is one of the most controversial accommodations. Approximately 23% (9 of the 39 states with policies) offer reading aloud with no restrictions, another 41% (16 of the 39 states with policies) offer reading aloud with some restrictions (examples include not reading the reading test aloud or only reading the directions aloud), and another 8% (three states) completely prohibit reading aloud. Equipment accommodations frequently mentioned by states are detailed in Tables 7 and 8.

Table 7 is a summary of presentation equipment accommodations, which involve providing specific equipment as part of the presentation of the test. Examples of these accommodations include providing magnifying equipment, or providing altered lighting or acoustics for taking the test. Frequently mentioned presentation equipment accommodations include magnification equipment, amplification equipment, templates, lighting or acoustic alterations, and using a computer or machine in the presentation.

Table 8 is a listing of all other types of presentation accommodations allowed by states (for those states with "Other" noted in Table 6), as well as specifications about tests to which accommodation policies apply. These "other" accommodations further reflect the variability in state policies on accommodations, with available accommodations including underlining verbs in instructions, providing adaptive or special furniture, using specific types of pencils, and making practice tests available.

Response Accommodations

Response accommodations are summarized in Table 9. This table includes many different ways students could respond to a test, for example, writing in test booklets, pointing, or using a tape recorder to record their responses. Again, the variability is very evident, particularly for the accommodation of providing a calculator, with 10 of the 39 states with policies (approximately 26%) allowing calculator use, another 10 (approximately 26%) allowing use with some restrictions (e.g., the IEP specifies calculator use as a goal), and two (approximately 5%) specifically prohibiting the use of calculators. The use of a proctor or scribe was the most frequently mentioned accommodation in this table, with nearly three-quarters of the states with policies allowing the accommodation (though two of states did have some restrictions on their use).

* = True only for certain tests within the state's assessment system. See Table 8 for specification of tests.