Prepared by Martha Thurlow, Ronald Erickson, Richard Spicuzza, Kayleen Vieburg, and Aaron Ruhland
August 1996
This document has been archived by NCEO because some of the information it contains is out of date.
Any or all portions of this document may be reproduced and distributed without prior permission, provided the source is cited as:
Thurlow, M., Erickson, R., Spicuzza, R., Vieburg, K., & Ruhland, A. (1996). Accommodations for students with disabilities: Guidlines from states with graduation exams (Minnesota Report No. 5). Minneapolis, MN: University of Minnesota, National Center on Educational Outcomes. Retrieved [today's date], from the World Wide Web: http://cehd.umn.edu/NCEO/OnlinePubs/MnReport5.html
Education is moving rapidly towards results-based accountability systems that measure the learning of all students. Included in the word "all" are students with disabilities; students with disabilities need to be involved in the assessments that form the basis for accountability. All students, including students with disabilities, need access to diplomas and other life opportunities that assessments provide. There is considerable evidence that this is not the current status of accountability systems and assessment programs in most states. Surveys of state directors of special education (Erickson, Thurlow, Seyfarth, & Thor, 1996) have demonstrated that participation rates are low or unknown. Policies regarding participation in graduation exams are variable (Thurlow, Ysseldyke, & Anderson, 1996), with some states requiring only the accrual of credits, some states requiring that students pass an exam, and some states requiring only that students meet their IEP goals.
For students with disabilities to participate in assessments, there is the need to determine which students are capable of taking the assessment without accommodations, which students are capable of taking the assessment with accommodations, and which students will need a different assessment (Ysseldyke, Thurlow, McGrew, & Vanderwood, 1994). Children with disabilities who require accommodations vary tremendously. For states with graduation exams, or states considering the adoption or development of such exams, accommodation policies are an important part of the accountability picture. The purpose of this report is to examine the policies overseeing the use of accommodations in those states that have high stakes assessments for students.
Eighteen states were included in this analysis of states' written accommodation guidelines: Alabama, Florida, Georgia, Hawaii, Louisiana, Maryland, Michigan, Mississippi, Nevada, New Jersey, New Mexico, New York, North Carolina, Ohio, South Carolina, Tennessee, Texas, and Virginia. The guidelines examined were those that had been compiled by the National Center on Educational Outcomes (NCEO) (Thurlow, Scott, & Ysseldyke, 1995). These were examined for this report specifically for graduation exams within the states (rather than across all tests used in the states). Perusal of the guidelines quickly reveals that states use a variety of terms (accommodation, adaptation, or modifications) sometimes to mean the same thing, sometimes to mean different things (see Thurlow, Ysseldyke, & Silverstein, 1995). In this report, we do not attempt to maintain any distinctions but instead use the term accommodations to include all changes that states allow for students with disabilities.
Each of the 18 states uses a different test for its graduation examination (see Table 1). Twelve states (Georgia, Louisiana, Maryland, Mississippi, Nevada, New Jersey, New Mexico, New York, Ohio, South Carolina, Texas, and Virginia) include a writing sample as part of their testing cycle. All states, with the exception of Nevada and North Carolina, have criterion-based tests. Nevada uses a norm-referenced test; North Carolina uses both criterion-based and norm-referenced tests.
To view Tables 1, 3, 4, 5, and 6, please purchase a copy of Report 5 from the Publications Office. Cost is $5.00
Each state also tests different subject areas. Of nine potential areas of testing (language arts, math, reading, writing, social studies, science, citizenship, English, and foreign language), only New York tests in English and foreign language. The subjects tested the most often were: math (17/18), reading (14/18), and writing (13/18). Hawaii was not included in this count; it has defined its graduation exam as measuring "essential competencies."
Due to the complicated nature of retrieving specific accommodation information from documents provided by the states, agreement data were collected for the two raters who coded information on allowed accommodations (see Table 2). Information was coded within 52 accommodation types (e.g., extended time, Braille version, in study carrel) within four categories (format/equipment, setting, scheduling, and response). Agreement also was calculated by state to determine the clarity of the accommodation information provided by each state (see Appendix A). Overall agreement percentages ranged from 76.7% to 98.5%.
| TABLE 2: Levels of Inter-Rater Agreement in Coding Accommodations | ||
Area Rated |
Mean |
Range |
Format |
88.6 |
62.5 - 100.00 |
Setting |
89.2 |
80.0 - 100.0 |
Scheduling |
93.1 |
75.0 - 100.0 |
Response |
90.2 |
69.2 - 100.0 |
Overall |
90.3 |
76.7 - 98.5 |
The actual number of accommodations also affected the agreement rates (i.e., the more accommodations used, the more likely there would be disagreements). Thus, reliability for one state was low (76.7%); this state also allowed a large number of accommodations (33 of 52 possible). By comparison, the state for which agreement was highest was one of the states that allowed a relatively small number of accommodations (10 of 52 possible). The number of accommodations allowed ranged from 3 to 36.
States that have graduation exams allow a variety of accommodations to be used when students with disabilities take the exams. In addition to listing specific accommodations, many of these states also have general considerations. One of the more common of these addresses the relationship between instructional accommodations and assessment accommodations.
The guidelines of seven states specifically state that testing accommodations must be consistent with accommodations used for the individual student in his or her typical classroom instruction. For example, Virginia's guidelines for testing students with disabilities state:
Accommodations should be chosen based on what the student generally uses during instruction as identified in the student's IEP or management tool. These accommodations are those which the student uses on a daily basis and are not just used for participating in the LTP (Virginia State Board of Education, 1993).
Guidelines from the other states have similar statements about the alignment of testing accommodations and accommodations implemented in daily classroom instruction.
In addition to lists of accommodations, other kinds of information in state guidelines include elaboration of information, deference to other sources, and discussions of the decision-making process. For example, North Carolina includes a "considerations and instructions" section following the description of each accommodation. This additional information provides useful clarification for using modifications. Ohio's guidelines indicate that it only allows those accommodations accepted by the test publisher. Georgia has provisions for out-of-level testing (as do some other states), and actively involves the student in decisions regarding accommodations. Several states require parental knowledge or participation in the decision-making process.
While most states separate information on eligibility for accommodations from their guidelines on accommodations, four states reiterate this point when presenting the accommodations guidelines. Florida's guidelines state:
Modifications are authorized, when determined appropriate by the school district superintendent or designee, for any student who has been determined to be an eligible exceptional student pursuant to Rules 6A-6.0301 and 6A-6.0331, FAC, and has a current individual educational plan, or who has been determined to be a handicapped person pursuant to Rule 6A-19.001(6), FAC (Florida State Board of Education Rules, 1994).
For analysis purposes, this report organizes accommodations into four groups: format/equipment accommodations, setting/administration accommodations, scheduling accommodations, and response accommodations. The percentage of states indicating that specific accommodations are acceptable is reported, but it does not account for those states that might have allowed the accommodation "with approval."
Format/equipment accommodations. This grouping included 15 separate accommodations that states allowed students to use in their high stakes assessments (see Table 3). The format accommodations listed most frequently by states included: Braille and large print (88.9%), word processor/typewriter (72.2%), sign language (61.1%), magnifying equipment and template (55.6%), and audio/video tape and amplification equipment (50.0%). The least frequently listed format accommodation was use of a pointer (16.7%).
Scheduling accommodations. This grouping included seven separate accommodations that states allowed (see Table 4). The scheduling accommodations listed most frequently by states included: extended time (66.7%), multiple sessions (50.0%), with breaks (44.4%), and over multiple days (33.3%). The least frequently listed scheduling accommodation was breaks away from test area (5.6%).
Setting/administration accommodations. This grouping included 14 separate accommodations (see Table 5). The setting/administration accommodations allowed most frequently by states included: individual administration (77.8%), interpretation of directions (66.7%), small group (66.7%), student's home (44.4%), separate room (38.9%), and administration by other (38.9%). The least frequently listed setting/administration accommodation involved adjusting the student's proximity to the test administrator (5.6%).
Response accommodations. This grouping included 12 separate accommodations (see Table 6). The response accommodations allowed most frequently by states included the use of: proctor or scribe (77.8%), machine (72.2%), test booklet (66.7%), student dictation (50.0%), and Braille writer (38.9%). The least frequently listed response accommodations were sign language (11.1%) and lined paper (11.1%).
Most states did not indicate whether the testing results of students using accommodations would be included in the local or state statistics. However, Florida's guidelines specifically state that "scores for the following exceptional students are not included in any classroom, school, district, region, or state averages: Deaf, Hard-Of-Hearing, Specific Learning Disabled, Physically Impaired, Emotionally Handicapped, and Educable Mentally Handicapped" (Florida State Board of Education Rules, 1994). Maryland and Mississippi were the only states among the 18 examined with high stakes exams for students to indicate that all scores would be included in reporting the final testing results.
Because state guidelines frequently are criticized for their vagueness (which contributes to variable implementation), an attempt was made to evaluate the guidelines in terms of clarity, inclusiveness, and organization. Each of the 18 state's accommodations guidelines were rated on a 5-point scale (see Table 7). The states with the highest ratings were Alabama (96.8% inter-rater agreement) and Georgia (93.4% inter-rater agreement), both with a rating of 4+. The states with the lowest ratings were Hawaii (96.9% inter-rater agreement), Michigan (96.9% inter-rater agreement), New Mexico (93.4% inter-rater agreement), Ohio (96.8% inter-rater agreement), and Texas (98.5% inter-rater agreement), all with a rating of 1. While the reliability for each of these states is high, those persons retrieving the information found that either the information presented was not clear, did not include enough information on how to include the largest number of students with disabilities, or the information provided was not well organized. The remaining states received ratings of 2 or 3.
| TABLE 7: Reviewers' Ratings of State Accommodation Guidelines | |||
|---|---|---|---|
State |
Rater 1 |
Rater 2 |
Average |
AL |
5 |
3+ |
4+ |
FL |
3 |
1 |
2 |
GA |
3+ |
5 |
4+ |
HI |
1 |
1 |
1 |
LA |
3 |
1 |
2 |
MD |
1 |
5 |
3 |
MI |
1 |
1 |
1 |
MS |
1 |
3 |
2 |
NV |
3 |
3 |
3 |
NJ |
3 |
3 |
3 |
NM |
1 |
1 |
1 |
NY |
3 |
3 |
3 |
NC |
3+ |
3+ |
3+ |
OH |
1 |
1 |
1 |
SC |
3 |
3 |
3 |
TN |
3 |
1 |
2 |
TX |
1 |
1 |
1 |
VA |
3+ |
3 |
3+ |
Note: Policies were rated on a 5-point scale in the areas of clarity, inclusiveness, and organization.
This rating process suggests that states should consider the manner in which information regarding accommodations is presented. The format of guidelines varied tremendously across states. During the process of analysis, several issues surfaced about the interpretation of allowed accommodations. In the area of format/equipment accommodations the issue was whether sign language and interpreter are the same accommodation (i.e., does allowing sign language as an accommodation imply that an interpreter is used; if so, does the use of an interpreter need to be stated or is it implied?). In the area of scheduling accommodations it was unclear whether multiple sessions automatically indicates over multiple days (i.e., can you have multiple sessions without allowing the test to be taken over multiple days?). In the area of setting accommodations the issue was whether interpretation for directions requires the use of an interpreter (i.e., sign language) or just having a person clarify directions. In the area of response accommodations, allowing students to point, dictate, or sign their responses would seem to necessitate the use of a proctor or a scribe.
Based on this analysis, we offer several recommendations for states considering the use of high stakes testing as part of graduation requirements:
Beyond these recommendations, we urge that states look at inconsistencies in policies across different states. These differences contribute to the lack of understanding of what test scores mean in states with graduation exams. Why is it considered acceptable in one state to allow the use of a proctor or scribe, while it is deemed unacceptable in another? The information on accommodations allowed by states also suggests that some accommodations are almost universally accepted (e.g., Braille version). For these accommodations, there is little need to conduct research on their score comparability, unless of course they are being excluded from aggregations of scores. Clearly, the information in this report suggests a serious need for research related to several accommodations (e.g., templates, separate testing rooms, and testing over multiple days).
Erickson, R.N., Thurlow, M.L., Seyfarth, A., & Thor, K. (1996). 1995 State special education outcomes: Longitudinal trends in how states are assessing educational outcomes for students with disabilities. Minneapolis, MN: University of Minnesota, National Center on Educational Outcomes.
Florida State Board of Education Rules. (February, 1994). Tallahassee, FL. Pp. 39-40.
Virginia State Board of Education. (November, 1993). Guidelines for testing students with disabilities in the literacy testing program, pp. 4-14.
Thurlow, M.L., Scott, D.L., & Ysseldyke, J.E. (1995). A compilation of states' guidelines for accommodations in assessments for students with disabilities (Synthesis Report 18). Minneapolis, MN: University of Minnesota, National Center on Educational Outcomes.
Thurlow, M.L., Ysseldyke, J.E., & Anderson, C.L. (1996). High school graduation requirements: What's happening for students with disabilities? (Synthesis Report 20). Minneapolis, MN: University of Minnesota, National Center on Educational Outcomes.
Thurlow, M.L., Ysseldyke, J.E., & Silverstein, B. (1995). Testing accommodations for students with disabilities: A review of the literature (Synthesis Report 4). Minneapolis, MN: University of Minnesota, National Center on Educational Outcomes.
Ysseldyke, J.E., Thurlow, M.L., McGrew, K.S., & Vanderwood, M. (1994). Making decisions about the inclusion of students with disabilities in statewide assessments (Synthesis Report 13). Minneapolis, MN: University of Minnesota, National Center on Educational Outcomes.
Reliability Across Four Accommodation Areas
For the appendix, please order a printed copy of this report from the Publications Office. Cost is $5.00.