Site Map

Conditions

Education

Behavior

Miscellaneous

Home

MARK ANSWERS IN TEST BOOKLET

In some situations, the typical multiple-choice “fill-in-the-bubble” separate sheet response format may not be an appropriate way for a student to complete a test. The “mark answers in test booklet” accommodation may be necessary for some students to demonstrate what they truly know and can do. This accommodation allows students to respond to test items directly in the test booklet rather than on a separate answer sheet.

Explanation

Burns (1998) suggested that if a student does not understand the “bubble-task,” is not able to record answers using a bubble format, or has a mobility or coordination problem, marking answers in the test booklet might be an effective accommodation. Bubbling-in answers on a separate answer sheet requires adequate student attention to ensure that the intended item is being marked. It also requires adequate motor coordination. If a student does not have these prerequisite abilities, his or her test performance may not reflect what the test was intended to measure.

State Use

Of the 48 states with statewide assessment programs, 28 allow the mark answers in test booklet accommodation. Five additional states allow it on a portion, but not all, of the required state assessments (Thurlow et al., 2000).

Empirical Research

Four empirical studies were identified that examined the effects of the “mark answers in test booklet” accommodation. Three of these studies showed no significant difference between performances of students receiving this accommodation and students completing the test under a standard response mode.

Rogers (1983) conducted a study of this accommodation with 8- to 16-year-old students with hearing impairments taking a multiple-choice spelling test. Results indicated that the separate answer sheet was a valid assessment strategy, and that the two methods had similar degrees of reliability. In other words, the standard and accommodated test formats functioned similarly for this group of students. Tolfa-Veit and Scruggs (1986) looked at differences in number of items marked “outside the box” for fourth grade students with and without learning disabilities. Results indicated that students with and without learning disabilities had similar percentages of items marked outside of the necessary area. Students with learning disabilities were found to complete significantly fewer items than students without learning disabilities in the specified time period; however, the two groups did not differ in the percent of items they answered correctly. Tindal, Heath, Hollenbeck, Almond, and Harniss (1998) similarly studied this accommodation with fourth grade students in special and regular education. In this study, students completed statewide reading and math tests in both accommodated and non-accommodated conditions. Results indicated that there were no significant differences between testing response formats for either group of students (those in regular education and those in special education).

In a study by Mick (1989), a modified test format including “marking answers in test booklet” was found to result in lower scores for students with educable mental handicaps and students with learning disabilities. A reading test was administered to these students under both a modified format (large print, unjustified lines, and writing answers directly in test booklet) and a standardized format. Although these results may suggest that marking answers in the test booklet puts students at a disadvantage, it is not clear which aspect(s) of the modified version accounted for changes in performance.

Overall, there is no empirical support for this accommodation if the criterion is increased test scores. Test scores for students taking the test with and without this accommodation are similar; only one study suggested otherwise. 

Controversy

Separate answer sheets help to facilitate the scoring process. Research has shown students to score similarly both with and without this accommodation. It might be argued, therefore, that the accommodation of marking answers in test booklets is unnecessary. However, the research has failed to address the needs of students with specific motor impairments, as well as those students with attention problems that might lead them to mis-mark answers when using a separate answer sheet. Because of this, this accommodation still seems warranted in certain situations with certain students.

Recommendations

• This accommodation should be allowed for students with motor coordination problems that hinder their ability to respond appropriately in a standard test format. Furthermore, this accommodation should be favored above other response accommodations (e.g. dictated response to a scribe) because it is less likely to result in test bias.
• More research should be done on this accommodation, especially involving students with attention problems and physical impairments.
• If a student does not know how or is not attentive enough to mark answers appropriately on a separate answer sheet, this accommodation should be provided; however, it may be that the student should be taught how to appropriately use the standard bubble sheet format for later testing purposes (Burns, 1998).
• There should be ample space in the test booklet for the student to respond when providing this accommodation (CEC, 2000).

READ ALOUD TEST ITEMS

Reading aloud a test to a student is an accommodation frequently used by students with learning disabilities. Students with physical or visual impairments may also benefit from the oral reading accommodation; however, it has been suggested that students with visual impairments should be tested with Braille or large-print when possible (Phillips, 1994). The read aloud accommodation generally refers to having a teacher or aide read test directions, test items, and test reading passages to students. Although this accommodation is often used in combination with other equipment (e.g., computer, video, cassette tape, etc.), for the purposes of this analysis, focus will be placed on read aloud accommodations as provided by test proctors and teachers.

Explanation

Reading is a prerequisite skill for demonstrating skills in a variety of academic areas. For instance, completing math word problems requires that students know how to read the questions. However, the intent of such an item may be to test math skills rather than reading skills. For this reason, it appears necessary to allow a read aloud accommodation for students with reading disabilities. In this way, students’ true abilities can be manifested on the test without error that is due to an inability to read quickly or accurately.

State Use

Of the 48 states with statewide assessment programs, 4 states allow the read aloud accommodation without limitations. One state prohibits its use, and 30 states allow it under certain conditions. The states that impose limitations on this accommodation generally allow it only on certain sections of the assessment, such as in the math, science, social studies, and writing sections, and prohibit its use on the reading sections (Thurlow et al., 2000).

Survey Research

The 64 teachers responding to the Gajria et al. (1994) survey of 100 teachers generally favored the read aloud accommodation. A total of 90.6% of the respondents were aware of the read aloud accommodation, 79.7% reported using it in their classrooms, and 89% thought that providing this accommodation maintained the integrity of the test. It was ranked 7th of 32 accommodations in terms of “ease of use,” and 5th in terms of effectiveness.

The Jayanthi et al. (1996) survey responses from 401 of the 708 general education teachers to whom survey were sent indicated that “reading test questions to students” was an easy accommodation to make. They also indicated that it was “helpful” to “very helpful” to use. Of the teachers that responded, 67.9% reported using this accommodation in their classrooms.

Hollenbeck et al. (1998) found that 56% of the 166 responding teachers (of 633 surveyed) had correct knowledge about the accommodation entitled “read math text.” A total of 21% reported using this accommodation. 

Empirical Research

Oral reading accommodation for math tests. Nine studies looked at the effects of the oral reading accommodation on math test scores. Five of these studies tested the hypothesis that the oral reading accommodation should differentially boost the scores of students with disabilities compared to students without disabilities. Three of these studies found that students with learning disabilities benefited significantly more from the accommodation than students without disabilities (Fuchs, Fuchs, Eaton, Hamlett, & Karns, 2000; Tindal, Heath, Hollenbeck, Almond, & Harniss, 1998; Weston, 1999). Although Johnson (2000) found that students with learning disabilities did differentially benefit from the accommodation, the increase in scores was not statistically significant due to a small sample size. Johnson (2000) also found that orally reading the test items had no differential effects for good versus poor readers among non-disabled students.

Further support on using this accommodation for students with learning disabilities is warranted from additional findings by Tindal et al. (1998). Tindal et al. (1998) found that even the lowest achievement-ranked students in general education did not benefit significantly from the oral reading accommodation. According to these findings, the oral reading accommodation appears to be a valid accommodation for students with learning disabilities. Calhoon, Fuchs, and Hamlett (2000) studied a sample of only students with learning disabilities, and found that providing a reader on a math test significantly increased scores. They also studied a computer read aloud accommodation, and found that this accommodation also increased scores of students with learning disabilities. No significant difference in scores was found between teacher-read and computer-read accommodations.

Other studies have looked at the read aloud accommodation in terms of overall test functioning. Pomplun and Omar (2000) found that a proposed two-factor model (writing and math factors) fit the data for all groups of students on a fourth grade math test, including students with learning disabilities receiving the read aloud accommodation, students with learning disabilities not receiving the accommodation, and students without disabilities. This further suggests that the read aloud accommodation is appropriate for students with disabilities on math tests. Additionally, Bielinski, Thurlow, Ysseldyke, Freidebach, and Freidebach (2001) found that although there were a few items that displayed differential item functioning for accommodated students on a statewide math test, overall, the accommodated version appeared to be measuring the same construct as the non-accommodated test for students without disabilities. In a similar study, Lewis and Green (1999) found there to be relatively few items displaying DIF for the accommodated group on a math test; however, no reference group of regular education students was provided in the study.

In contrast, Meloy, Deville, and Frisbie (2000) found that the read aloud accommodation increased the scores of both learning disabled and non-learning disabled students on math and science sections of the Iowa Tests of Basic Skills. They argued that the read aloud accommodation changed the construct being measured for most students relative to what is measured under standard conditions.

For the most part, the oral reading accommodation for math tests appears to be supported by research. However, the findings of Meloy et al. (2000) suggest that more research support is needed to validate the use of this accommodation

Oral reading accommodation for reading tests. Concern has been expressed regarding the use of the oral reading accommodation on reading tests. Phillips (1994) suggested that this accommodation results in the substitution of listening comprehension for reading comprehension, and therefore greatly alters what the test measures. Burns (1998) argued that the oral reading accommodation for reading tests undermines test validity, and could lead to students not getting necessary services.

Five studies on the use of the read aloud accommodation for reading tests were identified. Tippets and Michaels (1997) examined the factor structure of reading and language arts tests for students receiving the read aloud accommodation alone and in combination with other accommodations (e.g., read aloud + extended time, read aloud + dictated response, etc.), and compared this to the factor structure of the tests for non-accommodated students. Results indicated no difference between the structures, which suggests that the same underlying construct was being measured.

Contrary to this finding, Meloy et al. (2000) found that the read aloud accommodation benefited both learning disabled and non-learning disabled students on several sections of the Iowa Tests of Basic Skills, of which the greatest impact was found on the reading sections of the test. This suggests that the accommodation changes what the test measures, therefore is invalid. Bielinski (2000) also suggested that the read aloud accommodation changed the construct that statewide reading test intended to measure. He found numerous items displaying differential item functioning (DIF) when compared to results for students in regular education. However, there were also several DIF items for low performing students in regular education. Lewis and Green (1999) found few DIF items for accommodated students in a similar study; however, the study did not provide a comparison group of students without disabilities.

Finally, Barton and Huynh (2000) conducted a study of errors made by students with disabilities receiving the read aloud accommodation on a high school reading test. Their analysis suggested that there may be some items that are particularly difficult for students receiving the read aloud accommodation, such as items that include an index of entries or references that are difficult for students to listen to and remember if they are not skilled in following along as the proctor reads.

Clearly, more research needs to be done on the oral reading accommodation to determine how it affects what the test measures. 

Controversy

Research has primarily supported the use of the read aloud accommodation for students with disabilities on math tests. However, great concern has been expressed about the validity of using this accommodation on reading tests, and limited research has addressed this issue. Additionally, it is not always clear what the read aloud accommodation entails; in some cases students may be allowed to ask the reader to “re-read” certain sections, and in other cases students have the test read only once and must rely on auditory memory in answering the questions (Burns, 1998). Administering this accommodation in large groups of students can complicate things further; students may require different amounts of time to complete items, requiring the reader to adjust his or her pace accordingly. Concern is also evident in whether students should be given a copy of the test to follow along or if they should simply listen to the test being read. Some students may benefit from following along, whereas it may pose as a distraction for others (Burns, 1998). Finally, readers must be careful not to display inadvertent cues such as changes in voice inflection when correct answers are being read. Students may easily recognize these subtle hints, which may in turn alter the meaning of test scores. Overall, it appears that differences in how the read aloud accommodation is administered may have important effects on test scores. 

Recommendations

• The read aloud accommodation should be offered to students for whom it is determined will benefit from this accommodation on tests not specifically designed to measure reading achievement.
• More research needs to be done on using the read aloud accommodation on reading and language arts tests.
• Decisions should be made on an individual basis about whether the student can follow along when reading the test, or whether he or she should only listen to the test being read (Burns, 1998).
• In order to best reflect the non-accommodated condition, students should be allowed to ask the reader to re-read certain sections of the test.
• It is best to administer this accommodation individually to effectively accommodate each student’s pace.
• Students using the oral reading accommodation should be allowed approximately twice the normal amount of time to complete the test (Burns, 1998).
• Readers should be trained in how to effectively administer this accommodation (proper reading speed, avoidance of inappropriate voice inflection, etc.). They should know the pronunciation of words on the test and practice reading it in a straightforward and clear manner prior to test administration (CEC, 2000).

READ/REREAD/SIMPLIFY/CLARIFY TEST DIRECTIONS

A variety of adaptations are frequently made in how test directions are communicated to students. Directions may be read aloud or paraphrased, additional examples may be given, or the student may be allowed to ask to have directions repeated. Frequently, this accommodation is included along with the read aloud accommodation for the remainder of the test. This accommodation appears to be most appropriate for students with reading-related disabilities.

Explanation

In order for students to demonstrate what they know and can do on a test, they must have a good understanding of what the test requires them to do. A small misunderstanding in test directions can cause students to complete an entire test incorrectly. In general, tests do not intend to measure how carefully a person can follow directions. This accommodation helps to facilitate the understanding of test directions so that students can demonstrate their true knowledge and skills.

State Use

Of the 48 states with statewide assessment systems, 31 allow some form of the read/reread/ simplify/clarify directions accommodation (Thurlow et al., 2000). One state prohibits this accommodation, and four other states have limitations on its use. In one state that limits its use, clarifying directions is prohibited. In another state, the accommodation is allowed only on certain subtests.

Survey Research

This accommodation has been the subject of several surveys. The earliest was by Gajria et al. (1994), who surveyed 100 teachers and received responses from 64. They found that 90.6% of those surveyed were aware of the “read test directions and items” accommodation, 79.7% reported using it in their classrooms, and 89% thought that it maintained the integrity of the test. It was ranked 7th of 32 modifications in terms of “ease of use,” and 5th in terms of effectiveness.

Jayanthi et al. (1996) received surveys from 401 of 708 general education teachers. Results indicated that of the 24 accommodations studied, “give individual help with directions during tests” was rated as “easy” to “very easy” to provide. It was ranked first in helpfulness, and in terms of how many teachers used the accommodation (84.7% of the teachers who responded indicated that they used the accommodation).

Lambert et al. (1996) received responses from 121 of 171 teachers surveyed. “Allow a proctor to rephrase test questions” ranked sixth among several accommodations in terms of the percentage of teachers who had provided the accommodation in the past. Also, the average rating of willingness to provide this accommodation was 4.1 on a 1-5 rating scale, 5 being “very willing to provide.”

Hollenbeck, Tindal, and Almond (1998) surveyed 633 regular and special educators, of which 166 responded. They found that 51% of the teachers who returned surveys had correct knowledge about the “clarify directions” accommodation, and 16% reported using this accommodation.

Empirical Research

Very limited research on this accommodation was identified. Elliott, Kratochwill, and McKevitt (2001) included “support with understanding test directions” as part of a package of accommodations to students in a study of the effects of testing accommodations for students with and without disabilities. Several students without disabilities and all students with disabilities in this study were given an individualized accommodation package determined to be appropriate by their teachers. Another group of students without disabilities was provided a standard package of testing accommodations, which included “support with understanding test directions” as well as other accommodations (e.g., help reading words, verbal encouragement, extra time). Finally, one group of students without disabilities was not provided any accommodations. Accommodation packages were found to have positive effects for a large percentage of students with disabilities (63.4%), a medium percent of students without disabilities who received the teacher recommended packages (42.9%), and a smaller percent of students without disabilities receiving the standard package (20.0%). Because of the use of multiple accommodations in this study, however, it is difficult to determine the specific effects of the “support with understanding test directions” accommodation.

Controversy

Tests typically are not designed to measure how well students follow directions, but rather how well they read, write, calculate, etc. Logically, this accommodation should not change what the test measures. Teachers and policymakers tend to agree that this is a valid accommodation; overall, there is not much controversy. However, further empirical support should be sought. Furthermore, because this accommodation is “definitionally diverse,” it may be necessary to study its different manifestations separately (e.g., reading directions vs. simplifying directions).

Recommendations

• In general, test directions should be written as simply and clearly as possible on the test, thus avoiding the need for much additional clarification.
• When the purpose of the test is not to test the ability to follow directions, this accommodation should be allowed.
• This accommodation should always accompany the read aloud accommodation unless the test measures reading decoding skills.
• When clarifying test directions, it is important for the test administrator to clearly and accurately communicate how the test is to be completed.
• A set of guidelines should be developed to indicate appropriate and inappropriate procedures for this accommodation.
• This accommodation may require testing in a separate location in order to avoid distracting other students (CEC, 2000).

TEST BREAKS

Students often are provided breaks between subtests. In some circumstances, students are not given breaks between subtests, or the breaks they are given are very short. Rarely are students given breaks within subtests. In this summary, the “test break” accommodation is defined as allowing students to take the test with breaks when the normal administration of the test does not include breaks. The test breaks accommodation may provide students with various disabilities the opportunity to demonstrate what they truly know and can do on a test. 

Explanation

The use of magnification equipment, tape recorders, and earphones can cause fatigue, making frequent breaks a necessity (CEC, 2000). Also, students with attention problems may have difficulty concentrating on a test for the period of time that is typically expected of students. Fatigue and lowered concentration levels can limit students’ abilities to show what they really know on a test. When provided with more frequent or longer breaks, students with disabilities can experience a break from frustration and regain an optimal level of test concentration.

State Use

Of the 48 states with statewide assessment programs, 28 allow the test break accommodation. Five additional states allow this accommodation under certain conditions. For these five states, the accommodation is allowed on only some of the statewide assessments, or is allowed only between subtests (Thurlow et al., 2000). 

Survey Research

No survey research was identified that specifically addressed the test breaks accommodation. One study, however, examined the accommodation of testing over several sessions. Hollenbeck, Tindal, and Almond (1998) surveyed 633 regular and special educators and received responses from 166. They found that 45.2% of the responding teachers had correct knowledge about this accommodation, and 18% reported using this accommodation.

Empirical Research

No studies were identified that specifically addressed the test breaks accommodation. However, two studies were identified that examined the use of a very similar accommodation – testing on multiple days. DiCerbo, Stanley, Roberts, and Blanchard (2001) studied the effects of two-day and three-day administrations of reading comprehension tests on the scores of 939 third-grade students. All students completed a form of the Stanford 9 reading comprehension test, and a second form of the same test under either the two-day or three-day condition. Students were not able to return to previously completed test sections on the second or third days, and the overall amount of time students had to complete the test was equivalent. Results indicated that participants’ scores were 12 scaled score points higher for the divided-time administrations than for the one-time administration (ES=.25). There was also a significant time by reading comprehension ability interaction (p < .001). Middle and low ability readers experienced greater benefits from the multiple day administrations than high ability readers.

Walz, Albus, Thompson, and Thurlow (2000) examined the effects of allowing students to take a reading test over multiple days versus one day. Participants were 112 seventh and eighth grade students (48 students receiving special education services, and 64 non-special education students) taking the Minnesota Basic Standards reading test. Under the one day administration, students read and answered questions for three reading passages; in the multiple day administration, students read one passage and completed the corresponding questions each day for three days. Each student participated in both administrations of the test. Results indicated that a multiple-day test accommodation did not enhance the test scores of students with disabilities. Furthermore, students without disabilities performed better under the one-day administration.

Overall, it appears that more research needs to address the test breaks accommodation. No research was identified that specifically studied this accommodation, and studies of similar accommodations have differing results. Perhaps the age of the students is a factor (e.g., younger students may benefit more from this accommodation).

Controversy

Test breaks are suggested to provide students with disabilities a chance to regain concentration in order to perform optimally on tests. Furthermore, this accommodation does not appear to significantly affect what is being tested. However, no research has addressed whether students with disabilities perform better when provided the test breaks accommodation. It has been suggested that breaks within subtests may distract from the problem-solving rhythm a student has developed (Burns, 1998) and interfere with performance on a sequence of items (CEC, 2000). However, breaks between subtests are suggested to be beneficial, especially when students are being administered a test individually (Burns, 1998). It is clear that more research is needed to determine whether the test breaks accommodation is appropriate, and if so, how and when it should be used.

Recommendations

• Test breaks between subtests should be provided to students for whom it has been determined will benefit from this accommodation.
• Breaks within subtests should be avoided, unless absolutely necessary.
• Because it may be possible for students to find answers to previous test questions during breaks, students should not be allowed to return to previously completed test items following a break.
• The test break accommodation may require a separate setting (CEC, 2000).
• Lengthened test breaks may be necessary for some students (Burns, 1998).
• Test-developers should take into consideration the age and the sustained attention abilities of the students who will be tested.
• More research needs to address the use of this accommodation.

SUMMARY

This report on the research that has been conducted on the most frequently used accommodations reveals that there are not yet simple or conclusive answers to questions about the effects of accommodations. Yet, there are some trends. There were three accommodations included here that more than 90% of the states allow without limitations (Interpreter for Instructions, Large Print Edition of the Test, Read/Reread/Simplify/Clarify Directions). For these three, the primary concerns relate to the quality of the implementation of the accommodation and to the alignment of the accommodation to specific student needs.

The Read Aloud accommodation is clearly the most controversial of the most frequently allowed accommodations, with only four states allowing it without limitations. The research confirms this controversy, with results for reading tests mired in concerns about fidelity of implementation, validity issues, and logistical concerns, even though use of Read Aloud generally is supported for math tests.

It is important to continue to document what the research tells us, and to analyze the specific context of the studies, their methodologies, and their findings. Doing so will be facilitated by referring to NCEO’s searchable accommodations research database on its Web site (http://education.umn.edu/NCEO/AccomStudies.htm).

Site Map

Conditions

Education

Behavior

Miscellaneous

Home

Tourette Syndrome "Plus" © Copyright 1998 - 2005 Leslie E. Packer, PhD. except as noted.
All rights reserved 
This page last updated January 7, 2005. Reprint Policy. Privacy Policy. Contact.