Discussion
Innovation
Pick Conducting Preference
Assessments With Students With Significant Disabilities
Helen I. Cannella-Malone, Linsey M. Sabielny,
Elíseo D. Jimenez, and Megan M. Miller
Mr. Woods teaches eight middle-school- age students who all have significant intellectual and developmental disabili- ties and limited communication skills. Six of Mr. Woods's students make choices by physically reaching out and selecting an item or picture. His other two students have significant physical disabilities and are unable to physically select an item or picture if it is present- ed to them. Although Mr. Woods tries to build in choice-making opportunities throughout the school day, he struggles with developing systematic reinforcers for his students' behavior. He has often wondered if there were a way he could identify age-appropriate reinforcers for each of his students given that they have a wide range of strengths and challenges.
It is essential to identify the prefer- ences of students with significant intel- lectual, developmental, and physical disabilities (SIDPD) so that teachers can use those preferences to reinforce new behaviors. Students with SIDPD have demonstrated that they can indi- cate clear preferences when teachers conduct methodologically rigorous preference assessments. Moreover, teachers can use the items as rein-
16 COUNCIL FOR EXCEPTIONAL CHILDREN
forcers once students have identified them as preferred using these assess- ments. Many teachers have found the items useful as reinforcers to teach new behaviors and maintain learned behaviors (Cannella, O'Reilly, & Lancioni, 2005). By relying on prefer- ence assessments to assist in the identification of potential reinforcers, teachers can not only consider stu- dents' individual preferences, but teachers can also provide more effec- tive instruction for each student.
Teachers often ask why they couldn't simply ask others what the student likes, including parents, spe- cialists, and other caregivers. First, most people select common items for students to choose from—including many items that teachers have offered for student choices for many years, without variety. Second, in only offer- ing common items for choice, teachers may overlook idiosyncratic items. For example, we have worked with stu- dents whose most preferred activities included watching another person sneeze or watching someone trip, which we did not suggest as possible preferences in initial questionnaires. Instead, students identified these activ- ities as possible reinforcers during a
structured preference assessment when the student responded positively (e.g., laughed, looked toward the researchers and smiled) when the researcher just happened to sneeze or trip. Finally, researchers have suggested that many educators identify the wrong rein- forcers nearly 50 % of the time (Can- nella et al., 2005). Therefore, it is pure chance if items identified by people close to the student actually act as reinforcers. Bearing this in mind, how- ever, it is always worth communicating with parents and care providers as a starting point in identifying items that could be included in a systematic pref- erence assessment.
Given the evidence that conducting direct preference assessments can suc- cessfully identify reinforcers for stu- dents with SIDPD, researchers need to provide teachers guidelines for con- ducting these assessments. In this arti- cle, we describe methods for choosing the most appropriate assessment for a particular student, describe four com- monly used preference assessment pro- cedures, offer troubleshooting solu- tions, and discuss how teachers can use results to enhance educational pro- gramming for students with SIDPD
TEACHING EXCEPTIONAL CHILDREN | JULY/AUG 2013 17
What Does the Research Say About Preference Assessments?
Research clearly identifies prefer- ence assessments as a valid and effective way of identifying preferred stimuli for students with disabilities (Tullis, Cannella-Malone, Basbigill, Yeager, Fleming, Payne, & Wu, 2011). The variety of assessments allow for individualization of the preference assessment based on the student's needs. Teachers can choose from single-stimulus (Pace, Ivancic, Edwards, Iwata, & Page, 1985), paired-choice (Fisher, Piazza, Bowman, Hagopian, Owens, & Slevin, 1992), multiple stimulus without replacement (DeLeon & Iwata, 1996), and eye-gaze (Fleming, Wheeler, Cannella- Malone, Basbagill, Chung, & Graham Day, 2010) assessments. Within each of these types of assessments, teachers may make modifications for the student to make the assessment more effective. Additional research is needed regarding these modifications, espe- cially given the advancements occurring with technology. More research is also needed regarding the effective use of these assess- ments in the school and home envi- ronments by teachers and parents (Tullis et al., 2011).
(see box, "What Does the Research Say About Preference Assessments?").
Choosing Ihe Right Assessment
To select the appropriate assessment for each student, teachers can refer to Figure 1, which shows a preference assessment decision tree.
Assessment Procedures and Descriptions
Each of the following assessments is best suited to a particular type of stu- dent. Instructions include how many items can be assessed, approximately how long the assessment will take, how easy the assessment is to set up and conduct, what preparation is nee-
Figure 1. Decision Tree for Ciioosing tiie Appropriate Preference Assessment for Students With Significant intellectuai, Developmentai, and Physical Disabiiities
Can the student physically select an item?
Eye Gaze Assessment
YES
i Can the student choose
between two items?
Single Stimulus
Assessment
YES
i Can the student choose
between three or more items? Paired Stimulus
Assessment
YES
i Multiple Stimulus Without Replacement Assessment
essary, directions for conducting the assessment, and directions for deter- mining the preference hierarchy. The procedures described are based on the original assessment research, as well as our experiences conducting the assess- ments with students who have SIDPD.
Single-Stimulus Preference Assessment
The single-stimulus preference assess- ment was one of the first preference assessments developed for students with significant disabilities (Pace, Ivancic, Edwards, Iwata, & Page, 1985). It was designed particularly for students who do not choose from mul- tiple items, but do attend to a single item when it is presented. Up to 16 items can be assessed before the assessment becomes too cumbersome. As more items are added, the time to conduct the assessment increases. If
six items are included and each is presented five times, this will take at least 1 hour to complete. Although it is time consuming, benefits of this assessment are that the procedures are simple to implement, and choice mak- ing is not a prerequisite.
To conduct this assessment, the teacher should first prepare the data sheet (see Figure 2) by listing the items that will be assessed and defining "engagement" for the student. This assessment is based on examining stu- dent engagement with possible rein- forcers as a way to determine potential reinforcing strength. Engagement might include looking at the item, leaning toward the item, taking the item and playing with it, or a calming response (e.g., a student who engages in Stereo- typie behavior may stop engaging in the behavior when a particular item is presented). When defining engage-
18 COUNCIL FOR EXCEPTIONAL CHILDREN
ment, the teacher should be sure to use a description that is both observ- able and measurable, meaning that anyone observing the student would clearly be able to say that engagement occurred without additional input.
Next, the teacher should make sure that each item on the list is available and the student is comfortable [e.g., sitting at a table, on the floor, in a wheelchair, or in a Stander). Then, the teacher should present each item indi- vidually for 30 seconds. If the item is a power toy, it should be given to the student turned on; if the student is placed in a swing, it should be lightly pushed to move the swing, and so forth. If the student engages with the item, the teacher should score the first trial for that item with the number 1. If the student does not engage with the item, the teacher should score the first trial for that item with a zero. Once each item has been presented once, the teacher should present them all again for 30 seconds each in a different order. Each item should be presented a total of five times.
Although it is possible to conduct this assessment in one sitting, the stu- dent may get fatigued and stop engag- ing. If the assessment will be broken up across time periods, it is best to complete at least one presentation of each item in each session. During the next session, the teacher should follow the same procedures, changing the presentation order from the previous session. If 10 or more items are being assessed and breaks will be required, it is best to split the items into groups with like numbers of items. Using these general guidelines, decisions about how to break up the assessment should be based on the individual needs of the student. To determine the preference hierarchy once the assess- ment is complete, the teacher should sum the number of times the student engaged with each item, and then rank the items from the highest to lowest total. Items with higher totals are con- sidered more highly preferred than those items with lower totals.
Figure 2. Singie-Stimuius Preference Assessment Data Siieet
Items THal 1 THalZ Trial 3 TVia!4 Trials Total
Rank items from the highest total to the lowest total to obtain a preference hierarchy, where the higher number indicates a higher preference.
1.
2. _
3. _
4. _
5. _
Notes:
Note. Sample data sheet for a single-stimulus preference assessment that includes five items to be assessed. Add rows to the trials table and rank-order list as you add items to be assessed.
Paired-Stimulus Preference Assessment
A paired-stimulus preference assess- ment is appropriate for students who can choose between two items but do not reliably select items from a larger array [Fisher et al., 1992). A teacher can assess up to 16 items in a paired- stimulus assessment, though the num- ber of items assessed is generally clos- er to five. As this number increases, the complexity and duration of the assessment also increases, because each item has to be paired with every other item included in the assessment. For example, to assess six items, 15 pairs will be presented; to assess 16
items, 121 pairs will be presented. A teacher can expect to spend 45 minutes assessing six items and at least 2 hours assessing 16 items. Although the preparation is slightly complicated, the assessment procedures are easy to implement. A benefit of this assess- ment [and all subsequent assessments) is that it provides information on dif- ferential responding, which allows a teacher to compare preference in rela- tion to other items included in the assessment, something that is impossi- ble with the single-stimulus assess- ment.
To conduct this assessment, the teacher should first prepare the data
TEACHINC EXCEPTIONAL CHILDREN | JULY/AUG 2013 19
Figure 3 . Paired-Stimulus and Eye Gaze Preference Assessment Data Sheet
Items
1.
2.
3.
4.
5.
Number of Times Chosen
Trial 1
Trial 2
Trial 3
Trial 4
Trial 5
Trial 6
Trial 7
Trial 8
Trial 9
Trial 10
Items (circle selection)
Left
2
3
5
1
4
5
1
1
2
3
Right
1
2
3
3
2
4
5
4
5
4
None
None
None
None
None
None
None
None
None
None
Rank items from the highest total to the lowest total to obtain a preference hierarchy, where the higher number indicates a higher preference.
1.
2.
3.
4.
5.
Notes:
Noie. Sample data sheet for the paired-stimulus and eye gaze preference assessments that include five items to be assessed. Add rows to the items table, additional pairings to the trials table, and to the rank-order list as you add items to the assessment.
sheet (see Figure 3 ; also see Figure 4 for a data sheet involving multiple stimuli) by listing the items that will be assessed with a corresponding number. If assessing more than five items, the teacher will need to add pairs to the data sheet before beginning the assess- ment. The assigned number will help the teacher determine which items are to be presented and on which side to present each item during a given trial. The teacher should also determine the duration of access (with consumable items given shorter access and activi- ties longer access). This duration should remain constant.
Next, the teacher should gather all items to be included in the assessment and make sure the student is comfort- able. Before the first trial, the student should be allowed to individually sam- ple each item. The teacher should begin each trial by presenting two items, within reach of the student, approximately 1 foot apart and saying, "Pick one." Placement can be adjusted during the assessment if the student is having difficulty accessing the items (see Table 1 for troubleshooting sug- gestions) . If the student selects an item (i.e., physically takes, points to, or ver- bally labels one item), the teacher should provide 5-30 seconds of access while removing the second item and circling the corresponding number for the selected item for that trial. If the student does not select either item within 5 seconds, the teacher should remove both items and score the trial as "none." The assessment is complete once all pairs have been presented.
Although it is possible to conduct this assessment in one sitting, the stu- dent may get fatigued and stop engag- ing with the items. The teacher should use the procedures described previ- ously to make decisions about how to break up this assessment. To determine the preference hierarchy once the assessment is complete, the teacher should sum the number of times each item was chosen, then rank the items from the highest to lowest total. Items with higher totals are considered more highly preferred than items with lower totals. If a student selected several items an equal number of times, the
20 COUNCIL FOR EXCEPTIONAL CHILDREN
teacher can refer to the trial in which those items were paired together to determine which item should be ranked higher. If this assessment iden- tifies multiple low-preference items, those items should be documented and replaced with new items in subsequent assessments.
Eye-Gaze Preference Assessment
The eye-gaze preference assessment represents an adaptation of the paired- stimulus preference assessment and was the first specifically designed to identify reinforcers for people with multiple disabilities, particularly those who may lack the physical capabilities to reach out and select an item (Fleming et ai., 2010). As with the paired-stimulus preference assessment, up to 16 items can be assessed, but the time to conduct the assessment will increase exponentially. If the recommended six items are included, it will take several hours per student to complete.
Because eye gaze will vary
greatly across students, it is
importaut that the definitiou is
student specific, as well as
observahle and measurable.
To conduct this assessment, the teacher should first prepare the data- coilection sheet (see Figure 3) as described in the paired-stimulus assess- ment and define eye gaze for each stu- dent. Defining eye gaze for each stu- dent is the most difficult part of this assessment. Typically, eye gaze is defined as a 2-secünd directional gaze, initiated within 5 seconds of the pair of items being presented. However, pre- assessment sessions will likely be nec- essary to determine where to position the items and whether the student will maintain his gaze for 2 seconds. For example, a student's resting position might be with his head tilted to the right with chin to chest. In this case, the teacher should position the items farther to the right and near the stu-
Figure 4 . Multiple Stimulus Without Replacement (MSWO) Preference Assessment Data Sheet
Items Session 1 Session 2 Session 3 Session 4 Session 5 Total
Rank items from the lowest total to the highest total to obtain a preference hierarchy, where the lower number indicates a higher preference.
1.
2. ^
3.
4. _
5. _
6. _
Notes:
Note. Sample data sheet for the multiple stimulus without replacement (MSWO) preference assessment that includes six items to be assessed. Add rows to the trials table and rank order list as you add items to be assessed.
dent's lap to ensure that both items are visible. Tilting the student's wheelchair up or down, adjusting the headrest, and using or removing the student's lap tray may enhance a teacher's abili- ty to observe eye gaze. Because eye gaze will vary greatly across students, it is important that the definition is stu- dent-specific, as well as observable and measurable.
Following preparation, the teacher should present each pair of items in the order listed. To begin, the teacher should allow the student's eyes to
come to a neutral gaze, then present both items by placing or holding the items as identified during the pre- assessment. Next, the teacher should remove both items and re-present and manipulate each item individually for 5 seconds (i.e., sample the items), then remove and present both items together and say "Pick one." The teacher should watch the student's gaze to see which item he is looking at. A choice is defined as looking at an item for 2-3 seconds, depending on the definition of eye gaze created
TEACHING EXCEPTIONAL CHILDREN | JULY/AUG 2013 21
Tabie 1 . Treubiesiiooting Tabi< Suggested Soiutions
Witb tbe Probiem/Pitfaii and
Problem/Pitfall
The student grabs all of the assessment items
The student selects in a certain order
The student won't select any of the options
The student satiates on the assessment items
Several items are ranked equally
The student selects unavailable items/activities
The student demonstrates a side bias
The student will not select from an array of 6 or more items, but will choose from an array of 3
If a student is not able to sample the item, the assessment may be inaccurate
Behavior doesn't increase when items are presented contingently
Solution
Arrange items so they're not easily grabbed (in a cupcake tin) or have another person assist
Randomize the presentation order or introduce distractions
Possibly use a different selection response, a more basic assessment, or prompt sampling
Provide smaller samples of the items or break the assessment up over days
Items may be equally preferred, you can conduct another assessment to confirm
Make an inventory of the available options
Shift all choices to that side or present vertically
Decrease the number of items in the assessment
Allow the student to sample each item when selected
Increase rate of reinforcement
The reinforcing effects may not be pow- erful enough if the behavior is particu- larly challenging. It may be necessary to conduct another assessment with new items
during the pre-assessment. Once a choice has been made, the teacher should provide access to the chosen item for 30 seconds and remove the item not chosen. For each trial, the teacher should circle the choice made by the student on the data sheet. If a directional gaze is not initiated within S seconds, both items should be removed and then re-presented without sampling the items. If a directional gaze is again not initiated within 5 sec- onds, the teacher should remove both
method (i.e., eye gaze) is fatiguing for the student. Some students can toler- ate up to five trials per day, but some students will only complete one trial per day. To determine the preference hierarchy once all trials are complete, use the guidelines presented for the paired-stimulus preference assessment.
Multiple Stimulus Without Replacement
DeLeon and Iwata (1996) developed the multiple stimulus without replace- ment (MSWO) assessment in an effort to combine paired-and multiple-stimu- lus assessments. The MSWO is appro- priate for students who choose from an array of three or more items and can be used to assess up to 16 items. Although the assessment duration increases as items are added, the total duration is much shorter than the other assessments and will vary based on the number of items and trials presented. The average duration is approximately 20 minutes when six items are assessed across five sessions.
To conduct this assessment, the teacher should first list the items to be assessed on the data sheet (see Figure 4). The duration of access should also be determined (with consumable items given shorter access and activities longer access). This duration should remain constant.
Next, the teacher should make sure the student is seated comfortably with a tray or table in front of him to dis- play the items. The teacher should place all items in front of the student at the same time, generally in a straight line. The teacher should then tell the student to "pick one" and wait
In addition to using preferred items to increase behavior,
a teacher can use the identified items to provide his students
with activities for which they have demonstrated preference.
items, indicate "no choice" on the data sheet, and present the next pair. This process should be repeated until all pairs have been presented.
This assessment should be conduct- ed across days, because the selection
S seconds. Once the student selects an item (as defined in the paired-stimulus assessment), the array should be removed and the student should be allowed to engage with the item for the predetermined duration. On the
22 COUNCIL FOR EXCEPTIONAL CHILDREN
data sheet, the teacher should record a "1" in the first column next to the item the student chose. Next, the teacher should rearrange the remaiti- ing items and present the array again. This procedure should be repeated until all items have been selected or the student stops selecting items, and the teacher should record the order of selection for each item in the array (e.g., if item »2 is selected third, write a "3" in the row for item 2 in session 1). This is considered one session, and five sessions should be conducted before calculating the preference hierarchy.
All five sessions of the MSWO need not be conducted in one sitting. Although this is a relatively quick assessment, it is important to only conduct the assessment while the stu- dent is engaged, attending, and moti- vated. The teacher should use the pro- cedures described previously to make decisions about how to break up this assessment. To determine the prefer- ence hierarchy once five sessions have been completed, the teacher should sum the rankings of each item across the five sessions. In this case, the items with lower totals are considered more preferred than those with higher totals.
What Happens If Something Goes Wrong?
Ideally, teachers have identified the correct preference assessments to use with their students, the assessments were conducted without any problems, and the teacher identified several high- ly preferred items to use with his stu- dents. Several challenges, however, could arise throughout the assessment process. Table 1 depicts common issues and possible corresponding solutions.
The Assessments Are Finished! Now What?
A teacher now has several items that have been identified as highly pre- ferred for his students. So what can be done with them? Highly preferred items should be used during instruc- tion to reinforce students' behaviors that are in acquisition and to maintain
learned behaviors. We know that rein- forcing behaviors is the key to increas- ing them and that the reinforcement of new skills is most powerful when the reinforcer used is highly preferred. So, the teacher should use those items identified as highly preferred during instruction, by allowing access to the items contingent on desired behaviors.
In addition to using preferred items to increase behavior, a teacher can use the identified items to provide his stu- dents with activities for which they have demonstrated preference. Stu- dents with significant disabilities may have restricted interests or have few activities in which they engage inde- pendently. By conducting preference assessments and identifying preferred activities, teachers can broaden the number and type of activities available to the student.
Teachers should always remember the following as they begin to use pref- erence assessments. First, instruction will be enhanced with the use of con- tingently delivered reinforcers (i.e., only providing access to the reinforcer immediately following a behavior to be increased). Second, the identification of these reinforcers for students with SIDPD can most successfully occur using one of the preference assess- ments described in this article. Finally, a preference assessment will only be as good as the items included, so teachers should be sure to identify age-appropri- ate items.
After conducting systematic preference assessments with each of his students, Mr. Woods has successfully identified new reinforcers for all of his students. He used the MSWO with three of his students, the paired-stimulus preference assessment with two students, the single-stimulus preference assessment with one student, and the eye-gaze assessment with two students. Although he had to use various assess- ment methods with his students, he is thrilled that each student has made improvements since he started using the highly preferred items as reinforcers in his instruction.
References Cannella, H. 1., O'Reilly, M. F., & Lancioni,
G. (2005). Choice and preference assess- ment research with people with severe to profound developmental disabilities: A review of the literature. Research in Developmental Disabilities, 26, 1-15. http://dx.doi.Org/10.1016/j.ridd.2004.01 .006
DeLeon, I. G., & Iwata, B. A. (1996). Evalu- ation of a multiple-stimulus presentation format for assessing reinforcer prefer- encers. Joumal of Applied Behavior Analysis, 29, 519-533.
Fisher, W., Piazza, C. C , Bowman, L. G., Hagopian, L. P., Owens, J. G., & Slevin, I. (1992). A comparison of two approaches for identifying reinforcers for persons with severe and profound disabilities. Joumal of Applied Behavior Analysis, 25, 491-498. http://dx.doi.org/10.1901 /iaba. 1992.25-491
Fleming, G. V., Wheeler, G. M., Gannella- Malone, H. I., Basbagill, A. R., Ghung, Y. G., & Graham Day, K. (2010). An evalua- tion of the use of eye gaze to measure preference of individuals with severe physical and developmental disabilities. Developmental Neurorehabilitation, 13, 266-275. http://dx.doi.org/10.3109/175 18421003705706
Pace, G. M., Ivancic, M. T., Edwards, G. L., Iwata, B. A., & Page, T. J. (1985). Assess- ment of stimulus preference and rein- force value with profoundly retarded individuals. Joumal of Applied Behnvior Analysis, 18, 249-255. http://dx.doi.org /10.1901/jaba.l985.18-249
Tlillis, G. A., Gannella-Malone, H. I., Bas- bigill, A. R., Yeager, A., Fleming, G. V., Payne, D., & Wu, P. (2011). Review of the choice and preference assessment litera- ture for individuals with severe to pro- found disabilities. Education and Train- ing in Autism and Developmental Disa- bilities, 46, 576-595.
Helen I. Cannella-Malone (Ohio CEC), Associate Professor; Linsey M. Sabielny (Ohio CEC), Doctoral Student: Eliseo D. Jimenez (Ohio CEC), Doctoral Student: and Megan M. Miller (Ohio CEC), Doctoral Student, Department of Educational Studies, The Ohio State University, Columbus.
Address correspondence concerning this article to Helen I. Cannella-Malone, Depart- ment of Educational Studies, A324 PAES Building, 305 W. 17th Avenue, The Ohio State University, Columbus OH 43210 (e-mail: malone. 175@osu.edu).
TEAGHING Exceptional Ghildren, Vol. 45, No. 6, pp. 16-23.
Copyright 2013 CEC.
TEACHING EXCEPTIONAL CHILDREN JULY/AUG 2013 23
Copyright of Teaching Exceptional Children is the property of Council for Exceptional Children and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.