Research to Practice Summit

Embedding Prompting Strategies in Inclusive Preschool Classes ¹

Mark Wolery

• Investigator and Fellow
• Frank Porter Graham Child Development Center
• University of North Carolina at Chapel Hill

Although inclusive/mainstreamed and segregated programs produce similar developmental outcomes for young children with disabilities, inclusive programs result in more social and behavioral skills (Buysse & Bailey,1993). A large literature is available on inclusive preschool services (Peck, Odom, & Bricker, 1993; Wolery & Wilbers, 1994), but a recurring finding emerges: without specific intervention, many young children with disabilities in inclusive classes do not learn and/or use the skills that are seen as critical for their success (e.g., specific language forms, peer imitation, conversations with peers, social interactions with peers).

To implement successful instruction in inclusive classes, teachers must know four things:

1. what to teach,
2. what instructional strategies to use,
3. when to teach, and
4. whether their teaching is working.

They also must act on that knowledge.

Research across these four areas is uneven, and our knowledge is incomplete. In terms of knowing what to teach, the literature includes many recommendations about assessment practices, involving families in assessment activities, and assessment models. It contains much less information on how staff in inclusive programs actually determine what to teach. In terms of knowing when to teach, some research exists for selected goals-particularly language goals-and recommendations abound about when children are most receptive to teaching interactions. Much less knowledge exists on how early childhood teachers actually select when to engage children in learning opportunities. In terms of monitoring the effects of teaching, much is known about how it should be done, but less information exists on how usual programs monitor and adjust their programs based on the result of that monitoring In terms of knowing what practices to use, a great deal of research has occurred evaluating specific instructional strategies and practices. A segment of this research is addressed in this presentation.

Many of the studies evaluating instructional strategies and practices for young children with disabilities occurred in actual classroom settings, often in inclusive classes. Some of the research used classroom teachers as the persons who implemented the strategies, and some of the studies focused on children's individual IEP goals. Nearly all of the studies used single-subject experimental designs, direct observation of child behavior, and careful implementation of procedures. Most produced unambiguous results. As a limitation, the studies often focused on a narrow range of child outcomes and used only part-day implementation.

The following summary focuses on one type of intervention (teaching) strategies; specifically, response prompting procedures. These procedures originate from behavioral, direct instruction foundations. They involve the teacher providing instructional opportunities to children (individually or in groups), giving adult assistance, fading or removing that assistance, and delivering differential consequences based on children's responses. A number of strategies are subsumed under this broad category, including time delay, system of least prompts, simultaneous prompting, graduated guidance, most-to-least prompting, and naturalistic teaching procedures such as incidental teaching and the mand-model procedure. The summary statements below, exclude the naturalistic teaching procedures (Wolery, Ault, & Doyle, 1992).

1. The time delay procedures and the system of least prompts have been studied more extensively than simultaneous prompting, graduated guidance, and most-to-least prompting with preschoolers who have disabilities.
2. With accurate implementation, these response prompting procedures almost always result in learning-occasionally with individual modifications for selected children. To date, no disability by procedure interactions have been found; the procedures tend to be effective across children with different characteristics-given procedure-specific prerequisite behaviors exist.
3. The range of behaviors taught with response prompting procedures include discrete tasks (e.g., saying specific words, naming pictures, etc.), chained tasks (e.g., using a spoon, putting on a coat), and more socially oriented tasks (e.g., carrying on conversations with peers, peer imitation). The successfully instructed skills are found for the following domains: cognitive (play, and pre-academic skills), communication/language, motor, self-care, and adaptive.
4. In the studies, the instructional strategies have been implemented by research staff, teachers, teaching assistants, and therapists with a high degree of fidelity.
5. Some response prompting procedures (e.g., time delay) result in more rapid learning and more errorless learning than other procedures (e.g., most-to-least prompting, system of least prompts); however, the full range of comparisons have not been made.
6. For at least discrete responses, the procedures can be implemented in a range of ways without decrements in effectiveness: (a) one-on-one massed trial sessions, (b) dyads of children in massed trial sessions, (c) small groups of massed trial sessions with individual responding, (d) individually to children distributed throughout a given activity (e.g., circle time), (e) individually to children distributed throughout the day, and (f) individually or to groups of children during in-class transiions.
7. Consistently adding extra information (called instructive feedback) during the consequences of response prompting trials can produce additional learning and more rapid learning during later direct instruction.
8. For the time delay procedure with discrete tasks, sloppy implementation on the duration of the delay interval does not produce decrements in effectiveness or efficiency, but sloppy implementation of prompt delivery results in slower learning for some children and ineffective instruction for others.
9. Teaching with time delay can result in less frequent problem behaviors-when those problem behaviors are motivated by escape from difficult tasks.

In summary, these procedures are effective, often are more efficient than other procedures, are generalizable (useful in different settings, across children, and across skills), and are flexible can be used by variety of persons in a variety arrangements. Thus, they should be used widely.

References

Peck, C. A., Odom, S. L., & Bricker, D. D. (Eds.), Integrating young children with disabilities into community programs: Ecological perspectives on research and implementation. Baltimore: Paul Brookes.

Wolery, M., Ault, M. J., & Doyle, P. M. (1992). Teaching students with moderate and severe disabilities: Use of response prompting strategies. White Plains, NY: Longman.

Wolery, M., & Wilbers, J. S. (1994). Including children with special needs in early childhood programs. Washington, DC: National Association for the Education of Young Children.

¹ Presented at the Research to Practice Summit, July 30-31, 1998, Washington, D.C., sponsored by the National Early Childhood Technical Assistance System in collaboration with the Early Childhood Research Institute on Inclusion and the U.S. Department of Education's Office of Education Research and Improvement (OERI).