The goal of any behavioral program is to facilitate lasting change. A significant criticism of behavioral programs is that they work in the clinical setting but do not generalize once the clinical program is stopped. The authors suggest that behavioral programs often do not generalize because clinicians fail to plan for generalization to occur (Maag, 2004). A technology for planning and facilitating generalization has been in existence since the 1970's (Stokes and Baer, 1977, Stokes and Osnes, 1986 and Stokes and Osnes, 1989). The authors have created a form to prompt clinicians and researchers to systematically plan for generalization as part of their behavioral programming. A case study is given to demonstrate how the form can be used to increase the probability of behavioral changes transferring to other settings and maintaining over time. The Generalization Analysis Worksheet is designed to assure that clinicians and researchers program for generalization as part of any behavioral program they design. If the technology suggested by Stokes and Baer, 1977, Stokes and Osnes, 1986 and Stokes and Osnes, 1989 is routinely programmed into behavioral programs, behaviorists may finally be able to answer the criticism that behavioral programs do not generalize.
Children's behavior and academic achievement are closely linked (Algozzine, Wang, & Violette, 2010), children whose behavior interferes with their learning and social development are one of the most frequent referrals to school specialists (Bramlett, Murphy, Johnson, Wallingsford, & Hall, 2002; Chalfant & Van Dusen Pysh, 1989). Often these children are referred to a behavioral specialist, whose job is to help the classroom teacher create a behavioral management plan. The first step in this process is to help the child, through behavioral change techniques, to stop an interfering behavior and to replace it with a more appropriate behavior. The next step is to help the child to generalize the behavior, which was learned in a specific time and place, across settings and time. This paper focuses on this second step, which is often overlooked (Rutherford & Nelson, 1988), and presents strategies to generalize behavior through the use of the "Generalization Analysis Worksheet" (Appendix A). Generalization processes must be planned from the start of any behavior plan (Miltenberger, 2008; Maag, 2004), and unless they are, the initial success in changing behavior is often not maintained (Schloss & Smith, 1998). One of the most comprehensive models for facilitating the generalization of behavior was put forth by Stokes and Osnes (1989), who refined the generalization classification of Stokes and Baer (1977) and Stokes and Osnes (1986). Stokes and Osnes (1989) identified 12 strategies that behavioral research has shown to promote generalization. To prompt the use of these strategies we have operationalized them through the development of a worksheet that helps prompt behavioral change. To facilitate long-term behavioral change (generalization), we propose the use of this Generalization Analysis Worksheet. The first part of the paper describes each of the strategies and how they relate to the generalization of behavioral change in the schools, and the second part of the paper presents a case example that uses the worksheet.
Stokes and Osnes (1989) presented a comprehensive model of generalization that included 12 strategies that support generalization. These strategies are as follows:
1. Use the normal consequences or rewards that already exist in the child's environment. One needs to directly teach behaviors that are likely to come into contact with reinforcers (consequences) in the school or home. These are naturally occurring reinforcers that do not need to be programmed and include teacher attention, peer attention, and reward systems already in place. These can be used to generalize behavior if the behavior is similar to that which normally triggers the natural reinforcers. …