Research to Practice Summit

Engagement and Integration in Group Settings ¹

by Robin McWilliam

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

Findings and Implications: Engagement     Integrated Therapy

A number of studies have been conducted in inclusive settings, without actually investigating the effects of inclusion. This research adds to our knowledge about what features of inclusive environments are effective. Two aspects of group settings I have studied are engagement levels of children and how to provide specialized services to children.

Engagement is defined as the amount of time children spend interacting with the environment at different levels of competence. Engagement, therefore, has to do with both amount (i.e., the percentage of time spent) and level (i.e., the quality of how that time is spent). Researchers such as Greenwood and Carta at Juniper Gardens in Kansas and Sam Odom at Washington, Vanderbilt, and North Carolina have also studied engagement. Our work is complementary in that we see engagement as an index of the quality of children's environments, but I shall concentrate on my own work because it's somewhat more familiar territory. Although most of the research has involved observational methods, we have also used a summated rating scale and qualitative methods.

Integrated therapy and special education is getting increasing attention. This research has focused on the relative merits of providing specialized services to children in group settings in different ways. Our research has included the development of a continuum from "one-on-one pull-out" to "individualized within routines" and "pure consultation." We have sought to examine practitioners current practices, their perceptions of ideal practices, how children fare with different models, the extent to which teachers carry out interventions when therapists are not around, and so on. We have used multiple methods to understand service delivery-from single-subject methodology to survey to group designs.

List of Research Findings and their Implications for Inclusion

Engagement

• Global and observed engagement have been found to be fundamentally different, except for undifferentiated engagement, which was captured by both ratings and observational codes (McWilliam & Scarborough, 1997). This means that engagement is probably both a child characteristic, like temperament, and a class of observable behavior, like play levels. Teachers and parents would be well-advised to have some concept of both global and observed engagement when designing programs for their children.
• Global engagement consists of four factors: Competence, Social Engagement, Undifferentiated, and Attention (McWilliam, Snyder, & Lawson, 1994). We found that parents and professionals largely agreed on ratings. This means that ratings of children's global engagement are stable and could provide potentially useful alternative to temperament measures.
• Observations of child engagement have shown that raters accounted for less than 2% of the variance in the error of scores, whereas sessions accounted for most of the variance (McWilliam & Ware, 1994). This means that, at least for research purposes, observers should watch children eight times to get dependable estimates of their levels and types of engagement.
• The percentages of children engaged during frequent observations differed as a function of program type (focusing on criterion behaviors versus focusing on engagement) and classroom activities (group versus circle versus free play versus meals; McWilliam, Trivette, & Dunst, 1985). This means that programs should be specifically organized to foster engagement, especially to achieve more engagement in group activities and free play (i.e., engagement in circle and meals did not differ between program types).
• Children with disabilities have been found to spend less time interactively engaged with adults, attentionally engaged with peers, and in mastery-level engagement with materials than did children without disabilities; and they spent more time passively nonengaged (McWilliam & Bailey, 1995). This study also showed that there were differences in engagement between children with and children without disabilities even when controlling for developmental age. This means that there is something about the nature of disability that affects child engagement. The implication for practice is that a deliberate effort should be made to address engagement when working with children with disabilities.

Integrated Therapy

• The choice of whether therapists and consulting special educators use in-class versus out-of-class models of service delivery is based on
1. the discipline they represent,
2. the specific goals and techniques to be addressed,
3. the family's preferences (hypothetically),
4. and the child's characteristics-in this order (McWilliam & Bailey, 1994).

This means we should recognize that different team members are likely to have different views about the role of therapy and we should give families informed choices about service delivery models.

• Concepts of "needing" therapy have been found to be muddled, and most people say that "more is better" as long as the quality is high enough (McWilliam, Young, & Harville, 1996). This means we should decide on specialized services after taking the context into consideration (i.e., stop saying children need therapy based purely on their diagnosis) and we should understand the limitations of direct hands-on intervention that occurs in the absence of a caregiver to follow through between therapy sessions.
<
• A 6-model continuum of service delivery models exists: one-on-one pull out, small-group pull-out, one-on-one in classroom, group activity, individualized within routines, and pure consultation (McWilliam, 1995). This means that we can gradually move therapists (or children) along the continuum, with a realistic goal of serving most children through the individualized-within-routines model.

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).