About fifteen percent of all children show mild behavioral problems, and about seven percent show moderate to severe disorders (Cotler, 1986). However, not enough has been done to treat or prevent these problems (Levine & Perkins, 1987). Even less has been done to evaluate existing treatment programs (Felner et al., 1983).
Historically, evaluations of special education programs have examined deficits in student performance or capability, focusing on organismic, cognitive, or behavioral problems such as brain damage, perception difficulties, learning disabilities, and lack of motivation (Greenwood & Carta, 1987). Although these individual-level factors are important, they may lead to premature labeling and victimization. For example, young children have been disproportionately labeled as socially, cognitively, or behaviorally "deficient" (Drabman, Tarnowski, & Kelly, 1987). Hence, teachers may interact with students as if they were a bundle of deficiencies (labels), rather than considering settings and systems as part of the problem or solution.
Further, children who live in economically and educationally impoverished environments are at higher risk for a number of negative educational and social-emotional outcomes (Brand, Dubois, & Felner, 1990). Thus, ecological variables need to be investigated because they can either influence the onset of behavioral problems (Heller, 1990) or positively affect learning behaviors (Greenwood & Carta, 1987).
The interaction of academic materials (static features of the environment) with students and teachers (dynamic features of the environment) forms classroom ecologies. In an effort to describe these environments, the ecobehavioral approach to program evaluation has emerged in the field of applied behavior analysis. This approach links ecological factors, which may facilitate or hinder students' academic performance, with program outcomes (Greenwood & Carta, 1987). Thus, the ecobehavioral approach is useful for providing important information about the classroom climate and the effects of interventions (Greenwood et al., 1985).
One indication of positive effects of special education programs is that students engage more often in academic-related behaviors than in disruptive ones. For example, there is evidence that active classroom behaviors (e.g., asking or answering questions, reading, writing) are correlates of academic achievement, whereas disruptive behaviors are not (Rosenshine & Stevens, 1986). As Becker (1977) states, "at risk" students need instruction that enables them to perform academically at accelerated rates. In fact, these students need to learn more and faster just to obtain achievement comparable to more advantaged students. Therefore, the purpose of the present study was to examine ecological and student behavior variables that may be linked to the objectives of the program--increasing appropriate classroom behaviors and decreasing inappropriate ones. Here, the ecobehavioral approach was modified in order to describe patterns in the whole classroom rather than patterns in one student, as originally proposed by Greenwood and Carta (1987).
Setting and Program Description
The setting was a private residential treatment center located in a large midwestern city. The center serves behaviorally disturbed adolescent males between the ages of 12 and 18 years. For those whose educational needs cannot be met in a regular school setting, the center offers a special education program. This program aims to alleviate behavioral difficulties that impede academic performance, such as chronic task incompletion, acting out, behavior problems, or social interaction conflicts. Two important objectives are to increase academic-related performance and decrease disruptive behaviors in the classroom.
At the time the study was conducted, there were 84 emotionally disturbed adolescent males in the special education program. …