Classroom Observations of Child Aggressive and Disruptive Behaviors-What Do We See?

Aggressive and disruptive behaviors, as early as first grade, are confirmed antecedents for social and behavioral problems later in life (Kellam et al., 2008).  The Good Behavior Game (GBG), a classroom-based behavior management strategy (Barrish et al., 1969) is one of the few preventive interventions that are found to reduce early disruptive and aggressive behaviors. Accurately measuring child behavior is essential in determining the effectiveness of preventive interventions. Previous trials testing the effectiveness of GBG alone or in combination with other components (Ialongo et al., 1999) used a focal child approach, in which each child in a class was observed for a short, discrete time interval. At a given time point, only one child was observed; during an entire observation, each child was observed for 2-6 minutes depending on the size of the classroom and the length of the observation. This approach may fail to produce an adequate description of child behaviors since only the behavior of the focal child is recorded at any particular time point. To more accurately capture behaviors, we developed a new classroom observation approach that uses time sampling where all children are continuously observed minute by minute. In this paper we report on preliminary results obtained with data collected using the new observation protocol.

 1,027 children (girls: 48.8%; Hispanics: 75.0%) from 54 first-grade classrooms were observed 6 times during the year. With a few exceptions, each classroom was observed three times in the morning and three times in the afternoon. Length of observations varied but most observations were 60 minutes or longer. Every classroom was observed by two observers, each observing half the students (maximum 15). Each minute, instances of off-task, disruptive, verbal or physically aggressive behaviors and the child exhibiting the behavior were recorded.  Preliminary examination of the instrument suggests a high level of inter-rater agreement. Preliminary analyses suggest that about half of all children (morning: 57.1% of boys and 44.0% of girls; afternoon: 60.4% of boys and 43.1% of girls) were observed to be off-task at least once. Almost a third of boys were observed to have disruptive behaviors. Boys were more likely to exhibit off-task and disruptive behaviors than girls. Boys tended to exhibit more off-task and disruptive behaviors in the afternoon than earlier during the school day. In addition, children of Hispanic background were observed with less frequent off-task behaviors than non-Hispanic children while no difference in terms of disruptive behaviors was found. We will also examine whether the length of observations affected the recorded occurrence of child behaviors.  Implications for use will be discussed.