As part of the development of a classroom interaction observation instrument, this report presents the results of a comparative study of group interaction in contrasting educational environments (open and traditional) with children of different socioeconomic backgrounds (middle and lower classes). Additionally, problems of adequately assessing children's classroom interactions are explored. Background and rationale for the study are provided in the introduction. In Chapter I the Differentiated Child Behavior (DCB) System, a classroom observation instrument consisting of six major behavioral categories and 50 subcategories, is described. Classroom observation procedures are outlined. Chapter II describes the study. The primary component of the sample was a group of nine Open Lower classrooms in inner-city public schools. The DCB was administered three times to students in each of the 17 classrooms. Simultaneous observations by paired observers were conducted to assess reliability of the DCB. A series of teacher observations in 12 of the classrooms was conducted. Results are presented and discussed in Chapter III. Among the results, children in Open Lower classrooms asked three times as many higher level questions as did children in the Traditional Middle classrooms. Chapter IV explores the relation between DCB scores and the assessments of teacher behavior in terms of construct validity. A summary is provided and implications are indicated in Chapter V. (Author/RH)