IntroductionOne feature of previous research on teaching is that the vast majority of studies have been conducted in a more formal setting in which the teacher is the dominant actor. Rosenshine and Furst (1971) believe this research to be "chaotic, unorganized and selfserving." They argue that curriculum packages should be used more extensively in descriptive, correlational, and experimental studies of teaching because such packages offer some degree of control over the teaching -learning situation. With the increased availability of self-paced or "individualized" science curricula such as the Australian Science Education Project (ASEP) and ISCS, the need for systematic studies of teaching in classes using self-paced materials becomes even more pressing. At this point, we can only speculate about the nature and effects of different patterns of teacher and student behavior in "individualized" classroom settings. In this study, an attempt was made to (a) describe the nature of teacher-pupil-materials interactions occurring in classrooms using Australian Science Education Project (ASEP) materials, and (b) study relationships between these interaction patterns, changes in student achievement, and attitudes toward science in classes taught by teachers whose educational values, experience, and training differed.
Theoretical RationaleTheoretical positions (Bennis, 1966;Giaquinta, 1973) regarding the process of innovation in school systems would lead one to anticipate widespread adoption of State-supported, self-paced science curriculum materials in Australian schools, and for such a material-based innovation to retain substantial integrity of original meaning when used by teachers in different situations. Self-paced materials demand that the classroom game be played a particular way; students are constrained to work independently at their own pace through a more or less structured and sequenced set of experiments and activities, and the teacher is constrained to act as manager of materials and groups and as a guide and resource person. However, as Rogers and Shoemaker (1971) point out in their review of diffusion studies, the 489