<p>Back Ground Classroom processes can be thought of as communication processes. In everyday classroom, communication between teachers and children plays an important role. For the past two decades, researchers on classroom processes, specifically on teacher-student communication processes have investigated the relationship between teacher behavior and student achievement. These researchers have attempted to determine what teaching behaviors are most effective for enhancing the achievment of students. On the basis of this research, Gage (1978), Good (1979) and Rosenshine (1979) have suggested that direct instruction is the most effective approach for facilitating the mathematics achievement and reading achievement of elementary school students. In contrast, research using an Aptitude-Treatment interaction approach has suggested that the effectiveness of direct instruction depends on differences in student's aptitudes. (Peterson, 1979; Peterson & Janichi, 1979, Janichi & Peterson, 1981) Furthermore, Good (1979) and other researchers remarked that the series of these researches have neglected the effect on children's attitude and suggested that there exists a necessity to examine this problem in wider framework. Purpose This paper examined the effect of teaching behavior or children's cognitive and affective components in terms of their aptitude and the treatment they were given. The following hypotheses were to be tested. Hypothesis 1 : Children of external control show more positive cognition toward dircet instructional behavior than children of internal control do. Hypothesis 2 : Children of external control show more positive affect toward direct instructional behavior than children of internal control do. Method To test the hypotheses above, an experimental research was conducted on both direct instruction and open classroom. In as much as open classroom is considered to be the opposite curricular mode to direct instruction, it is predicted that the result will be contrary to the hypotheses, and thus giving more concrete evidence for the hypotheses. (direct instruction) 1) Subjects and date of administration Thirty-nine third-grade students (21 males and 18 females) in an elementary school in Yokohama city. November 9th, 1983. 2) Procedure The teacher was given enough explanation about the experiment before it being conducted. She was given the manual along with the instructions to read it, and several meetings were held with her to respond to questions that she had about the meaning of certain teaching behaviors and to react to any difficulties that she might have. In the experiment she covered two units in twenty hours altogether from the end of September to the beginning of November. 3) Matters to be measured (1)Sex (2) Number of Juku school attending (3) Locus of control (4) Cognitive component of attitude (5) Affective component of attitude (open classroom) 1) Subjects and date of administration The same subjects were used as in the experiment on direct instruction. November 24th, 1983. 2) Procedure Personal learning was employed, each student determining his/her own goal in accordance with his/her ability and attaining it without any instruction by the teacher. The unit was to be covered in six hours. 3) Matters to be measured The same matters were measured as in the experiment on direct instruction. Results and Discussion Contrary to the expectation, both hypothesis 1 and 2 , as a whole, were not supported enough by the results of this experiment. Hypothesis 1 : In the case of direct instruction, more of the students showed positive cognition to the teachers behavior regardless of locus of control, while more of them showed negative cognition when treated with the open classroom. That the students at their age tend to consider teacher's direct instruction more preferable or necessary could be a possible explanation of the result.</p><p>(View PDF for the rest of the abstract.)</p>