Direct instruction in fourth and fifth-grade classrooms

Herbert Alan Sloan, Purdue University


The relationship was examined between teacher behaviors in fourth and fifth grade classrooms associated with the Direct Instruction Model and student achievement in mathematics, reading, science, and social studies; student self-concept; and climate. The first stage focused on teacher implementation of questioning and instructional behaviors following direct instruction training where the experimental unit was the teacher. Two observation instruments were developed by the author to measure levels of teacher instructional and questioning behaviors associated with the Direct Instruction Model. Two null hypotheses tested for teacher implementation of these behaviors. In the second stage, with the class as the experimental unit, the effect of teacher behaviors on student outcomes was tested by three null hypotheses. The design of this experiment was a pretest and posttest control group model with random assignment of fourth and fifth grade teachers in the South Bend Community School Corporation, by school units, to treatments. Five teachers were assigned to the control group, and five teachers were assigned to the experimental group. The greatest limitation noted in this study was the reliance on teacher volunteers which resulted in a small control and experimental group sample size in the first stage of the experiment and correspondingly large standard deviations in the pre and posttest means of the two groups. The analyses of data were done with a one-way analysis of covariance and the Pearson Product-Moment Correlation Coefficient. The.05 level of significance was specified for this study. Tests of Hypotheses 1 and 2 did not support the idea that following training, teachers in experimental group successfully demonstrated Direct Instruction behaviors to a greater degree than teachers in the control group. Tests performed for Hypotheses 3, 4, and 5 did support significant gains in academic achievement, self-concept, and classroom climate in experimental classrooms. Upon closer examination of operational hypotheses following each of the five hypotheses, patterns emerged that supported, to a limited degree, a linkage between questioning behaviors related to the Direct Instruction Model and student outcomes in the areas of self-esteem and classroom climate. Results from this study supported the efficacy of developing new, reliable, and valid teacher observation instruments, but at the same time called into question reliance on the teacher training methods that were used.




Kline, Purdue University.

Subject Area

School administration|Education

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server