| This study examined two assumptions underlying school reform. The first assumption is that classroom practices do change, and the second is that changes in educational practices will lead to improved student learning and achievement. Both assumptions have been called into question based on the works of Cuban (1993) and Coleman et al. (1966).;Before the two assumptions can be tested, a reliable measure of classroom practices must be used. This study begins by examining the reliability of the School Observation Measure (SOM) using the framework of generalizability theory. Results indicated that the SOM is reliable, and greater numbers of SOMs conducted at schools increased the reliability of the SOM ratings. A range of reliability estimates was provided when specified numbers of SOMs were completed at schools.;The study also described classroom practices currently in use at schools, based on approximately 1,100 SOMs conducted in 137 schools in the southeast United States. Schools were categorized by elementary versus high school, and rural versus urban status. The most frequently used classroom practices included direct instruction, independent seatwork, higher-level instructional feedback, teacher as coach, multi-age grouping (high schools), ability groups, higher-level questioning, workcenters (elementary schools), cooperative learning (elementary schools), and experiential, hands-on learning. The frequency of these practices was practically the same for both elementary and high schools. Examination of the differences between school location and level indicated significant differences between high schools and elementary schools, rural and urban schools, and significant differences for the two-way interaction term.;The effects of school size, student-teacher ratio, and socioeconomic status were also analyzed. Results at elementary schools indicated that school size was negatively related to multi-age grouping, workcenters, and student discussion. Student-teacher ratio was positively related to team teaching, cooperative learning, multi-age grouping, and others. Degree of poverty was negatively related to project-based learning, experiential, hands-on learning, individual instruction, sustained reading, the computer for instructional delivery, and student attention.;Finally, the relationship between classroom practices and student achievement was examined. Regressions using classroom practices and student achievement at the school level indicated that project-based learning and student self-assessment were positively linked to student achievement. |
