Abstract
Although some theoretical models for bilingual language learning exist, they are not adequate to explain events that take place in a classroom, nor aid teachers in curriculum development. A four-dimensional pedagogical model for transitional English bilingual classrooms developed by Lara-Alecio & Parker (I 994) attempts to explain these areas that are lacking in other models. As a first step toward empirically validating the model, an instrument for measuring model variables was developed. The purpose of this study was to develop a reliable observation instrument to adequately code classroom events and structures according to this pedagogical model. Hispanic students, exiting the fifth grade from a large urban school district, were observed in classroom activities within a summer school program. This program, funded by a Title VII federal grant, aimed to teach mathematic skills while transitioning students from Spanish to English instruction. To develop the instrument, two researchers gathered data from classrooms and created a tool that coded all classroom activities according to operational definitions of each model dimension. After a period of piloting and revising, a formal test of interrater reliability was conducted. Using Cohen's Kappa to determine interrater agreement, the formal reliability test scored above 0.90 in all categories except Language of Instruction and Curriculum Area, which were unable to be calculated. Although limited by small sampling of calssrooms and instructional time, this first study indicates optimism toward use of this pedagogical model for classroom observation.
Bruce, Karen Leigh (1995). Measuring bilingual classroom activities according to a four-dimensional model. Master's thesis, Texas A&M University. Available electronically from
https : / /hdl .handle .net /1969 .1 /ETD -TAMU -1995 -THESIS -B78.