A MODEL FOR ENHANCING ENGLISH LANGUAGE COMPETENCY AS MEDIUM INSTRUCTION FOR STUDENT TEACHERS: A CASE STUDY OF STUDENTS MAJORING IN DIFFERENT DISCIPLINES

Abstract

This study aimed to 1) design and develop a model to enhance the competence of student teachers of non-English majors in using English for instruction, and 2) test the effectiveness of the model. The research was conducted in two phases. The first involved the creation and development of the competence enhancement model, and the second tested its effectiveness with student teachers of non-English subjects. The pilot study involved a sample group of student teachers from the Faculty of Education at Pibulsongkram Rajabhat University. Statistics used include mean, standard deviation, content analysis, and t-test dependent. The findings revealed that: 1. The developed model comprised five components: 1) the model’s principles, 2) the model’s objectives, 3) the model’s content, 4) the learning process, and 5) assessment and evaluation. The quality evaluation of the model indicated high effectiveness across four aspects: 1) utility, 2) feasibility, 3) appropriateness, and 4) accuracy. Overall, the model was rated at a high-quality level, with mean scores of 4.31 and a standard deviation of .193. 2. The model implementation led to significant improvements in participants’ competence in using English for instruction. Post-training knowledge score averaged 14.05, showing a statistically significant difference at the 0.05 level. During the teaching practicum stage, the overall mean score across all subcomponents was 2.10, and in the instrument development stage, the overall mean score across all subcomponents was 1.83. The English language competency for instruction was evaluated twice in the supervision stage. In the first supervision, based on video-recorded teaching, the mean score was 2.04; in the second supervision, conducted at the school site, the mean score increased to 2.33. Regarding the participant’s self-perceived competence in using English for instruction, the mean scores and standard deviation after training were higher than those before training across all components. Furthermore, the study of the After-Action Review (AAR) showed improvement at every step of implementing the model.