Linking Instructional Supervision to Teacher Performance in the Context of School Management

Abstract

This study investigates the effects of instructional supervision on teacher performance in public primary schools in Pulaupanggung District, Tanggamus Regency. Instructional supervision is conceptualized as a multidimensional process comprising planning, implementation, and follow-up activities aimed at improving instructional quality. Using a quantitative, non-experimental ex post facto design, data were collected from 75 teachers selected through cluster random sampling across 32 schools. A structured questionnaire with a four-point Likert scale was employed to measure instructional supervision and teacher performance in lesson planning, instructional implementation, and learning evaluation. Data were analyzed using simple linear regression techniques. The results indicate that instructional supervision planning does not have a significant effect on teacher performance across all instructional dimensions. In contrast, the implementation of instructional supervision shows a significant influence on teacher performance in lesson planning and learning evaluation, while follow-up supervision demonstrates a significant effect specifically on learning evaluation. However, neither implementation nor follow-up supervision significantly affects instructional implementation. These findings suggest that the effectiveness of instructional supervision depends more on direct supervisory engagement and continuous feedback than on formal planning procedures. The study contributes to the literature by integrating perspectives from educational administration and performance management, emphasizing the importance of supervision as a managerial–pedagogical mechanism. Practically, the results highlight the need for school leaders to prioritize meaningful supervision implementation and systematic follow-up to enhance teacher performance. Despite its contributions, the study is limited by its correlational design and reliance on self-reported data, indicating opportunities for future research using mixed-methods approaches.