Representation of Confirmation Bias in Geometry Learning in Islamic Boarding School Environment: Teachers' Perspective of Mathematics Learning at MTs Sarji Ar-Rasyid Situbondo

This study aims to uncover the representation of confirmation bias in geometry learning within Islamic boarding school (pesantren) environments, particularly from the perspective of mathematics teachers at MTs Sarji Ar-Rasyid Situbondo. Geometry learning demands adequate visualization skills, conceptual understanding, and mathematical representation. However, instructional practices in pesantren settings remain predominantly teacher-centered, relying heavily on conventional lecture-based methods. Although various innovative approaches have been widely recommended in mathematics education literature, their implementation in pesantren environments remains considerably limited. This gap motivated the present study to examine how teachers' beliefs in the effectiveness of conventional methods can give rise to confirmation bias in instructional decision- making. This study employed a descriptive qualitative approach, with data collected through classroom observations and semi-structured interviews with two mathematics teachers and several students. Data analysis was conducted using the Miles and Huberman model, encompassing the stages of data reduction, data display, and conclusion drawing. The findings reveal that teachers tend to sustain lecture-based methods, perceiving them as the most effective approach based on prior teaching experience, environmental conditions, and students' habitual learning patterns in the pesantren. The most significant finding demonstrates that confirmation bias operates through teachers' cognitive frameworks, in which they interpret students' limited comprehension, diminished focus, and learning fatigue as consequences of internal student factors or pesantren conditions, rather than as potential indicators of weaknesses inherent in the lecture method itself. Although teachers are familiar with technology-based approaches such as GeoGebra, their implementation remains suboptimal due to limited facilities, deeply ingrained teaching habits, and concerns regarding their practical effectiveness. These findings affirm that confirmation bias substantially influences how teachers maintain conventional practices while simultaneously dismissing more interactive learning alternatives. The implications of this study underscore the critical importance of professional development programs oriented toward deep pedagogical reflection and openness to instructional innovation, in order to enhance the quality of geometry learning within pesantren environments.