Bridging the Divide: Unpacking the Perception-Practice Gap in Blended Learning Implementation among Natural Science Educators in Ethiopian Applied Universities

Blended learning has emerged as a transformative pedagogical approach in higher education, offering potential solutions to resource constraints and enhancing educational quality. Ethiopia's higher education sector has witnessed significant policy developments, including the e-Learning for Strengthening Higher Education (e-SHE) program and the National e-Learning Policy, aimed at integrating blended learning across all public universities. However, despite policy support and positive educator perceptions, implementation remains inconsistent, revealing a perception-practice gap that is poorly understood, particularly in natural science education within applied universities. This study examined the perception-practice gap in blended learning implementation among natural science educators in Ethiopian applied universities, investigating educators' perceptions, current practices, contributing factors to the gap, and strategies for bridging it. A convergent mixed-methods design was employed, integrating quantitative survey data from 387 natural science educators across four Ethiopian applied universities and qualitative data from 48 semi-structured interviews. Quantitative data were analyzed using descriptive and inferential statistics, while qualitative data were analyzed using thematic analysis following Braun and Clarke's six-phase framework. Findings revealed a substantial perception-practice gap, with educators holding strongly positive perceptions of blended learning (perception means ranging from 3.91 to 4.21) yet actual practices remaining restricted to basic functions such as material sharing and communication (practice means ranging from 1.76 to 3.87). The largest gap was observed in laboratory instruction (mean difference = 1.79, d = 1.56). Multiple barriers contributed to this gap: infrastructural (unreliable internet connectivity, M = 4.56), institutional (insufficient training, M = 4.45), pedagogical (weak conceptual clarity, M = 4.02), and individual (workload concerns, M = 4.15).