The sudden transition to remote learning in response to COVID-19: lessons from Malaysia

The paper addresses how Universiti Teknologi MARA (UiTM) students accepted and experienced the rapid transition from in-person to remote learning during COVID-19. It situates remote learning within synchronous and asynchronous modalities and highlights the Malaysian context where the Movement Control Order (MCO) forced universities to move online. Prior to the pandemic, UiTM used blended learning with an LMS for content and assessment, and students routinely evaluated lecturers. The study identifies a gap in Malaysian higher education research linking learning activities (LA) with lecturer professionalism (LP), course impression (CI), and facilitating conditions (FC). The research question explores the extent to which LP, CI, and FC influence students’ enjoyment of remote LA pre- and mid-pandemic, and tests a model of remote learning acceptance.

The study builds on literature indicating that lecturer professionalism—encompassing competence, support, IT skills, and attitudes—enhances student satisfaction and remote learning success (Vayre & Vonthron, 2017; Wiranto & Slameto, 2021). Infrastructure and facilitating conditions (devices, connectivity, tools, and support) are linked to online engagement and effectiveness (Teo et al., 2020; Roman & Plopeanu, 2021). Course impression reflects perceived knowledge gains, relevance, assessment design, and confidence (Almaiah et al., 2016; Yuan et al., 2021). Pandemic-era constraints, such as poor connectivity and family environments, further shape remote learning experiences (Kamble et al., 2021; Maqsood et al., 2021). The authors propose five hypotheses: H1 LP→CI, H2 LP→FC, H3 LP→LA, H4 CI→LA, and H5 FC→LA. The study model adapts Danielson’s (2007) framework for teaching, adding CI and FC to capture Malaysian higher education context.

Design: Quantitative study using a validated institutional teaching evaluation instrument operationalizing the proposed model; the model has informed UiTM policy since 2010. Instrument: Online evaluation form with 24 items on a 4-point forced Likert scale (1=strongly disagree to 4=strongly agree): Section A CI (4 items), Section B LP (7 items), Section C LA (11 items), Section D FC (2 items). Unit of analysis and data collection: UiTM undergraduate and postgraduate students complete Student Feedback Online (SuFO) within the LMS UFUTURE between weeks 11–16 each semester, evaluating each course/lecturer enrolled. Data periods: October 2019–February 2020 (pre-pandemic) and March–August 2020 (mid-pandemic). Data processing and analysis: Outlier and unengaged response screening (responses with SD<0.30 removed). Descriptive statistics in SPSS; inferential analysis using PLS-SEM. Measurement model assessed via outer loadings, composite reliability (CR), average variance extracted (AVE), and discriminant validity (HTMT). Multicollinearity checked via VIF (threshold <5). Structural model assessed for R², path coefficients, effect sizes (f²), and significance via bootstrapping (5,000 subsamples; significance at p≤0.01). Importance-Performance Map Analysis (IPMA) used with LA as the target construct to guide managerial priorities.

Sample and descriptive results: Pre-pandemic, 92,752 students evaluated 7,216 lecturers; mid-pandemic, 95,747 students evaluated 6,647 lecturers. After cleaning, analyzable responses totaled 194,559 (pre) and 221,366 (mid). Pre-pandemic, 95.4% undergraduates and 4.6% postgraduates participated; mid-pandemic, 95.1% undergraduates and 4.9% postgraduates. Measurement model: Outer loadings exceeded 0.7 for all indicators except LP3 and LA8, which were removed. CR for LP, CI, FC, and LA exceeded 0.7 and AVE exceeded 0.5 in both datasets, indicating good reliability and convergent validity; HTMT values were <0.90, indicating discriminant validity. VIF ranged from ~1.4 to 2.8, below the threshold of 5. Structural model: LA showed strong R² of 0.692 (pre) and 0.722 (mid). All hypothesized paths were positive and significant at p≤0.01. Effect sizes: LP had a large effect on CI and LA, and a medium effect on FC; CI→LA and FC→LA had small effects. Path estimates (examples): Pre: LP→CI β=0.519 (t=238.987), LP→FC β=0.351 (t=128.105), LP→LA β=0.682 (t=343.034), CI→LA β=0.164 (t=81.336), FC→LA β=0.121 (t=81.912). Mid: LP→CI β=0.532 (t=260.694), LP→FC β=0.345 (t=146.913), LP→LA β=0.706 (t=419.021), CI→LA β=0.164 (t=94.651), FC→LA β=0.106 (t=78.806). IPMA: LP had the highest total effects and performance: pre total effect 0.810 with performance 84.54; mid total effect 0.830 with performance 85.45. FC performance decreased from 78.8% (pre) to 73.13% (mid). LP fell in the “keep up the good work” quadrant. Outcomes linkage: Despite remote learning, undergrad GOT decreased slightly by 1.5% in 2020 vs 2019; postgraduate coursework GOT decreased by 1.7%. Academic performance remained strong: 83.1% of undergraduates and 48.7% of coursework postgraduates attained CGPA 3.00–4.00.

Findings demonstrate that lecturer professionalism is the dominant predictor of students’ enjoyment of remote learning activities both pre- and mid-pandemic. LP substantially shapes students’ course impressions and perceived learning activities, consistent with evidence that teacher support, pedagogical quality, and technological self-efficacy enhance distance learning. While facilitating conditions and course impressions contribute positively, their effects on LA are comparatively small in the model. The IPMA underscores LP as a high-importance, high-performance area, suggesting UiTM’s emphasis on lecturer commitment and support translated into sustained student engagement and satisfaction during the transition online. However, decreased FC performance mid-pandemic reflects student challenges with connectivity, devices, and home environments, especially among B40 students, potentially dampening motivation and participation. Pedagogical adaptations (flipped classroom, UDL-aligned self-instructional materials, varied synchronous/asynchronous assessments) and broad use of LMS and communication platforms likely supported positive perceptions and steady academic outcomes (GOT and CGPA), aligning with literature on the role of technological and instructional quality in e-learning satisfaction and performance.

The study validates a simplified, high-predictive-power model of remote learning acceptance at UiTM, showing that lecturer professionalism, together with course impression and facilitating conditions, explains substantial variance in enjoyable learning activities, with even stronger predictive metrics mid-pandemic. Students reported greater enjoyment of remote learning, and overall academic outcomes (GOT targets and CGPAs) remained stable during the pandemic. Practical implications include prioritizing LP through professional development in digital pedagogy, communication, assessment, and well-being; enhancing the primary LMS (UFUTURE) with robust tools; and focusing less on CI and FC when the goal is enjoyment of remote learning. As UiTM transitions to post-pandemic hybrid models, policies should be refined to support students’ emotional and learning needs, promote personalized and flexible assessments, and sustain communities of practice to bolster engagement. Future work should expand the model to include factors such as personal attributes, emotional intelligence, self-efficacy, satisfaction, and motivation to better predict enjoyable remote learning and academic performance.

Student perceptions during the pandemic may have been influenced by fear, stress, uncertainty, depression, distractions, and financial hardship, warranting cautious interpretation. Connectivity and device limitations (notably among B40 students), unconducive home environments, and technological unpreparedness among some lecturers and students were noted. Data derive from institutional evaluations and two semesters (pre- and mid-pandemic), which may limit generalizability. The instrument excluded some indicators (LP3, LA8) for validity reasons. Future studies should capture transition and post-pandemic perceptions and incorporate additional constructs (e.g., personal competence, emotional intelligence, self-efficacy, satisfaction, motivation).