Adaptive Learning Analytics for Business Education Platforms in SMEs

Thesis Abstract

The rapid digitization of business education platforms in small and medium-sized enterprises (SMEs) has exposed a critical gap in delivering personalized learning that aligns with diverse SME workforce competencies and operational contexts. Adaptive Learning Analytics (ALA) offers a data-driven approach to tailor content, pacing, and assessment by leveraging learner interactions, performance histories, and domain-specific metrics. The study aims to evaluate how ALA can optimize learning outcomes, engagement, and knowledge transfer within SME-facing business education platforms. Specifically, the objectives are (1) to design an adaptive learning framework that integrates domain ontologies, learning analytics dashboards, and intent-aware recommender systems; (2) to assess the impact of adaptive pathways on learning gains across finance, marketing, and operations micro-credentials; (3) to examine how managerial support and organizational readiness moderate the effectiveness of ALA; (4) to explore learner experiences, perceived usefulness, and motivation through qualitative inquiry; and (5) to develop a scalable implementation blueprint for SMEs with cost-effective technology stacks. The study adopts a mixed-methods design, combining quasi-experimental evaluation with an explanatory sequential approach. The population comprises 420 employees across 14 SMEs in the manufacturing and services sectors that adopt a unified business education platform. A stratified random sample of 240 learners will be assigned to an adaptive learning condition (AL) and 180 to a control condition using non-adaptive, fixed-path curricula over a 12-week program. Data collection instruments include platform log files capturing clickstreams, time-on-task, question response accuracy, and learning path progression; standardized assessments measuring domain-specific competencies pre- and post-intervention; the System Usability Scale (SUS) for user experience; and semi-structured interviews with 24 learners and 8 SME managers. Validation will involve content experts, pilot testing, and triangulation across quantitative and qualitative sources. Data analysis will proceed in three stages. First, descriptive statistics and reliability analyses will summarize engagement and performance indicators. Second, inferential analyses will test hypotheses regarding learning gains and engagement differences between AL and control groups, employing ANCOVA to control for baseline ability and hierarchical linear modeling to account for nested data within SMEs. Regression analyses will identify predictors of learning improvement, including time-on-task, adaptability of learning paths, and prior domain knowledge. Third, qualitative data will undergo thematic analysis, guided by the Technology Acceptance Model and Self-Determination Theory, to elucidate learner motivation, perceived autonomy, competence, and relatedness. Mixed-methods integration will occur through joint displays and meta-inferences to interpret how quantitative outcomes relate to learner narratives and manager perspectives. Expected findings include statistically significant higher post-test gains and retention rates for the AL group, with effect sizes in the small to moderate range (Cohen’s d = 0.35–0.60) across micro-credential domains. Adaptive pathways are anticipated to reduce time-to-competence and increase sustained engagement, particularly when organizational readiness—measured by leadership support, ICT infrastructure, and training—meets a defined threshold. Qualitative insights are expected to reveal enhanced perceived usefulness and autonomy, while potential challenges may include data privacy concerns and initial adoption resistance. The study will also identify critical design features, such as real-time competency mapping, transparent rationale for path adjustments, and intuitive analytics dashboards for managers. Contributions to knowledge encompass (1) an empirically validated, domain-informed ALA framework tailored to SME business education contexts, (2) robust evidence on the effectiveness of adaptive versus non-adaptive pathways in improving competency gains and learner engagement, and (3) a practical, scalable implementation model detailing governance, data governance, cost considerations, and change-management strategies for SMEs. The theoretical contribution integrates the Technology Acceptance Model with Self-Determination Theory to explain adoption dynamics and motivation in adaptive learning environments. The study concludes that adaptive learning analytics can meaningfully enhance business education outcomes in SMEs when underpinned by domain ontology alignment, transparent adaptation mechanisms, and organizational readiness. Recommendations include adopting modular, open-standards-based architectures, investing in data privacy protections, providing managerial sponsorship and training, and conducting iterative pilot deployments to refine adaptive rules and assessment alignment.