2026 ASEE Annual Conference & Exposition

This research brief presents a single, fully analyzed case study drawn from a larger, ongoing research project exploring evidence-based instructional design in graduate STEM courses.

Accelerated eight-week online graduate STEM courses are increasingly common, yet many are created by compressing 15-week designs without intentional redesign of early cognitive learning supports. In technically demanding courses such as database systems, this often increases cognitive load and contributes to early disengagement. These challenges are further intensified in the era of Generative Artificial Intelligence (GenAI), where students may rely on AI tools without sufficient instructional scaffolding for responsible use.

This research brief reports examines the redesign of the first two weeks of an online graduate database course across two existing course iterations (n = 25 per semester). Guided by the Community of Inquiry framework and theories of self- and co-regulation, the redesign introduced reflective activities supporting responsible GenAI engagement, low-stakes formative “Learning Quests,” and collaborative error-analysis assignments.

Data sources included standard end-of-semester course evaluation ratings, student learning artifacts, early assessment outcomes, and instructor debrief reflections. Results showed observable increases in early peer interaction, improved student perception ratings (overall mean 4.13 → 4.75), and stronger early comprehension compared to the prior iteration. Based on these preliminary findings, the paper proposes a Cognitive Learning Redesign Framework for Early Course Weeks to support cognitive presence and metacognitive regulation in accelerated graduate STEM courses.