What the study found
The study found that five factors—relevance, clarity, detail, examples, and preferences—each had a positive influence on how useful students thought supplementary materials were. Detail and preferences were the most impactful factors, while matching materials to learning styles had only a minimal effect.
Why the authors say this matters
The authors conclude that software engineering education may benefit more from preference-driven personalization than from rigid learning-style categories. They also suggest that the adaptive algorithm showed potential, but that its performance varied across topics and needs refinement.
What the researchers tested
The researchers studied 96 undergraduates in an Introduction to Software Engineering course at a large public university in the southeastern United States. They used an adaptive experiment with contextual Thompson Sampling, a method that chooses options based on prior outcomes, to select supplementary materials for nine course topics using student ratings, gender, and topic; surveys measured learning styles, perceived usefulness, and the factors affecting usefulness.
What worked and what didn't
All five factors were positively associated with usefulness, with detail and preferences having the strongest influence. Visual learners rated materials higher when the materials matched their preferences, while verbal learners were less influenced. No differences related to gender were found.
What to keep in mind
The abstract does not describe detailed limitations beyond noting that the adaptive algorithm varied across topics and needs refinement. The findings are limited to the students, course, and setting described in the study.
Key points
- Five factors—relevance, clarity, detail, examples, and preferences—positively influenced perceived usefulness.
- Detail and preferences were the most impactful factors.
- Matching materials to learning styles had minimal effect overall.
- Visual learners rated materials higher when preferences were aligned; verbal learners were less influenced.
- No gender-related differences were found.
- The adaptive algorithm showed potential but varied across topics.
Disclosure
- Research title:
- Preferences and detail mattered most in supplementary materials
- Authors:
- LAWAL Olarotimi Badru, Jeffrey C. Carver
- Publication date:
- 2026-03-30
- DOI:
- 10.1145/3806053
- OpenAlex record:
- View
Get the weekly research newsletter
Stay current with peer-reviewed research without reading academic papers — one filtered digest, every Friday.