From Real-World Projects to Research-Oriented Learning: Continuous Improvement of a Master-Level Course in Software Engineering Education
This longitudinal mixed-methods study demonstrates how a master-level software engineering course at a German university successfully evolved from practice-oriented to research-oriented learning over six years while maintaining positive student perceptions through authentic projects, external collaboration, and structured scaffolding.
Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Imagine a university course as a cooking class.
In a standard cooking class, students might follow a recipe step-by-step to make a perfect cake. They learn the techniques, but they are just applying known rules to a known problem.
This paper describes a specific "Master's level" software engineering course that started as a practical workshop (like a cooking class where students build a real restaurant menu for a local diner) and slowly transformed over six years (2019–2025) into a culinary research lab.
Here is the story of that transformation, explained simply:
The Big Question
The teachers wanted to know: If we make the class harder and more like "scientific research," will the students hate it?
Usually, when you add more complexity, more uncertainty, and more "real-world" pressure to a class, students get stressed and give bad reviews. The teachers wanted to see if they could evolve the course into a high-level research environment without losing the students' enthusiasm.
The Journey: From "Apprentice" to "Researcher"
Think of the course evolution like upgrading a video game:
- Level 1 (2019): Students were Apprentices. They worked on real projects for real companies (like insurance or car manufacturers). They used standard tools to solve practical problems. It was hands-on and useful, but not deeply scientific.
- Level 2 (2021–2022): The game got harder. Students had to stop just "doing" and start asking "why." They had to treat their projects like scientific studies. They had to collect data, interview people, and justify their methods.
- Level 3 (2023–2025): The game became a Research Lab. Students weren't just solving one problem; they were running complex experiments, studying multiple companies at once, and dealing with very new, tricky topics (like how AI affects team dynamics or how neurodiversity works in tech teams). The work became much more uncertain and demanding.
The Surprise Result
The teachers were worried that as the course got harder (moving from Level 1 to Level 3), the students would complain that it was too much work or too confusing.
But that didn't happen.
Even though the course became significantly more difficult and research-heavy, the students still rated the course highly. They didn't hate the extra work; in fact, they often loved the challenge.
Why Did It Work? (The Secret Sauce)
The paper identifies four "ingredients" that kept the students happy while the course got harder:
- Real-World Relevance (The "Why"): The projects weren't fake classroom exercises. Students were working with actual companies on real problems. It felt like they were doing something that mattered in the real world, not just passing a test.
- The "Safety Net" (Scaffolding): Even though students were given a lot of freedom (autonomy), the teachers didn't just leave them alone. They provided a strong structure. Think of it like a safety harness for rock climbing. The students could climb high and take risks, but they knew the teacher was there with a rope, ready to help if they slipped.
- The Coach, Not Just a Lecturer: The teacher acted more like an Agile Coach or a mentor than a traditional professor. They were available, gave quick feedback, and helped students navigate the confusion. When the work got hard, the students knew they had a guide to talk to.
- Clear Structure: The course had clear milestones (like "sprints" in software development). Students knew what was expected of them at each step, which reduced the anxiety of the unknown.
The Catch (It wasn't perfect)
The paper is honest: The students did feel the stress. They complained about the heavy workload, the time it took to collect data, and the confusion that comes with doing research. They didn't find it "easy."
However, they felt the stress was worth it. They valued the freedom to explore, the chance to work with real companies, and the feeling that they were learning something truly valuable for their future careers.
The Bottom Line
The paper concludes that you can turn a practical project course into a serious research course without scaring students away.
But you can't just dump a bunch of hard research tasks on them and hope for the best. You have to wrap those hard tasks in a supportive environment where the work feels meaningful, the teacher is accessible, and the path is clear. When you do that, students don't just tolerate the difficulty; they embrace it.
In short: You can take students from "following a recipe" to "inventing a new cuisine," as long as you give them a great chef to guide them and a real kitchen to practice in.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.