Teaching Practically Relevant Research Problem Formulation in Software Engineering with Lean Research Inception

1. Problem Statement

The core problem addressed is the persistent gap between academic Software Engineering (SE) research and industry needs. Many SE studies suffer from low practical relevance because research problems are often formulated based on overly simplified assumptions or overlook real-world industry challenges.

• The Gap: While literature emphasizes bridging the industry-academia divide, existing educational approaches often focus on project execution or empirical methods, neglecting the initial stage of research problem formulation.
• The Consequence: Students struggle to articulate problems that are theoretically sound yet practically applicable, leading to research that lacks clear context, stakeholders, or impact.
• The Need: There is a need for a structured pedagogical strategy to teach students how to formulate research problems that are aligned with professional practice from the outset.

2. Methodology

The authors conducted a case study within the context of a Capstone Project course in a Brazilian Software Engineering undergraduate program.

• Participants:
  • 60 Students: Final-year SE students from three different campuses (Vassouras, Maricá, and Saquarema).
  • 7 Faculty Advisors: Instructors supervising the projects, acting as proxies for professional practice.
• Intervention (Lean Research Inception - LRI):
  • The study applied Lean Research Inception (LRI), a structured approach grounded in agile principles (Design Thinking, Lean Startup, Lean Inception).
  • Core Artifact: The Problem Vision Board, a visual template containing seven attributes: Practical Problem, Context, Implications/Impacts, Stakeholders, Evidence, Objective, and Research Questions.
  • Process:
    1. Phase 1 (Outline): Students used the Problem Vision Board to draft their research problems.
    2. Phase 2 (Alignment): Advisors (acting as industry practitioners) collaborated with students to challenge assumptions, validate relevance, and refine the problem.
    3. Phase 3 (Formulation): The problem was consolidated based on the alignment.
  • Note: The study focused strictly on the first three phases (Formulation), excluding later assessment and decision-making phases to maintain focus on the educational objective.
• Data Collection & Analysis:
  • Quantitative: Two distinct questionnaires (Likert scale) for students and advisors covering reflection, clarity, communication, ease of use, and adoption intent.
  • Qualitative: Open-ended questions analyzed via content coding to identify patterns and insights.
  • Triangulation: Quantitative and qualitative data were combined to validate findings.

3. Key Contributions

The paper makes three primary contributions to SE education:

1. Reframing Problem Formulation: It positions research problem formulation as a distinct, teachable, and learnable competence in SE education, rather than an innate skill.
2. Pedagogical Strategy (LRI): It introduces and empirically evaluates Lean Research Inception as a scaffold for the early stages of research design, specifically within Capstone courses.
3. Empirical Evidence: It provides data from a real-world educational setting involving both students and faculty, demonstrating that structured tools can improve the quality of problem definition and industry alignment.

4. Results

The study yielded positive perceptions from both students and advisors regarding the efficacy of LRI.

Student Perceptions (N=60):

• Reasoning & Reflection: 60% agreed that the board helped them reflect more effectively on the problem.
• Clarity & Definition: 61.7% agreed it made their problems clearer and better defined; 31.6% partially agreed.
• Communication: 50% agreed it facilitated communication with advisors; 28.4% partially agreed.
• Ease of Use: 60% found it easy to use; 25% partially agreed.
• Adoption Intent: 58.3% agreed they would use the board in future projects.

Advisor Perceptions (N=7):

• Clarity & Structure: 57.1% agreed that problems formulated with LRI were clearer and better structured than usual.
• Practical Alignment: 42.9% agreed the problems were better aligned with practical needs (28.6% partially agreed).
• Communication: 57.1% agreed the board facilitated communication and reduced ambiguity during supervision.
• Ease of Use: 71.4% found the framework easy to understand and use.
• Recommendation: 85.7% of advisors expressed a strong intention to recommend the approach for future projects.

Qualitative Insights:

• Students reported the board helped transform "vague ideas" into "concrete problem statements" and organized their thoughts holistically.
• Advisors noted that the structured attributes forced students to explicitly articulate context and stakeholders, making supervision more objective and productive.
• Suggestions for improvement included adding more explicit step-by-step guides and visual refinements for complex relationships between attributes.

5. Significance

• Bridging the Gap: LRI offers a scalable mechanism to align academic research with industry realities by forcing early consideration of practical context and stakeholders.
• Pedagogical Shift: It shifts the educational focus from how to execute research to how to define relevant research, addressing a critical bottleneck in SE education.
• Scalability: The approach is low-cost, uses open-source tools (Miro templates), and can be integrated into existing Capstone curricula without requiring massive infrastructure changes.
• Future Impact: By improving the quality of problem formulation at the undergraduate level, LRI has the potential to influence the long-term relevance and rigor of future SE research.

Limitations: The study is limited to a single university in Brazil, relies on self-reported perceptions (potential social desirability bias), and lacks a control group. Future work aims to replicate the study across different institutions and explore the full five-phase LRI process.