Scientific Inquiry and Framing a Researchable Problem

Start with a researchable problem, not a broad topic

Scientific research is not simply the collection of information about an interesting topic. It begins with a question for which evidence can be sought through declared procedures, followed by choices that reduce the chance that conclusions merely reflect our expectations or the way data were collected. This diploma therefore links the question, design, analysis, interpretation and reporting rather than treating them as separate tasks. Scientific rigor emphasizes that the quality of an inference depends on disciplined design, methodology, analysis, interpretation and reporting, not on polished writing alone.

From topic to problem: state the gap between what is known and what needs to be known

A broad topic can be converted into a research problem by writing four layers. First, the context: where does the issue occur and for whom? Second, the initial observation or evidence: what pattern deserves investigation? Third, the knowledge gap: what can current information not explain or estimate? Fourth, the significance of closing that gap: what decision or understanding would improve with better evidence? In the fictional case, the problem could become: “In the organization’s introductory units, completion drops before the third activity, but current usage records do not explain whether the drop is associated with content difficulty, study timing, or other factors.” This wording does not claim a cause; it identifies a phenomenon and a gap.

A sound problem statement is not a request to prove management’s preferred explanation. If the starting statement is “learners drop out because the videos are long,” a possible explanation has been promoted to fact before evidence is collected. A better formulation separates observation from explanation: “completion falls at a defined point; what factors are associated with that decrease?” A stronger design may later test the effect of a specific factor, but the strength of the claim must match the design. Describing a pattern is different from explaining an association, and both are different from claiming that changing one factor caused a change in the outcome.