Module 2 Article: The Hardest Part — Framing the Right Research Problem

The big idea

Most research failure is not bad data analysis; it is a badly framed problem. A vague problem yields a vague study; a sharply bounded problem yields a doable one. This article shows how to move from a fuzzy interest ("social media and mental health") to a researchable question, and how to diagnose what is really causing it before leaping to a solution.

Why a sharp problem matters

A problem that is too broad leads to:

• scope creep (the project grows forever),
• unfocused data collection (everything and nothing measured),
• inconclusive conclusions (no single question answered).

A sharp problem is:

• feasible (within time, skill, budget, ethics),
• answerable (with available data/methods),
• meaningful (fills a real gap).

From broad interest to researchable question: iterative refinement

Start broad and add constraints step by step:

1. Topic: "Social media and mental health."
2. Population: "among college students."
3. Exposure: "using Instagram > 2 hours/day."
4. Outcome: "anxiety score (GAD-7)."
5. Comparison: "non-users of Instagram."
6. Time: "over 6 months."
⇒ "Does >2h/day Instagram use predict changes in anxiety
   (GAD-7) among college students over 6 months, vs non-users?"

This is iterative refinement — each pass sharpens the frame. It is rarely linear; you loop back as feasibility and literature force new constraints.

Identifying sources of a research problem

Valid problems come from:

1. A gap in the literature (existing studies disagree or omit something).
2. A practical problem reported by practitioners.
3. Inconsistent findings across prior studies.
4. New technologies/theories that create new questions.
5. Replication needs (verify a landmark finding in a new context).

Warning: a "problem" that is really just a technology or a hobby is not a research problem until you connect it to a gap.

Root-cause diagnosis before solution

Too often a "problem" is a symptom. Two tools map the real causes.

The 5-Whys

Ask "why?" about the answer you get, five times, to reach a root cause.

Q1: Why do students report high stress?     → Heavy workload.
Q2: Why heavy workload?                     → Many assignments due together.
Q3: Why do assignments cluster?             → No coordination of deadlines.
Q4: Why no coordination?                    → Each course planned alone.
Q5: Why planned alone?                      → No shared calendar/policy.
Root cause: lack of institutional coordination.

Fishbone (Ishikawa) diagram

Map causes under categories (People, Methods, Machines, Materials, Environment, Policy). For student stress:

                 ┌─ Course load (Methods/People)
Student Stress ←┤
                 ├─ No gym (Machines/Materials)
                 ├─ Exams (Methods)
                 └─ Hostel noise (Environment)

Framing aim, objectives, and deliverables

• Aim (broad): To evaluate whether a supervised daily exercise plan reduces stress among MCA students.
• Objectives (specific, measurable, in order): O1. Measure baseline stress (PSS-10) in a pilot (n = 20). O2. Randomise 60 students to exercise vs control (8 weeks). O3. Compare post-intervention PSS-10 change (t-test, effect size). O4. Report feasibility for a larger trial.
• Deliverables: final report, de-identified dataset, analysis script, poster.

Feasibility and scope analysis

Always ask, before designing:

• Time: can the 8-week programme fit the academic calendar?
• Resources: is a researcher available to supervise? A gym free?
• Ethics: can we get consent and protect minors/ill participants?
• Data: is PSS-10 valid in this language/culture (translation/back-translation)?
• Scope: one college, two arms — bounded enough to finish.

A feasibility table with pass/fail per dimension keeps the project grounded.

Exam angle

For "meaning and identification of a research problem":

1. Define: a researchable gap stated as a question.
2. List two (or more) valid sources (gap, practice, inconsistency, new tech).
3. Show iterative refinement in 2–3 steps.
4. Contrast aim (broad) with objectives (specific) with deliverables.
5. State the feasibility dimensions (time, resource, ethics, data, scope).

Common mistake to avoid

Writing the problem statement as a topic or a complaint ("social media is bad") instead of as a question. Examiners penalise a problem that is just an opinion.

See also

• Module 3 (design) for testing the refined question.
• Glossary.md (framing, feasibility, scope, root cause).