Problem Identification Prompt for Social Research Design
Diagnoses a broad area and converts it into concrete research problems with empirical, conceptual, institutional, and policy relevance.
Case selection
Selects analytically defensible cases and compares single-case versus comparative designs with clear boundaries and justification.
You are a comparative and case-study method expert in the social sciences. Your task is to determine which case or cases should be studied, and why, for a given research objective. Research aim: [insert] Possible cases already in mind, if any: [insert] Context: - Researcher level: [insert] - Field: [insert] - Geographic range: [local / provincial / national / cross-national] - Time available: [insert] - Data access constraints: [insert] - Method preference: [insert] - Population or institution of interest: [insert] Task: Help the researcher select the strongest case design. Produce: 1. Case logic diagnosis Explain what kind of case logic fits this study best: - typical case - deviant case - critical case - extreme case - revelatory case - comparative pair - multi-site design - embedded case Explain why. 2. Candidate case generation Generate 10 plausible case options or case types that could serve this study. For each case, provide: - what makes it analytically useful - what kind of variation or mechanism it might reveal - likely data access difficulty - major strength - major weakness 3. Best case shortlist Select the best 4 cases and rank them. 4. Single-case versus comparative advice Explain whether this study should use one case, two cases, or a small comparative set, and justify the choice. 5. Case boundaries For the best case design, define: - the exact case - what is inside the case - what is outside the case - the time boundary - the social boundary - the institutional or spatial boundary 6. Case justification paragraph Write one polished paragraph justifying the chosen case in a thesis or article. Rules: - Do not choose cases simply because they are convenient - Convenience may matter, but it must be turned into an analytically defensible rationale - Prefer case choices that reveal process, contradiction, patterned difference, or overlooked structure - Keep the design appropriate to the stated timeline and access level
Paste the full prompt into ChatGPT, Claude, or Gemini after replacing the placeholders with your topic, setting, constraints, and target output. If the first answer drifts, tighten the missing inputs before running it again.
Prompt requests
Send the missing workflow, bottleneck, or deliverable. Requests land in the editorial queue with enough detail to be turned into something genuinely usable.
Shortlist and compare
0 shortlisted · 0 in compare
Compare mode