---
name: big-research-design
description: Develop research questions, compare approaches and design method-appropriate studies or arguments, including experimental, qualitative, theoretical, interpretive and mixed-method work.
---

# Questions and approaches

Start with what the researcher wants to understand and why it matters. Distinguish exploratory dialogue from a request for a finished protocol. Compare materially different directions by research value, evidence, feasibility, cost and ethics when useful. The researcher selects consequential directions.

Select method modules by paradigm and task. Hypotheses, statistical power, interventions and falsification are useful for relevant designs, not universal requirements. State inferential aims and how the chosen approach supports them.

For quantitative confirmation, specify provenance, units, comparisons, metrics, uncertainty, preprocessing, leakage controls and budget before observing outcomes. For qualitative work, specify access, consent, sampling rationale, analytic approach and reflexivity. For interpretation, specify primary materials, editions, context and rival readings. For theory, specify definitions, assumptions, propositions and audit strategy. Mixed methods also require an integration rationale.

Save candidate comparisons and the researcher's actual choices. Create editable tasks, dependencies, expected artifacts and applicable checks. Budget is an explicit local record, not measured API billing. When resources are inadequate, propose a smaller valid pilot or record the blocker without pretending the original study was performed.