Research Plan Writing

Write standalone research plans for anthropological and qualitative research that articulate what is being studied, why it matters, how it will be done, and why anyone should believe it can be done. A research plan is a forward-looking argument for feasibility, rigor, and ethical legitimacy — the foundational document from which grant proposals, dissertation prospectuses, and fieldwork clearance applications are derived.

The audience for a standalone plan varies: it may be the researcher themselves (as a thinking tool), a committee, a department fieldwork review, or a collaborating partner. The plan should be precise regardless of audience, but register and emphasis adapt to context.

Quick Reference

Task	Reference
Full research plan guide (sections, architecture, audience adaptation)	Read references/research-plan-guide.md

Workflow

Step 1: Identify What the User Needs

Determine the entry point:

• Writing from scratch. The user has a topic or research idea and needs to develop it into a structured plan. Load the full reference file and work through sections in order.
• Revising an existing plan. The user has a draft and wants feedback or help improving specific sections. Identify which sections need work and load the relevant guidance.
• Audience adaptation. The user has a working plan and needs to adapt it for a specific context (departmental review, community partner, fieldwork clearance). Focus on register, emphasis, and format adjustments.

Step 2: Gather Context

Before generating any content, collect these inputs:

Required:

1. Research topic and site(s). What is the project about and where will research occur?
2. Epistemic stance. Which theoretical orientation(s) does the researcher work within? Ask for primary and secondary.
3. Intended audience(s). Who will read this plan? This determines format, register, and emphasis. Common audiences: self (thinking tool), advisor, department fieldwork review, community partner, collaborator.
4. Stage of development. Starting from scratch, refining a draft, or adapting to a different audience?

Helpful but not required:

• Theoretical framework or key interlocutors
• Preliminary fieldwork or pilot data
• Field configuration (single site, multi-sited, digital, archival, hybrid)
• Career stage (affects scope and ambition calibration)
• Language competencies relevant to the field site
• Timeline constraints
• Whether the plan will later be adapted into a grant proposal or prospectus

Step 3: Load Reference

Load references/research-plan-guide.md for the full section-by-section architecture, guidance by epistemic family, formatting principles, and audience adaptation notes.

Step 4: Generate Output

Follow the section structure in the reference file. The plan uses a front-loaded clarity architecture: state the core argument up front, justify it in the middle, demonstrate feasibility at the end.

Ten sections (adapt order and emphasis to audience):

1. Overview — compress the full argument into 0.5-1 page
2. Research problem and questions — name the phenomenon, establish stakes, state empirically researchable questions
3. Conceptual and empirical context — focused literature argument for the gap, not a comprehensive survey
4. Field site(s) and access — justify "why here," specify access pathways, include contingency planning
5. Methods and data generation — argue for each method in terms of what evidence it generates and why that evidence is necessary; specify sampling and recruitment logic
6. Analysis plan — specify analytic approach, documentation practices (audit trail), how analysis and data collection interact, what a "finding" looks like
7. Reflexivity and positionality — analytical statement about how the researcher's location shapes access, interpretation, and relationships; concrete management steps
8. Trustworthiness and transparency — credibility, transferability, dependability, confirmability strategies built into the design
9. Ethics and governance — consent strategy, confidentiality, potential harms, data governance, community obligations, dissemination ethics
10. Feasibility — preparation completed, timeline, budget, personnel

Key generation principles:

• Methods as inferential design. Do not list techniques. Argue for each method: what evidence it generates, why that evidence is necessary for the questions, how it relates to other methods. Specify sampling logic and recruitment pathways.
• Analysis plan is mandatory. This is the most common weakness in qualitative research plans. Specify the analytic approach, documentation practices, how analysis and fieldwork interact, and what form findings will take. "I will use thematic analysis" is insufficient.
• Reflexivity is analytical, not confessional. Address how positionality shapes access, perception, and interpretation. Specify concrete steps to manage these effects. Integrate with methods and ethics, not siloed.
• Trustworthiness is designed in. Credibility, transferability, dependability, and confirmability are practices built into the plan, not claims made after the fact.
• Ethics spans the full lifecycle. Address consent modality, data governance (FAIR + CARE for Indigenous data), community obligations, dissemination ethics, and processual ethics — not just IRB compliance.
• Feasibility is a realism test. Include pre-fieldwork preparation, month-by-month timeline, realistic budget, and contingency planning. Reviewers treat these as credibility signals.
• Front-load clarity. A reader who stops after the overview should understand the project's intellectual core.

Step 5: Quality Check

Before presenting output, verify:

• Research questions are empirically researchable and bounded
• Conceptual context is an argument about the gap, not a literature survey
• Field site choice is justified ("why here" is answered explicitly)
• Each method is argued for, not just listed
• Sampling logic and recruitment are specified
• Analysis plan names the approach, documentation practices, and what constitutes a finding
• Reflexivity section is analytical, not autobiographical
• Trustworthiness strategies are concrete practices, not abstract claims
• Ethics addresses consent modality, data governance, harms, and community obligations substantively
• Contingency plan is present (what if access fails?)
• Timeline includes pre-fieldwork workstreams
• Scope is realistic for the stated timeline and resources
• Epistemic stance is reflected consistently in language throughout
• Output matches the intended audience's expectations

Parameters

• Epistemic stance: All 42 stances are relevant. The stance shapes how methods are described, what theoretical vocabulary is appropriate, and how "contribution" is framed. See DESIGN.md for the full list.
• Genre/audience: Self (thinking tool), advisor, departmental fieldwork review, community partner, collaborator, general standalone plan.
• Compression: Working draft (expansive, exploratory) through polished plan (10-20 pages, tightly argued).
• Risk posture: Low-risk, vulnerable populations, high-surveillance, politically sensitive. Higher risk postures require more detailed ethics and contingency sections.
• Formality register: Working draft (informal), advisor-ready (polished but frank), review-ready (formal, defensible), community-facing (accessible, mutual-benefit emphasis).
• Field configuration: Single site, multi-sited, digital, archival, hybrid, comparative. Affects site justification, methods description, and feasibility framing.

Guardrails

• Do not generate without knowing the intended audience. A plan written as a personal thinking tool has a different register and emphasis than one written for a departmental fieldwork review. Format varies significantly by audience. Ask first.
• Do not produce vague analysis plans. The analysis section must specify the analytic approach, documentation practices (audit trail, memoing, codebook evolution), and what constitutes a finding. "I will analyze the data thematically" is a failure mode, not an analysis plan.
• Do not produce boilerplate ethics. Ethics must address the specific consent modality, data governance choices, community obligations, and potential harms relevant to the project. If the risk posture is elevated, ethics protections must match.
• Reflexivity must be analytical. Do not produce confessional autobiography. Positionality discussion should specify how the researcher's location affects access, interpretation, and relationships, and what concrete steps will manage those effects.
• Flag scope-feasibility mismatches. If the proposed research scope exceeds what is realistic for the stated timeline, resources, or career stage, flag it explicitly and suggest narrowing.
• Do not conflate the research plan with a grant proposal or prospectus. A research plan is audience-flexible and foundational. If the user actually needs a funder-specific document, redirect to the grant-proposal skill. If they need a committee-facing prospectus, redirect to the dissertation-prospectus skill.

Common Failure Modes

Failure mode	Prevention
Methods as wish list — lists techniques without inferential logic	Argue for each method: what evidence, why necessary, how methods relate
Vague analysis plan — "I will use thematic analysis"	Specify approach, documentation practices, analysis-fieldwork interaction, finding form
Ethics as afterthought — perfunctory paragraph at the end	Make ethics a workflow step throughout; address consent, governance, harms, community obligations
Missing reflexivity — no attention to researcher positionality	Include analytical positionality section; integrate with methods and ethics
No contingency planning — assumes everything goes as planned	Include Plan B for access disruption; state what claims survive under alternative strategies
Scope exceeds feasibility — too many questions, sites, or methods for timeline	Flag mismatches; suggest narrowing to what is realistic for available time and resources
Generic social science prose — could be from any discipline	Attend to ethnographic specificity, cultural context, relational processes, and disciplinary voice
Literature survey instead of argument — lists sources without engagement	Structure literature section as an argument about the gap this project addresses

Examples

Example 1: Standalone research plan, sociocultural, single site

Input: "I need to write a research plan for my project on how street vendors in Lima negotiate municipal regulation. I'm an interpretivist drawing on practice theory. This is mainly for myself and my advisor to clarify my thinking."

Output approach:

• Load references/research-plan-guide.md
• Set audience to advisor-ready (polished but frank)
• Set epistemic stance to interpretivist + practice theory
• Set field configuration to single site
• Set risk posture to standard (public spaces, not high-risk)
• Methods: participant observation with vendors and municipal inspectors, semi-structured interviews (~30-40), document analysis of regulatory frameworks
• Analysis: iterative coding with practice-theoretical categories; contrast between official regulatory logic and everyday negotiation practices
• Reflexivity: researcher position as foreign academic studying informal economy; access mediated through vendor associations
• Feasibility: 12-month fieldwork, Spanish language competency, prior preliminary visit

Example 2: Research plan for departmental fieldwork review

Input: "My department requires a fieldwork plan before I can go to the field. I'm studying how climate adaptation policies are negotiated between Indigenous communities and government agencies in northern Canada. I work within political ecology and Indigenous methodologies."

Output approach:

• Load references/research-plan-guide.md
• Set audience to review-ready (formal, defensible)
• Set epistemic stance to political ecology + Indigenous methodologies
• Set field configuration to multi-sited (communities + government offices)
• Set risk posture to enhanced (Indigenous communities, politically sensitive)
• Emphasis on risk management and ethical clearance (departmental review priorities): detailed consent strategy (community-level + individual, processual), CARE principles for data governance, community research agreements, contingency for access disruption
• Methods: foreground community authority and reciprocity; collaborative research design with community partners
• Timeline: include pre-fieldwork preparation (community consultation, ethics clearance, research agreements)

Example 3: Research plan for digital/hybrid ethnography

Input: "I want to plan a study of how Chinese diaspora communities use WeChat to maintain transnational political identities. This will be mostly digital ethnography but with some in-person interviews. I'm coming from a linguistic anthropology perspective."

Output approach:

• Load references/research-plan-guide.md
• Set epistemic stance to linguistic anthropology + digital anthropology
• Set field configuration to hybrid (digital platforms + in-person)
• Set risk posture to enhanced (political speech, surveillance risk, cross-border data)
• Site justification: explain why WeChat specifically (platform affordances, community adoption), why this diaspora community, why digital + in-person is necessary
• Methods: digital participant observation in WeChat groups, discourse analysis of message interactions, semi-structured interviews with community members (~20-25); address platform ethics (terms of service, consent in group contexts, researcher visibility)
• Data management: address cross-border data flows (China's data regulations), encrypted storage, de-identification challenges with digital trace data
• Analysis: discourse-centered approach to register, code-switching, and political stance-taking in digital interactions