Fieldwork Methods & Data Collection

Cross-references: For full IRB protocol narratives, use the irb-protocol skill. For informed consent documents, use the informed-consent skill. For upstream decisions about which methods to use and why, use the methodology-selection skill. For research question development, use the research-question skill.

Quick Reference

Workflow

Step 1: Identify What the User Needs

Determine the entry point:

• Designing a complete research protocol from scratch. The user has selected their methods and needs to build out the full data collection system: instruments, sampling, training, data management, and pilot testing. Load all three reference files and work through the complete protocol design process.
• Writing an interview guide. The user needs a semi-structured, unstructured, life history, or key informant interview guide for a specific study. Load the data-collection-methods-guide for interview design guidance.
• Writing a focus group guide. The user needs a moderation guide including ground rules, question flow, activity design, and composition recommendations. Load the data-collection-methods-guide for focus group design guidance.
• Designing observation protocols. The user needs a participant observation protocol, structured observation checklist, or field note system. Load the data-collection-methods-guide for observation design guidance.
• Developing a sampling and recruitment strategy. The user needs help selecting and justifying a sampling approach (purposive, snowball, theoretical, maximum variation, criterion, or other), determining sample size, and designing recruitment procedures. Load the protocol-design-guide for sampling strategy guidance.
• Creating a data management plan. The user needs a DMP covering storage, transcription, de-identification, backup, retention, and destruction. Load the data-management-guide for DMP structure and requirements.
• Planning researcher training. The user is managing a research team and needs a training plan covering instrument use, interview technique, observation skills, reflexivity practices, and quality assurance. Load the protocol-design-guide for training plan guidance.
• Adapting instruments for a specific context. The user has existing instruments that need adaptation for a new field site, population, language, or cultural context. Load the data-collection-methods-guide for adaptation principles plus the protocol-design-guide for pilot testing.

Step 2: Gather Context

Before generating any content, collect these inputs:

Required:

1. Research methods. What data collection methods will be used? Interviews (semi-structured, unstructured, life history, key informant), focus groups, participant observation, structured observation, or a combination? Different methods require different instruments and protocols.
2. Field site characteristics. Where will fieldwork take place? Urban, rural, institutional, community, digital, multi-sited? Site characteristics shape instrument design, logistics, and data management requirements.
3. Participant population. Who are the participants? What are their demographic characteristics, power positions, vulnerabilities, languages, and literacy levels? Population characteristics affect question wording, interview format, observation focus, and sampling approach.
4. Study duration and timeline. How long is the fieldwork period? Is this a single intensive phase or multiple visits? Timeline shapes sampling targets, pilot testing feasibility, and data management volume.

Important but can be inferred: 5. Sampling approach. Purposive, snowball, theoretical, maximum variation, criterion, convenience, or other? If unspecified, recommend based on research questions, population characteristics, and epistemological stance. 6. Recording plans. Audio recording, video recording, handwritten notes, typed notes, photographs, or a combination? Recording decisions affect instrument design, consent requirements, and data management needs. 7. Data storage infrastructure. What tools and platforms are available? Institutional servers, encrypted cloud storage, local encrypted drives? Infrastructure affects data management plan design. 8. Team composition. Solo researcher or team? If a team, what is the experience level? Team size and experience affect training plan design and quality assurance protocols.

Helpful but not required:

• IRB status and any specific requirements or constraints from the review process
• Prior instruments from related studies that could serve as starting points
• Language and translation needs (multilingual fieldwork requires adapted instruments and transcription protocols)
• Community engagement history and existing relationships with field sites
• Epistemological stance (if not already established through the methodology-selection skill)
• Funder requirements for data management

Step 3: Load Appropriate References

• Load all three reference files when the user needs a complete protocol design covering instruments, sampling, training, data management, and pilot testing. This is the default for protocol-from-scratch tasks.
• Load references/data-collection-methods-guide.md alone when the user needs help drafting a specific instrument: interview guide, focus group guide, observation protocol, or field note template. This is the default for instrument-drafting tasks.
• Load references/protocol-design-guide.md alone when the user needs help with sampling strategy, recruitment planning, researcher training, or pilot testing without also needing new instruments.
• Load references/data-management-guide.md alone when the user needs a standalone data management plan covering storage, transcription, de-identification, backup, and retention.
• Load protocol-design-guide + data-collection-methods-guide when the user needs instruments plus sampling and training but already has a data management plan.
• Load data-collection-methods-guide + data-management-guide when the user needs instruments plus a data management plan but does not need sampling or training guidance.

Step 4: Generate Content

Follow the protocol design process, adapting the sequence to the user's entry