fieldwork-methods

name

fieldwork-methods

description

Use this skill whenever a user needs help designing fieldwork data collection instruments or protocols for qualitative or anthropological research. Triggers include: "interview guide," "interview protocol," "focus group guide," "observation protocol," "field notes," "field note template," "fieldwork protocol," "data collection instruments," "sampling strategy," "purposive sampling," "snowball sampling," "data management plan," "DMP," "transcription protocol," "researcher training," "pilot testing," "semi-structured interview," "life history interview," "key informant interview," or "participant observation protocol." Covers interview guides, focus group guides, observation protocols, field note systems, sampling and recruitment, training, pilot testing, and data management. Do NOT use for IRB protocol narratives (use irb-protocol skill), consent documents (use informed-consent skill), or methodology selection (use methodology-selection skill).

Fieldwork Methods & Data Collection

Design fieldwork data collection instruments, protocols, and processes for qualitative and anthropological research. This skill treats fieldwork instruments as both technical tools and relational practices: an interview guide is simultaneously a cognitive scaffold for the researcher and a conversational contract with the participant; an observation protocol is both a data capture system and a disciplinary lens that shapes what is noticed and what is missed. Good instruments balance structure with flexibility, providing enough scaffolding to ensure rigor and comparability across contexts while leaving enough space for the emergent, iterative, and relational dynamics that define ethnographic work.

Data collection design is inseparable from data management planning. Decisions about how data will be recorded, stored, transcribed, de-identified, and retained shape what can be collected and how it can be used. A beautifully designed interview guide loses its value if recordings are lost to poor backup practices, or if transcription choices strip context that the analysis depends on. This skill therefore covers the full arc from instrument design through data management, treating them as a single integrated system rather than separate concerns.

The skill also addresses the human infrastructure of fieldwork: training researchers to use instruments consistently without rigidly, building reflexivity into data collection practice, and planning pilot tests that improve instruments before the stakes are highest. Fieldwork quality depends as much on the preparation and adaptability of the researcher as on the design of the instruments themselves.

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

Task	Reference
Protocol structure, sampling strategies, pilot testing, training and reflexivity	Read references/protocol-design-guide.md
Interview guides, focus group guides, observation protocols, field note systems	Read references/data-collection-methods-guide.md
Data storage, transcription, de-identification, backup, retention and destruction	Read references/data-management-guide.md

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 point:

1. Define the data collection strategy. Based on research questions and methods, specify what data will be collected, from whom, through what instruments, in what sequence, and over what timeline. Map each research question to the data collection activities that will address it.

2. Design instruments. Draft the specific instruments needed:

   • Interview guides: opening sequence, domain-based question modules, probe sets, closing and member-checking questions, adapted to interview type (semi-structured, unstructured, life history, key informant)
   • Focus group guides: ground rules, warm-up activities, core discussion questions, structured activities or ranking exercises, closing synthesis
   • Observation protocols: observation domains, spatial and temporal sampling strategies, behavioral categories (if structured), sensitizing concepts (if ethnographic), recording format
   • Field note templates: header metadata, running description format, analytic memo sections, reflexivity prompts

3. Plan sampling and recruitment. Select and justify the sampling strategy. Specify target sample size with rationale (information power, theoretical saturation, or other justification). Design recruitment procedures including access strategies, screening criteria, and contingency plans for recruitment challenges.

4. Develop the training plan. For team-based research, design a training protocol covering instrument familiarization, mock interviews or practice observations, inter-rater reliability procedures (if applicable), reflexivity exercises, and ongoing quality assurance mechanisms. For solo researchers, design a self-preparation plan with pilot exercises.

5. Create the data management plan. Specify data storage (encryption, access controls, location), transcription protocol (verbatim, edited, conventions, quality checks), de-identification procedures (pseudonyms, contextual detail management, deductive disclosure prevention), backup schedule and redundancy, retention period and destruction procedures.

6. Design the pilot testing approach. Plan cognitive pre-testing of instruments, pilot interviews or observation sessions, debrief procedures, revision criteria, and the timeline for incorporating pilot feedback before full data collection begins.

Step 5: Generate Output

Produce one or more deliverables depending on user needs:

• Interview guide. Complete guide with opening sequence, domain-based question modules, probe sets, and closing. Formatted for use in the field with clear visual hierarchy and space for notes.
• Focus group guide. Moderation guide with ground rules, question flow, activities, timing estimates, and facilitation notes. Includes composition recommendations and logistics checklist.
• Observation protocol. Structured or semi-structured observation guide specifying what to observe, when, where, and how to record. Includes spatial and temporal sampling strategies where relevant.
• Field note template. Standardized template with header metadata, running description format, analytic memo sections, and reflexivity prompts. Adapted to the specific study context.
• Sampling plan. Strategy document specifying sampling approach, selection criteria, target sample size with justification, recruitment procedures, and contingency plans.
• Recruitment strategy. Detailed plan for identifying, contacting, and enrolling participants, including gatekeepers, access strategies, and scripts or materials.
• Data management plan. Complete DMP covering storage, transcription, de-identification, backup, retention, and destruction. Suitable for inclusion in an IRB application or funder report.
• Training plan. Researcher training protocol covering instrument use, interview and observation technique, reflexivity practices, and quality assurance procedures.
• Pilot testing plan. Pre-testing strategy specifying what will be piloted, with whom, how feedback will be collected, and what revision criteria will be used.
• Complete research protocol. Full protocol document integrating all of the above into a single coherent system. Includes a protocol summary, data collection matrix, and implementation timeline.
• Adapted instruments. Existing instruments revised for a new context, with documentation of what changed and why.

Step 6: Quality Check

Before presenting output, verify:

• Instruments are clearly linked to research questions — every question or observation domain can be traced to a specific research question it helps answer
• Interview questions are open-ended and non-leading — no yes/no questions in the main guide, no embedded assumptions, no double-barreled questions
• Observation protocols balance structure with flexibility — enough categories to ensure systematic coverage, enough openness to capture the unexpected
• Focus group guides include ground rules, timing, and facilitation notes — not just a list of questions
• Sampling strategy is justified with a clear rationale — not just "we will interview 30 people" but why 30, why these people, and how they will be selected
• Data management plan covers the full lifecycle — from recording through storage, transcription, de-identification, backup, retention, and destruction
• Training plan addresses key competencies — instrument use, interview technique, reflexivity, and quality assurance, not just logistics
• Pilot testing is planned with specific procedures — what will be tested, with whom, how feedback will inform revisions
• Ethical considerations are embedded throughout — not siloed in a separate section but integrated into instrument design, sampling, training, and data management
• Instruments are culturally appropriate — language, format, and process reflect participants' norms, literacy levels, and communication styles
• Protocols specify flexibility and iteration points — where and how instruments can be adapted during fieldwork without compromising rigor
• Field note system includes reflexivity prompts — not just "what happened" but "what did I notice, what did I miss, how did my presence shape the interaction"

Parameters

• Output type: Interview guide, focus group guide, observation protocol, field note template, sampling plan, recruitment strategy, data management plan, training plan, pilot testing plan, complete research protocol, adapted instruments. Determines scope and format of the deliverable.
• Methods covered: Semi-structured interviews, unstructured interviews, life history interviews, key informant interviews, focus groups, participant observation, structured observation, mixed. Different methods require different instrument structures and design principles.
• Field context: Urban, rural, institutional, community, digital, multi-sited, cross-cultural. Context shapes instrument adaptation, logistics, and data management requirements.
• Population: General adult population, youth, elders, professionals, marginalized communities, Indigenous communities, multilingual populations. Population characteristics affect question wording, interview format, observation focus, and sampling approach.
• Document stage: Initial draft, revision after pilot testing, adaptation for new context, revision after IRB feedback. Stage determines how much scaffolding to provide versus how much refinement to focus on.
• Protocol scope: Single instrument, multiple instruments, complete protocol with all components. Scope determines which reference files to load and how many deliverables to produce.

Guardrails

• Protocols must balance structure with flexibility. Rigid protocols that script every moment of data collection are inappropriate for ethnographic work. But fully unstructured approaches risk missing critical data and undermine comparability. Help users find the right balance for their methods, questions, and field context. Interview guides should have clear domains and probes but leave room for follow-up; observation protocols should specify what to attend to but not constrain what can be noticed.
• Interview questions must be open-ended, not leading. Do not generate questions that embed the researcher's assumptions ("How has climate change affected your farming?") or that can be answered with yes/no ("Do you think the program was effective?"). Frame questions to invite the participant's own framing: "Tell me about changes you have noticed in your farming over the years" or "What has your experience with the program been like?"
• Observation protocols are guides, not scripts. Participant observation cannot be fully pre-specified because ethnographic attention must be responsive to what is happening. Observation protocols should sensitize the researcher to key domains and provide a recording format, not dictate a rigid sequence of behaviors to watch for.
• Sampling must be justified, not arbitrary. Every sampling decision needs a rationale grounded in the research questions and epistemological approach. "We will interview 30 people" is not a sampling strategy. Require a named sampling approach, clear selection criteria, a justified target number, and a contingency plan for recruitment challenges.
• Data management is not optional. Every protocol must include a data management component. Do not generate instruments without also addressing how the data they produce will be recorded, stored, transcribed, protected, and eventually retained or destroyed. Treat data management as integral to research quality, not as administrative overhead.
• Instruments must be culturally adapted. Do not generate generic instruments that ignore the cultural context of the field site and participant population. Question wording, interview format, observation focus, group composition, and data recording methods must all reflect the specific context. Flag when instruments need translation, visual adaptation, or procedural modification for the target population.
• Pilot testing should be standard practice. Encourage pilot testing for all instruments. Untested instruments risk wasting limited fieldwork time on questions that do not work, observation categories that miss the action, or recording procedures that fail in the field. Help users design realistic pilot testing plans even when time and access are constrained.
• Route IRB and consent tasks to the correct skills. When a user's request shifts from instrument design to IRB protocol writing or consent form drafting, redirect to the appropriate skill. This skill covers the data collection system; the irb-protocol and informed-consent skills cover the regulatory and ethical documentation that accompanies it.

Common Failure Modes

Failure mode	Prevention
Interview questions that are leading or closed-ended	Write open-ended questions that invite the participant's own framing; include probe sets for follow-up rather than embedding assumptions in the main questions
Observation protocols that are too rigid for ethnographic work	Design protocols that specify domains and recording formats but leave room for emergent attention; distinguish structured observation checklists from ethnographic observation guides
Sampling plan that is unjustified or arbitrary	Name the sampling approach, specify selection criteria, justify the target sample size using information power or theoretical saturation reasoning, and include contingency plans
No data management plan accompanying instruments	Require a data management component for every protocol; treat storage, transcription, de-identification, backup, and retention as inseparable from instrument design
Instruments not adapted to the field context or population	Review instruments against field site characteristics, participant literacy and language, cultural communication norms, and power dynamics; flag where adaptation is needed
No pilot testing plan before full data collection	Design a pilot testing protocol specifying what will be tested, with whom, how feedback will be collected, and what criteria will trigger instrument revision
Training plan absent for team-based research	Include a training component whenever multiple researchers will use the instruments; cover instrument familiarization, practice sessions, inter-rater alignment, and reflexivity
Field notes that lack structure or reflexivity	Provide field note templates with header metadata, running description sections, analytic memo space, and reflexivity prompts — not just blank pages

Examples

Example 1: Complete interview guide for health research

Input: "I need an interview guide for my study of community health workers' experiences navigating between biomedical and traditional healing systems in rural Guatemala. I'll be doing 25-30 semi-structured interviews in Spanish and K'iche', audio-recorded. Participants are mostly women with varying literacy levels."

Output approach:

• Load data-collection-methods-guide for interview design; load protocol-design-guide for sampling and pilot testing
• Design a semi-structured interview guide with four to five thematic domains: (a) professional trajectory and training, (b) daily practices and routines, (c) encounters between biomedical and traditional approaches, (d) community relationships and trust, (e) challenges and aspirations
• Each domain gets an opening grand-tour question, three to four specific probes, and follow-up prompts
• Opening sequence designed to establish rapport through narrative: "Tell me how you came to do this work"
• Questions worded to avoid biomedical framing bias; probe sets allow exploration of participant's own categories and frameworks
• Include bilingual considerations: key concepts that may not translate directly, space for K'iche' terms and participant explanations
• Add a sampling plan: purposive sampling stratified by community and experience level, with snowball recruitment through health center networks
• Design pilot testing: three pilot interviews across communities, debrief with local research assistant on question clarity and cultural appropriateness, revise before full data collection

Example 2: Observation protocol for educational ethnography

Input: "I'm studying how teachers in an urban middle school negotiate standardized testing pressures with culturally responsive pedagogy. I'll be doing participant observation in four classrooms over six months. I need an observation protocol and field note system."

Output approach:

• Load data-collection-methods-guide for observation design and field note systems
• Design a semi-structured observation protocol with key domains: (a) instructional practices and curriculum choices, (b) teacher-student interactions and discourse patterns, (c) assessment-related activities and talk, (d) cultural responsiveness markers, (e) institutional context (announcements, testing schedules, administrative intrusions)
• Include spatial and temporal sampling strategy: observe each classroom at varied times across the week; rotate between focused observation sessions (single domain) and holistic sessions (full classroom ecology)
• Design a field note template: header (date, classroom, period, activity type, number of students), running description section (chronological narrative), thematic annotations (linked to observation domains), analytic memo section (emerging patterns and questions), reflexivity prompt ("How did my presence affect the classroom today? What did I attend to and what might I have missed?")
• Include guidance on positioning: where to sit, when to participate versus observe, how to handle students' questions about the researcher's role
• Design a pilot observation week: observe each classroom once using the protocol, assess whether domains capture the relevant action, revise before systematic data collection begins

Example 3: Multi-method protocol for community-based research

Input: "I'm designing a community-based participatory research project on food sovereignty with three Indigenous communities in the Pacific Northwest. We'll use interviews, focus groups, and participant observation. I need the full protocol: instruments, sampling, training for community co-researchers, data management, and pilot testing."

Output approach:

• Load all three reference files for complete protocol design
• Design three coordinated instruments:
  • Semi-structured interview guide for individual food sovereignty narratives (20-25 per community): land and resource access, food practices and knowledge transmission, institutional relationships, visions for food sovereignty
  • Focus group guide for community-level discussion (two per community, six to eight participants each): collective food history mapping exercise, current challenges and assets discussion, priority-setting activity, action planning
  • Participant observation protocol for food-related community events, gardens, markets, and gatherings: spatial and social organization, knowledge sharing practices, resource distribution, institutional interactions
• Sampling plan: purposive sampling within each community using maximum variation on age, gender, and role (elder, youth, food producer, community leader); recruitment through community advisory boards; target justified through information power framework
• Training plan for community co-researchers: instrument familiarization workshops, paired practice interviews with feedback, observation practice at community events, reflexivity discussions, ongoing peer debrief sessions
• Data management plan: community data sovereignty provisions (CARE principles), encrypted storage with community-controlled access, verbatim transcription with community review, de-identification protocols that protect individual identity while preserving community attribution where desired, backup to community-designated secure location, retention and destruction governed by community research agreement
• Pilot testing: pilot one interview and one focus group in each community, debrief with community advisory boards, revise instruments based on feedback before full data collection