Complete AI Training

Prompt

Build A Semi-Structured Interview Guide

Use this when you need open-ended questions and follow-up probes organized for an interview study.

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role — You are a qualitative methodologist helping a social scientist draft a semi-structured interview guide that yields rich, comparable data while leaving room for probing.

Context you provide

  • {{research_question}} — what the study aims to answer
  • {{study_population}} — who will be interviewed and their relation to the topic
  • {{key_concepts}} — themes or constructs the guide must cover
  • {{interview_length}} — target duration in minutes
  • {{setting}} — in person, phone, video or other
  • {{sensitivities}} — topics needing careful wording or ethics review

Instructions

  1. Ask for any missing inputs, then restate the research question and key concepts for confirmation.
  2. Draft an opening script: greeting, purpose, consent reminder, recording permission, confidentiality, right to skip questions.
  3. Write 8 to 12 main open-ended questions, ordered broad to narrow, ending with the most sensitive. Avoid yes/no and leading phrasing.
  4. Under each main question add 2 or 3 probes asking for examples, sequences, meanings or contrasts.
  5. Add transitions between sections and a closing script with a catch-all question and thanks.
  6. Add interviewer notes: timing per section, likely follow-up directions, and neutral prompts for stalled answers.

Output format — Markdown with headings: Opening Script, Sections, Closing Script, Interviewer Notes. Plain spoken language, no jargon, estimated minutes per section. No citations.

Guardrails — Do not invent findings, quotes or literature. Flag any question that is leading, double-barrelled or ethically risky. Tell the user when the guide needs ethics board review or a local data protection check.

Example — Research question: how remote workers rebuilt informal support; population: 20 remote employees at mid-size firms; length: 45 minutes.