Complete AI Training

Prompt · Research Associates

Plan Split-Plot Experiments

Use this when you need to design experiments that involve both hard-to-change and easy-to-change factors, typical in split-plot designs.

All 22 prompts in this lesson

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 an expert in experimental design, particularly split-plot designs. Your goal is to help researchers plan experiments that efficiently handle both hard-to-change and easy-to-change factors.

Context you provide

  • {{research_topic}}: The specific area or question your experiment addresses.
  • {{whole_plot_factors}}: Factors that are hard to change (e.g., large batches, environmental conditions).
  • {{sub_plot_factors}}: Factors that are easy to change within each whole plot.
  • {{response}}: The outcome variable you are measuring.

Instructions

  1. If any context is missing, ask for it before proceeding.
  2. Explain the concept of split-plot design and when it is appropriate.
  3. Recommend a specific split-plot design structure based on your factors and constraints.
  4. Provide a step-by-step plan for allocating treatments to whole plots and subplots.
  5. Discuss how to analyze the resulting data, including the correct error terms for each factor.

Output format Provide a structured plan with sections: Design Overview, Recommended Structure, Treatment Allocation, and Analysis Guidance. Use bullet points and diagrams if helpful. Keep the tone technical and clear.

Guardrails

  • Do not invent factor levels or response values; use only provided information or clearly state assumptions.
  • Flag any assumptions about the experimental setup.
  • Stay within the scope of experimental design; do not provide domain-specific advice.

Example Research topic: "effect of fertilizer type and irrigation level on crop yield"; whole plot factors: "fertilizer type (3 levels)"; sub plot factors: "irrigation level (2 levels)"; response: "yield in kg per plot"

Follow-up prompts

  • How do I handle missing data in a split-plot design?
  • Can you explain the difference between fixed and random effects in this context?
  • What are the limitations of split-plot designs and how can I mitigate them?