aamc.org does not support this web browser.

Write Better Questions

When working with AI platforms, the quality of your questions directly affects how helpful the responses will be. Think of it like the difference between asking a new employee "Can you help with admissions?" versus giving them specific tasks and context about your needs. Clear, specific questions lead to more valuable insights by:

  1. Better understanding. Clear questions lead to clearer explanations, much like giving specific instructions to a colleague.
  2. Finding connections. Structured questions help AI identify patterns and relationships across different AAMC resources.
  3. Quality checks. Detailed questions enable AI to spot potential gaps or inconsistencies in your understanding.

Original (Less Effective) Prompt

"I need help understanding AI terminology for medical school admissions. Can you explain the terms from the AAMC documents?"

Improved Prompt

"I am attaching these AAMC documents about implementing AI in medical education selection:

'AAMC Principles for Responsible AI.pdf' 
'Essential AI Terms and Definitions for Implementing AI in Selection.pdf' 
'Guide to Assessing Your Institution's Readiness for Implementing AI in Selection.pdf'
'Guide to Evaluating Vendors on AI Capabilities and Offerings.pdf'

Using the 'Essential AI Terms and Definitions for Implementing AI in Selection.pdf,' please:
     1. Provide a specific example of how each term applies to medical school admissions.
     2. Explain how understanding this term helps support human decision-making in the selection process.

Requirements:
     o Keep explanations clear and accessible for admissions and selection staff.
     o Use consistent terminology from the AAMC document.
     o Focus on practical applications in medical education.”

Why it's better

  • Names specific source document.
  • Provides structured, numbered requests.
  • Clearly states requirements.
  • Asks for practical examples.
  • Uses consistent AAMC terminology.

Write Better Questions (Follow-up)

Sometimes your initial question may not yield all the information you need, or you might want to delve deeper into a specific aspect of the AI platform's response. Crafting effective follow-up questions ensures you extract more targeted, practical insights tailored to your institution's selection process. This iterative approach helps clarify ambiguities and fill in gaps, leading to better-informed decisions.

Original (Less Effective) Prompt

"How does model complexity affect accuracy?"

Improved Follow-up

"Your explanations of model complexity and predictive accuracy were helpful. I’d like to delve a bit deeper:

1. Relationship in Admissions: Can you clarify how model complexity and predictive accuracy specifically relate in medical school admissions, using examples or citations from AAMC definitions?

2. Tradeoffs: Could you outline the tradeoffs between using more complex models and achieving high accuracy? Please include any practical scenarios or potential pitfalls we might encounter.

3. Interpretability: How might these tradeoffs affect our ability to interpret applicant results? What practical strategies or adjustments could help mitigate these challenges?

Please provide detailed reasoning for each point, and if there are any limitations or additional considerations, include those as well.”

Why it's better

  • Breaks the follow-up into clear, actionable questions.
  • Requests AAMC citations and practical examples for grounding the response.
  • Prompts deeper explanations by addressing uncertainties with detailed rationale.
  • Specifies needed details, reducing ambiguity and aligning with your selection process needs.