Apply
Group of people in a computer lab

Strategies for Constructing Effective Prompts

  1. U of R Home
  2. Generative AI Hub
  3. Strategies for Constructing Effective Prompts

Prompt writing is crucial to the effective use of GenAI for inquiries and learning activities. Prompting is the process of instructing GenAI; it is the skill and practice of writing inputs (“prompts”) for a GenAI tool.

The CLEAR framework* provides a useful guide for constructing effective prompts.

CLEAR stands for:

Concise, Logical, Explicit, Adaptive, Reflective

Concise Prompts

Clear and precise prompts guide the GenAI model’s understanding leading to a more relevant response.

Intial Prompt

"Explain the universe."

Improved Prompt

"Provide an overview of the Big Bang theory and its significance in understanding the origin and expansion of the universe."

Limiting each prompt to one request or question is recommended so as not to overwhelm the tool.

Logical Prompts

Providing context within prompts helps the GenAI model create meaningful output.

Intial Prompt

“Tell me about programming”

Improved Prompt

“Explain the concept of loops in programming languages and their importance in executing repetitive tasks”

To provide context, try defining the role you would like the GenAI tool to assume or describe your role or level of knowledge. For example:

  • You are a first-year student of computer science.
  • Act as an expert in programming languages.
  • Assume I am 10 years old.
  • Explain the concept of loops in programming languages.
Explicit Prompts

Including a specific structure or style in your prompt may assist the GenAI tool in meeting user expectations in terms of the format and length of the generated responses.

Intial Prompt

“What are the causes of climate change?”

Improved Prompt

“List and explain in 10 sentences, the primary human-induced factors contributing to climate change, including the impact of carbon emissions and deforestation? Use lay terms”.

To help you get more accurate results, consider specifying the following in your prompts:

  • Background information related to the topic (what AI can base its response on)
  • Length of response (number of sentences, paragraphs, words, or characters)
  • Format (a list, paragraph, chart)
  • Structure (essay, blog post, poem)
  • Style (formal or informal)
  • Target audience (who this is intended for)
  • Examples (provide an example of what you are looking)
Adaptive Prompts

Experiment with general and then more specific questions or statements to assist the GenAl model in generating a response that matches the desired output.

Intial Prompt

“How does technology affect society?”

Improved Prompt

“Discuss both the positive and negative impacts of technology on various aspects of society, including education, healthcare, and social interactions.

Reflective Prompts

Reflect on the steps you followed and learn from your experience to evaluate and improve the accuracy, coherence, and utility of the generated responses.

Intial Prompt

Any

Improved Prompt

Identify what could be refined in your prompt and try again.

  • Rephrase your initial prompt.
  • Give GenAI more information.
  • Change your approach.

Meta-prompting

Another technique to improve your prompt writing is by using meta-prompting techniques, which is especially helpful when you are not sure what you need to ask.

For example: "I want to learn about computer programming, but I don't know where to start. Can you ask some questions to help me figure out where I need to start and what questions I need to ask first?"

*Created by Dr. Leo Lo, Dean of Libraries at the University of Virginia, in the article “The CLEAR path: A framework for enhancing information literacy through prompt engineering". The Journal of Academic Librarianship, 49(4), 2023.


 

The section on prompting was adapted from “Maximizing learning with effective GenAI prompt writing: An instructor’s resource for students” by Pletnyova, A., Anselmo, L. and Keyhani, M., Taylor Institute for Teaching and Learning, University of Calgary, licensed under CC BY-NC-SA 4.0.