In the fast-evolving world of artificial intelligence (AI), a new term has also emerged, generating many conversations, short courses, and article posts. This term is prompt engineering, and it is an essential skill when interacting with any AI. The ability to craft effective prompts—clear, purposeful, and specific—plays a key role in using AI to its full potential.

At a closer glance, the concept of prompt engineering aligns closely with the engineering design process described in the Next Generation Science Standards (NGSS). This blog post explores the significance of prompt engineering, its parallels with the engineering design process and practices outlined in the NGSS, and the characteristics of well-crafted prompts.

What Is Prompt Engineering?

Prompt engineering refers to the practice of crafting input questions, queries, or instructions that are used with AI systems to guide their outputs. The quality of a prompt helps to generate and direct the response that is provided. A well-constructed prompt steers AI responses toward desired outcomes, while a fuzzy or unclear prompt may result in a less-than-informative response. When using large language models like ChatGPT, Gemini, and other AI tools, the quality of prompts significantly impacts the relevance, accuracy, and clarity of the responses.

Imagine asking an AI tool, “Write an essay on climate change.” While this prompt is functional and a direct query, it is too broad, leaving room for vague, rambling, or unfocused responses. Compare that to the following: “Write a 500-word explanation for high school students explaining the main causes of climate change, using three examples supported by the scientific explanation as to how each example contributes to climate change.”

The latter prompt provides parameters, criteria, and limitations akin to specifications in how an engineering prompt might be written. This level of detail ensures the AI delivers a more tailored and meaningful output.

Characteristics of Effective Prompts

Crafting effective prompts requires an understanding of both the task at hand and AI’s capabilities. For example, some AI programs focus more on the design of a graphic output, while others focus more on narrative outputs. Below are key characteristics of good prompts.

1. Clarity. A clear prompt eliminates ambiguity. When asking an AI to generate information, specificity is paramount. Often in this area, the term that is used at the start is very similar to terms used in developing objectives from Bloom’s Taxonomy. For example,

• Unclear. “Explain photosynthesis.”
• Clear. “List the steps in the process of photosynthesis in simple terms suitable for an eighth-grade science class.”

The second example is a lower-level output asking for the steps. The user could change the desired output of a prompt (not necessarily this one) with terms like compare and contrast, visualize, etc.

2. Context and Scope. Context and Scope ensure the AI understands the focus and depth of detail to include. Other attributes in this category include information about the audience that the prompt is to be written for, the format of the output, or the purpose.

• Instead of “Tell me about the planets,”
•  “Develop a concise, 200-word description of the eight planets in the solar system for a middle school astronomy textbook. Make and explain connections among planets, including key concepts such as patterns, types, and location in the solar system. Ensure the description encourages students to analyze and interpret the data to make comparisons.”

3. Constraints, Parameters, or Limitations
Adding constraints, such as word limits, formatting styles, or perspectives, helps refine responses. Parameters or limitations will set the boundaries for the response. For instance,

• “Write a persuasive essay about renewable energy, 200–300 words long, focusing on solar and wind power, and conclude with a call to action.”

4. Purpose-Driven
Purposeful prompts align the output with specific goals. For example,

• Academic Goal. “Generate a 10-question quiz for high school students about Newton’s Laws of Motion.”
• Creative Goal. “Write a nonfiction short story, using vivid imagery and dialogue, about a young scientist discovering a new planet.”

As you read and learn more about prompt engineering, you will find that multiple sources indicate that a prompt should have a certain number of words or contain a certain number of categories. Ultimately, the length of the prompt will depend on the required task.

The Connection to the NGSS Engineering Design Process

Appendix I of the NGSS highlights the engineering design process (EDP), which encourages students to solve real-world problems through structured, iterative approaches. The EDP includes several stages.

1. Defining the Problem. Clearly understanding the challenge and constraints.
2. Developing Solutions. Brainstorming and prototyping potential solutions.
3. Optimizing Solutions. Testing, refining, and improving the design based on feedback.

The Prompt Engineering Process has similar stages.

1. Determine the Task. Clearly understanding the task that is being asked, as well as the parameters and limitations desired in an answer. (Input)
2. Considering the Response. Examining the response provided by AI and if the answer meets the needs. (Output)
3. Optimizing the Output. Asking AI for clarification, modification, and alterations to the output if necessary, based on needs. (Refinement or Revision)

Prompt engineering involves defining the clear problem to be tackled along with the constraints. Designing AI prompts incorporates the thinking process, and ultimately, the result produced by AI. The final step is iterating based on the AI’s responses by modifying or including more clarifications or characteristics. By framing prompt creation as a process of iterative design, educators and learners can view it not merely as a technical skill, but also as a cognitive tool for critical thinking and problem solving.

For example, a student practicing the EDPs in science might craft a question and modify that question as data is added. Similarly, users who learn prompt engineering can better refine questions to elicit better outputs, reflecting an iterative learning process. Both activities demand clarity of purpose, attention to detail, and modification based on information—practices that are important in both science and AI.

Prompt Engineering Meets the Classroom

Educators can harness the principles of prompt engineering to transform teaching and learning. Here’s how.

1. Modeling the Engineering Design Process
   Teachers can introduce prompt engineering as a classroom activity aligned with the NGSS EDP. For instance, students working on a project about renewable energy can use AI to first brainstorm ideas, then refine their prompts iteratively, much like testing and improving prototypes in the EDP.
2. Encouraging Critical Thinking
   Prompt engineering inherently requires students to think critically about their objectives. If an AI-generated response doesn’t meet expectations, students analyze why and refine their prompts, which cultivates problem-solving skills.
3. Fostering Creativity
   By asking students to design creative prompts, educators can encourage them to explore novel ideas. For example, a prompt like “Imagine you’re an engineer designing a bridge for a city prone to earthquakes. Describe your design in detail” merges creativity with real-world application.

For additional connections to three-dimensional instruction, check out the October issue of this blog series.

Conclusion
Prompt engineering is more than a technical exercise: It bridges human ingenuity and AI potential. Its alignment with the NGSS EDP underscores its value as a pedagogical tool that fosters critical thinking, creativity, and problemsolving. By crafting prompts with clarity, specificity, constraints, and purpose, students and professionals alike can harness AI’s capabilities to achieve meaningful outcomes. In teaching this art, we prepare the next generation to work with AI and lead it toward a future of limitless possibilities.

Christine Anne Royce, Ed.D., is a past president of the National Science Teaching Association and currently serves as a Professor in Teacher Education and the Co-Director for the MAT in STEM Education at Shippensburg University. Her areas of interest and research include utilizing digital technologies and tools within the classroom, global education, and the integration of children's literature into the science classroom. She is an author of more than 140 publications, including the Science and Children Teaching Through Trade Books column.
 

Valerie Bennett, Ph.D., Ed.D., is an Assistant Professor in STEM Education at Clark Atlanta University, where she also serves as the Program Director for Graduate Teacher Education and the Director for Educational Technology and Innovation. With more than 25 years of experience and degrees in engineering from Vanderbilt University and Georgia Tech, she focuses on STEM equity for underserved groups. Her research includes AI interventions in STEM education, and she currently co-leads the Noyce NSF grant, works with the AUC Data Science Initiative, and collaborates with Google to address CS workforce diversity and engagement in the Atlanta University Center K–12 community.
 

Note: This article is part of the blog series From Chalkboards to AI, which focuses on how artificial intelligence can be utilized within the classroom in support of science as explained and described in A Framework for K–12 Science Education and the Next Generation Science Standards. 

The mission of NSTA is to transform science education to benefit all through professional learning, partnerships, and advocacy.