Creating a Performance vs Learning Matrix

A Practical Guide for Educators

Image generated with MidJourney

In my previous post, I discussed the tension between performance gains and learning outcomes when integrating AI in education. The research is clear: unrestricted AI use poses risks to learning, while thoughtfully implemented AI can enhance performance without compromising essential skill development. But how do we move from understanding this challenge to actually implementing solutions in our classrooms?

The Performance vs Learning Matrix is a practical tool that helps educators make intentional decisions about AI integration. Through careful analysis of various academic subjects and ongoing discussions with educators, I've developed a structured approach that any educational team can use to create their own matrix.

Why Create a Matrix?

Before diving into the "how," let's address the "why." When schools attempt to implement AI policies without careful consideration of learning outcomes, they often fall into one of two traps: either completely banning AI (which doesn't prepare students for the future) or allowing unrestricted AI use (which risks significant learning loss). A well-developed matrix helps navigate between these extremes by:

• Clearly identifying which skills require independent mastery

• Defining appropriate AI use for specific tasks

• Creating consistent expectations across grade levels and courses

• Providing a framework for teaching AI skills intentionally

• Supporting thoughtful technology integration

The Matrix Development Process

Creating a Performance vs Learning Matrix is a significant undertaking that requires thoughtful consideration and collaborative effort. However, when approached systematically, the process helps teams think deeply about their curriculum while creating clear guidelines for AI integration. The work naturally falls into three main steps, each building upon the previous one to create a comprehensive framework for decision-making around AI use in your course or department.

Here is a three-step approach for creating an effective matrix. Let's walk through each step.

Step 1: Categories and Skills Definition

The first step is to meet as a department or course team to establish your matrix's framework. This meeting typically takes 2-3 hours, but it's time well spent. However, the efficiency of this process largely depends on your team's previous work in clarifying learning outcomes. If your department or course team has already done the important work of defining clear learning outcomes, you'll find this process much more straightforward. If not, expect this process to take longer - possibly requiring multiple sessions. The work of defining learning outcomes typically requires extended, inclusive conversations and careful consideration of educational objectives. Without clarity on what students need to learn and master, it becomes nearly impossible to make informed decisions about how AI might support or diminish these goals.

Whether you're starting with established outcomes or defining them as part of this process, begin by identifying the major categories that encompass your course content and skills. For example, in Algebra I, these categories include Mathematical & Problem-Solving Skills and Equation Solving. In World History, they include Historical Research & Analysis Skills, Historical Writing & Communication, and Project & Assessment Work.

The key is to be comprehensive while maintaining clear organization. Under each category, list specific tasks and skills students should master. Be specific - avoid vague actions like "understand" or "know." Instead, use observable actions that can be clearly assessed.

Step 2: The Learning vs Performance Decision

This is where the real work begins. For each skill or task in your matrix, your team needs to decide: Is the goal learning or performance? This isn't always an easy distinction, but it's crucial for determining appropriate AI use.

Learning goals involve skills students must master independently for future success. For example, in this sample English 9 matrix, writing thesis statements is a learning goal because it's a fundamental skill students need to develop independently. In contrast, formatting citations is a performance goal - once students understand citation principles, AI can help with the mechanical aspects of proper formatting.

After determining the goal, decide whether AI should be allowed and, if so, how it should be used. Document your reasoning - this helps maintain consistency and provides justification for your decisions.

Step 3: Planning AI Instruction

Here's where many schools stumble: they identify where AI is allowed but don't plan how to teach students to use it effectively. Looking at this Physics matrix, we see that AI is allowed for complex data analysis - but students won't inherently know how to use AI effectively for this purpose. It requires explicit instruction.

Review all tasks where AI is allowed and compile a list of AI skills you'll need to teach. For example, if AI is allowed for research support, students need to learn:

• How to write effective search prompts

• How to evaluate AI-generated sources

• How to verify information

• How to document AI assistance

• How to use AI ethically in research

Implementation Considerations

As you develop your matrix, keep these key points in mind:

1. Consistency is Crucial Your matrix should align vertically within departments and horizontally across grade levels. For example, if 9th grade English prohibits AI for thesis development, 10th grade shouldn't suddenly allow it without appropriate scaffolding.

2. Start Conservative It's easier to gradually allow more AI use than to pull back permissions once given. This Art matrix is particularly conservative with core skill development while allowing AI use for research and organization.

3. Plan for Evolution Technology changes rapidly. Your matrix should be a living document that evolves with both AI capabilities and your understanding of its educational impact.

4. Focus on Skills First Always begin with the skills students need to develop, then consider how AI might support or hinder that development. This Italian I matrix exemplifies this approach, protecting core language acquisition while allowing AI support for cultural exploration.

Moving Forward Together

Creating a Performance vs Learning Matrix takes time and thoughtful consideration, but it's a valuable investment in your students' future. It provides a framework for intentional AI integration that protects learning while leveraging the benefits of AI for appropriate tasks.

This is community work. I encourage you to explore the example matrices I've created and adapt them for your needs. Share your experiences, modifications, and insights. What categories have you added? How have you modified the implementation for your specific context? What AI skills have you identified as essential for your students?

Let's continue this conversation and learn from each other as we navigate this new educational landscape.

Thanks for reading Fullstack Educator! Subscribe for free to receive new posts and support my work.