Course Learning Outcomes
Overview
Most instructors have a clear sense of what they want to teach. But there’s an equally important question to answer before the semester begins: What do you want students to be able to do once the course is over?
Course learning outcomes answer that question. A learning outcome is a concise, measurable statement describing what students should be able to demonstrate by the end of a course. Unlike broad goals or general objectives, learning outcomes name a specific, observable skill or ability—something you and your students can actually see and assess.
Writing clear outcomes before you finalize your syllabus pays off throughout the semester. Outcomes can guide your selection of readings, activities, assignments, and rubrics. They help students understand what success looks like and take ownership of their learning. And they create a shared reference point you can return to as the course evolves.
Research shows when instructors communicate learning outcomes clearly, students can better approach their learning (Bembenutty, 2011) and make more informed decisions about how to engage with course material (Wiggins & McTighe, 2005). Transparency about expectations benefits both sides of the classroom.
Most courses work well with five to ten outcomes which is enough to capture the course’s key takeaways without overwhelming students or locking you into an unworkable checklist.
In This Guide
- How to write a good learning outcome
- Using Bloom’s Taxonomy to calibrate expectations
- Writing outcomes in an AI environment
- Aligning outcomes with your assignments and program
- Practices that deepen student engagement with outcomes
- Practice in action
How to Write a Good Learning Outcome
Effective learning outcomes share two features: they use action verbs, and they describe something observable.
A useful sentence frame is: By the end of this course, students will be able to [action verb] [specific skill or knowledge].
The action verb is where many instructors get stuck. Words like “understand,” “learn,” or “appreciate” are common in course descriptions, but they describe internal states that are hard to observe or assess. How do you know when a student truly “understands” something?
More precise verbs make the expectation visible. Compare:
- Students will understand political ideologies.(hard to assess)
- Students will compare and contrast US political ideologies regarding social and environmental issues.(observable, assessable)
The second version names a concrete intellectual action, which you can design an assignment around that students can prepare for.
Using Bloom’s Taxonomy to Calibrate Expectations
Bloom’s Taxonomy organizes cognitive skills from foundational to complex, and its associated action verbs are a practical tool for writing outcomes at the right level for your course. The table below also notes where generative AI is capable of performing each level to assist you in designing outcomes and assessments.
| Level | What It Involves | Example Verbs | AI & Implications for Outcomes |
|---|---|---|---|
| Remembering | Recalling facts and basic concepts | define, list, recall, identify | High AI capability. AI can retrieve and state factual content reliably. Embed recall-level outcomes within larger tasks that require student judgment or context. |
| Understanding | Explaining ideas or concepts | explain, summarize, classify, compare | High AI capability. AI can produce fluent explanations and summaries. Strengthen outcomes by specifying the context or disciplinary lens students must bring to the explanation. |
| Applying | Using knowledge in new situations | apply, calculate, solve, demonstrate | Moderate AI capability. AI handles routine or formulaic application well. Outcomes are more durable when they require students to apply knowledge to specific, situated problems AI cannot fully anticipate. |
| Analyzing | Breaking down information to examine relationships | analyze, differentiate, distinguish, categorize | Moderate AI capability. AI can identify patterns but struggles with nuanced disciplinary judgment. Outcomes that require students to justify their reasoning process, not just name a conclusion, are stronger. |
| Evaluating | Making judgments based on criteria | evaluate, justify, recommend, critique | Limited AI capability. Genuine evaluation draws on values, disciplinary standards, and lived experience that AI lacks. These outcomes tend to be the most AI-resilient. |
| Creating | Producing something new | design, formulate, develop, integrate | Limited AI capability for original work. AI can generate drafts, but cannot produce work grounded in a student’s own disciplinary perspective or situated experience. Specify what makes the work distinctively the student’s. |
An introductory course might emphasize remembering, understanding, and applying. An advanced seminar might focus on analyzing, evaluating, and creating. Neither is better. What matters is that the level of your outcomes matches what you actually teach and assess.
Here’s what calibrated outcomes look like across levels, using a nutrition course as an example:
- Remembering: Learners will recall federal nutritional guidelines for planning meals.
- Applying: Learners will apply food safety principles to evaluate quality management decisions in a real-world case scenario.
- Evaluating: Learners will recommend a meal plan for a specific patient profile and defend their reasoning by drawing on course frameworks.
- Creating: Learners will integrate knowledge of metabolism, cultural food practices, and chronic disease to design an individualized nutritional therapy plan.
Notice how the higher-level outcomes specify context, judgment, or perspective that students must bring themselves. This makes them meaningful whether or not students have access to AI tools.
Writing Outcomes in an AI Environment
Generative AI can now perform many tasks that course assignments have traditionally asked students to do: drafting explanations, recalling facts, summarizing texts, solving standard problems. This doesn’t make Bloom’s Taxonomy obsolete, but it does raise a useful question to ask about every outcome you write:
Does this outcome require students to develop something that AI cannot produce on their behalf?
If a student could submit AI-generated work that genuinely meets the outcome, then that’s a signal worth paying attention to. It doesn’t necessarily mean the outcome is wrong, but it may mean you need to reconsider how it’s assessed.
A few practical moves:
- Anchor lower-order outcomes in larger tasks. Remembering and understanding still matter: foundational knowledge is real. But consider embedding those outcomes in assignments that require students to apply, critique, or extend that knowledge in ways specific to your course context.
- Specify what makes the work the student’s own. Outcomes like “students will analyze” or “students will create” become more durable when they name the disciplinary lens, the specific data set, the community context, or the personal or professional experience that students must draw on.
- Distinguish product from process. AI can produce a polished draft. It cannot demonstrate the thinking that led there. Outcomes focused on students’ reasoning process such as how they made decisions, what they considered and rejected, why they arrived at a conclusion, are difficult for AI to fulfill.
Research suggests the risk cuts both ways: AI tends to support deeper learning when students engage critically with its output, but research suggests that unstructured or prolonged AI use can lead students toward passive acceptance of AI-generated content, which may limit the development of higher-order thinking skills (Zhao et al., 2025). Outcomes that require students to interrogate, extend, or situate AI-generated content can turn this dynamic into a learning opportunity.
For more on designing courses and assessments in an AI environment, see CATLR’s Teaching with AI resources at learning.northeastern.edu/ai.
Aligning Outcomes with Assignments and Program Goals
Learning outcomes are most useful when they connect to everything else in your course. This is the principle behind backward design: once you know what students should be able to do, you can work backward to choose assessments, activities, and materials that build toward those outcomes (Wiggins & McTighe, 2005).
Consider a business course with this outcome: Learners will collaborate effectively on a team to create a marketing campaign for a specific regional client. A well-aligned course would teach collaboration and campaign development explicitly, give students structured opportunities to practice both, and include assignments that offer feedback on how well students are meeting the outcome. The specificity of “a specific regional client” also makes the work harder to outsource entirely to AI.
When outcomes are aligned, students see a coherent course where the readings, discussions, and assignments all point in the same direction.
It’s also worth checking whether your outcomes connect to your program’s broader goals. Understanding how your course fits into the program sequence helps you calibrate what students are bringing in and what they’ll need for future courses. If your course carries NUpath attributes, confirming alignment with those outcomes ensures students achieve the learning breadth expected at Northeastern.
Practices That Deepen Student Engagement with Outcomes
Share outcomes early and revisit them often. Include course learning outcomes in your syllabus so students can refer to them throughout the semester. Before introducing a new concept or assigning a major project, point back to the relevant outcome. This helps students understand why they’re doing what they’re doing and how individual assignments connect to the larger arc of the course (Cuevas, Matveev, & Miller, 2010).
Consider co-creating outcomes with students. When course structure allows, spend time on the first day of class by having students to weigh in on or even help draft learning outcomes. Such an activity can significantly increase motivation and investment. Students who have a voice in shaping expectations are more likely to take ownership of meeting them.
Treat outcomes as living documents. Revisit your outcomes each time you revise the course. If you’ve changed major assignments, updated course content, or shifted the focus of the course, check that your outcomes still reflect what you’re actually teaching. This is especially worth doing now: as AI tools change what students can produce and how they engage with course material, outcomes that were appropriate two or three years ago may need rethinking.
Check the Practice in Action page to see examples of learning outcomes across disciplines.
References and Additional Reading
Anderson, L. W., & Krathwohl, D. R. (2001). A Taxonomy for learning, teaching and assessing: A Revision of Bloom’s Taxonomy of Educational Objectives. New York: Longman.
Bembenutty, H. (2011). Self-regulation of learning in postsecondary education. New Directions for Teaching and Learning, 126, 3-8.
Cuevas, N. M., Matveev, A. G., & Miller, K. O. (2010). Mapping general education outcomes in the major: Intentionality and transparency. Peer Review, 12(1), 10-15.
Gonsalves, C. (2024).Generative AI’s impact on critical thinking: Revisiting Bloom’s taxonomy. Journal of Education for Business, 99(4–5).
Wiggins, G., & McTighe, J. (2005). Understanding by Design (Expanded). Alexandria, US: Association for Supervision & Curriculum Development (ASCD).
Zhao, Y., Yue, Y., Sun, Z., Jiang, Q., & Li, G. (2025). Does generative artificial intelligence improve students’ higher-order thinking? A meta-analysis based on 29 experiments and quasi-experiments.Understanding by Design Journal of Intelligence, 13(12), 160.
Interested in exploring further? Check out all of our CATLR Teaching Tips or upcoming events. Then meet with a CATLR consultant to discuss your questions and ideas! To schedule a consultation, email [email protected] or call +1.617.373.3157.