How to Assess Student Learning Without Relying on Traditional Tests

Tests are a useful assessment tool. They're not the only one, and for many learning goals, they're not the best one. A student who can identify a literary device on a multiple-choice test may or may not be able to use it in their own writing. A student who can write a correct equation may or may not understand what the equation models. Tests measure recall and recognition well. They measure many other things poorly.

Teachers who rely primarily on tests for assessment have a limited view of what their students know and can do. Alternative assessments — used alongside or instead of tests — fill in the picture.

The Question Before the Tool

Before choosing an assessment, answer one question: what do I actually want to know? Not what can I easily grade, but what evidence would tell me whether students understand this.

For some goals, a well-designed test question is the right evidence. For others, the evidence would be a piece of student-created work, a verbal explanation, an observable performance, or a student-generated product. Matching the assessment to the actual learning goal produces better information than defaulting to whatever format is easiest to administer.

Performance Tasks

A performance task asks students to demonstrate knowledge or skill by applying it to a real or realistic problem. The key word is "apply" — not recall, but use.

Examples:

• Students design an experiment to test a claim, not just recall steps of the scientific method
• Students write to an authentic audience (a letter to a local official, a how-to guide for a younger class) rather than a generic essay
• Students solve a real-world math problem that requires selecting the appropriate operation, not just practicing a known procedure
• Students create a museum exhibit about a historical period rather than summarizing facts

Performance tasks produce evidence of transfer — whether students can use what they've learned in a new context. This is the most important evidence for many learning goals, and tests rarely capture it.

Portfolios

A portfolio is a curated collection of student work over time, with student reflection on what the collection shows. The reflection is as important as the work itself — it requires students to evaluate their own learning, which is a high-order thinking skill that most assessments don't demand.

Portfolios are particularly useful for showing growth over time (which a snapshot test can't do), for capturing learning goals that are difficult to test (creative thinking, voice in writing, problem-solving approach), and for giving students agency in demonstrating what they've learned.

The challenge of portfolios is the evaluation burden. Clear criteria for what a strong portfolio looks like and what reflections demonstrate genuine insight reduce the subjectivity.

Observation and Conference

Two of the most information-rich assessments available — and two of the most underused — are observation during work time and brief one-on-one conferences.

Observing a student as they work tells you things a test never could: where they hesitate, what they do when they're stuck, whether they self-correct, how they handle ambiguity. A teacher who circulates deliberately during independent work, asking a few targeted questions, gathers formative data that shapes instruction in real time.

A brief conference — three to five minutes with a student, asking them to explain their work — surfaces understanding and misconceptions faster and more accurately than most written assessments. "Walk me through your thinking" produces more information about what a student understands than any multiple-choice format.

The limitation is time: you can't conference every student every day. Build these into your practice as targeted, regular events rather than occasional additions.

Projects and Products

Student-created products — models, videos, podcasts, artwork, research papers, websites — can demonstrate a wide range of learning goals that don't fit the test format. The product is tangible evidence that the student can apply knowledge in a sustained, complex task.

The key to making projects work as assessment is designing them so the evidence connects clearly to the learning goal. A beautiful poster about ecosystems that required only copying and design work is an art project. A poster that required students to research, synthesize information, make decisions about what to include, and explain their choices in a brief presentation is a genuine assessment.

Keeping Grades Connected to Learning

When you shift toward alternative assessments, you face a grading challenge: how do you assign grades to things like portfolios, projects, and conferences in a way that is fair, defensible, and meaningful?

The answer is explicit criteria applied consistently. A rubric that describes what excellent work looks like, what adequate work looks like, and what inadequate work looks like gives students clear targets and gives you a defensible basis for your grades. The rubric should be shared before the assessment, not after — students need to know what they're aiming for.

Your Next Step

Identify one upcoming unit where your primary assessment is a test. Ask: what does the test measure, and are there things I want students to know or be able to do that the test won't capture? If yes, design one additional assessment — not more work, just a different window into the learning. A brief performance task, a reflection, a conference with a small group. One additional data point gives you a meaningfully fuller picture of what students understand.