Formative vs summative assessments: when to use each

Most teachers can define the difference between formative and summative assessment. Fewer teachers use them intentionally as distinct tools. Getting the distinction right — and designing for each purpose — is one of the highest-leverage things you can do for your grading workflow.

The core distinction

Formative assessment happens during learning. Its purpose is to diagnose what students understand so teaching can adjust. It's ongoing, often informal, and typically not graded (or weighted lightly).

Summative assessment happens after learning. Its purpose is to evaluate what students have mastered. It's high-stakes, formal, and weighted heavily in grading.

Both are assessments. Both can look like "tests." They serve very different goals and should be designed differently.

What good formative assessment looks like

Formative assessment succeeds when it surfaces specific misconceptions fast enough for you to act on them. Characteristics:

Short — a 5–10 minute check-in, not a 50-minute test
Low-stakes — ungraded or lightly weighted so students don't panic
Frequent — weekly or even daily
Actionable — reveals specific gaps, not just aggregate scores
Immediate — results available the same day so you can adjust tomorrow's lesson

Examples:

Exit tickets: 3–5 questions at the end of a lesson
Mid-unit quizzes: 10 questions covering the week's material
"Explain this concept in your own words" prompts
Think-pair-share discussions with observed outputs

What good summative assessment looks like

Summative assessment succeeds when it produces reliable, defensible evidence of what students have learned. Characteristics:

Comprehensive — covers the unit or term's material
Standardized — consistent format so scores are comparable
Controlled conditions — supervised, time-bounded, no external help
Rubric-driven — for open-ended questions, clear scoring criteria
Formal — students prepare for it; it counts

Examples:

Unit tests at the end of a topic
Midterms and finals
Major projects with clear rubrics
Standardized state/district assessments

The common mistake

The mistake most teachers make isn't conceptual — it's operational. They run a formative-looking quiz but weight it like a summative. Or they run summative-style tests weekly, exhausting students and themselves.

Signs you've blurred the line:

Students are anxious about low-stakes check-ins
You're grading weekly in a way that feels like midterm grading
Item analysis from a test reveals specific gaps — but it's too late to address them (it was supposed to be summative)
You're not adjusting your teaching based on assessment data (you're running summative tests but expecting them to serve formative purposes)

How to structure a unit assessment-wise

A typical 3-week unit might look like:

Week 1:

Daily exit tickets (3 questions, 5 minutes, ungraded)
One formative quiz at end of week (10 questions, counts 2% of unit grade)

Week 2:

Daily exit tickets continue
Mid-unit check (longer, 20 questions, counts 5% of unit grade)
Reteaching based on what Week 1-2 data surfaces

Week 3:

Final review
Summative unit test (50 items, mixed format, counts 20% of unit grade)

The formative assessments are cheap, informative, and don't carry grade weight. The summative assessment is weighted and comprehensive. Both serve their purpose; neither does both jobs.

Make formative data visible to students

Students often don't distinguish between "this counts" and "this is practice." Tell them explicitly. Post or email: "This is a check on your learning — it counts toward the course goal of knowing X, not your grade." Many students adjust how they engage when the stakes are clear.

Auto-grading helps both

One reason teachers collapse formative and summative is that running both feels like twice the grading load. Auto-graded check-ins make the formative work sustainable: