Formative vs Summative Assessment — How They Differ + When To Use Each

Formative assessment happens during learning to inform teaching — short, low-stakes checks like exit tickets or mini-whiteboards that help you adjust the next lesson. Summative assessment happens after learning to measure what was retained — higher-stakes tasks like unit tests, end-of-course exams, or final projects that produce a grade. The same activity can serve either purpose depending on what you do with the results: if the data changes your teaching, it's formative; if the data goes on a report, it's summative.

Yes. A short weekly quiz can be formative if you use the results to reteach and you don't enter the grade in a permanent gradebook. The same quiz becomes summative the moment it counts toward a report card or qualification. Many teachers run the same item twice — once formatively mid-unit (to diagnose gaps) and once summatively at the end (to certify learning). What matters is the function, not the format.

Embed at least one quick formative check in every lesson. The research consensus from Dylan Wiliam's work on assessment for learning is that frequent, low-stakes checks every 10–20 minutes are more powerful than occasional big tests. Exit tickets at the end of class, hinge questions mid-lesson, or a 30-second mini-whiteboard show-me all qualify. The cost is low; the payoff is that you start the next lesson with real evidence of what students actually understood.

There is no universal rule — it depends on your school's policy and your jurisdiction. As a rough working ratio, many secondary teachers weight summative assessments at 70–90% of the final grade, with the remainder coming from coursework, participation, or selected formative tasks. Under NCEA in New Zealand, almost all reported grades come from summative internal and external standards. In English schools under GCSE / A-Level, summative terminal exams typically dominate the final grade. Check your school's assessment policy before designing your weightings.

Almost always, yes — a final exam is by definition end-of-course, high-stakes, and produces a grade that is reported. The one edge case is when a final exam is administered as a mock or practice paper purely to identify what to reteach before the real exam. In that case the same paper functions formatively. Once the exam counts toward a transcript or qualification, it's summative.

Rubrics can serve both. Used during a task — for example, a student self-checks against the rubric before submitting a draft — the rubric is formative because it surfaces gaps the student can still fix. Used to grade a finished submission, the rubric is summative because it produces the mark. The best practice is to share the rubric up front and use it formatively as a learning tool, then use the same rubric to grade the final piece.

Hinge questions are one of the strongest formative moves in maths. Mid-lesson you pose a multiple-choice problem where each wrong answer maps to a specific misconception (e.g. forgot to distribute, swapped numerator and denominator, dropped a negative). Students show their answer on mini-whiteboards. You read the room in 10 seconds and either move on, reteach to the whole class, or pull a small group. Other strong examples: error analysis (find the mistake in this worked solution), two-minute exit tickets with one problem, or low-stakes weekly retrieval quizzes.

Treat summative as the destination and formative as the navigation. Plan summative tasks first — the unit test, the project, the externals — so you know what students must be able to do. Then back-design frequent formative checks to keep you informed about progress toward that target. A healthy weekly rhythm in most classrooms looks like five to ten formative touchpoints and zero to one summative grade. The summative tells you whether you arrived; the formative tells you whether you're on course.

No, but it has to be low-stakes. You can record formative checks in a tracker for your own planning — many teachers do — but if the score is going into a report card, it has crossed into summative territory. The reason matters: students take risks and reveal genuine confusion only when the cost of being wrong is low. Heavy grading on every check kills the feedback loop you're trying to create.

The distinction is usually traced to Michael Scriven in 1967, who introduced the terms in the context of programme evaluation. Benjamin Bloom adapted them for classroom use shortly after. The modern revival of formative assessment in schools is largely thanks to Paul Black and Dylan Wiliam, whose 1998 paper Inside the Black Box argued that consistent, well-designed formative assessment is among the most effective interventions a teacher can make.

Observation is almost always formative. Walking the room, listening to group discussion, noticing who's stuck on which step — these are real-time diagnostic signals that should reshape the next ten minutes of your lesson. Observation only becomes summative in specific structured situations: an oral exam, a performance assessment with a rubric, or a moderated practical (for example, a lab skill or a music performance) where you're certifying competence.

Usually summative — a finished project graded against a rubric produces a mark for the report. It can be made partially formative by giving feedback against the rubric on a draft (which the student then revises) before the final mark. That two-stage workflow — draft with formative rubric feedback, then final summative grade — is one of the highest-leverage assessment patterns for project work.