---
name: formative-assessment
description: Use when asked to design ongoing, in-the-moment checks for understanding during instruction — as distinct from [[summative-assessment]], which evaluates mastery after a unit or course is finished rather than while it's still being taught.
---

# Formative Assessment

Formative assessment is the practice of checking student understanding
while instruction is still happening, so a teacher can adjust before
moving on rather than finding out too late. It's diagnostic and
low-stakes by design — its purpose is to inform the next instructional
move, not to produce a grade — which is what separates it from
[[summative-assessment]], which measures mastery once teaching for a
unit is done.

## Key components

- **Frequent, low-stakes checks** — quick, recurring probes of
  understanding (questioning, quick writes, polls, thumbs-up checks)
  woven through a lesson rather than concentrated at its end.
- **A specific target** — each check is tied to a specific piece of
  the day's objective, so the result says something actionable rather
  than a vague overall impression.
- **A visible response plan** — a formative check is only useful if
  the teacher has already decided what happens on a weak result:
  reteach, regroup, slow down, or move on.
- **Fast turnaround** — formative data has to come back quickly enough
  to change the very lesson it was collected in, not days later.
- **Low or no grading weight** — keeping stakes low is what makes
  students respond honestly rather than performing for a score, which
  is what makes the data trustworthy.

## Adjusting the lesson, not just collecting data

Formative assessment stops being formative the moment its results stop
changing anything — a quick poll or exit question that gets recorded
but doesn't actually alter the next five minutes of instruction is
just ungraded busywork wearing the name. The value is entirely in the
decision it enables: if two-thirds of the class gets a check wrong,
the plan for the rest of the period changes, whether that means
re-explaining with a different example, pairing students up to teach
each other, or abandoning the planned next activity. That means a
formative check has to be built with its decision rule already in
mind — what result triggers what response — rather than designed first
and interpreted afterward. A single technique, like the
[[exit-ticket]], is one recurring instance of this practice, not the
whole of it; formative assessment also happens through questioning,
observation, and quick informal tasks spread across a lesson.

## Common pitfalls

- **Checking without adjusting** — collecting formative data but
  running the next activity exactly as planned regardless of the
  result defeats the purpose of checking at all.
- **Too infrequent to matter** — one check per class period, especially
  late in the period, catches a misunderstanding too late to do
  anything about it that day.
- **Grading it like summative work** — attaching real grade weight to
  a formative check pushes students toward safe, performed answers
  instead of honest signals of where they actually are.
- **Checks not tied to the objective** — a generic "any questions?"
  produces silence, not data; a specific, targeted prompt is what
  actually reveals whether the objective landed.
- **No decision rule in place** — collecting a result with no
  pre-decided threshold for what triggers a reteach leaves the teacher
  improvising under time pressure instead of executing a plan.

## Learn more

- [[exit-ticket]] for one specific, end-of-lesson formative technique.
- [[lesson-plan]] for where formative checks get built into a single
  session's structure.
- [[summative-assessment]] for the end-of-unit counterpart that
  measures mastery rather than informing the next lesson.
- [[rubric-design]] for scoring criteria when a formative check is
  more than a simple right/wrong.