Use formative assessment examples that produce evidence you can act on before the next lesson, not a pile of activities you never review.
The most useful formative assessment examples are short checks that produce evidence you can act on before the next lesson: an exit ticket you can sort into three piles, a whiteboard flash that shows who missed the middle step, or a one-sentence claim you can reteach tomorrow. A list of forty activities is not the job. Matching the check to the thinking you need, then using what you see, is.
Teachers search this query because the period is already full. You do not need another bank of icebreakers. You need a check that fits today's learning target, takes minutes, and still tells you whether to reteach, extend, or move on. NWEA treats formative assessment as a planned process for eliciting and using evidence during learning, not a quiz brand and not a grade. This page is built that way: examples first, then the decision rule, then a collect-sort-act workflow you can run after the bell.
If the bottleneck is reviewing written responses before tomorrow, keep the teacher in charge. GradeWithAI grading drafts scores and comments against your rubric and leaves them as drafts until you release them, including work that arrives through Google Classroom. The check still belongs to you. The model does not decide who understood the lesson.
Quick Answer
Use one formative check per learning target, not one per activity. Pick the example that shows the thinking you actually need.
Edutopia's practical rule still holds: formative checks generally need to be checked, not graded. If you cannot name the instructional move the evidence will trigger, skip the activity.
What Counts as a Formative Assessment Example
A formative assessment example is any planned move that elicits evidence of learning during teaching so you and the student can change what happens next. The Council of Chief State School Officers' 2018 definition, which NWEA adopts, is the cleanest version: a planned, ongoing process used by students and teachers during learning to elicit and use evidence, improve understanding of intended outcomes, and support students as self-directed learners.
Three tests keep a cute activity from pretending to be formative.
It has a learning target. "Do a Kahoot" is not an example. "See who can still set up a two-step equation after yesterday's reteach" is.
It produces visible evidence. A thumbs-up that every student copies from the person in front is not evidence. A whiteboard that shows the setup, or a sentence that names the variable, is.
Someone uses it before the window closes. Carnegie Mellon's Eberly Center draws the line this way: formative assessment monitors learning so instructors can improve teaching and students can improve learning, usually at low or no point value. A midterm, a final paper, or a recital is summative unless you still have time and a plan to act on it.
NWEA is explicit that formative scores are not appropriate for calculating grades or making placement decisions. If the mark will live in the gradebook as a verdict, you have changed the job. Students start performing for points. You stop seeing the messy thinking you needed.
This page is not a second copy of classroom assessment techniques. CATs are a specific, often college-oriented toolkit. Formative assessment examples are the broader classroom set: elementary through secondary, paper or digital, 30 seconds or 10 minutes. If you want the CAT catalog, use that post. If you want to pick a check for tomorrow's lesson, stay here.
20 Formative Assessment Examples by What You Need to See
The current search results mostly stack activities. The useful version groups them by the evidence they produce. Use two or three from different rows across a week. Do not run all twenty.
Readiness and recall
1. Bell ringer. Three to five short items as students sit down: a yesterday skill, a vocabulary match, or a one-sentence warm-up. Evidence: who walked in ready, and which prerequisite is still missing. Skip it when the first ten minutes are already a lab safety briefing.
2. Question or problem of the day. One item aimed at today's main idea. Revisit it at the end so students can revise. Evidence: how understanding moved during the period, not only whether the first try was right.
3. Pre-test or pretest probe. A short, ungraded baseline before a unit. Evidence: who already has the skill, who is missing a prerequisite, and whether your planned pacing is honest. Do not average it into the marking period.
4. Mini-whiteboard responses. Students solve or write, then hold up the board. Evidence: a class snapshot in ten seconds, including the wrong setup, not only the wrong answer. When boards are unavailable, use paper or a shared doc. This is the fastest process check that still shows work.
5. Polls and fist-to-five. Hands, cards, or a digital poll. Evidence: confidence and, if the item is well written, a misconception pattern. Do not treat a high confidence score as mastery. Students can be confidently wrong.
Process and reasoning
6. Checks for understanding mid-lesson. A planned pause: one oral question, one written step, or one digital reply. Evidence: whether to keep going or stop now. Build the pause into the plan so it is not the thing you skip when you run long.
7. Documented problem solutions. Students write every step of a multi-step math, science, or logic task. Evidence: the exact broken move — skipped unit, reversed operation, copied neighbor. Heavier to review than a poll. Use it when the process is the learning target.
8. Spot the mistake. You present a worked example with an intentional error. Students mark the error and explain why. Evidence: whether they can recognize a common misconception, which is different from being able to produce a clean solution from scratch.
9. Write-pair-share. Silent write, partner talk, then a few public shares. Evidence: individual first thinking plus the language students use when they have to explain. Circulate during the pair minute. That is the assessment, not the whole-class recap.
10. Jigsaw teach-back. Expert groups learn one piece, then teach a home group. Evidence: whether a student can explain a chunk well enough that a peer can use it. Skip it on day one of new content.
Misconception and transfer
11. Exit tickets. One prompt in the last two minutes: a sentence, a sketch, or a single problem. Edutopia still recommends sorting paper tickets into three piles — got it, sort of, did not — and letting pile size decide tomorrow. That sort is the analysis. Entering 32 scores is not.
12. One-minute brain dump. Students write without stopping on the main idea, the muddiest point, or the takeaway. Evidence: vocabulary they reach for, and the confusion they will not volunteer out loud.
13. Low-stakes quiz. Five to ten targeted items. Evidence: skill-level data you can reteach the same week. Keep points tiny or unused. A quiz that can sink a marking period is no longer formative, even if you named it a "check-in."
14. Performance task, kept short. A two-minute oral explanation, a labeled diagram, or a one-paragraph lab claim — not a week-long project. Evidence: application, not recall. If the product takes longer to make than you have to review, it will become homework you never formatively use.
15. Invented dialogue. Students write a short conversation between two figures, characters, or concepts. Evidence: whether they understand a relationship, not a memorized date. Skip it when the target is a procedure.
Self-direction and classroom evidence you already have
16. 3-2-1 countdown. Three things learned, two questions, one next step. Evidence: what stuck, what is still open, and whether students can name a next action. Good closer. Weak as your only check in a skill-heavy lesson.
17. Know-wonder-learn. Students record prior knowledge, questions, and later learning. Evidence: misconceptions at the start of a unit and growth when you return to the chart. Best for inquiry work, not a short skills block.
18. Structured peer feedback. Students comment using a short rubric or TAG (tell, ask, give). Evidence: the author's work and the reviewer's grip on the criteria. Unstructured "be nice" review is social time, not assessment.
19. Student reflection against criteria. "What did I master, and what still needs work?" only works if students have seen success criteria. Spot-check a sample. Do not trust every self-score.
20. Observation and homework-as-admission-ticket. Watch work during a lab, or review last night's practice for patterns rather than points. Observation needs a roster note or it evaporates. Do not turn this into a completeness grade that hides the skill.
Homework and written reflections are the examples that break the time promise. If reviewing 30 open responses will keep you from acting tomorrow, tighten the prompt or draft the first pass with a teacher-reviewed AI grader. Do not paste a named roster into a consumer chatbot. Student work is an education record; see student data privacy.
How to Choose the Right Example
Start with the learning target, not the tool you like. The question is not "What is a fun formative?" It is "What thinking do I need to see in the next eight minutes?"
Match the format to the cognition. Vocabulary recall can live on a whiteboard flash. Multi-step reasoning cannot. A multiple-choice exit ticket will tell you whether students picked the definition of photosynthesis. It will not tell you whether they can explain why it matters. If the target is explanation, collect a sentence.
Match the grouping to the decision. A whole-class poll answers "Do I keep going?" in seconds. Individual tickets answer "Who needs the small group tomorrow?" They take longer to read. Do not collect individual writing if you only needed a pace check.
Rotate enough to see past the medium. Students who write poorly may explain the water cycle clearly out loud. Students who talk well may collapse on paper. Classroom assessment techniques make the same point in a college frame: the method should reveal the learning, not the student's preferred performance style. Across a unit, mix written, oral, visual, and observed work.
Budget the review, not only the activity. A one-minute student task that creates 40 minutes of evening reading is not efficient. It has moved the work off the clock. Design the response format so patterns are visible: one sentence, a labeled diagram, a numbered list, or a multiple-choice item whose distractors are real misconceptions.
Name the instructional move in advance. Before you hand out the ticket, write the if-then: if more than a third miss the setup, reteach with a new model; if five students miss the same step, pull a group; if almost everyone is clean, skip the warm-up tomorrow and extend. Without that sentence, the example is decoration.
A short quiz is a legitimate example when the items are targeted and the review is fast. It is the wrong example when you need to see a process, a misconception explanation, or a transfer to a new context.
A Classroom Workflow: Collect, Sort, Act
The example is only the first third of the loop. NWEA's five practices are clarifying goals, eliciting evidence, engaging self- and peer assessment, using actionable feedback, and adjusting the next step. Most classrooms collect. Fewer sort. Fewer still change tomorrow's plan.
1. Put the target on the board in student language
"By the end of this lesson you should be able to set up a two-step equation and say what the variable stands for." Ask one student to restate it. Early confusion here is cheaper than confusion on the ticket.
2. Collect one piece of evidence, not five
Run the example that matches the target. Stop when you have enough to decide. A second check in the same period is useful only when the first one changed instruction and you need to see whether the reteach landed.
3. Sort before you comment
For paper, use Edutopia's three piles. For whiteboards, scan for the same three groups while boards are still up. For digital work, skim for the repeated broken step, not for a perfect comment on every file. Pile size is the class diagnosis. Individual cards in the "did not" pile are the small-group list.
4. Give feedback that can be used, then make time to use it
The Education Endowment Foundation's feedback toolkit rates feedback as high impact for very low cost, with an average of about six months' additional progress across a large evidence base. The same review is blunt: effects vary, some feedback makes learning worse, and comments about the person tend to be the weakest kind. Focus on the task, the subject, or the student's strategy. "Great job" and "try harder" fail that test.
EEF also notes that written marking is only one delivery method and that workload has to be watched. Verbal feedback during the whiteboard scan often beats a paragraph written at 10 p.m. that the student reads after you have already moved on. Feedback that arrives after the next lesson has started is historical commentary.
Give one next step in a short sentence. Then give students a chance to apply it — a revised ticket, a second problem, or tomorrow's opener. Feedback with no attempt window is a note to the file.
5. Change one thing in the next plan
Reteach the missed step with a new model. Pull the "did not" pile for eight minutes. Drop the warm-up you no longer need. If nothing in tomorrow's plan changed, the example did not earn its minutes.
When written responses are the right evidence
Some targets require sentences: claims, explanations, lab conclusions, short constructed response. Those are slower to review. Use grading tips that cut busywork: one criterion, a tight prompt, a sample you score first.
If you use GradeWithAI on those written checks, treat the output as a draft. The current pricing page lists a Free plan with 25 AI grading requests a month, Google Classroom and Canvas sync, Google Forms grading, handwritten support, AI rubric generation, and unlimited Kleo. Pro is $20 a month for unlimited AI grading requests, automated submission grading, AI detection on every submission, and priority support. Schools add Microsoft Teams and Schoology, an admin dashboard, SSO/SAML, and bulk users. None of those plans remove your review step. Use the same draft-and-review habit described in how teachers can use AI in the classroom.
Subject and Grade Worked Examples
Abstract lists fail when the period starts. These are complete loops: target, example, evidence, next move.
Grade 3 reading. Target: retell a story with beginning, middle, and end. Example: exit ticket — three labeled boxes, one sentence each. Evidence: middles that jump or endings that are only "they were happy." Next move: tomorrow's mini-lesson models a middle sentence that names the problem, then students revise yesterday's ticket.
Grade 7 math. Target: set up a two-step equation from a word problem. Example: mini-whiteboards, one new story problem. Evidence: students who compute a number with no equation, or who treat the variable as the answer instead of the unknown. Next move: if more than a third miss the setup, do a spot-the-mistake on a wrong setup before any new practice.
Grade 9 biology. Target: connect a method to a claim in a lab conclusion. Example: one-sentence claim plus one sentence of data. Evidence: conclusions that repeat the procedure or ignore the graph. Next move: small group on "what the data do not show"; everyone else gets a new data set, not more vocabulary.
Grade 11 history. Target: explain how two documents disagree. Example: invented dialogue, eight lines, between the two authors. Evidence: speeches that agree politely or ignore the conflict. Next move: highlight one pair of contradictory lines, then rewrite the weakest dialogue.
Any grade, language support. Target: show content knowledge without making writing the gate. Example: oral write-pair-share or a labeled sketch. Accept that evidence for the formative decision; teach the written form later.
Ordinary loops that change tomorrow beat a gallery walk you never look at.
Common Mistakes
- Collecting without using. A stack of tickets in your bag is storage, not assessment.
- Grading the formative. Points change the student move from "show me your thinking" to "protect my average."
- Using the same two examples all year. Exit tickets miss students who think better aloud. Observations miss the misconception that only appears in writing.
- Choosing a tool because it is digital. A poll that asks a yes/no confidence question tells you mood. A paper sentence can tell you the skill.
- Writing feedback about the person. EEF's review is clear that person-level comments are the weak form. Describe the work.
- Giving feedback after the window closed. If tomorrow's lesson already moved on, you assessed for the gradebook, not for learning.
- Treating confidence signals as mastery. Thumbs-up and emoji ratings are useful for climate. They are not a substitute for work that can be wrong.
- Pasting student writing into a consumer chatbot. Formative work still identifies students. Use a school-approved workflow and keep the teacher responsible.
- Running a performance task that you cannot review. A three-day project used "formatively" is usually a late summative with extra stickers.
Who should not use a given example: skip peer feedback until criteria are public; skip jigsaw on day one of new content; skip documented solutions when you only need a pace check; skip graded quizzes when you still need honest wrong answers.
Frequently Asked Questions
What are the best formative assessment examples?
The best formative assessment examples are the ones that match the thinking you need and that you can actually review in time to change instruction. For most lessons that is an exit ticket, a whiteboard flash, a spot-the-mistake item, or a one-sentence explanation. Variety across a unit matters more than a longer list.
How many formative assessment examples should I use in one lesson?
One planned check per learning target is enough. Add a second only after the first one changed what you did and you need to see whether the reteach worked. Three activities in one period usually means you collected theater, not evidence.
Should formative assessment examples be graded?
Usually no. Carnegie Mellon describes formative assessments as low-stakes or no-stakes. NWEA says formative scores are not for grades or placement. A tiny point value can raise effort on a quiz. A mark that can sink a report card will hide the thinking you needed.
What is the difference between formative and summative assessment?
Formative assessment monitors learning during teaching so you and the student can still change course. Summative assessment evaluates learning against a standard at the end of a unit or course. The same product — a quiz, a lab, a paper — can be either, depending on whether you still have time and a plan to use the evidence.
Are classroom assessment techniques the same as formative assessment examples?
They overlap. Classroom assessment techniques are a named set of quick checks, often associated with college teaching. Formative assessment examples is the broader classroom query: any planned check that produces usable evidence. Use the CAT post when you want that catalog. Use this page when you want to pick a check by purpose.
How do I use formative assessment examples without adding hours of grading?
Tighten the response format, sort instead of scoring, and comment on one next step. If the review will not finish before the next lesson, the example is too heavy.
How does GradeWithAI help with formative assessment examples?
It helps when the example produces writing you would otherwise review too late: short constructed responses, homework-as-admission-tickets, reflections, or quiz explanations. You supply the rubric. It drafts comments. You keep or change them before students see anything. It does not replace a whiteboard scan, an observation, or your decision about tomorrow's lesson.
