Feedback Without Grades: A Formative Assessment Playbook

TL;DR. Grades change student behaviour from learning to performing. Feedback without grades — done right — keeps the learning benefits of frequent assessment while shedding the motivational damage. This playbook covers when, how, and which formats actually work.

The problem with grading every quiz

A growing body of research shows that putting a grade on student work, even good work, narrows attention to the grade itself. Students stop reading written comments. Risk-taking drops. The classroom shifts from a learning environment to a performance environment.

This doesn’t mean grades are bad — they’re necessary for high-stakes assessment. But for the day-to-day, week-to-week feedback loop, grades are often the wrong tool.

What “formative assessment” actually means

Formative assessment is any assessment used to *inform* teaching and learning — as opposed to *summative* assessment, which evaluates and grades.

A formative quiz: students take it, see what they got wrong, adjust their study. Score isn’t recorded.

A summative quiz: counts toward the final grade.

The cognitive benefits of frequent retrieval practice come from *the act of being tested*, not from being graded. A 10-question Friday quiz that no one grades produces nearly identical learning outcomes to one that gets entered in the gradebook — without the demotivating side effects.

Five formats that work

1. Self-scored quiz with discussion

Hand out a 10-question quiz. Students take it silently for 10 minutes. Then walk through each question aloud — “What did people get for #3?” — and discuss reasoning for 2 minutes per question.

No grades. Discussion is the entire point. Surfaces misconceptions in real time.

2. Two-stars-and-a-wish written feedback

For written work: identify two specific strengths (“stars”) and one specific area to develop (“wish”). No letter grade. Students read because the feedback isn’t competing with a number.

3. Peer feedback with a rubric

Students grade each other’s work against a rubric — but the result goes only to the student, not the gradebook. They get peer perspective; the teacher doesn’t add another grade.

4. Confidence-rated quizzes

Each question has a confidence slider (1–5). Score = correctness × confidence. Wrong-but-confident answers cost double. Right-but-tentative answers don’t fully count.

This trains metacognition (do I actually know this?) rather than guessing. Don’t put it in the gradebook.

5. Quick check-ins (3-question exit tickets)

3 questions at the end of class. Students self-grade. Teacher collects to see class-level patterns, not individual scores. Adjusts tomorrow’s lesson based on what landed.

The weekly cadence

A pattern that balances formative and summative without overloading:

| Day | Activity | Graded? |

|---|---|---|

| Mon | Intro new material; discussion | No |

| Tue | Hands-on practice; informal feedback | No |

| Wed | Confidence-rated quiz; self-score | No |

| Thu | Peer feedback session | No |

| Fri | 10-question quiz, scored privately | No |

| End of unit | Substantial summative assessment | Yes |

The single graded event is the unit summative — every 2–3 weeks. Everything else is for learning, not grading.

How to tell students

Be explicit. “This quiz doesn’t count toward your grade — it’s for you to see what you still need to study.”

This sounds obvious but matters: students assume everything is graded unless told otherwise. Saying it out loud changes how they engage.

When grading IS the right tool

For summative work — unit tests, semester exams, term papers — grades remain the right tool. The point isn’t to abolish grading. It’s to stop grading the day-to-day formative work that doesn’t benefit from a number.

Tools that support this workflow

SimpleQuizMaker has built-in support for self-scored quizzes (results visible to student, not to teacher gradebook by default). See the analytics dashboard for how class-level patterns surface without per-student grades.

For confidence-rated MCQs, you can add a confidence question (1–5) after each item; we sum correctness × confidence at the end.

For peer feedback, students can take each other’s quizzes through a shared link without needing accounts.

Related reading

Build a formative-friendly quiz →

Common mistakes when going gradeless (and how to avoid them)

Most teachers who try ungraded formative assessment and abandon it hit one of the same four walls. Knowing them in advance is most of the fix.

1. Removing grades but keeping the grade-shaped ritual

If the ungraded Friday quiz still looks and feels exactly like a test — silent room, collected papers, no discussion — students treat it as a test anyway, and you lose the point. The format has to change with the stakes. Pair every ungraded quiz with an immediate, visible learning move: whole-class discussion, self-scoring, or a two-minute "what will you review this weekend?" reflection. The retrieval itself does the cognitive work, as the research on the testing effect shows — the discussion afterwards is what makes the shift legible to students.

2. Going gradeless silently

If parents and students discover mid-term that "half the quizzes didn't count," it reads as chaos, not pedagogy. Announce the system in week one: which assessments are formative (feedback only), which are summative (graded), and the rough ratio. One sentence in the syllabus and one repeated line in class ("this one is for you, not for me") prevents most pushback.

3. Letting feedback quality drop because there is no number

A grade is at least *some* signal. If you remove it and replace it with a generic "good effort," students get less information than before. Every ungraded assessment needs a concrete replacement signal: which questions were missed, what misconception each wrong answer points to, and one specific next step. Question-level explanations do a lot of this work automatically — when you build quizzes with an AI quiz generator, each question ships with an explanation students see the moment they answer, so the feedback loop closes in seconds rather than days.

4. Drowning in the extra workload

Ungraded does not mean unprepared. If writing weekly formative quizzes by hand takes 40 minutes each, the system collapses by October. Generate the draft from your actual teaching material — pasting notes or uploading a handout to create a quiz from a PDF turns quiz creation into a review-and-edit task instead of a writing task. Reserve your hand-crafting energy for the summative assessments that genuinely need it.

A four-week starter plan

You do not need to convert your whole gradebook philosophy at once. A contained pilot:

Teachers running this on SimpleQuizMaker can see class-level patterns without exposing individual scores; the free plan covers 5 AI generations and up to 100 student submissions per month, which comfortably fits a one-class pilot — details are on the for teachers page.

Frequently Asked Questions

Does ungraded formative assessment actually improve test scores?

The evidence on frequent low-stakes retrieval practice is strong: students who are quizzed regularly retain more than students who only re-read or review, and the benefit comes from the act of retrieving, not from the score being recorded. Removing the grade does not remove the learning benefit — it removes the performance pressure that makes students avoid revealing what they don't know.

How do I stop students from blowing off quizzes that don't count?

Make the quiz immediately useful to them. Discuss answers right after, let students see exactly which concepts they missed, and reference quiz content in the graded unit assessment so the connection is obvious. Most students disengage from ungraded work only when it feels disconnected from anything that matters.

What ratio of formative to summative assessment should I aim for?

A workable default is one graded summative event per unit (every two to three weeks) with two to four ungraded formative checks per week. The exact ratio matters less than the principle: daily and weekly checks inform learning, and only substantial end-of-unit work carries a grade.

Can I still record formative results for my own planning?

Yes — and you should. "Ungraded" means the score doesn't enter the gradebook or affect the student's mark, not that the data disappears. Keep class-level results to spot weak topics and plan reteaching; just don't attach consequences to individual formative scores.