How EduGenius's Assign Feature Turns a Generated Quiz Into a Gradable Assignment
Assign is the step in EduGenius that turns something you generated — a quiz, a worksheet, a flashcard set, almost any format — into a distributable, trackable classroom assignment. You open it directly from your content history, work through a short wizard, and publish a link and QR code students can open with no account. Objective items score themselves; free-response items get graded by you, or optionally by AI.
Quick Answer: Assign is a four-step wizard (Delivery, Sections, Assessment, Publish) that converts any generated content into an assignment with a no-login student link and QR code. Logged-in learners also see it inside their own account, matched to their Teaching Spaces section and sorted by due date. A teacher-facing report tracks average score, on-time completion, and which specific wrong answer a class converged on per question. Multiple-choice and other objective items auto-score on the server; free-response work can be graded manually or, on a paid plan, by Gemini — a 403
PAID_PLAN_REQUIREDresponse blocks the AI path on a free plan, and AI grading is rate-limited to 5 requests a minute. Creating and sending the assignment itself costs nothing. Answer keys and grading rubrics are stripped out of everything a student's browser ever receives.
Generating a quiz is the easy half of the job. Getting it in front of a specific group of students, finding out who actually finished it, and knowing which question tripped up half the class is the part that usually falls apart into a spreadsheet, a screenshot, or a pile of paper. Assign exists to close that specific gap — not as a separate app, but as the next click after you already have a piece of generated content in hand.
This piece walks through what the feature actually does end to end: the wizard, the no-login delivery mechanism, the two real grading paths, the analytics a teacher actually gets back, and where the feature draws honest, published limits.
What Assign Actually Turns a Generated Quiz Into
Assign takes one piece of already-generated content and produces one shareable, trackable assignment — a specific record with its own link, its own submissions, and its own report, separate from the generated content it started from.
The Four-Step Wizard: Delivery, Sections, Assessment, Publish
Opening Assign on a piece of content in your history starts a short wizard with four tabs: Delivery, Sections, Assessment, and Publish. Nothing is created until you hit Publish on the final step — everything before that is just composing the assignment's settings.
- Delivery — choose whole-class (one open link and QR code for everyone) or specific sections (a unique link and QR per Teaching Space section).
- Sections — if you picked section delivery, pick which Teaching Space and which sections to target; this step is skipped entirely for whole-class delivery.
- Assessment — optionally attach a rubric, set a due date, and decide whether late submissions are still accepted.
- Publish — a one-line summary of what you're about to send, then a single button that creates the assignment.
Works on Any Generated Format, Not Just Quizzes
Assign isn't restricted to multiple-choice quizzes. It works on any generated content session — a worksheet, a flashcard set, a short-answer set, a mixed-format packet — because it operates on the content session as a whole rather than on one specific format's data shape. Whatever you generated is what students receive.
| Wizard step | What it captures |
|---|---|
| Delivery | Whole-class open link/QR, or per-section links/QR |
| Sections | Which Teaching Space and which of its sections to target |
| Assessment | Optional rubric, optional due date, late-submission toggle |
| Publish | Final confirmation; creates the assignment and its share link(s) |
Once published, the result screen hands you the link and QR code immediately, along with a toggle to stop accepting new submissions at any point and an optional free-text roster field for classes not yet set up in Teaching Spaces.
How Students Actually Get and Complete the Work
The student side of Assign is built around one deliberate constraint: nobody needs an EduGenius account to open and complete an assignment.
No Login Required: Link or QR Code
A student opens the link — typed, pasted, or scanned from the QR code — and lands directly on the assignment. They type a name (and, if you've turned it on, a roll number or student ID), answer the questions, and submit. There's no sign-up form and no password anywhere in that path. That link resolves through an unguessable, randomly generated token rather than a predictable ID, so an assignment can be shared broadly without exposing anyone else's content.
Logged-In Learners See It In-App Too
Students who do have an EduGenius account get a second, complementary path: assigned work they're eligible for shows up inside their own learner dashboard automatically, cross-referenced against whichever Teaching Spaces section they belong to. It's sorted by due date and flagged with a plain todo-or-submitted status, so a logged-in learner doesn't have to go hunting for a link a teacher posted somewhere.
The Answer Key Never Leaves the Server
Before anything reaches a student's browser, EduGenius strips every answer-bearing field out of the content — correct answers, explanations, sample answers, marking rubrics, and even the teacher's private grading instructions where those exist. Scoring for objective items happens against the original, unsanitized content on the server, never against a copy sent to the client. A student inspecting the page's network traffic can't recover the key that way, because it was never sent.
Why Fast, Trackable Feedback Loops Matter
None of this is EduGenius-specific, but it's the research backdrop that makes a fast generate-to-assign-to-report loop worth building rather than a nice-to-have UI flourish.
Feedback Is One of the Strongest Levers on Learning — When It's Done Right
In their influential review of the education research on feedback, Hattie and Timperley found that feedback sits among the more powerful single influences on student achievement of anything they examined — well above the general threshold researchers use to separate a typical year of growth from an above-average one. Their caveat matters just as much as the headline: the effect swings from strongly positive to actively harmful depending on the type, specificity, and timing of the feedback, not just whether feedback happens at all (Hattie & Timperley, 2007).
Formative Assessment's Payoff Depends on the Loop Actually Closing
Black and Wiliam's foundational review of classroom formative assessment reached a similar conclusion from a different angle: assessment that's actually used to adjust teaching and give students concrete next steps produces measurable achievement gains, but only when the information comes back and gets acted on — a quiz that's graded weeks later, or never handed back at all, doesn't function as formative assessment no matter how well-designed the questions were (Black & Wiliam, 1998).
Timing Itself Is Not a Trivial Detail
A classic meta-analysis on feedback timing found that in ongoing classroom instruction and testing, immediate feedback tended to outperform delayed feedback, even though some laboratory-style paired-learning tasks showed the opposite pattern (Kulik & Kulik, 1988). The practical read for a real classroom: the faster a teacher can see who missed what, the sooner that information is useful for the next lesson, not just for a single student's grade.
| Research finding | Source |
|---|---|
| Feedback ranks among the largest single influences on achievement studied, though effect direction and size depend heavily on type and timing | Hattie & Timperley, 2007 |
| Formative assessment produces achievement gains specifically when results are used to adjust teaching and give students next steps | Black & Wiliam, 1998 |
| Immediate feedback tends to outperform delayed feedback in classroom instructional/testing settings | Kulik & Kulik, 1988 |
| AI-generated writing feedback was broader in coverage but rated less accurate and specific than teacher feedback in a head-to-head comparison | Steiss et al., 2024 |
That last line is worth sitting with before the next section, because it's the exact tension Assign's grading design is built around: AI can help close a feedback loop faster, but it isn't presented as a drop-in replacement for a teacher's judgment.
Two Real Grading Paths: Auto-Score, AI-Assist, or Manual
Once submissions start arriving, EduGenius handles objective and free-response items differently, and it's worth being precise about which is which.
Objective Items Score Themselves
Multiple-choice, true/false, and similarly structured items are scored automatically, server-side, the moment a submission comes in — no teacher action required. The score and a per-question correct/incorrect breakdown are available immediately in the assignment's report.
Free-Response Items: AI Grading (Optional) or Manual
Open-ended, free-response answers don't have a single correct string to match against, so they need either a person or a model to read them. EduGenius supports both:
- Manual grading — you read the submission and grade it yourself, exactly as you would on paper. Always available, on every plan, at no cost.
- AI grading (optional) — a request to grade a batch of ungraded submissions sends each one to Gemini along with the assignment's brief and any rubric you attached, and gets back a suggested score and two-to-five sentences of feedback per student. It's explicitly a draft — feedback stays out of a student's view until you publish it.
A single AI grading request handles a batch, up to 20 submissions per call, so grading a full section doesn't require triggering the model one student at a time.
The Paid-Plan Gate and Rate Limit, Explained Honestly
AI grading is not a free-tier feature. A teacher on a free plan who tries to trigger it gets a PAID_PLAN_REQUIRED response with a 403 status — a real, server-enforced gate, not a UI suggestion that can be worked around. The endpoint is also rate-limited to 5 requests per minute per teacher, regardless of plan, to keep a single account from hammering the model.
Two things are true at the same time here, and both matter: AI grading costs a paid plan and is rate-limited; creating and publishing the assignment itself, delivering it to students, and manually grading every submission are all free, on every plan, with no rate limit tied to the free path.
The Submission Report: Real Analytics, Not Just a Delivery Pipe
A link that students can open is only half a useful assignment tool. The other half is knowing, afterward, whether it actually worked — and for whom.
Average Score and On-Time Completion
Every assignment's report computes a running average score across all scored submissions and, when a due date was set, the percentage of submissions that came in on or before it. Average time spent is tracked too, where that data is available, giving a rough sense of whether an assignment ran long for the group that attempted it.
Wrong-Answer Clustering: Finding Where the Class Got Stuck
This is the detail that separates Assign's report from a plain gradebook column. For every objective question, the report counts how many students got it wrong and ranks questions by how many students missed them — surfacing not just that a question was hard, but which specific question, ordered from most-missed down. A teacher scanning that list after a quiz can see, in one glance, exactly where reteaching would help the whole class rather than guessing from a single average score.
The report also carries per-submission detail — student name, score, submission status, and (when AI grading has run) the drafted feedback pending your review — so the aggregate numbers and the individual student view live in the same place rather than two disconnected tools.
Honest Limits, and How Assign Layers on Teaching Spaces
Assign publishes real, fixed limits rather than leaving capacity open-ended, and it's built to work with or without Teaching Spaces already set up.
Roster and Section Caps
A free-text roster (the "who is this for" list you can attach to a whole-class link) is capped at 60 names. Targeting specific Teaching Space sections when publishing an assignment is capped at 20 sections in a single publish. Both are fixed, published constraints — not soft suggestions that quietly stop working past a certain point.
Whole-Class Delivery Works Without Any Teaching Space at All
It's worth being direct about this, because it's easy to assume otherwise: Assign does not require a Teaching Space to function. Whole-class link-and-QR delivery works the moment you publish, with zero Teaching Space setup beforehand — that's the original, always-available path.
Teaching Spaces sections are an optional layer on top, useful specifically when you want per-section tracking (a separate link and roster per class period, say) or when you want assigned work to show up automatically inside a logged-in learner's dashboard, matched to the section they've joined. If you've never set up a Teaching Space, Assign still works; it just delivers to everyone through one shared link instead of several section-specific ones.
Pro Tips
- Generate first, assign second, every time. Assign works from your content history, so review and tweak a generated quiz or worksheet before you publish it as an assignment — the link a student opens reflects exactly what's in the session at that moment.
- Set a due date if you want the on-time stat to mean anything. On-time completion percentage only calculates when a due date is attached; skip it and that one report figure stays blank.
- Turn off "accepting submissions" once you've moved on. The toggle stops new submissions instantly without deleting the assignment or its existing data, which is cleaner than leaving an old link quietly open all term.
- Scan the wrong-answer cluster before reteaching, not just the average score. A 78% class average can hide one question 90% of the class missed — the cluster view is what actually tells you where to spend the next lesson.
- Use AI grading as a first draft you review, not a final answer. Published research on AI-generated writing feedback found it covers more ground but is less precise than a teacher's read (Steiss et al., 2024) — which is exactly why AI-graded feedback stays a draft until you publish it.
- Pair Assign with Compete mode for review sessions, and save the routine as a content recipe if you generate-then-assign the same shape of quiz every week.
What to Avoid
- Don't assume AI grading is included on every plan. A free-plan account hits a real
PAID_PLAN_REQUIREDblock, not a soft warning — confirm your plan before you build a grading workflow around it. - Don't expect AI grading to run instantly on a large batch. It's rate-limited to 5 requests a minute and 20 submissions per call; a big section needs a couple of batches, not one instant pass.
- Don't rely on the free-text roster for a class over 60 students. That field is capped at 60 names; a larger group needs a Teaching Space section instead, which doesn't share that ceiling the same way.
- Don't publish a section-targeted assignment expecting one shared link. Section delivery mints a distinct link and QR per section by design — post the right one to the right group, or use whole-class delivery instead if you actually want one link for everyone.
- Don't treat AI-drafted feedback as final until you publish it. It's deliberately held back from students as a draft specifically so a teacher can review, edit, or override it first.
- Don't post a student-facing assignment link somewhere public and permanent. The link doesn't require login, which is the point for a classroom, but that also means anyone with it can open the assignment — share it the same way you'd share a Teaching Spaces join code, not on the open internet.
Key Takeaways
- Assign converts any generated content — quiz, worksheet, flashcards — into a trackable assignment through a short Delivery/Sections/Assessment/Publish wizard, launched directly from your content history.
- Students need no account: a link or QR code is the entire access path, and the same assignment surfaces automatically inside a logged-in learner's dashboard when it matches their Teaching Spaces section.
- Answer keys and grading instructions are stripped server-side before anything reaches a student's browser; scoring runs against the original, unsanitized content.
- Grading has two real paths: objective items auto-score instantly; free-response work is graded manually (free, always available) or optionally by Gemini on a paid plan, rate-limited to 5 requests a minute.
- A free plan cannot use AI grading — it returns a real
PAID_PLAN_REQUIRED403 — while creating, publishing, and manually grading an assignment stays free on every plan. - The report goes beyond a delivery pipe: average score, on-time completion percentage, and a ranked list of which specific wrong answer a class converged on per question.
- Roster and section limits are real and published: 60 names in a free-text roster, 20 sections per publish — not soft suggestions.
- Teaching Spaces is an optional layer, not a requirement: whole-class link/QR delivery works with zero Teaching Space setup; sections add per-group tracking and in-app delivery to logged-in learners on top of that baseline.
Frequently Asked Questions
Does Assign work on formats other than multiple-choice quizzes?
Yes. Assign operates on the generated content session as a whole, so worksheets, flashcard sets, short-answer sets, and mixed-format packets can all become assignments the same way a quiz can — there's no format restriction built into the feature.
Do students need an EduGenius account to complete an assignment?
No. A student opens the shared link or scans the QR code, enters a name, and submits — no sign-up, no password. Students who do have an account also see assigned work automatically inside their own dashboard, matched to their Teaching Spaces section.
Is AI grading free?
No. AI grading specifically requires a paid plan; a free-plan account attempting it receives a PAID_PLAN_REQUIRED error. Creating and publishing the assignment itself, and grading submissions manually, remain free on every plan.
How fast can I run AI grading on a full class?
AI grading is rate-limited to 5 requests per minute, and each request can grade up to 20 submissions in one batch. A typical class fits in one or two requests; a very large batch needs a couple of minutes spread across a few calls.
Can a student see the correct answers before submitting?
No. Answer keys, sample answers, and grading rubrics are removed from the content before it's ever sent to a student's browser, and objective scoring happens server-side against the original content — not a copy the student can inspect.
What does the wrong-answer clustering feature actually show?
For each objective question in an assignment, the report counts how many submissions got it wrong and ranks questions from most-missed to least-missed. It's designed to show you exactly which question tripped up the group, not just an overall average score.
Do I need a Teaching Space set up before I can use Assign?
No. Whole-class link-and-QR delivery works immediately with no Teaching Space required. Teaching Spaces sections are an optional add-on for per-section tracking and automatic in-app delivery to logged-in learners, layered on top of that baseline whenever you choose to use them.
How many students or sections can one assignment reach?
A free-text roster caps at 60 names, and targeting Teaching Space sections when publishing caps at 20 sections in a single assignment — both real, fixed, published limits.
Related Reading
- What Are Teaching Spaces in EduGenius? Organizing Classes, Sections, and Rosters
- How to Export an AI-Generated Quiz to Google Forms With EduGenius
- How EduGenius's Compete Mode Turns Practice Into a Timed Challenge
- How EduGenius's AI Coach Balances Chat With Actual Practice
- How to Save and Reuse Content Recipes in EduGenius
- How the EduGenius Chrome Extension Turns Web Content Into Teaching Materials