Quizzes and grading

Quiz lessons are central learning tools — they check whether the content was really understood, give learners direct feedback, and contribute to course completion. This page explains how to configure quizzes, what to watch out for with AI-assisted free-text grading, and how to set the minimum score sensibly.

Quiz types at a glance

Type	What for	Grading
Multiple choice	Questions with answer options — one or several correct (several marks turn the question into a multiple-selection one)	Automatic — a question counts when it is answered completely correctly
Fill in the blanks	Complete words in a text	Automatic — points per correctly filled blank
Matching	Pair up terms	Automatic — points per correct pairing
Drag & drop	Place elements into zones	Automatic — points per correct position
Sorting	Put elements into the right order	Automatic
Free text	Open answer, graded by AI	AI-assisted, using your criteria (see below)
H5P quiz	Various H5P quiz types	Automatic — see Creating H5P content

Creating multiple-choice questions

1. In your course, create a new lesson of the type Multiple choice, then open Edit content.
2. Formulate the question — a clear, unambiguous question.
3. Create the answer options (3-5 makes sense).
4. Mark the correct answer(s) with a check.
5. Optionally add an explanation per question — it is shown to learners after submitting, perfect for turning wrong answers into a learning effect.
6. Add more questions and Save.

The lesson score results automatically from the share of questions answered completely correctly — there is no weighting of individual questions.

Tips for good multiple-choice questions

• Plausibly wrong options (distractors) — learners should really think, not obviously guess right
• Every distractor a typical misconception — if the wrong option also exposes wrong assumptions, people learn even when they pick the wrong one
• No "all answers are correct" trick questions — they confuse more than they test
• One question = one learning objective — no compound constructs

Free-text grading with AI

AI grading is useful for tasks that don't fit "A/B/C/D" schemes — e.g. reflections, professional justifications, one's own application examples.

Creating a free-text lesson

1. Choose the lesson type Free text, then open Edit content.
2. Write the task — what is to be assessed.
3. Store a model solution — the AI compares against it.
4. Define the grading criteria — e.g. "Factual correctness", "Completeness of the argument", "Own examples or reflection". The criteria are free keywords; there is no percentage weighting (you can write it into the text for orientation).
5. Set the maximum score (e.g. 100).
6. Save.

The settings at a glance

Per question (each question has its own values):

Field	What it's for
Task	Required field. The text that learners see and answer.
Model solution (optional)	A sample answer the AI orients itself on. The more concrete it is, the more accurate the grading.
Grading criteria	Keywords the AI grades by. Add with Enter, remove the last one with Backspace.
Min. words / Max. words	Guideline for the length of the answer. Learners see the minimum as a hint at the word counter; a shorter or longer answer is not rejected.
Maximum score	The question's top score. The AI awards points between 0 and this value.

For all questions of the lesson (box Global settings):

• System prompt for AI (optional) — additional instructions to the AI that apply to every question of the lesson. You don't need it; without an entry, the AI grades with its default settings. Use it if the AI should grade in a particular tone or with a particular strictness, for example: "Grade generously, the participants are first-year apprentices. Give the feedback in plain language."

Which AI model does the grading is determined by the platform — you can't choose it in the lesson.

What the AI grading delivers

Per submission, the learner receives:

• Score relative to the maximum score
• Strengths of the answer
• Suggestions for improvement
• Overall feedback as a coherent text

When AI grading makes sense and when it doesn't

Makes sense	Use with care
Reflection, application questions, "explain in your own words"	Highly critical assessments where the result has consequences (certificate, certification)
Fast feedback in large classes	Specialist knowledge the AI doesn't reliably know
Exercises without grade pressure	Answers that are meant to be creative — the AI tends to favour "normal" answers

For critical assessments, enable the course toggle "AI gradings require approval by an authorised person" — a passed AI grading then only counts towards completion once you have approved it.

Reviewing and overriding AI gradings

There is a dedicated area for this in the main navigation: AI gradings.

1. The list first shows the submissions waiting for approval. These are only submissions from courses with an approval requirement. Via the status filter you also see all other AI-graded submissions of your company. Submissions from courses without an approval requirement appear there as KI-bewertet: they already count, but you can override them. You can also filter by score range, among other things (handy for borderline cases, e.g. 60-70%).
2. Per submission you see the answer, the AI score, and the AI feedback.
3. Approve — you confirm the AI grading. Or Override — your score and your feedback replace the AI grading (the original AI score remains internally traceable).

If a lesson has several questions, each answer appears separately in the list. The lesson is only completed once the answers to all questions have been reviewed and reach the minimum score.

Who may approve

The AI gradings section is visible to trainers, content managers and company administrators — not only the person who created the course. Whoever checks first, approves. For courses with an approval requirement, everyone who may approve gets a notification via the bell as soon as a new grading is waiting. There is one notification per course; as long as it is unread, no further one is added for that course. Clicking it opens the list.

Setting the minimum score sensibly

Whether a quiz lesson has to be "passed" is controlled by the minimum score of the lesson (in the lesson form). Learners have to reach it in order to complete the lesson — below that it says "Try again".

Score	What it means
0% (form default)	No hurdle — the lesson counts as completed on submission
50%	Very relaxed hurdle — only the basics have to be there
70%	A solid middle ground, tolerates gaps
80-90%	Demanding, typical for mandatory training with a compliance angle
100%	Perfect — only sensible for very short quiz lessons, otherwise frustrating

Learners may repeat failed quiz lessons as often as they like — the result of the passed attempt is what gets stored. There is no attempt limit and no waiting period between attempts.

Frequently asked questions

How do I stop learners from simply guessing answers in multiple choice? - More answer options (5 instead of 3) reduce the guessing success rate to 20% instead of 33% - Explanations per question: at least people learn something when guessing wrong - Set the lesson's minimum score higher - Use a free-text or a drag & drop lesson instead of multiple choice, where guessing works less well

Learners complain about the AI grading — that it's too strict/too lenient. First step: review your model solution and the criteria. Are they really clearly formulated? Are acceptable variations missing? Beyond that, you can override individual gradings under AI gradings or enable the approval requirement for the course.

How do I see whether a quiz question is too hard/too easy? At lesson level, the Course details tab under Administration → Analytics & Reports (visible from the Content Manager role upwards) shows attempts, completions, and average scores — a lesson with a conspicuously low average score probably contains questions that are too hard or unclear.

Can I import quiz questions from an Excel/CSV file? Quiz questions have to be entered, or they can be generated with AI. For many similar questions (e.g. vocabulary), an H5P quiz with a bulk-input mode is worthwhile.

A lesson is marked as "completed" even though learners didn't reach the full score. "Completed" means: the lesson's minimum score was reached, i.e. passed. 80% with a 70% minimum score = passed. If you want to be stricter, set the minimum score higher.