QuizMagic’s Bloom’s taxonomy quiz generator lets you go beyond “easy, medium, or hard” and target the specific kind of thinking you want students to demonstrate, such as recall, application, analysis, integration, or original creation. Instead of hoping the AI produces questions at the right cognitive depth, you tell it exactly which levels you want and how many questions each level should contain.
This guide covers all three cognitive frameworks available in QuizMagic: Bloom’s Taxonomy, SOLO Taxonomy, and the Mixed (Bloom’s plus SOLO) mode, with step-by-step configuration instructions, worked examples, and practical tips for getting the best results from each framework.
Good news: All three cognitive frameworks are available to every user, Free and Premium, at no extra cost. No upgrade is required to use Bloom’s, SOLO, or Mixed.
Why Cognitive Levels Matter for Quiz Design
Most AI quiz generators produce questions that cluster at the Remembering level of Bloom’s Taxonomy because recall questions are the easiest to extract from source text. The result is a quiz full of “What is the definition of X?” and “Name the three Y” questions that test whether students memorized content, not whether they understood it.
This matters because Remembering-level questions are the easiest to answer correctly without genuine understanding. A student who reads a summary the night before a test can often score 70 percent on a recall-only quiz without being able to apply a single concept the next week. Conversely, a quiz that includes Applying, Analyzing, and Evaluating questions identifies students who understand deeply and students who only superficially recognize terminology.
QuizMagic’s Bloom’s taxonomy quiz generator solves this by building the cognitive framework into the generation step. You set the cognitive level before the AI writes a single question. As a result, every generated question reflects the level of thinking you actually want to assess, not the level the AI defaults to under no instruction.
What each framework measures
| Framework | Levels available | Best for |
|---|---|---|
| Bloom’s Taxonomy | 6 levels: Remember, Understand, Apply, Analyze, Evaluate, Create | Targeting the type of thinking a question demands |
| SOLO Taxonomy | 4 selectable levels: Unistructural, Multistructural, Relational, Extended Abstract | Targeting the structural complexity of the required answer |
| Mixed (Bloom’s + SOLO) | 11 curated combinations | Combining cognitive process with answer complexity for advanced rubric design |
Note on SOLO: A fifth SOLO level — Prestructural, exists in academic literature, but it represents a failure to engage with the question at all. QuizMagic does not generate Prestructural questions because no teacher intentionally writes items to test misunderstanding.
Which framework should you use?
Use Bloom’s Taxonomy when you want questions distributed across the thinking-skills ladder, for example, 4 Remembering, 3 Applying, and 3 Analyzing questions in a single quiz. This is the most common framework for classroom formative and summative assessments.
Use SOLO Taxonomy when you care about the structural complexity of what students must produce, for example, a quiz that starts with single-fact items and builds toward multi-concept synthesis. SOLO is particularly well-suited for rubric design and writing assessments.
Use Mixed (Bloom’s + SOLO) when you want both dimensions simultaneously. A combination like “Apply + Relational” forces students to use a concept in a new situation and connect it to other ideas, which is a meaningfully higher bar than Apply or Relational alone.
Step 1: Choose Your Source
The Bloom’s taxonomy quiz generator works with every input source QuizMagic supports. Choose the source that contains the material you want to assess.
| Source type | How to use it |
|---|---|
| Topic | Type a subject directly — for example, “Photosynthesis, Grade 10” or “Contract Law: Offer and Acceptance” |
| Text | Paste your notes, handout content, or any block of source material |
| File upload | Upload PDF, Word (.docx), PowerPoint (.pptx), Excel (.xlsx), plain text (.txt), or image files (JPG, PNG, GIF, BMP, WebP) |
| YouTube URL | Paste any public or unlisted YouTube link with captions available |
| My Uploads | Select a file you previously uploaded without re-uploading it |
| Multi-Source | Combine different sources like files, text, and YouTube. |
For the most curriculum-specific questions, upload your actual source material, like your lecture PDF, your slide deck, or your chapter handout, rather than typing a topic. The AI generates questions from your content, not from a generic knowledge base about the subject.
Step 2: Configure Basic Quiz Settings
Before selecting a cognitive framework, configure the standard quiz settings. These interact with your framework choices.
| Setting | Free plan | Premium plan |
|---|---|---|
| Number of questions | 1 to 10 | 1 to 100 |
| Difficulty | Easy, Medium, Hard | Easy, Medium, Hard |
| Quiz type | MCQ, True/False, Fill in the Blanks, Short Answer, Mixed, Essay | All types |
| Output language | Auto-detect or choose a specific language | Same |
Language note: Quiz content is generated in your chosen language. However, cognitive level labels and descriptions remain in English regardless of the output language setting. For example, the badge on a question always shows “Apply” rather than its translation in the target language.
The Difficulty setting in Step 2 acts as a global default. When you configure cognitive frameworks in Step 3, you can override this with a per-level difficulty setting, which means a single quiz can include Easy Remembering questions alongside Hard Analyzing questions.
Step 3: Select Your Cognitive Framework
In the quiz generator, locate the Cognitive Framework dropdown. You will see four options.
None (Standard Mode) is the default. In this mode, the AI selects question types and cognitive levels naturally from the source material. This tends to produce questions clustered at the lower Bloom’s levels, as noted above.
Bloom’s Taxonomy activates the Bloom’s level distribution panel.
SOLO Taxonomy activates the SOLO level distribution panel.
Mixed (Bloom’s + SOLO) activates the combination panel with 11 curated pairings.
Select the framework that matches your assessment goal, then proceed to the relevant configuration step below.
Step 4a: Configuring the Bloom’s Taxonomy Distribution
When you select Bloom’s Taxonomy, the distribution panel shows all six levels with a count field and a difficulty selector for each.
The six Bloom’s levels explained
| Level | What it asks students to do | Sample action verbs | Example question stem |
|---|---|---|---|
| Remember | Recall specific facts, terms, definitions, or formulas from memory | define, list, recall, identify, name | “What are the four stages of mitosis?” |
| Understand | Explain ideas in their own words, classify, summarize, or interpret | explain, summarize, classify, describe, interpret | “In your own words, explain why the cell membrane is described as selectively permeable.” |
| Apply | Use knowledge or a principle in a new or unfamiliar situation | solve, demonstrate, use, execute, implement | “Given this customer complaint scenario, which company policy applies, and why?” |
| Analyze | Break material into components, identify relationships, compare or contrast | compare, contrast, differentiate, examine, organize | “Compare the economic causes of World War I with those of World War II. What structural similarities do you find?” |
| Evaluate | Make and defend a judgment based on evidence, criteria, or standards | evaluate, justify, defend, critique, argue, assess | “Evaluate the claim that the Industrial Revolution improved living standards for the majority of the working class. Justify your position with evidence.” |
| Create | Produce something original by combining knowledge in a new way | design, construct, compose, formulate, propose, hypothesize | “Design an experiment to test whether plant growth rate is affected by the color of light. Outline your method and predict your results.” |
Configuration rules
- Set the count for each level you want included. Levels with a count of zero are excluded from the quiz.
- The sum of all level counts must equal your total number of questions. The panel shows a running total and alerts you if the counts do not match.
- Each level has its own difficulty selector, independent of the global difficulty setting. This means you can generate Easy Remembering questions alongside Hard Analyzing questions in the same quiz.
- The per-level count maximum is the lower of 50 or your total question count.
Worked example: end-of-unit biology quiz (20 questions)
A teacher covering cellular respiration for Year 11 students might configure the following distribution for a summative quiz.
| Level | Count | Difficulty | Rationale |
|---|---|---|---|
| Remember | 4 | Easy | Establish baseline recall of terminology |
| Understand | 4 | Medium | Check conceptual comprehension |
| Apply | 5 | Medium | Test transfer to novel cell scenarios |
| Analyze | 4 | Hard | Require comparison between respiration and photosynthesis |
| Evaluate | 3 | Hard | Defend a claim about metabolic efficiency |
| Create | 0 | — | Excluded from this quiz |
| Total | 20 |
This distribution produces a quiz that tests genuinely different cognitive depths rather than twenty variations on the same recall task.
Step 4b: Configuring the SOLO Taxonomy Distribution
When you select SOLO Taxonomy, the distribution panel shows four generative levels.
The four SOLO levels explained
| Level | What it measures | What a student response looks like |
|---|---|---|
| Unistructural | The student identifies one relevant aspect of the topic | A single correct fact stated without connection to anything else |
| Multistructural | The student identifies several relevant aspects, but treats them as separate items | A list of correct facts with no explanation of how they relate |
| Relational | The student integrates multiple aspects into a coherent understanding | An explanation that connects ideas and shows how they interact |
| Extended Abstract | The student generalizes the understanding to a new context or produces an original insight | An answer that goes beyond the given information to identify broader principles |
The same configuration rules apply as for Bloom’s: the sum of level counts must equal the total number of questions, and each level has its own difficulty selector.
True/False tip: SOLO levels work particularly well for declarative-statement quizzes. The generator includes a True/False adaptation instruction when SOLO is active, so each level is expressed as a statement students can mark true or false rather than as an open-ended prompt. This makes SOLO-aligned True/False quizzes significantly more cognitively targeted than standard True/False question sets.
When to choose SOLO over Bloom’s
SOLO is the right choice when you want to assess the depth of integration in student understanding rather than the type of thinking. A teacher assessing whether students have moved from surface knowledge (Multistructural) to genuine conceptual understanding (Relational) benefits more from SOLO framing than from Bloom’s. SOLO is also more natural for rubric design in subjects like Literature and Philosophy, where the quality of the argument matters more than the cognitive process used to produce it.
Step 4c: Configuring the Mixed (Bloom’s + SOLO) Distribution
Mixed mode pairs a Bloom’s level (what the question demands) with a SOLO level (how complex the answer must be). QuizMagic offers 11 curated combinations that make pedagogical sense together.
The 11 available combinations
| Combination | What it produces |
|---|---|
| Remember + Unistructural | Recall one specific fact |
| Remember + Multistructural | Recall several items as a list |
| Understand + Unistructural | Explain one concept clearly |
| Understand + Multistructural | Describe multiple aspects without connecting them |
| Understand + Relational | Explain the connections between concepts |
| Apply + Multistructural | Use multiple procedures in sequence |
| Apply + Relational | Connect theory to a practical context |
| Apply + Extended Abstract | Apply knowledge to a genuinely novel scenario |
| Analyze + Relational | Compare or contrast while explaining the relationships found |
| Analyze + Extended Abstract | Identify generalizable patterns across what was analyzed |
| Create + Extended Abstract | Produce original work that extends meaningfully beyond the given material |
How to configure Mixed mode
Click Add Combination to add a row to the distribution panel. Select a combination from the dropdown, enter the question count for that combination, and set the difficulty. Add additional rows for each combination you want to include. The total across all rows must equal your total question count.
Advanced use case: Mixed mode is particularly powerful for project-based and research-grade assessments where you want to assess both process (“how are you thinking?”) and output complexity (“how integrated is your response?”). The combination “Analyze + Extended Abstract” is especially effective for capstone and portfolio assessments.
Step 5: Generate the Quiz
Click Generate Quiz. Several things happen behind the scenes that are worth understanding.
How your framework configuration becomes questions
Your level distribution is converted into a structured instruction that is sent to the AI alongside your source material. Each question the AI generates is tagged with a specific Bloom’s level, SOLO level, or combination, which determines the badge displayed on the question card and the data stored for analytics and export.
Handling large quizzes
For quizzes over approximately 34 questions, the generator splits the work into batches of roughly 17 questions each. Importantly, your cognitive distribution is split proportionally across batches. If you configured 30 percent Analyzing questions in a 60-question quiz, each batch produces approximately 30 percent Analyzing questions, so the final quiz matches your intended distribution rather than front-loading one level.
Mixed quiz types with cognitive frameworks
If you select Quiz Type as “Mixed” (combining Multiple Choice, True/False, Fill in the Blanks, and Short Answer) alongside a cognitive framework, both settings compose cleanly. The AI generates questions at the specified cognitive levels in the question formats specified by the Mixed quiz distribution.
Step 6: Review and Refine
After generation, cognitive level badges appear on each question card in the quiz editor. These badges show the assigned Bloom’s level, SOLO level, or combination for each question.
What you can do in the editor
Toggle badge visibility. The “Show Cognitive Levels” control in the results header lets you hide or show all badges simultaneously.
Review the distribution. Before saving, confirm that the question set reflects the distribution you configured. Pay particular attention to higher-level questions (Analyzing, Evaluating, Creating, and Extended Abstract). These are the most demanding to generate accurately and occasionally benefit from manual refinement.
Fallback behavior: If you select a cognitive framework but do not configure any level counts, the generator falls back to standard mode silently. No error appears. If your quiz does not show cognitive level badges after generation, check whether you saved your distribution configuration before clicking Generate.
Cognitive Levels in Sharing and Exports
Cognitive level data travels with the quiz through every sharing and export surface.
| Surface | Cognitive badges shown |
|---|---|
| Teacher quiz preview and editor | Yes |
| Saved Quizzes dashboard | Yes |
| PDF and Word exports | No |
| Smart Sharing teacher analytics dashboard | Yes |
| Student quiz-taking view (live session) | No — badges are hidden while students take the quiz |
| Student post-submission review | No |
Hiding cognitive level badges from students during the quiz prevents the label from acting as a hint about what kind of answer is expected. A student who sees “Evaluate” knows they need to make a judgment and defend it, which defeats the purpose of assessing whether they can recognize that a question requires judgment.
Tips and Best Practices
Pair difficulty and cognitive level intentionally. “Easy + Create” produces a creative task with a low technical barrier. This is well-suited for younger students or introductory creative exercises. “Hard + Remember” tests obscure recall of precise technical terminology. This is useful for certification exam preparation. The combination tells the AI more than either setting alone.
Use a scaffolded Bloom’s distribution for most classroom quizzes. A 10-question quiz with 3 Remember, 4 Apply, and 3 Analyze questions tests the same content at three cognitive depths and produces more diagnostic data than 10 questions at a single level.
Use SOLO for capstone assessments. A SOLO-sequenced quiz that starts at Unistructural and ends at Extended Abstract functions as a learning ladder. Students can see their own progression through the levels and receive feedback specific to where their understanding currently sits.
Do not spread small quizzes too thin. A 10-question quiz distributed across all 6 Bloom’s levels gives you roughly 1 to 2 questions per level. That sample is too small to draw reliable conclusions about any individual level. For a 10-question quiz, choose 2 to 3 levels and concentrate questions there.
Use the Additional Notes field to guide the AI. Before generating, add specific instructions in the Additional Notes field: “Focus on real-world examples from manufacturing contexts” or “Generate questions based on the case studies in Section 4, not the theoretical introduction.” The AI reads these notes alongside your source material and adjusts accordingly.
Combine with anti-cheating for high-stakes use. When using higher-level Bloom’s questions (Analyzing, Evaluating) for graded assessments, enable Smart Sharing with anti-cheating monitoring. Higher-order questions are significantly harder to answer with a generic AI tool, but combining them with copy/paste blocking and tab-switch detection closes both vectors.
Frequently Asked Questions
Is the Bloom’s taxonomy quiz generator available on the Free plan? Yes. Bloom’s Taxonomy, SOLO Taxonomy, and the Mixed (Bloom’s + SOLO) framework are all available on both Free and Premium plans. No upgrade is required to use any of the three frameworks.
Does the cognitive framework work with every quiz type? Yes. Multiple Choice, True/False, Fill in the Blanks, Short Answer, Mixed, and Essay all support cognitive frameworks. True/False questions receive a special adaptation when SOLO is active, so that SOLO levels are expressed as declarative statements that students can mark true or false.
What happens if I select a framework but do not configure any level counts? The generator falls back to standard mode silently and ignores the framework. No error or warning appears. If your quiz does not show cognitive level badges after generation, check whether your distribution counts were saved before you clicked Generate.
Can I assign a different difficulty to each cognitive level? Yes. Each level in the distribution panel has its own difficulty selector, independent of the global difficulty setting you configured in Step 2. This allows you to mix Easy Remembering questions with Hard Analyzing questions within the same quiz.
What is the maximum number of questions per level? The maximum per level is the lower of 50 or your total question count. For example, if you are generating a 10-question quiz, no single level can contain more than 10 questions.
Are cognitive level labels translated when I generate in another language? No. Quiz content is generated in your selected language, but cognitive level labels, descriptions, and badges remain in English regardless of the output language.
Do students see the cognitive level badges while taking the quiz? No. Badges are hidden from students during a live quiz session through the Smart Sharing feature.
What happens to the cognitive distribution when the generator processes large quizzes in batches? For quizzes over approximately 34 questions, the generator splits the work into batches and divides your configured distribution proportionally across those batches. The final quiz reflects your intended distribution rather than concentrating certain levels in one part of the quiz and other levels in another.
What is the Mixed (Bloom’s + SOLO) framework best used for? Mixed mode is most useful for advanced rubric design, project-based assessments, and any context where you want to assess both the type of thinking (Bloom’s) and the structural complexity of the response (SOLO) simultaneously. The combination “Analyze + Extended Abstract” is particularly effective for capstone and portfolio assessments at the university level.