Explore Webb's four DOK levels with classroom examples and a printable task-demand toolkit. Distinguish cognitive complexity from difficulty when planning.
Main, P. (2023, May 11). Webb's DOK Levels: A Teacher's Guide to Depth of Knowledge. Structural Learning. https://www.structural-learning.com/post/webbs-depth-of-knowledge
What are Webb’s Depth of Knowledge levels?
Webb's Depth of Knowledge (DOK) is a four-level framework for judging the cognitive complexity of a task: recall, skill or concept, strategic thinking, and extended thinking. It asks how much reasoning and transfer a task demands, not how hard it sounds. For teachers, it helps match questions and assessments to the thinking you actually want.
A task can sound demanding and still ask learners only to recall, or sound modest and ask them to reason from evidence. Telling the two apart changes the questions you set, the support you give and what you accept as an answer.
Try this tomorrow
Year 8 science, worked example: sugar dissolving. Run the practical you already planned, then change the question you ask about it. At DOK 2, ask "Explain, using the particle model, why sugar dissolves faster in warm water." At DOK 3, ask "Which factor would you change first to speed up dissolving? Plan a fair test and justify your variables."
Look for: at DOK 2, learners link the speed to moving particles. At DOK 3, learners choose a variable, say what they will hold constant and defend the choice with what they saw in the practical.
You are looking at tomorrow's task and asking one question: what thinking will this demand of the learner? Webb's Depth of Knowledge gives that question four possible answers, and the answer changes what you plan, not how hard you make it. This guide shows you how to judge a task's DOK level, change the question without throwing the activity away, and hold the demand when learners struggle.
The chart below sets the four levels side by side rather than as a staircase. A lesson moves between them, and what learners do with the content sets the level of a task, whatever its place in the chart.
Webb’s four Depth of Knowledge levels, with a climate change task at each. Examples are possible demands, not fixed ratings.
Level
Name
What learners do
One example task
Level 1
Recall and reproduction
Retrieve information from memory: terms, definitions, historical facts or a routine procedure.
Define climate change, or name the greenhouse gases.
Level 2
Skills and concepts
Show they understand a concept or skill: explain it, interpret data to support a claim, or summarise the key ideas from a text.
Explain a cause, or read a graph.
Level 3
Strategic thinking
Reason with what they know: argue from sources, choose and justify a method, or judge which of two explanations fits the evidence better.
Decide which of two policies would cut more emissions, and defend the choice with data.
Level 4
Extended thinking
Run an investigation, design or argument that draws on several sources over several lessons, and revise when the first idea fails.
Investigate a local question over several lessons and report to someone who can act on it.
DOK measures the thinking a task demands, not how hard it feels: Level 1 recalls or follows a routine procedure, Level 2 processes concepts, Level 3 reasons strategically, Level 4 sustains synthesis and reflection (Webb, 1997).
The verb does not set the level: "Describe the water cycle" is Level 1 when learners repeat taught stages and Level 3 when they use rainfall data to explain a flood.
Keep the task, change the question: ask for a response that requires reasoning with evidence, then check that the instructions and support preserve that demand.
Norman Webb built the Depth of Knowledge model in 1997 at the University of Wisconsin-Madison, as a way for review panels to check whether a state test asked for the same depth of thinking as the standards it claimed to assess (Webb, 1997). It does more than check whether a task is difficult. The model has four levels. DOK and Bloom's taxonomy answer different questions: Bloom's names the type of thinking, such as analyse or create, while DOK rates how much reasoning and transfer the task demands, so a high Bloom's verb can sit at any DOK level.
For example, "describe the water cycle" may be Level 1 when a learner repeats taught stages. It becomes Level 3 when they use rainfall data to explain why a local flood happened. This guide helps teachers plan questions, activities and assessments that move learners from recall towards strategic and extended thinking, without mistaking difficulty for depth. This article is part of our thinking frameworks guide, which reviews 42 models and shows when to use each one.
Teachers use the model in three ways. It gives a department a shared language for the demand of a task, it shows where a unit asks only for recall, and it helps staff check that classroom tasks prepare learners for the assessment they will face. In a survey of 528 senior high school teachers in Ghana, teachers saw DOK as useful for planning tasks and assessments, and they named large classes, test-driven teaching, few resources and little training as the obstacles (Adade, 2026).
The Four Levels of Webb's Depth of Knowledge
Webb's Depth of Knowledge helps teachers design and review tasks by cognitive demand (Webb, 1997; Webb, 2002). Each level names the demand of a task, so one topic can appear at every level.
Watch for this
The levels are not a staircase. Choose the demand that serves the curriculum goal, and teach the knowledge learners need for it. A lesson may move between recall, conceptual processing and reasoning without requiring a fixed march through all four levels.
The four levels describe increasing cognitive complexity. The first level is recall: learners retrieve information from memory. This may be terms, definitions, historical facts or a routine procedure, including one with several steps.
At the second level, learners must show they understand a concept or skill. They might explain a concept, interpret data to support a claim, or summarise the key ideas from a text.
Level three asks learners to reason with what they know. They might argue from sources, develop a research question, choose and justify a method, or judge which of two explanations fits the evidence better. There may be more than one defensible approach, even when the final answer is fixed.
The fourth level asks for extended thinking: an investigation, design or argument that draws on several sources over several lessons. Learners plan, gather, synthesise and present, and they revise when the first idea fails.
Take climate change. At Level 1, learners define it or name the greenhouse gases; at Level 2, they explain a cause or read a graph. At Level 3, they decide which of two policies would cut more emissions and defend the choice with data. At Level 4, they investigate a local question over several lessons and report to someone who can act on it. Hands-on work sits at every level too: sorting animal cards into mammals and birds is Level 1, building a water-cycle model is Level 2, and designing a prototype to cut canteen waste, then justifying the materials, is Level 3.
Save or print the chart, then judge tomorrow's task against all four levels before you choose the question.
Where the planning toolkit is on this page
The Webb Task-Demand Toolkit further down provides four level descriptors, a filled classroom example and a blank task audit. Use these to review the full task and expected response. Routine application can remain DOK 1; the decisions, reasoning and evidence required determine the demand.
Where DOK came from, and the wheel that grew around it
The Webb label is misleading. Unlike the SOLO taxonomy, which describes the structure of learners’ responses, DOK describes the demand of a task and makes no claim about how learners acquire knowledge. It is an alignment method: the question it asks is whether test items reflect the cognitive demand of the intended curriculum (Webb, 2007).
Compare the thinking expected, the task taught and the evidence the assessment asks for.Read the figure description
Three points form a triangle: curriculum expectation, learning opportunity and assessment evidence. The triangle asks teachers to compare the thinking expected, the task taught and what the assessment would show. A task label alone does not prove alignment. This is a planning adaptation.
Teachers can still use DOK for lesson and assessment design, but its strongest use is alignment. It helps staff ask a simple question. Are learners being taught and assessed on recall, routine application, strategic reasoning or extended investigation at the point where the curriculum expects that demand?
Used well, DOK helps teachers design better assessments. Used badly, it becomes a tick-box language of higher-order tasks, even when the underlying work is still recall. That difference matters for curriculum leaders.
DOK grew from US standards and assessment alignment work. Teachers elsewhere can use the same task-analysis question without treating it as a national requirement. A UK department can compare a curriculum objective, an assessment question and the responses it expects.
Understanding the DOK Wheel
The DOK wheel is the most shared DOK image online, and Webb never made it. It files each verb at one level, which is the misreading Webb has warned against: a wheel hides whether a learner must use one fact, connect two ideas, justify a strategy or sustain an investigation.
A safer test is to cover the verb and ask what comes after it. Describe the character is Level 1 if learners recall a taught description and Level 3 if they use several quotations to defend a judgement about motive.
One verb, two levels
Cover the verb. Ask what comes after it.
Level 1Describe the characterLearners recall a taught description.
Level 3Describe the characterLearners use several quotations to defend a judgement about motive.
In PE, keep the same throw and change the evidence requested. Naming the technique and defending a tactical choice can require different thinking. Judge the full question and expected explanation rather than the activity alone.
Cover the verb. Ask what comes after it.
DOK vs Bloom's taxonomy
Benjamin Bloom's 1956 taxonomy sorts the type of thinking learners use. Webb's DOK (Webb, 1997) looks at how deeply they must engage with the content. The Cognitive Rigor Matrix, a grid that pairs each Bloom's process with each DOK level, crosses the two: learners may analyse at DOK 2 when they compare two familiar texts, and analyse at DOK 4 when they combine several sources to build and defend an original argument (Hess et al., 2009).
What Bloom’s taxonomy and Webb’s DOK each look at
What you compare
Bloom’s taxonomy
Webb’s DOK
What it looks at
The type of thinking learners use.
How deeply learners must engage with the content.
What it helps teachers check
It helps teachers name the type of thinking attached to a content aim.
Whether classroom assessments match the cognitive demand of the curriculum standards, and whether learners must use knowledge in the way the standard expects.
When used together
Names the cognitive process.
Checks the depth and context of the task.
The trap
Reading a high Bloom’s verb as a high DOK level.
A create instruction can still require routine recall when the method and information are supplied.
Another difference is assessment alignment. Bloom's Taxonomy helps teachers name the type of thinking attached to a content aim. Webb's DOK helps teachers check two things. Do classroom assessments match the cognitive demand of the curriculum standards? And must learners use knowledge in the way the standard expects?
Webb's DOK should be implemented through task analysis and moderation, not by choosing a verb from a wheel. Teachers should compare examples together and agree the evidence needed for each level. Then they should revisit their ratings after looking at learner responses.
The two work together when each does its own job: Bloom's names the cognitive process, DOK checks the depth and context of the task. The trap is reading a high verb from a Bloom's taxonomy verbs list as a high DOK level. A create instruction can still require routine recall when the method and information are supplied. A poster made from a template, with every fact supplied, is creating at Level 1.
How Solo Taxonomy Complements DOK Levels
Biggs and Collis (1982) developed SOLO to describe the structure of an observed response. DOK concerns the task's intended cognitive demand. Together they can prompt questions about the task and the work produced, but neither diagnoses a learner's ability or the cause of a gap.
The same reasoning task can receive a brief statement or a connected explanation. Inspect the response and its factual accuracy before deciding what to teach next. A task label records the intended demand; the learner's response shows the thinking that occurred. If your school already uses Marzano's new taxonomy, run the same check on the task and the response before trusting any label.
Teacher voices
The questions teachers raise
These posts raise practical questions about task demand, preparation and explanation. Use them to review what a task asks learners to do.
Is DOK a ladder from easy to hard?
Webb's Depth of Knowledge isn't a ladder where Level 1 is easy and Level 4 is hard.
Alice Keeler challenges the idea that DOK levels simply rank difficulty. When planning, judge the reasoning a task requires, alongside the knowledge and support learners will need.
Have learners practised this kind of reasoning?
I was very wrong. Most of the class got it wrong and it opened my eyes.
Robert Kaplinsky reports introducing a higher DOK problem on an assessment. His experience is a reminder to plan opportunities to practise the required reasoning before using a task to assess it.
What makes a number placement defensible?
5th grade students at YES are creating a decimal number line to the thousandths
Stella Miramontes reports a Grade 5 decimal-number-line activity in which learners justify why their number belongs in a particular place. The explanation gives the teacher something to inspect beyond the placement itself.
Difficulty is not depth
Common misconception
Difficulty is how many learners get a task right. Depth is the kind of thinking it requires. The two are independent, and treating DOK as a difficulty sticker is the commonest way it goes wrong.
A recall task can be difficult while remaining recall. Ten definitions instead of five add work without necessarily changing the cognitive demand. Check what learners must do with the information. In a validity study of online higher-education IT assessments, task difficulty was significantly related to Bloom-based complexity but not to Webb's DOK level (Hayes et al., 2021). The setting was higher education, so it supports the principle without settling it for school tasks.
Adding items can add work while leaving the kind of thinking required unchanged.Read the figure description
Two boxes list five taught definitions and ten taught definitions. A bracket links both to retrieval. Below it, one question concerns who can complete the task in context; the other asks what thinking the task requires. Adding items does not automatically add reasoning.
A curriculum map can label tasks DOK 3 and 4 while the enacted lesson asks mainly for recall. Stein, Grover and Henningsen (1996) analysed 144 mathematics tasks and found that intended high cognitive demand tended to decline during implementation. This observational study used its own task-demand framework and tested no Webb DOK intervention. In a Year 7 science lesson, collect an explanation of which model best fits a new observation and check the reasoning learners actually use.
Using DOK in your classroom on Monday
Teachers using Webb's DOK should first check what thinking the curriculum and assessment actually require. DOK works best in planning and moderation meetings, where staff can inspect tasks and learner work together. It should not become a learning-walk checklist used to judge a teacher during a 20-minute visit.
The move that works needs no new task. Keep the one you already have and change the question you ask about it.
The move in three steps
Keep the taskKeep the decimal number line, the source, the practical.
Ask a different questionAfter learners have done it, ask for a reasoned decision supported by subject evidence.
Check the demandCheck whether that response actually requires non-routine reasoning.
A Year 5 class placing decimals on a number line may use recall or conceptual processing, depending on the numbers and support. Asking why a placement is correct can stay at DOK 2. A non-routine comparison of competing explanations, requiring learners to choose evidence and defend a strategy, may add strategic demand.
Keep the number line and change the evidence learners must use to explain a decision.Read the figure description
A zero-to-one line marks 0.35 between 0.25 and 0.5. The first prompt asks learners to place 0.35. Amir’s claim says that 0.35 is greater than 0.5 because 35 is greater than 5. The second prompt asks learners to decide whether he is right and explain why the idea is tempting. The image states that 0.35 is less than 0.5. The DOK demand remains conditional on prior teaching and the reasoning required.
Try this tomorrow
Year 5 maths, worked example: decimals on a number line. After learners place 0.35 on a 0 to 1 line (DOK 1), pose a disagreement: "Amir says 0.35 is bigger than 0.5 because 35 is bigger than 5. Decide whether he is right and explain why his idea is tempting." The disagreement can move the task to DOK 3, as in the maths row of the table below.
Look for: in a primary class, learners who compare the two decimals by place value, name the slip in Amir's reasoning and use the number line to show where 0.35 and 0.5 sit.
Webb's DOK helps teachers align teaching, activities and assessment. Differentiation should not mean giving some learners a lower DOK level by default. Adjust vocabulary, models, talk frames, worked examples and time instead, so more learners can reach the intended cognitive demand (Webb, 1997).
One task at four depths: maths, reading, science
Every guide lists the levels. What teachers ask for is the same task at each level, side by side, so the difference is visible. The three subject columns below start from one classroom task and rewrite it four times. Nothing about the topic changes; what changes is what the learner has to do with it.
One task at DOK levels 1 to 4 in maths, reading and science
Level
Maths: fractions, decimals and percentages
Reading: the class novel
Science: dissolving
DOK 1: recall
Write three quarters as a decimal.
Who tells the story, and where does it open?
Name two substances that dissolve in water.
DOK 2: skills and concepts
Order three quarters, 0.7 and 65% from smallest to largest and show how you converted each one.
Summarise what the narrator learns in chapter three in two sentences.
Explain, using the particle model, why sugar dissolves faster in warm water than in cold.
DOK 3: strategic thinking
Two learners disagree about whether three quarters or 0.7 is larger. Decide who is right and explain why the other answer is a tempting mistake.
Using three quotations, argue whether the narrator can be trusted.
Plan a fair test to find which factor most affects how fast sugar dissolves, and justify your choice of variables.
DOK 4: extended thinking
Over a week, collect three real prices or measurements given as fractions, decimals or percentages. Compare when each form is useful, develop and test a guide for younger learners, then revise it using their responses and reflect on its limits.
Compare how this novel and one other text you have read handle an unreliable narrator. Develop an interpretation across both texts and a critic's view, test it against counter-evidence, then revise the argument and explain what changed.
Investigate over several lessons how a local water company removes dissolved substances. Integrate evidence from at least two sources, compare possible methods, test and revise a recommendation, and explain its limitations.
The examples illustrate possible demands, not fixed ratings. DOK 1 can include a routine multi-step procedure. DOK 2 adds conceptual processing; DOK 3 requires non-routine reasoning. DOK 4 needs sustained analysis, synthesis, iteration and reflection. Time, several sources or an audience on their own leave a task short of Level 4.
An extended inquiry needs connected evidence, testing, revision and reflection.Read the figure description
A possible canteen food-waste inquiry has six labelled boxes. They name data collection, comparison of at least two sources, a recommendation, testing options, revision and explaining limitations. Solid lines connect the boxes. Dashed side lines link revision with data collection and development with testing. The centre reads “Review as evidence changes”. The footer names sustained, iterative and reflective work. Time, sources and an audience alone do not establish Level 4.
The verbs barely matter: explain appears at Level 2 and Level 3, and the reading column uses no verb from any wheel. To move a task you already teach up one row, keep it and add the next level's question after it. Keep the recall and add the reasoning on top; learners need both in the same lesson, and the recall is where the reasoning gets its material.
Classroom adaptation
Check the task behind the verb
The post distinguishes Bloom's taxonomy from DOK. The planning suggestion below adapts that distinction; it does not report a complete teacher lesson sequence.
Reported distinction
Try in your planning
Bloom’s taxonomy and DOK describe different aspects of a task.
Choose one question. Check the reasoning, evidence and connections needed to answer it. A verb such as “create” can still lead to a routine response.
Depth of Knowledge is subject specific. In maths, DOK 3 may involve selecting a strategy, representing a problem and justifying each step. In English, DOK 3 may involve weighing quotations, authorial intent and context to defend an interpretation. The same verb carries different demand because each subject has a different knowledge structure (Webb, 1997; Hess et al., 2009). Webb's own descriptors for reading, writing, mathematics, science and social studies sit in an unpublished 2002 paper, and the examples below draw on them.
DOK examples by subject and level, drawn from the paragraphs below and the level descriptions above. Possible demands, not fixed ratings: duration, several sources or an audience alone do not establish Level 4.
Level
English
Maths
Science
DOK 1
Recall specific details from a text, such as the main characters or the setting.
Recall number facts and basic formulae.
Recall facts about the laws of physics or ecosystems.
DOK 2
Apply knowledge of literary devices to the text, for example by identifying symbols or interpreting metaphors.
Apply those facts and formulae to problems such as finding the area and volume of three-dimensional shapes.
Interpret data to support a claim.
DOK 3
Compare viewpoints or infer why a character acts as they do, and defend the reading with quotations.
Analyse data to predict a trend and justify the model chosen.
Explain how a change in the environment affects a species, using data they have to select, or use evidence from one experiment to predict the result of a related one.
DOK 4
Write a longer piece that draws on more than one text and takes a position a reader could argue with.
Apply maths to a real situation over several lessons, for example designing and costing a footbridge.
Design and carry out investigations into real questions, such as a new energy source for the school or the local effects of climate change, and report what they found.
English Language Arts DOK Examples
In English, teachers can use DOK to build reading and writing activities at all four levels. At Level 1, learners recall specific details from a text, such as the main characters or the setting. At Level 2, they apply their knowledge of literary devices to the text, for example by identifying symbols or interpreting metaphors.
Level 3 tasks ask learners to compare viewpoints or infer why a character acts as they do, and to defend the reading with quotations. Level 4 tasks are longer pieces of writing that draw on more than one text and take a position a reader could argue with.
Mathematics DOK Examples
In maths lessons, teachers can use DOK to challenge learners to apply their understanding of concepts to real problems. At Level 1, learners recall number facts and basic formulae. As tasks grow more complex, they apply those facts and formulae to problems such as finding the area and volume of three-dimensional shapes.
Learners at Level 3 analyse data to predict a trend and justify the model they chose. Learners at Level 4 apply maths to a real situation over several lessons, for example designing and costing a footbridge.
Science DOK Examples
In science classes, teachers can use DOK to challenge learners to apply their knowledge to real-world phenomena. At Level 1, learners could be asked to recall facts about the laws of physics or ecosystems.
Vygotsky (1978) argued that learners can manage harder thinking when a teacher or a peer supports them through talk. At Level 3, learners explain how a change in the environment affects a species, using data they have to select. They can also use evidence from one experiment to predict the result of a related one. At Level 4, learners design and carry out investigations into real questions, such as a new energy source for the school or the local effects of climate change, and report what they found.
Creating DOK-Based Assessments
When planning assessments, teachers should map the intended DOK level against the learning objective before they write questions. A balanced end-of-unit assessment may include recall and routine application. It should also show where learners must justify a strategy, use evidence or sustain an investigation (Webb, 1997).
Maths exams often test several levels. Level 1 can include applying a formula to a routine problem. Level 2 adds conceptual decisions; Level 3 requires non-routine reasoning and justification. Level 4 needs sustained synthesis and reflection, rather than time alone. Expect a shock the first time you add a Level 3 item.
Use the responses to plan the next explanation, example or scaffold. A difficult first attempt is a reason to examine what learners need, rather than automatically removing the task.
Formative and summative uses of DOK need different maps. Formative checks ask what evidence the teacher needs today to adjust the next explanation, task or scaffold. Summative maps ask whether the final assessment matches the cognitive demand of the curriculum standard. Black and Wiliam (1998) showed why assessment evidence should change teaching, not merely record performance.
DOK question stems by level
DOK question stems are the openings that prompt a level of thinking, such as "Explain how..." or "Decide whether...". Use them to draft questions quickly, then cover the verb and ask what comes after it, because the same stem can set a recall task or a reasoning task depending on what learners were taught first. For more on questions that stretch thinking, see our guides to higher-order questioning and higher-order thinking skills.
DOK question stems by level, with the check that sets the level
Level
Stems to start from
Check before you set it
DOK 1: recall
What is ...? Name ... List the ... from the last lesson. Use the method we practised to work out ...
Is every part of the answer something learners were taught, or a routine they practised?
DOK 2: skills and concepts
Explain how ... works. Compare ... and ..., then say what they share. Summarise ... in two sentences. Show what happens to ... when ... changes.
Must learners process a concept, such as choosing how to represent it or explaining a relationship?
DOK 3: strategic thinking
Decide which of ... and ... is better and defend your choice. Plan a fair test to find out ... and justify each variable. Two learners disagree about ...: who is right, and why is the other answer tempting?
Is there more than one defensible route, and must learners choose and defend theirs?
DOK 4: extended thinking
Investigate ... over several lessons, using at least two sources, then present a recommendation. Design and test ..., revise it after the first trial and explain what changed.
Does the work need sustained analysis, synthesis, revision and reflection, beyond time or several sources?
Use the one-page set of question stems to draft a question at each level, then cover the verb and check the task.
AI-Generated DOK Questions
AI tools can draft question stems, but treat the output as a candidate task. Check the content, instructions, support and acceptable response yourself. Ask whether a learner could complete it by copying supplied information or applying a taught procedure before labelling it strategic reasoning. Give the tool the learning objective, the year group and what learners already know, ask for one task at each level, then run the cover-the-verb check on every one.
Calibrating Formative Assessment to DOK Levels
Match formative assessment checks to the intended demand of the lesson. For DOK 1, a brief retrieval quiz works well. For DOK 2, ask learners to explain the relationship between two concepts using a graphic organiser.
DOK 3 could see learners critique worked examples and pinpoint the flawed step. DOK 4 uses peer review of projects against agreed criteria. Between levels, a hinge question can show whether learners are ready to move from recall to processing. Dylan Wiliam's 2011 book Embedded Formative Assessment makes the same point: match the check to the cognitive demand, rather than only asking whether learners understood.
Using DOK in Special Education
Use DOK to review the task, not to categorise learners receiving SEND or EAL support. Match the intended demand to the curriculum and the learner's agreed provision. Vocabulary support, worked examples, talk prompts and additional time may improve access, but do not guarantee that every learner will reach DOK 3 or 4.
In practice, DOK is not neutral across culture, language or neurology. A learner may fail a DOK 3 task because of the language load, the executive-function demand or an unfamiliar context. That does not always mean they cannot reason strategically. When teaching special education learners, separate the thinking demand from the barriers you can remove.
DOK Implementation Best Practices
Activities that look complex may still require only recall, while a short question can require strategic reasoning. Use department moderation to compare task evidence and agree the DOK level. Note where prior knowledge or language demand may change what learners actually experience.
Teacher voices
Real public posts on X, quoted word for word. We have not edited or endorsed them.
@TolentinoTeach Unfortunately many teachers are required to fill them up by admin: ie posted objectives and standards, jargon like dok level for each lesson, etc.
DOK isn’t about making the task harder. It’s about changing the thinking.
Same activity. Different questions. Bigger thinking.
After reading @alicekeeler breakdown of some common DOK myths, I wanted to bring the idea into PE.
Don’t raise the DOK. Raise the thinking. https://t.co/XroQRsoEAR
@misterminsoo As a retired Literacy teacher I can confidently tell you that is the equivalent of a 3rd grade “right there”, Depth of Knowledge Level 1 and Bloom’s Taxonomy Knowledge (most basic) level question- I had to go to college to know all that🤦🏼♀️🤦🏼♀️🤦🏼♀️
@yeswarriors1 Great job! 5th grade students at YES are creating a decimal number line to the thousandths and they go one step further to deepen their understanding by justifying why their number goes in that place. Ms Lopez has her DOK question stems to to choose from ✔️🤩 https://t.co/3h5Gb5Cgpm
The Depth of Knowledge Toolkit below gives you the four level descriptors, a filled Year 8 / Grade 7 science example and a blank task-demand audit, as printable pages for your classroom and staff room. Use the audit page in a planning meeting to record, task by task, what learners must do.
Classify these eight tasks
Teacher-training courses often set this as a quiz, and the quiz items turn up in search. Here are eight tasks of the kind they use. Decide the level of each before you read the key, and remember that DOK is about what the task requires of the learner, not the verb it opens with (Webb, 1997; Hess et al., 2009).
Set the eight tasks out on a wall and let colleagues place them at Levels 1 to 4 before they read the key.
List the capital cities of five European countries.
Convert 45 minutes into a fraction of an hour.
Explain why the distributive property is useful when multiplying 7 by 98 in your head.
Sort ten materials into conductors and insulators, then say what the conductors have in common.
Using evidence from the passage, decide whether the writer wants us to like the main character.
Design a plan to test the physical properties of three rock samples, and justify your choice of tests.
Write a forty-word summary of the water cycle in your own words.
Over four weeks, investigate how food waste in the school canteen could be cut. Use at least two sources and your own data, and present a recommendation to the site manager.
Answer key: the provisional level for each of the eight tasks
Task
Level
Why
Task 1
Level 1
Straight recall. It can be hard if the learner does not know them, and it is still recall.
Task 2
Level 1
One routine procedure with one right answer. Webb files a simple algorithm at Level 1 however many learners get it wrong.
Task 3
Level 2
The learner explains how the distributive property supports this calculation. A conceptual explanation may be Level 2; one expected answer alone does not determine the level.
Task 4
Level 2
This may be Level 2 if learners identify a conceptual pattern and explain it. Sorting by a supplied or memorised property could remain Level 1. Inspect the prior teaching and expected response.
Task 5
Level 3
There is more than one defensible answer and the learner must choose evidence and argue from it.
Task 6
Level 3
The verb is design, which most wheels file at Level 4. The task is one planned investigation with a justified method, which is strategic thinking. It would need sustained analysis, synthesis, iteration and reflection to support Level 4. Carrying out a routine test over several lessons is not enough.
Task 7
Level 2
Summarising in your own words is skills and concepts. It becomes Level 3 only if the learner must decide what matters most and defend the choice.
Task 8
Level 4
Potentially Level 4 if learners integrate the evidence, test options, revise their recommendation and reflect on limitations. Four weeks, multiple sources and an audience alone do not establish extended thinking.
If your ratings differ
If your ratings differ from this provisional key, compare the full task, prior teaching and evidence required. Wyse and Viger (2011) found some inconsistency between item-writer and committee ratings, but did not test these eight examples. Bring contrasting ratings to a moderation discussion and identify which information would resolve the disagreement.
Teacher planning mini app
Check the task demand
Use Webb's DOK as a moderation lens. Choose a tentative level yourself, then record what learners actually have to do.
This mini app does not classify a task. DOK is judged from the thinking learners must sustain, not a verb, a label or extra time.
Written by the Structural Learning Research Team
Reviewed by Paul Main, Founder & Educational Consultant at Structural Learning
Limitations and Critiques of Webb's DOK
Webb's DOK is useful, but it is not a universal measure of lesson quality. Schools should use it to improve task design, curriculum mapping and assessment moderation (Webb, 1997).
Raters can place the same task at different levels. Wyse and Viger (2011) studied item writers in a Midwestern US alternate-assessment system and found some misconceptions and inconsistencies with committee ratings. That study examined how raters read the framework and measured no learner achievement. Agreement can be strong: in a Ghanaian technical-education mathematics study, raters coding curriculum and exam items to Webb's levels reached a kappa of 0.83 (Folson and Awuah, 2025). Compare task-specific evidence in moderation, and treat a label as a judgement that needs checking.
Checkbox compliance. When schools mandate DOK alignment, teachers sometimes relabel existing tasks instead of redesigning them. A worksheet that asks learners to "analyse" may still require only recall if the information sits on the same page. A label alone cannot guarantee the cognitive demand (Hess et al., 2009).
Motivation and context. DOK says little about motivation, classroom relationships, cultural knowledge or the social conditions that support extended thinking. A Level 4 inquiry can still fail if learners feel they do not belong, lack language access or lack secure foundational knowledge. Combine DOK with formative assessment, explicit teaching and scaffolded discussion (Black and Wiliam, 1998).
Evidence about effects. DOK describes cognitive demand and causes no learning by itself. None of the DOK studies cited in this guide tests an intervention on learner outcomes. In the Ghanaian study, alignment between curriculum and exam did not strongly predict pass rates (Folson and Awuah, 2025). Use the framework to review alignment, then inspect what learners actually produce.
Different study contexts. The evidence comes from US rating and mathematics-task studies, an Indonesian school study, a Ghanaian curriculum study and a higher-education assessment study. Rahmawati and colleagues (2021) studied 47 Grade 12 learners in a polymer and STEM project, and the largest response category combined Bloom's understanding level (C2) with DOK 1. That qualitative design cannot establish a causal DOK benefit. An ambitious project format is a reason to inspect the responses, and gives no proof of their depth.
DOK in projects, online lessons and staff training
Three planning tables for uses beyond the single lesson. Each is the same lens, what the task requires, applied to a project, an online sequence and a staff training session.
A visible product or activity format does not establish the thinking learners must sustain.Read the figure description
A poster, online lesson, group discussion and long project point to one inspection panel. It asks what knowledge is supplied, what decisions learners make, what evidence they use and what thinking they must sustain. A format alone does not set a DOK level.
A project across four levels
Take a challenge to cut local air pollution. A DOK 1 task lists the common causes of local air pollution. A DOK 2 task compares how well different public transport options reduce emissions. A DOK 3 task asks learners to propose a strategy to encourage cycling and justify it with evidence. A DOK 4 task has them design and run a survey of local transport habits, analyse the data and present a policy recommendation to the school council.
A project across four levels: task description and learner action example
DOK Level
Challenge task description
Learner Action Example
DOK 1: Recall & Reproduction
Identify basic facts or definitions related to the challenge.
Learners list common causes of local air pollution.
DOK 2: Skills & Concepts
Apply information, compare, or classify within the challenge context.
Learners compare the effectiveness of different public transport options for reducing emissions.
DOK 3: Strategic Thinking
Formulate a plan, justify a solution, or analyse complex aspects of the challenge.
Learners propose a strategy to encourage cycling in their community, justifying their choices with evidence.
DOK 4: Extended Thinking
Design and execute an investigation, synthesise multiple sources, or create an original solution over time.
Learners design and conduct a survey on local transport habits, analyse the data, and present a comprehensive policy recommendation to the school council.
Online lessons
Online lessons drift toward recall unless the sequence plans for more. The table gives one strategy and one learner activity at each level.
Online lessons at each DOK level: lecture strategy and learner activity
DOK Level
Online Lecture Strategy
Learner Activity Example
Level 1: Recall
Micro-lectures with embedded comprehension checks.
Learners answer multiple-choice questions recalling key terms from a short video.
Level 2: Skills/Concepts
Interactive simulations or guided problem-solving demonstrations.
Learners choose a suitable way to represent the data and explain what a pattern means.
Level 3: Strategic Thinking
Case study analysis in virtual breakout rooms or structured debates.
Learners collaboratively analyse a historical document, identifying biases and forming arguments.
Level 4: Extended Thinking
Long-term research projects or complex problem-based learning scenarios.
Learners design and present a solution to a real problem, drawing on several sources over several weeks.
Staff training
When you plan professional development, DOK levels can help you set the right level of challenge. A DOK 1 task may ask teachers to recall key safeguarding policies. A DOK 2 task could ask them to use these policies in a practice scenario. At DOK 3, teachers may review how well different assessment strategies work in their subject. They would then support their choices with clear evidence.
Staff training at each DOK level: teacher professional development task
DOK Level
Teacher Professional Development Task Example
DOK 1
Recall the main components of the school's behaviour policy.
DOK 2
Explain how to differentiate a lesson for learners with specific learning needs.
DOK 3
Analyse learner data to identify patterns in misconceptions and propose targeted interventions.
DOK 4
Design, implement, and evaluate a new whole-school literacy initiative over a term.
Choose one task and three learner responses for your next planning conversation. Compare the reasoning they show before deciding what to teach or ask next. Use metacognition routines to ask learners why they chose an approach and where it needs revision.
Free Resource Pack
The pack is the Webb Task-Demand Toolkit: four level descriptors, a filled Year 8 / Grade 7 science example and a blank task audit, on two printable A4 pages. Choose one curriculum goal for your next planning session. Review the response you want learners to produce before choosing a classroom activity or digital planning tool.
Frequently asked questions
What does DOK stand for in education?
DOK stands for Depth of Knowledge. It is Norman Webb's four-level scale for how much thinking a task or test item demands, built in 1997 to check whether state assessments matched their standards (Webb, 1997). Teachers now use the same four levels to plan questions and check that classroom tasks ask for the reasoning the curriculum expects.
How many DOK levels are there, and what are they?
There are four. Level 1 is recall and reproduction: a fact, a definition or a routine procedure, including routine procedures with several steps. Level 2 is skills and concepts: apply, classify, summarise, explain a relationship. Level 3 is strategic thinking: reason with evidence, justify a choice, plan an investigation. Level 4 is extended thinking: an investigation or project that draws on several sources over several lessons.
Is there a printable DOK levels chart?
Yes. The Free Resource Pack above is a two-page A4 PDF, the Webb Task-Demand Toolkit, with four level descriptors, a filled Year 8 / Grade 7 science example and a blank task audit. The levels chart near the top of this page sets out the same four levels in one view.
What is a DOK level 3 question? Give me an example.
A Level 3 question makes the learner reason and justify, usually with evidence they have to choose. In history: Two accounts of the 1926 General Strike disagree about who ended it. Decide which is more reliable and say why. In geography: Choose the best site on the map for a new flood defence and defend it against one alternative. In PE: Watch the two rallies and justify one change to the losing player's positioning.
The tell is that there is more than one defensible route and the learner has to defend theirs.
What are DOK question stems?
DOK question stems are sentence openings that prompt a level of thinking, such as "Explain how..." for Level 2 or "Decide which... and defend your choice" for Level 3. The stems table above gives examples for all four levels. A stem helps you draft, but the level comes from the task, so cover the verb and ask what the learner must do with the content.
How is Webb's DOK different from Bloom's taxonomy?
Bloom's names the type of thinking (remember, understand, apply, analyse, evaluate, create). DOK names how deeply the learner must engage with the content to complete the task. They are two axes, not two versions of one ladder, which is why Hess et al. (2009) crossed them into the Cognitive Rigor Matrix. A task can ask learners to create something and still sit at DOK 1 if all it needs is recall.
Is the DOK wheel accurate?
No. The wheel of verbs is a planning aid that Webb never made and has warned against. It files each verb at one level, but the same verb sits at different levels depending on what the learner must do with the content. Use the wheel to find question openers if you like, then judge the level from the task, not the verb.
Which verbs go with each DOK level?
Verbs help you draft a question, but none fixes a level. Explain appears at Level 2 and Level 3, describe at Level 1 and Level 3, and create can stay at Level 1 when the method and information are supplied. Judge the level from what comes after the verb: what learners must do with the content and what they were taught first.
Does a higher DOK level mean a harder task?
No. Difficulty is how many learners get it right; depth is the kind of thinking required. Recalling the capital of Kazakhstan is hard and still Level 1. Explaining why two familiar fractions are equal is easy for many and still Level 2. Plan the depth you want first, then adjust difficulty with vocabulary, scaffolds and time.
How do I work out the DOK level of a task?
Read the full task and expected response, including what has already been taught or supplied. Ask whether it requires routine recall, conceptual processing, non-routine reasoning or sustained synthesis and reflection. Use the eight examples above for moderation; a verb or automatic verdict is insufficient.
Webb, N. L. (2002). Depth-of-Knowledge levels for four content areas. Unpublished paper, Wisconsin Center for Education Research, University of Wisconsin-Madison.
Paul Main is an educator and metacognition researcher who founded Structural Learning in 2017. With a psychology degree from the University of Sunderland and a career spent helping schools embed thinking skills, he bridges the gap between educational research and classroom practice. Fellow of the RSA and Chartered College of Teaching, with 128+ Google Scholar citations. More at paulmain.com.
