Error hunt
Check understandingError hunt is a structure in which students look for the one error in a worked solution. The board shows a solution with numbered lines, and exactly one line is wrong. First, each student writes down alone the number of that line, and in one sentence why. Pairs compare and write the correct line, the class shows its numbers at once, and then the screen marks the error. Whoever got it right stands up and explains it to someone who missed it. Checking work becomes a skill the class practices, not something only the teacher does.
One problem · ± 14 min
Put it on the boardOne screen: per problem, how many lines are on the board and which one holds the error, and whether the class switches partners after the reveal. The worked solution itself goes on your board. Only you see it; the class sees it only when you press Start.
This exact screen goes on your board. The timer runs per step, the reveal and the explaining wait for you, and you stay in control: pause, add time or move on to the next step. After the class shows its numbers, the screen marks the wrong line, so you never have to give it away.
What the class sees on the board
7 steps · one problem, ± 8 min
For you: Say there is one error, not where. Check that everyone writes a number and a reason.
Group size
alone, then in pairs · the whole class
Duration
± 14 min
incl. briefing and look-back
Level
grades 4–12, built for middle school
Materials
a worked solution on the board, a mini whiteboard or slip of paper per student
Format
alone → pair → class → new pair
Purpose
finding an error and explaining why
Check one thing before the first time: does the error break a rule? A minus sign that does not flip when a term moves across the equals sign, minutes read as hundredths of an hour, a subscript changed instead of a coefficient: those give the class something to explain. The class will find an arithmetic slip (7 × 8 = 54) too, but the conversation afterwards lasts five seconds.
The flow at a glance
± 14 min totalThe minutes above are the default times, not a script: whoever prepares the structure sets the time for each step. Steps 4 and 6 have no clock: the reveal and the explanation to the class take as long as they take, and you move on. With more problems, steps 1 to 6 repeat; the look back comes once, at the end.
Step-by-step instructions
7 steps
Find the error
2.5 minalone, in silenceA worked solution with numbered lines is on the board, and exactly one line is wrong. You say there is one error, not where. Everyone writes on a mini whiteboard or a slip of paper the number of the wrong line, and in one sentence why.
Talk to your partner
3 minin pairsSame line? Same reason? Whoever has a different number explains why they chose that line. Then the pair writes the correct line together.
Show your number
0.5 minthe whole classOn your signal everyone holds up their number at the same time. Look around: which numbers do you see? You say nothing yet.
Here is the error
1 minthe whole classThe screen marks the wrong line. Everyone sees whether their number was right; the explaining comes next. This step has no timer: you move on. If everyone got it right, nobody switches this round.
Switch partners
2 minwith a new partnerWhoever had the right line stands up, picks someone who is seated and sits next to them. They explain why it is this line and how it should read; the one who stayed seated says what made their own line look wrong.
Explain it to the class
2 minthe whole classThe screen draws a number: that student's pair explains, and the one who missed the line at first does the talking. Then you confirm, and add only what was missing. This step has no timer.
Look back
1.5 minin pairsAfter the last problem: what did your partner show you that you didn't see yourself, what was hardest, and what will you do differently when you check your own work? Two or three pairs give one answer.
Several problems in a row? The new pair from the switch goes on to the next problem. In a crowded room, or the first time, choose «Stay seated» in the preparation: nobody walks, and the pair checks or fixes its own line.
Which learning strategy does the class train?
Retrieve from memory
Students pull the material from memory without looking in the book. Retrieval is itself a learning event: what you bring back stays better than what you reread. Source: Dunlosky et al. (2013), practice testing, the top rating of ten techniques.
On the boardFrom memory, book closed · You pull it from memory without looking. What you retrieve yourself stays.
Also trainable with this structure
- Ask why it is so · With your number you write one sentence why: a number without a reason is a guess.(natural sciences · text subjects)
- Teach each othersupporting · Whoever had the right line explains it to someone who missed it, not the teacher.
- Show what you understandsupporting · All numbers up at once: the teacher sees at a glance how many students found the error.
Why does this work?
Finding an error in someone else's work is not the same as solving a problem. You have to know how it should go, line by line, and see where it goes off. Error hunt adds three things that turn that search into learning.
Julie Booth and colleagues (2013) gave students in algebra classes worked examples with and without errors, and asked them to explain each one. The incorrect examples especially helped conceptual understanding: whoever explains why a step is not allowed says out loud the rule that stays silent in a correct step. Kelley Durkin and Bethany Rittle-Johnson (2012) found the same with decimals: students who compared an error with a correct example kept fewer misconceptions than those who compared two correct ones. That is why everyone writes down not only the number in step 1, but also why.
The number is on paper before the screen shows anything. Andrew Butler and Henry Roediger (2008) showed that feedback after your own answer makes the correct answer stick better and lets a wrong one linger less. Janet Metcalfe (2017) describes in her review that an error you make and then see corrected does not harm learning but helps it, and that errors made with confidence are corrected best. The condition is that the correction really comes: that is why the screen marks the line, and you stay quiet until then.
Noreen Webb (1989) brought together the research on small-group work: students who give elaborated explanations learn from giving them. So whoever missed the error gets an explanation, not just the answer. The switch in step 5 puts the two together: whoever had the right line goes to someone who missed it, and has to be able to really explain it. Then the one who missed it at first speaks to the class. If they can explain it now, the switch worked.
The problem: one error, and not a careless one
The content is yours: a worked solution with numbered lines, exactly one of which is wrong. The lines after it follow correctly from that wrong line. That way the class looks for the place where it went wrong, not for a wrong answer at the bottom.
Pick an error against a rule. A minus sign that does not flip when a term moves, a unit converted the wrong way, a subscript changed instead of a coefficient. An error like that has a why, and the why is the lesson. The best source is errors you grade again and again: the one half the class made on the last test, written out neatly and anonymously.
On the board and on the screen. You write the solution on the board, with the line numbers. In the preparation you only enter how many lines there are and which one holds the error: students write down a number, and after they show it the screen marks that line. If you want to keep the solution, you can add it; the class never sees that text on the screen.
Ready-made. For math, physical science, chemistry and life science there are preparations ready, each with a worked solution and one error against a rule. You will find them below with the examples.
When it fits, what to watch for
When it fits
- After teaching, before independent practice. The class has just learned the rule. Finding an error against it shows whether they really understand it, and the conversation about it sharpens the rule more than one more problem would.
- With an error you grade again and again. Put that error in a worked solution and let the class find it. Whoever has found and explained it once in someone else's work spots it faster in their own.
- For test review. Two or three problems in a row with the errors that came up often. With every round each student sits with a different partner, and whoever missed an error gets it explained by someone who saw it.
- When you want to see who gets it. When the numbers go up, you see at a glance how many students have the right line. Two numbers that come up often show which line also looked suspicious: a misconception still lives there.
- Outside science and math too. A sentence with one grammar error, an argument with one weak step, a timeline with one event out of order: wherever a worked solution follows rules, the structure works.
What to watch for
- When the content is still new: start with a correct solution. Große and Renkl (2007) found that incorrect solutions helped mainly students who already knew the material reasonably well; students with little prior knowledge learned more from correct examples only. With new content, show a correct solution first, and run the error hunt a lesson later.
- A careless slip: pick another error. Everyone finds an arithmetic slip, and then there is nothing to explain. Pick an error against a rule, so the why becomes the conversation.
- More than one error: take one out. With two errors half the class writes one number and half the other, and the screen marks only one. Make sure every other line is right, including the lines after the error.
- In a crowded room: stay seated. If half the class cannot get between the desks, choose «Stay seated» in the preparation. Nobody walks, and the pair checks or fixes its own line. The order stays the same: a number first, then the reveal.
- With English learners. Writing a reason takes a lot of language. Let the reason be one word or an arrow next to the line, and shorten step 1. The number and pointing at the line work in any language.
One structure, several functions
Examples by context
- Solve: 2(x − 3) + 4 = 3x − 1
- 2x − 6 + 4 = 3x − 1
- 2x − 2 = 3x − 1
- 2x − 3x = −1 − 2
- −x = −3
- x = 3
- A cyclist rides 6 miles in 20 minutes. What is the speed in mph?
- t = 20 min = 0.2 h
- v = d ÷ t
- v = 6 mi ÷ 0.2 h
- v = 30 mph
- Balance: H₂ + O₂ → H₂O
- Left: 2 H and 2 O. Right: 2 H and 1 O.
- The right side needs one more O: H₂ + O₂ → H₂O₂
- Left: 2 H and 2 O. Right: 2 H and 2 O.
- The equation is balanced.
- Parents: Aa × Aa (A = yellow pea seeds, dominant; a = green seeds)
- Gametes from each parent: A or a
- Possible offspring: AA, Aa, Aa, aa
- Green seeds: Aa and aa, so 3 out of 4
- Chance of green seeds: 75%
Variations
By age
Common mistakes
Frequently asked questions
What exactly is the Error hunt structure?
A structure in which students look for the one error in a worked solution with numbered lines, and explain why it is an error. Seven steps: Find the error (each student alone, in silence: the number and one sentence why); Talk to your partner (same line, same reason? write the correct line together); Show your number (everyone at once); Here is the error (the screen marks the line); Switch partners (whoever got it right picks someone seated and explains); Explain it to the class (a drawn number, and the one who missed it at first does the talking); Look back (on working in pairs). In education research, learning from wrong solutions is called erroneous or incorrect worked examples.
How long does Error hunt take?
With the default settings the whole structure takes ± 14 minutes, briefing and look-back included.
Whoever prepares the structure sets the time for each step. Finding and talking are the longest steps; showing the numbers takes half a minute. The reveal and the explanation to the class have no clock: they take as long as they take, and you click on. With more problems, the first six steps repeat; the look back comes once, at the end.
Why do students write a line number, not the correction?
Because a number is quick, the same for everyone, and the whole class can show it at once. When the numbers go up you see at a glance how many students found the right line, and which line also looked suspicious. The correction comes after, in the pair, where there is time to write it properly. A number without a reason does not count: that is why one sentence why goes with it.
Why does the screen mark the error, and not the teacher?
So you can stay quiet until the end. Whoever explains the error decides for the class what there was to understand; the screen only marks the line, and the why stays with the students. You confirm only after the explanation to the class, and add only what was missing.
Why do only those who got it right walk?
Because they have something to tell, and the others something to hear. Whoever got it right goes to someone still seated and explains; that way every student who missed the error gets their own explanation, from a classmate who just found it. At most half the class walks, and whoever is seated goes nowhere. If nobody should walk in a crowded room, choose «Stay seated» in the preparation.
What if everyone had the right line?
Then tap «Everyone got it right» at the reveal: the switch drops out of that round, because there is nobody to explain it to. Let a drawn pair tell the class the why. And take note: the error was too easy for this class, so next time pick an error against a harder rule.
What if almost nobody had the right line?
Then there is nobody to switch with, and that says something: the class cannot see this error yet. Explain it yourself, with the rule next to it, and put the same kind of error in an error hunt again a week later. To avoid it, start new content with a correct solution first.
Won't students learn the error from a wrong solution?
Not if the error is found and corrected. Metcalfe (2017) describes how an error you make and then see corrected does not harm learning but helps it. The condition is that the correction really comes, and the reveal and the explanation to the class make sure it does. So never leave a wrong solution up without the class knowing which line was wrong, and why.
Where do you find good solutions with an error?
For math, physical science, chemistry and life science there are ready-made preparations with the examples on this page. The best source is your own graded tests: the error that came up often, written out anonymously and neatly. For ideas, ask Microsoft Copilot for "common errors in [topic] for [grade] students, with the rule each one breaks". Write the solution out yourself and check every line: a second error that slips in by accident ruins the round.
How is it different from Predict – Observe – Explain and Which approach?
All three have each student commit to something alone before anyone talks. In Predict – Observe – Explain the class predicts what a demonstration will do; in Which approach? it sorts problems by method, without solving them. In Error hunt the solution is already on the board, and the class looks for where it breaks. Choose Error hunt when the class knows a rule and you want to know whether they recognize it when it is broken.
Where does the structure come from?
Working with wrong solutions comes from the research on learning from worked examples. Große and Renkl (2007) had students find and fix errors in worked probability problems; Durkin and Rittle-Johnson (2012) and Booth and colleagues (2013) showed in classes on decimals and algebra that incorrect examples with explanation strengthen understanding. That students learn by explaining to each other is what Noreen Webb (1989) described in her review of group work. LudensFlow put it in the order of cooperative learning: first alone, then in pairs, then with the class, and a switch that gives the explanation to the student who needs it.
Checking as a skill: which structure fits?
| Structure | Purpose | Group | Duration | When it fits |
|---|---|---|---|---|
| Error hunt | find an error and explain why | alone, then pairs | ± 14 min | the class knows a rule: do they spot it broken? |
| Which approach? | recognize the type of problem, without solving | alone, then pairs | ± 16 min | the class knows the methods, not when to use which |
| Predict – Observe – Explain | test a prediction against a demonstration | alone, then pairs | ± 17 min | an outcome the class can reason about |
| Pair Coach | solve out loud, the partner coaches | pairs | ± 16 min | practice with a set of problems |
| Numbered Minds | make sure together that everyone understands | 2–6 | ± 16 min | one right answer, everyone accountable |
| 1 goes, 3 stay | explain to another team | teams of 4 | ± 20 min | every team has something to explain |
Sources
- C. S. Große & Alexander Renkl, Finding and fixing errors in worked examples: Can this foster learning outcomes?, Learning and Instruction 17(6) (2007), 612–634: incorrect next to correct solutions helped learners with more prior knowledge; learners with little prior knowledge learned more from correct examples only.
- Kelley Durkin & Bethany Rittle-Johnson, The effectiveness of using incorrect examples to support learning about decimal magnitude, Learning and Instruction 22(3) (2012), 206–214: comparing an error with a correct example led to more understanding and fewer misconceptions than comparing two correct ones.
- Julie L. Booth, Karin E. Lange, Kenneth R. Koedinger & Kristie J. Newton, Using example problems to improve student learning in algebra: Differentiating between correct and incorrect examples, Learning and Instruction 25 (2013), 24–34: incorrect examples with explanation especially helped conceptual understanding in algebra classes.
- Andrew C. Butler & Henry L. Roediger III, Feedback enhances the positive effects and reduces the negative effects of multiple-choice testing, Memory & Cognition 36(3) (2008), 604–616: feedback after one's own answer made the correct answer stick better and a wrong one linger less.
- Janet Metcalfe, Learning from errors, Annual Review of Psychology 68 (2017), 465–489: errors that are corrected do not harm learning and can help it, provided the correction really follows.
- Noreen M. Webb, Peer interaction and learning in small groups, International Journal of Educational Research 13(1) (1989), 21–39: students who give elaborated explanations in a group learn from giving them.
Compiled by Oksana Oliinyk · Last updated 27 October 2026
Was this page useful?
After your answer we ask three short questions. Half a minute, and it genuinely helps us.