Exam error log: turn a mistake into useful practice

1. Give your exam error log one clear job

Choose a small group of returned questions you can inspect carefully. Three questions from one practice session are a reasonable starting point, although the number should fit your available time. Include a correct answer that you guessed or cannot explain. A lucky choice and a reasoned solution can earn the same mark while revealing different study needs.

Keep the original question reference and your attempt available. You need to see the instruction, the information supplied and the step you actually took. If returned papers must remain unaltered for a review or appeal, use a separate permitted copy or a reference in your notebook. Do not photograph restricted examination material or publish another student’s work.

Cornell’s advice on reviewing a returned test encourages students to consider both mistakes and preparation. Use that broad reflection to choose one change you can observe on the next task. This article’s particular entry format is our own practical design; the university does not endorse it or Lirno.

2. Write a short entry that you can actually reuse

Use six fields: question reference, my decision, evidence of the problem, checked correction, next task and result. Write them as short headings on paper. A reference such as “practice sheet 3, question 5” is usually enough to find the original; copying every sentence into the log adds work without necessarily clarifying your decision.

For “my decision”, identify what you did, not what kind of person you think you are. “I divided by the final amount” is usable evidence. “I am terrible at percentages” is not a description of a mathematical step. In the next field, explain what conflicts with the instruction, calculation or source. Keep an uncertain diagnosis explicitly uncertain.

The correction should include a reason and its source: a teacher’s comment, an approved worked example or a calculation you can check independently. In “next task”, choose something you can attempt without seeing the correction. In “result”, record what you could explain and where you still needed help. Leave room for an unresolved question instead of forcing every entry to look complete.

3. Separate the visible mistake from its possible cause

A wrong answer is an outcome, not a reliable diagnosis. Perhaps you misunderstood a concept, selected an unsuitable method, lost a minus sign, misread a command or ran out of time. More than one explanation can apply. Start with the earliest step you can show is wrong, then ask what you understood when you made that choice.

If you can solve a similar question slowly but not under time pressure, compare the actual working before calling the problem “timing”. You may be omitting a check, or the method itself may still be unfamiliar. If you cannot explain why the first step is valid even with the notes open, revisiting the underlying idea is more useful than racing through another full paper.

Use a label only when it changes the action. A reading error suggests marking the requested quantity before working. A missing concept suggests a focused explanation and an easier example. A transcription error suggests comparing the copied value with its source. If the instruction itself is unclear, write a precise question for the teacher rather than invent a hidden rule.

4. Work through a percentage example without copying

Consider an original practice problem: a club’s membership rises from 40 to 50. Find the percentage increase. A student calculates 10 divided by 50 and reports 20%. The arithmetic matches that division, but the denominator does not match the requested comparison with the original membership.

An entry could say: “I used the final total as the base. The increase is 10 members, and the original total is 40, so 10 ÷ 40 × 100 = 25%.” Check by finding 25% of 40: it is 10, which takes the starting membership to 50. This check reconnects the percentage to the situation instead of merely repeating the same division.

Now close the correction and try a new case: a class collection rises from 60 books to 75. Explain which value supplies the base before calculating. The increase is 15 and the percentage increase is again 25%. To avoid memorising that answer, also consider 60 to 72: the increase is 12 and the percentage increase is 20%.

The useful log entry is the rule about the original comparison value, not a list of percentages to remember. If percentage change is itself unfamiliar, work through the guide to choosing the original value before returning to this reflection. The log can point to a lesson; it does not replace the lesson.

5. Use the same method for written answers

Imagine a fictional history question asks you to compare two accounts, but your response summarises each separately. Both summaries might be accurate while the answer still fails to make the requested comparison. The log should name the missing relationship: one supported similarity or difference, connected to the exact focus of the question.

Write a next task that requires that relationship. For example, use two permitted short extracts and draft one sentence that states a difference, followed by evidence from each. Check whether both pieces of evidence concern the same point. Do not improve the answer merely by adding “however” between unrelated observations.

Feedback about writing is often open to several defensible responses. Record the assessment criterion and ask whether your revision meets it; do not treat a model paragraph as the only possible wording. When the command itself is the obstacle, the guide to exam command words helps you identify what the task asks you to do before adding more subject knowledge.

6. Check explanations before you trust them

A confident explanation can still be wrong. Compare it with the original question, the course material and any teacher feedback. In a numerical task, test a substitution, estimate a sensible range or check units where appropriate. In a written task, return to the text and criterion. Two websites repeating the same statement are not necessarily independent confirmation.

If your reasoning and the marking appear to conflict, retain both and write the exact point of disagreement. Ask for clarification with the relevant step visible. Do not silently change the original attempt or assume that every lost mark proves a conceptual error. Equally, an answer key without working may establish a final value without explaining how to reach it.

Lirno is optional here. Where school rules permit, you can capture the complete task, verify the recognised text and ask for a hint or explanation about the specific step. Make your own attempt before comparing it. AI may misread information or reason incorrectly; Lirno does not guarantee correctness, grades, mastery or permission. It is free to download or start, while some AI usage levels or advanced features may require Premium. Never upload personal or restricted material just to fill the log.

7. Choose the next exercise by what it needs to reveal

An immediate repeat can check whether you understood a correction, but it may also be easy because the answer is still visible in memory. Later, try a permitted fresh question that changes the numbers, source or wording while retaining the decision you wanted to improve. There is no universal interval that guarantees the right result for every subject.

Keep the new task small enough to expose the step. If the difficulty was identifying the percentage base, explain that choice in two short problems before taking another whole test. If it was comparing evidence, practise one paired claim before writing an entire essay. A smaller task can reveal what you understand without the noise of several unrelated demands.

Record whether you worked independently, used a prompt or still needed a full example. These are descriptions of support, not grades for your character. Lehigh’s past-exam reflection resource offers another way to examine preparation and performance. Use it to formulate a concrete question when your own records do not reveal the obstacle.

8. Keep the notebook useful when time is limited

At the end of a study session, select one entry that affects an upcoming assignment or a recurring difficulty. Check the correction, attempt its next task and write a brief outcome. If a problem is resolved across different attempts, keep its reference available but reduce how often you revisit it. Do not promise yourself that one correct repetition means permanent mastery.

If many entries point to a topic you have never learned, stop cataloguing every wrong answer and schedule time for the basic lesson. If several entries depend on unclear feedback, take a small selection to your teacher. The notebook should help you choose the next learning action; it should not become another large assignment that crowds out actual practice.

A finished entry can be modest: “I chose the original amount independently; I still need help distinguishing an increase from a percentage-point difference.” That observation is more useful than a page of copied solutions. Keep the system free, brief and adaptable. Its value comes from the checked explanation and the next honest attempt, not from the number of pages you accumulate.