Taking a full practice test creates useful evidence. Reviewing only the wrong answers throws most of that evidence away. Correct guesses, slow correct answers, and repeated moments of uncertainty can be just as important as obvious misses.
A strong review has four passes: reconstruct the attempt, classify every costly question, repair the underlying decision, and schedule a retest. The goal is not to admire the explanation. The goal is to make the same error less likely next time.
Before reviewing: protect the evidence
During the test, mark questions that felt uncertain if the testing interface allows it. Do not stop to diagnose them. Immediately after the test, take five minutes to record:
- Where you felt rushed or mentally tired.
- Which section felt easier or harder than expected.
- Any questions you guessed on, even if the guess was correct.
- Any timing choice you would change.
Then take a real break. Review later that day or the next morning, while the decisions are still recognizable but frustration is lower.
Pass 1: Re-solve without the answer
Return to every wrong, guessed, flagged, or unusually slow question. Hide the correct answer and try again without a clock.
This separates two very different problems:
- You can solve it now: timing, attention, interpretation, or test procedure likely caused the miss.
- You still cannot solve it: the underlying content or method needs instruction and focused practice.
Write enough work to see the decision path. “Careless” is not a diagnosis. Name the action: copied the sign incorrectly, answered the wrong question, eliminated an option without evidence, or spent four minutes protecting an approach that was not working.
Pass 2: Classify the cost
Give each reviewed question one primary category. Keep the categories simple enough that you will use them consistently.
| Category | What it means | Likely response |
|---|---|---|
| Knowledge | You did not know the rule, concept, vocabulary, or formula. | Learn the concept, then complete a short focused set. |
| Method | You knew the content but chose an inefficient or unreliable approach. | Compare methods and rehearse the cleaner one. |
| Interpretation | You misunderstood the task, evidence, graph, or wording. | Practice paraphrasing the question before solving. |
| Execution | Your setup was sound, but arithmetic, notation, or answer transfer failed. | Add a specific check at the vulnerable step. |
| Pacing | You rushed, stalled, or ran out of time. | Set a decision rule for skipping, returning, or estimating. |
Use “content” only when you can name the content. “Algebra” is too broad; “rewriting a quadratic to expose its roots” is actionable.
An error log is valuable only when its labels tell tomorrow’s version of you what to do.
Pass 3: Write the correction as a decision
For each meaningful mistake, complete this sentence:
“Next time I see _____, I will _____ because _____.”
Examples:
- Next time a reading question asks for evidence, I will predict the answer from the passage before comparing choices because attractive choices often add an unsupported claim.
- Next time a system of equations looks algebra-heavy, I will check whether substitution or graphing exposes the relationship faster because my first method cost too much time.
- Next time I reach the final ten minutes, I will complete the questions I can enter quickly before returning to the hardest one because one stubborn problem should not hide several available points.
This turns review into a reusable cue. Copying the official explanation may show what worked once; writing a decision tells you what to recognize later.
Pass 4: Repair, then retest
Group the error log by skill and behavior. Choose the two or three patterns that cost the most points—not the ten topics that feel most interesting.
- Repair: review the underlying concept or procedure.
- Practice: complete a small set focused on that exact pattern.
- Mix: solve it again among unrelated questions, where the method is not announced.
- Retest: use a timed section to see whether the correction survives pressure.
Do not immediately retake the identical full exam. Familiarity can inflate the result. Use fresh questions for the next measurement, then revisit selected old questions later to confirm the method still holds.
Do not ignore correct answers
Review a correct answer when it was guessed, took much longer than nearby questions, or depended on reasoning you could not explain. Label it “fragile correct.” Those questions are warnings: the score survived this time, but the process may not survive a different version.
How long should review take?
A complete review can take as long as the test itself, especially early in preparation. That is not wasted time. One thoughtfully reviewed exam usually teaches more than racing through another full test before the first set of errors has changed anything.
If time is limited, prioritize:
- Wrong answers you can now solve.
- Clusters of misses in the same skill.
- Questions lost to pacing near the end of a section.
- Correct guesses and unusually slow correct answers.
Your review checklist
- I reviewed wrong, guessed, flagged, and slow questions.
- I re-solved before reading the explanation.
- I named the specific cause, not just “careless.”
- I wrote a next-time decision.
- I chose no more than three priority patterns.
- I scheduled focused practice and a fresh timed check.