How to review an MCAT full length

When to review

Not immediately after finishing. You have just spent a long morning making decisions and the review you do in that state is mostly re-reading. Leave it until the next day, while you can still remember what you were thinking on individual questions — that memory is the thing review runs on, and it does fade.

There is no correct number of hours. You will see the rule that review should take at least as long as the test did; treat it as someone's habit rather than a finding. A focused pass that reaches a decision beats a thorough one that reaches a list.

What to look at first

Sort the whole test into four piles before examining anything. The order matters, because attention runs out and you want it spent on the questions that carry information.

1. Wrong, and you thought you were right. Start here, always. These are the questions where your reasoning felt sound and was not, which makes them the only place a genuinely invisible habit shows up. Everything else you already had some signal about.
2. Wrong, and you knew you were guessing. Quicker to handle. Usually these are content you have not learned or a stem you could not parse, and both are legible once you look. Do not spend the same effort here as on the first pile.
3. Right, but shaky. You hesitated, changed your mind, eliminated your way there, or cannot now say why the answer was right. These are misses that have not happened yet, and skipping them is why accuracy can look stable while a score sits still.
4. Right, and solid. Skip. You can state the reason and the reason is the real one. Reviewing these is the part of "review every question" that consumes the afternoon and returns nothing.

That fourth pile is usually the largest, and giving yourself permission to leave it alone is what makes the rest of this possible.

Within the first pile, the fastest way in is the option you actually chose. MCAT distractors are engineered — each wrong option is written to catch one specific error — so the one you selected is evidence about what you did, even when you cannot remember doing it. The question to ask is not why is C right but what would someone have had to do to find my option attractive. The per-question review procedure works through that one question at a time.

One is an event, three is a pattern

You will now have a pile of individually explicable misses and an instinct to fix all of them. Most of them do not need fixing, and the useful part of a full length is deciding which ones do.

The working threshold:

One occurrence is an event. Two is a coincidence. Three is a pattern.

Below three, note what happened and carry on. Every named failure will happen to you once for a reason that never repeats — you misread a word, you were tired at question 190, the question was genuinely ambiguous. Rebuilding your study plan around a single bad question is the most common way a full length makes preparation worse instead of better.

The threshold is mostly permission to ignore things. It converts an overwhelming pile into a small number of real findings, and small numbers are the only kind you can act on.

Why one full length is not a trend

A separate and stricter rule applies to direction. To say something is getting better or worse you need three separated points in time — not three occurrences inside one sitting.

Three text not consulted misses in one exam tells you about that exam. It might tell you about those particular passages. The same three spread across three weeks tells you about you. Only the second one is a trend, and only the second one justifies changing what you study.

This is the most common error in full-length review, and it runs in both directions. A score five points below your last one is not a decline, and a score five points above it is not progress. One test is one observation. Sitting with that is uncomfortable and it will stop you from redesigning your preparation every fortnight.

The findings that cross sections

The score report hands you a section breakdown, and the natural conclusion is I am weak in B/B. It is almost always the wrong conclusion to draw from a full length, because a section is a subject grouping and the things worth finding are not organised by subject.

Group by how the reasoning failed instead. Two misses in different sections that failed the same way are one finding, not two. A biochemistry miss where the most-rehearsed rule arrived before you finished the stem and a physics miss where exactly the same thing happened are one habit — and filed by section they look unrelated, so each gets its own topic review and the habit gets none.

This gives you a useful hierarchy. A pattern living inside one topic is probably a content gap, and content gaps are the easy case: go and learn the content. A pattern spanning three sections is a mechanism, and mechanisms are what actually cost you points across a whole test.

When you have a choice of explanations, prefer the one that reaches furthest across the exam. The names for what you are looking for are in why “careless” is never the answer.

You will see advice to find the one main reason for lost points in each section. That is better than a topic tally and it still stops at the section boundary — which is the boundary the finding usually crosses.

What you can stop working on

Review is almost always additive: a list of things to start doing. It should also be subtractive, and this is the part nobody does.

Look for what was a problem on your last two tests and is not a problem on this one. Something you were drilling has stopped producing misses. That is a finding, it is as real as any failure, and the correct response is to stop spending time on it.

This matters more than it sounds, because study time is fixed. Every topic you keep reviewing out of habit after it has been fixed is time not spent on the thing that is still costing you points. A weakness that is fading on its own needs no plan item at all.

The same three-point rule applies before you trust it: something absent from one test is absent from one test.

Name what is working

A review that is entirely failures does not get repeated. That is a practical problem rather than a motivational one — the process has to survive being run again next month, and one that produces only evidence of inadequacy will not.

So name what went right with the same specificity you used for what went wrong. Not CARS felt better. Something like: on three passages I went back to the text before committing, and all three were correct — that is a behaviour you can identify and keep doing deliberately.

Vague credit is useless in the same way vague blame is. Both fail the test of implying a next action.

Turning review into one decision

The endpoint is not a list. A list of twelve things to fix is a list of zero things you will do, and it is what most full-length reviews produce.

Find the single mechanism that explains the most misses, and let everything else be evidence for or against it. Then decide one thing that changes before the next test. One.

Some tests genuinely do not support a single root. If the misses do not ladder up to anything — if they are scattered, one-off, and unconnected — say that and do not manufacture one. An invented root sends you off drilling a habit you do not have, and the honest reading of some exams is that they were a spread of unrelated events plus a bad morning.

Where this gets hard

Everything above is doable with the exam in front of you. The difficulty is the three-point rule, because the third occurrence usually lands weeks after the first, in a different section, on a different test — by which time the first two are in a document you have stopped opening. Cross-test patterns are the findings that matter most and the ones a person doing their own review is least equipped to see, not through lack of discipline but because nobody holds nine weeks of questions in their head at once.

Reading one test is a single instance of the larger question of why a score stalls — the same triage and the same threshold, applied across every session rather than to one exam in isolation.

That gap is what CLIVEA is built around. Your tutor is in the session while you study, so the analysis is written as you work rather than becoming homework afterwards, and every session is read together with every session before it — which is where a third occurrence becomes visible as a third occurrence.

Your AAMC full-lengths remain your own, and they should. AAMC writes the exam, so their official material is the last word and nothing replaces it. CLIVEA is the months before that — the practice where habits are still forming and still cheap to change.