How to Review Your UCAT Mock Scores

The Score Is the Least Useful Part

A mock exists to answer one question: what exactly should you practise before the next one? The overall score cannot answer that. A 2700 built on a strong VR and a collapsing DM needs completely different work from a 2700 built the other way around, and even within one subtest, the number does not tell you whether you lost marks to method, timing or panic.

By all means track the number, and use our UCAT Score Calculator to see what it would mean for the universities you care about. But if you spend ten minutes looking at the score and ten minutes reviewing the questions, you have the ratio backwards. Reviewing a mock properly takes about as long as sitting it. If that sounds excessive, consider that a mock reviewed properly is worth more than two mocks reviewed badly, and mocks are a finite resource. There are only so many high-quality ones available before test day.

Review Every Question, Not Just the Wrong Ones

Go through the whole paper, including the questions you got right. You are looking for three kinds of question, and only one of them is marked wrong in the results screen.

The first is the genuine error, which we deal with below. The second is the lucky guess: right answer, wrong or absent reasoning. Mark these as errors in your review, because on test day the coin lands the other way. The third is the expensive correct answer, the question you got right in 110 seconds. That question cost you a mark somewhere else in the subtest, and it deserves the same attention as a wrong answer, because the fix (a cleaner method, or the discipline to flag and move on) is exactly the same kind of fix.

For every question in the first two categories, do it again untimed before you read the explanation. If you get it right untimed, your problem is speed or pressure, not knowledge. If you still get it wrong, the problem is method, and now the explanation is worth reading. This one distinction, can I do it at all versus can I do it in time, should shape everything you practise next.

Sort Your Errors Into Four Buckets

Every miss goes into one of four buckets, and each bucket has a different cure.

Method errors. You did not know how to approach the question, or your approach was wrong. Cure: go back and learn the technique for that question type, then drill it untimed until it is reliable. Speed work on a broken method just produces fast mistakes.

Timing errors. You knew how, but ran out of clock, either on that question or because an earlier question ate the time. Cure: timed drills on that question type, plus an honest look at your flag-and-skip discipline.

Careless errors. Misread the question, took the wrong number from the table, clicked the wrong option. These feel random but usually are not. If your careless errors cluster in the last third of a subtest, the problem is fatigue. If they cluster on a question type, something about your process for that type invites the slip. Cure: find the pattern first, then fix the process.

Judgement errors. Mostly a VR and DM phenomenon: you understood everything and still picked the wrong option, usually on an inference or a strongest-argument question. These need the most careful review, because the fix is recalibrating how you weigh options, and that only comes from studying many of your own wrong calls side by side. Our guide on VR accuracy covers the common miscalibrations in that subtest.

Keep an Error Log

A spreadsheet with one row per miss: mock number, subtest, question type, which of the four buckets, and one sentence on what actually happened. The sentence matters. "QR, percentages, careless, took percentage of old value instead of new" is useful eight mocks later. "Got it wrong" is not.

The log earns its keep after three or four mocks, when the patterns become undeniable. Students are consistently surprised by what theirs shows. The subtest they worry about is often fine, while a single question type they never think about is quietly costing six marks per sitting. You cannot see that pattern inside one mock, which is precisely why the log exists.

The testCould someone else read your error log and know what you should practise this week? If yes, it is working. If it is a list of scores, it is a diary, not a tool.

Turn the Review Into a Plan

The output of a mock review is a short, written plan for the gap before your next mock. Something like: drill Venn diagram questions untimed until accuracy is above 90%, do ten minutes of mental maths daily, practise VR with a hard 30-second cap on True/False/Can't Tell questions. Three items is plenty. A plan with eight priorities is a plan with none.

Then space your mocks far enough apart to act on the plan. A mock every two or three days sounds industrious, but there is no time to fix anything between sittings, so the same errors repeat and the score flatlines, which is demoralising for no benefit. For most of your preparation, one mock a week with real targeted work in between beats three mocks a week, with the frequency only rising in the final stretch before your test date.

If you have done a few mocks and the pattern still is not clear, or the same bucket keeps filling up no matter what you do, that is exactly the situation where a tutor pays for itself. Reviewing someone's last three mocks and telling them precisely where the marks are leaking is most of what we do in a first session.