Correctness vs. reasoning quality in GAMSAT preparation

You finish a practice set, mark it, and look at your percentage. The wrong answers go on the review list. The right answers get a tick and you move on. That approach treats correctness as the only signal that matters, and it collapses situations that need to be kept separate.

What correctness alone cannot show

Consider a Section 1 question asking which statement the passage most strongly supports. You read four options, eliminate two that clearly reach beyond what the text establishes, and choose between the remaining pair. One feels closer to the passage's argument. You select it, check the answer, and you were correct.

That outcome tells you the answer was right. It does not tell you whether the method that got you there can be applied consistently to other questions.

Four outcomes, not two

Return to that same question and consider four ways the attempt could have gone.

First: you identified what the question was actually asking, checked each option against the passage's reasoning rather than its subject matter, recognised that one option required an assumption the passage did not support, and chose the option whose inference the passage licensed. You recorded high confidence, and when you review the question you can reproduce the justification. This is the outcome preparation aims for.

Second: you eliminated the two weakest options, felt uncertain between the survivors, and chose one because its phrasing sounded more like something the passage would say. You recorded low confidence. The answer was correct, but check whether the distinction you relied on is one you could apply consistently.

Third: you identified the question's demand correctly, compared each option to the passage at the right level of inference, and selected an option that genuinely was supported by the text. The correct answer was the adjacent option, requiring one further distinction you did not make. You were incorrect, but the reasoning was one step from working.

Fourth: the reasoning did not match what the question was asking, or rested on an assumption the passage did not support. You were incorrect, and the approach needs to change.

A percentage collapses the first two into one category and the last two into another. The distinction it loses is the one that tells you what to work on.

Two correct answers, different evidence

A correct answer you recorded high confidence on, where you can state why each alternative was weaker, is different evidence from a correct answer you recorded low confidence on, where you cannot say why the option you chose was justified. Both were correct. What the review record shows about how you reached them is not the same.

Verantrix treats correctness as one part of the evidence used to understand an attempt. That is the position we hold: a right answer is meaningful, and it is not the only thing worth looking at.

Inspecting a correct answer

Before moving on from a question you answered correctly, check whether the reasoning behind it was repeatable. Not every correct answer needs this level of scrutiny, but the ones reached with low confidence or after several re-readings are worth the time.

How confident were you before checking the answer? A correct answer you recorded low confidence on is worth a second look.

Can you state why the option you chose was justified by the passage, at the level of inference the question required? Not whether you remember choosing it, but whether you can reproduce the reasoning that supports it.

Can you say why each of the other options was weaker? If you cannot, that is the part worth revisiting.

Would you make the same distinction on a similar question, under the same time pressure, if the subject matter changed? A method that depends on familiarity with the passage's content will not transfer.

These are diagnostic questions, not a learning method. They help you separate the correct answers reached by a process worth reinforcing from the ones that happened to land correctly.

What this means for preparation

Correctness remains meaningful. A correct answer is better than an incorrect one, and accuracy is one piece of evidence about whether preparation is working. The problem is treating it as the only piece.

Consider a candidate who completes ten Section 1 practice passages over two weeks. Seven correct, three incorrect. They review the three wrong answers carefully, identify what they missed, and move on. The seven correct answers receive no further attention.

That review treats the three incorrect answers as the only diagnostic evidence available. But of the seven correct, suppose the review record shows that four were answered with high confidence and no answer changes, two were answered with low confidence, and one took twice as long as the others and had the answer changed once before submission.

Those three are worth reviewing alongside the three incorrect ones. Whether the reasoning was repeatable is a separate question from whether it produced the right answer this time.

Richer practice evidence reveals whether a reasoning pattern is repeating, whether confidence aligns with correctness, and whether the errors that cost marks are recurring or scattered. That evidence helps you distinguish questions answered by a repeatable method from those that happened to produce the right answer.

Next step

If you want to assess where your reasoning is actually breaking rather than where your percentage says it is, Verantrix records which option you selected, how confident you were, how long you took, whether you were correct, and whether you changed your answer. That evidence, reviewed across attempts rather than one question at a time, is what diagnosis works from.