Study guide
Find the topic you keep getting wrong
An overall percentage hides the one topic that costs you marks every time. Here is how to find it and what to do about it.
The average is the wrong number
Most people look at a score, see a percentage, and move on. A percentage is an average, and an average of uneven parts tells you very little. Ninety percent across a whole reviewer can mean you know nine topics well and one badly, or that you know all of them equally and are near the top everywhere. Those need completely different amounts of work.
The same applies across reviewers. A healthy average across eight reviewers can be hiding one subject where you score below chance because you have never properly learned it.
Read the misses, not the score
Hachi keeps a full record of what you chose on every question, not just whether it was right. Review Answers shows your answer next to the correct one with the explanation attached, and it stays available after the session ends.
- Group the misses by topic rather than by reviewer. Questions carry the topic they came from, so the same weakness usually shows up in more than one reviewer.
- Look at the shape of the error. Missing every question on one narrow topic is a gap in the material. Getting one option wrong across a whole topic is usually a wording problem.
- Count the repeats. A question you have missed on three separate attempts is not bad luck and is not a schedule problem.
Distinguish not-known from not-understood
These look identical in a score and need opposite treatment.
Not known means the information was never in front of you, or it was there once and did not survive. The fix is retrieval: quiz the topic until it sticks.
Not understood means the information is there and you keep choosing the wrong option anyway. That is usually a sign you are pattern-matching on the shape of the question rather than reading it. The fix is slower: read the explanation for those questions slowly and restate the rule in your own words until the distractor stops being tempting.
One useful diagnostic: if you would have picked differently had you read the question twice, it is a reading problem rather than a knowledge gap, and a smaller question set on that topic will fix it faster than a larger one.
Fixing the weak topic
Build a reviewer for the weak topic alone rather than re-taking the big one. Generation makes this easy: give the generator your notes for that one section, ask for a focused question count, and you have a short, targeted reviewer in a couple of minutes.
Then take it until you are consistently correct, and go back to the original reviewer a week later to check the fix held. A weak topic that has been repaired is one of the few changes in revision that reliably moves an overall score.
Try it
Every technique here works on a reviewer you already have, or on one built from your notes in a couple of minutes. The features page covers how, and the FAQ covers the parts people usually get stuck on.
More guides
- Active recall beats rereading, and you can prove it on yourself
Rereading feels like studying because it is easy and it feels familiar. It is also close to useless on an exam. Here is the difference, and what to do instead.
- A revision schedule you will actually keep
Revising everything the night before is a way of feeling organised. Spacing is what makes it work. Here is a schedule you can keep, including for the nights when it slips.
- Turning a past paper into a reviewer you can revise from
Exam Paper mode transcribes a paper that already exists instead of inventing one. That changes what you can trust, and how you should use it.
- Running a study group that actually turns up
Shared material goes stale fast unless someone owns it. Here is how to share reviewers and groups without ending up with four slightly wrong copies.
- Revising with no signal
Most of Hachi runs on your device, which means a bad connection is much less of a problem than it sounds. Here is exactly what stops working.