What the LSAT tests
LSAT preparation is not mainly about memorising legal knowledge. The exam tests skills used in law-school work: logical reasoning, analytical reasoning, including logic games on some administrations, and reading comprehension. It also includes an unscored experimental section and a separate writing sample.
The central task is to choose one best answer under tight time limits. Working shown does not earn partial credit, so your preparation must develop two abilities at once:
- solving the problem accurately;
- recognising when a method is taking too long and moving on.
You should also check the current structure, administration details and official guidance on the LSAC website. The format and delivery arrangements can change, so use official information for anything that affects registration or test-day planning.
A useful LSAT method is consistent across sections:
- Identify exactly what the question asks.
- Extract only the information that controls that task.
- Make deductions or map the argument before looking for a preferred answer.
- Eliminate choices for a stated reason.
- Move on when the question has become a poor use of time.
The aim is not to have an interesting interpretation. It is to justify the single best choice from the passage, stimulus or formal structure in front of you.
Build an exam-specific study board
Start with your own material: diagnostic work, error notes, official preparation resources and explanations of questions you have attempted. Divide it by skill rather than by the order in which you encountered it. A useful board has separate areas for logical reasoning, analytical reasoning, reading comprehension, pacing and review.
Your notes should record the difference between a knowledge problem and a process problem. For example, if you missed a necessary-assumption question because you treated a sufficient condition as necessary, that is a conditional-logic error. If you understood the question but spent too long testing a fourth answer choice, that is a pacing error.
When you add a rule, write a small example beside it. “If A, then B” is less useful than “If a book is shelved in Section 1, it must be blue; a blue book need not be in Section 1.” This prevents you from memorising labels without understanding their direction.
MySummaries can turn that material into a revision board. A board on this topic ends up looking like this:
- Assumption — find what must be true for the conclusion to follow
- Must be true — prove the choice from every permitted case
- Strengthen — add support without needing to prove the conclusion
- Weaken — damage the link, not merely disagree with the conclusion
- Translate each rule before testing answer choices
- Place the most restrictive rule first
- For could-be-true questions, find one legal arrangement; for must-be-true, try to break each choice
Mark the conclusion, premises and any viewpoint or concession. Ask whether the evidence supports the exact conclusion stated, not a stronger version.
For each passage, label the main point, paragraph roles, author attitude and any change in viewpoint. Answer from the passage rather than outside knowledge.
- If A → B, then not-B → not-A; do not reverse A → B into B → A
- Chain rules only when their directions match
- A sufficient condition guarantees the result; a necessary condition is required for it
| Problem | Repair |
|---|---|
| Re-reading stimulus | Underline conclusion and premise once |
| Sunk cost | Flag after a defined check and continue |
| Unproved choice | State the rule or line that proves it |
The board should change as you work. Add the exact wording of a missed question, the tempting wrong answer and the reason it failed. Do not write “careless mistake” unless you can describe the behaviour that caused it.
Reduce the board to a core checklist
Before each practice session, use a short checklist rather than rereading every note. It should contain the operations you need under pressure. A good checklist is brief enough to scan in seconds and specific enough to prevent a known error.
Use the core before timed work, then revisit only the item connected to an error. This keeps revision active. If your checklist grows to several pages, it has stopped being a checklist and become another set of notes.
Practise logical reasoning by question task
Logical reasoning questions reward accurate classification. Before analysing the options, identify the task. A question asking what must be true requires proof across all permitted cases. A question asking what could be true needs one valid case. A strengthen or weaken question concerns the relationship between evidence and conclusion, not whether you personally agree with the conclusion.
For argument questions, write a compact structure:
- conclusion;
- supporting premises;
- missing assumption or logical gap;
- effect required by the question.
Then test answer choices against that structure. The strongest distractor is often relevant but does not affect the central gap, or it changes the scope of the argument. “This is plausible” is not a sufficient reason to keep an answer.
Handle analytical reasoning systematically
On administrations that include logic games, translate the rules into a diagram or list before evaluating choices. Separate global rules, which apply throughout the setup, from local rules introduced by the question. Derive the most restrictive consequences first: fixed positions, exclusions, blocks, ordering chains and limited slots.
For a could-be-true question, one valid arrangement is enough. For a must-be-true question, a candidate choice needs to survive every valid arrangement. Try to construct a counterexample rather than relying on how natural a choice sounds. For a cannot-be-true question, find the contradiction and identify the rule that causes it.
If the setup is taking too long, mark the point where the constraint map became unclear. Review that map later. Do not allow one stubborn question to consume the time needed for several questions you could answer reliably.
Read passages for structure, not detail alone
In reading comprehension, your first pass should establish the passage’s architecture. Identify the main point, the purpose of each paragraph, the author’s attitude and any shift between viewpoints. Details matter, but they matter in relation to that structure.
When an answer choice uses a strong word such as “always”, “only” or “never”, verify that the passage supports that strength. When two choices appear compatible, return to the relevant lines and ask which one answers the task more directly. Do not import general knowledge to repair an answer that the passage does not support.
Your notes need not reproduce the passage. A short map is usually more useful:
- paragraph 1: common view or problem;
- paragraph 2: evidence or criticism;
- paragraph 3: author’s qualification or conclusion.
Use marked practice to diagnose the method
Do not measure progress only by a raw percentage. After a timed set, classify every missed or uncertain question. Useful categories include wrong task identification, missed premise, invalid conditional reversal, unproved inference, weak elimination, reading-detail confusion and pacing.
Review the question before reading an explanation. State why your choice seemed attractive, then identify the first step at which your reasoning stopped being justified. Rewrite the solution in a compact form that another candidate could follow.
A marked paper should show whether you lost marks through content, method or time. This example uses an argument question with a necessary assumption task:
A marked written paper on this topic ends up looking like this:
A town report argues: The new bus lane reduced average journey times last month, so the lane should be made permanent. Which assumption is required by the argument?
3/4The reduction in journey times was caused by the bus lane rather than by an unrelated change in traffic or roadworks.
You identified the causal gap and chose a relevant assumption. The answer needed to make clear that the comparison was not distorted by a temporary change affecting journey times.
Missed
−Separate correlation from the claimed cause
−Test whether an alternative explanation would defeat the conclusion
Model answerStimulus/Passage Structure — Premise: average journey times fell after the bus lane was introduced. Conclusion: make the lane permanent. Task Type — Necessary assumption. Argument Core — The argument needs the fall to be attributable to the lane, or at least not wholly explained by an unrelated temporary factor. Answer Choice Evaluation — Reject choices that merely say shorter journeys are desirable; they do not connect the evidence to the cause. Time/Pacing Notes — Identify the causal link before comparing wording.
The point of marking is to produce the next action. If you reversed a conditional rule, write a new card. If you knew the rule but failed to apply it under time pressure, practise a short set with that operation made explicit. If you spent too long, rehearse the decision to flag and return.
Build a timed routine
A practical weekly routine should contain different kinds of work rather than repeated full papers. Adjust the number of sessions to your schedule, but keep the roles distinct:
- one untimed session to learn or repair a method;
- two short timed sets focused on one question type;
- one mixed timed set to practise switching tasks;
- one review session that converts errors into cards or concise notes;
- one longer practice session under realistic conditions, followed by detailed marking.
At the start of a session, choose a measurable objective: “classify assumption questions correctly” is better than “study logical reasoning”. During the session, record the questions you flagged. Afterward, compare the answer you gave with the reason you gave. An accurate answer based on an invalid reason is still a weakness.
Use a simple pacing rule that you can apply without debate. For example, after one structured attempt and one check for a missed constraint, flag the question and continue. The exact threshold should come from your timed results, but it must be decided before frustration begins.
Add audio review only after writing the method
Audio can help consolidate a board, but it should not replace solving. First write the structure of an argument or the rules of a setup. Then listen to a short explanation and pause when the speaker reaches the conclusion, key inference or elimination step. Say the next step before continuing.
An examiner-style lecture is most useful when it follows the mistakes in your own material. It should explain why a tempting choice fails, not simply announce the correct choice. A focused lecture from the board might look like this:
A lecture on this topic follows the board’s most important reasoning moves:
Transcript · tap any word to jump there
Start with the task, because the same stimulus can support different operations. If the question asks what must be true, a choice is acceptable only if every permitted case supports it; one counterexample defeats it. If it asks what could be true, one legal case is enough.
Now mark the argument. The conclusion is the claim the premises are trying to establish, not necessarily the final sentence. Ask what would have to be true for that movement to work. In a causal argument, test whether another cause could explain the evidence.
Finally, eliminate actively. A choice can be relevant and still fail because it answers a different task, changes the scope or makes an invalid conditional move. When the next check will not produce a justified answer, flag the question and protect the remaining time.
Listen for the language of proof: “must”, “could”, “only if”, “unless”, “at least” and “exactly”. Precise quantifiers are often where a plausible answer becomes an unsupported one.
Review progress by error pattern
At the end of each week, group errors by cause and choose the highest-impact repair. If most errors are from task classification, spend less time doing full mixed sets and more time naming the task before solving. If accuracy is strong but completion is weak, add timed sets and practise flagging. If reading questions are accurate only when untimed, shorten your passage notes and rehearse paragraph roles.
Keep a record of:
- accuracy by question type;
- questions answered, flagged and guessed;
- recurring invalid inferences;
- time lost to rereading or overchecking;
- whether the same error returned after review.
Improvement is shown when the error disappears in a new question, not when an old explanation looks familiar. Re-test the same skill with unfamiliar material and require yourself to explain every surviving answer choice.
How MySummaries helps
MySummaries lets you build an LSAT revision board from your own notes and practice material, then turn recurring errors into flashcards, marked written drills and examiner-voice lectures. For this exam, the useful workflow is to keep question tasks, argument structure, conditional rules, analytical setups, reading maps and pacing decisions on one board, then review the items your timed work shows are still unreliable.