Should You Read LSAT Questions Before the Passage?
Read the passage first, once and closely, then answer the questions. Reading the questions first sounds efficient, but it turns every question into a separate hunt through a passage you have not yet understood, and it usually costs more time than it saves. The only common exception is the last few minutes of a section, when you are out of time and need to salvage points from a set you cannot finish properly.
Should I read the questions before the passage on the LSAT?
No, in almost every case. Read the passage first. The reason is simple: most Reading Comprehension questions test your understanding of the whole passage, not your ability to find one sentence. Main point questions, primary purpose questions, author attitude questions, and function questions all require a sense of the passage as a whole. You cannot answer them well from fragments you collected while hunting for something else.
Reading the questions first also changes how you read. Instead of building a picture of the main point, the viewpoints, and the structure, you read with a checklist in mind. You notice details that match a question stem and skip the sentences that connect them. Then, when a question asks about the author's overall position, you have no picture to draw on, so you reread. That second read is where the time goes.
Our guide on how to improve LSAT Reading Comprehension lays out the passage-first routine in full: one careful read for structure, a one-sentence summary, then questions answered from that structure with detail checks in the passage. This page covers the questions-first debate specifically, and the exceptions where it can help.
What reading the questions first actually costs you
1. You lose the main point
The main point is the single most tested idea in Reading Comprehension. It shows up directly in main point and primary purpose questions, and it quietly decides many others, since wrong answers often contradict it. A questions-first read rarely produces a clear main point, because you never read the passage on its own terms. You read it as a set of hiding places for answers.
2. You reread more, not less
The pitch for questions-first is that it saves a read. In practice it usually adds one. You skim for the first question, fail to find a clean answer because you lack context, reread a wider chunk, and repeat that for the next question. A single careful first read, by contrast, lets most questions become lookups: you know roughly where each idea lives, so you verify it in seconds.
3. You fall for trap answers more easily
Many wrong answers are true statements from the passage that do not answer the question asked. They are easy to pick when you found the matching sentence but never built the passage's argument. A detail that supports a critic's view looks correct until you remember the author disagreed with that critic. That check requires the whole-passage view that questions-first reading skips.
4. Comparative reading gets worse
Comparative reading sets have two shorter passages and questions about their relationship. Questions-first is especially weak here, because relationship questions ask how the passages agree, disagree, or differ in emphasis. You cannot judge a relationship between two texts you have only hunted through. Our comparative reading guide explains the read-for-relationship routine: read both passages for structure first, then answer.
The three approaches, compared
| Approach | How it works | When it helps |
|---|---|---|
| Passage first (recommended) | One careful read for main point, viewpoints, and structure, then the questions | Nearly always: the default for full sets |
| Questions first | Read the question stems, then hunt the passage for each answer | Almost never on a full set; see the timed exceptions below |
| Skim first, then reread | Quick skim of passage and questions, then a careful reread | Rarely: it is two reads disguised as a shortcut |
The two exceptions: when questions-first can salvage points
There are two situations where the normal rule bends. Both involve running out of time, and both are salvage plans, not study methods.
Exception 1: The last few minutes of a section
If you reach a final passage set with only a few minutes left, a full careful read is no longer possible. In that case, glance at the questions and count the types. If most are global questions, like main point and primary purpose, skim the passage for its gist and answer those. If most are specific-reference questions, with line numbers or quoted phrases, skip the passage and work those questions directly, reading only the lines each one points to. Pick the approach that matches the set in front of you, and answer every question: there is no guessing penalty on the LSAT.
Exception 2: A set you have decided to sacrifice
Some test takers deliberately leave their hardest set for last. If time runs out on that set, the same salvage logic applies: global questions from a quick skim, or specific-reference questions answered from their lines. This is a damage-control plan for a set you already chose to deprioritize, not a reason to read questions first on the sets you plan to finish.
Neither exception argues for questions-first as a habit. A method that only works when you have already lost the time battle is not a method. It is a backup plan.
What about Logical Reasoning: stem first or stimulus first?
Logical Reasoning works the other way. Read the question stem first, then the stimulus. The stem names the job before you read a word of the argument: strengthen, weaken, find a flaw, draw an inference. Knowing the job changes how you read, because you know what the answer has to work on: the conclusion, and the gap in its support where the type asks about one. Reading the stimulus first, with no job in mind, usually means reading it twice, once to understand it and again once you know what the question wants. Our Logical Reasoning question types guide lists what each type asks, so you can recognize the job from the stem in seconds.
How to test which method works for you
Do not take anyone's word for this, including this page's. Run a small experiment with two timed sections, one per method, under the same conditions.
- Section one: passage first. Take a timed 35-minute Reading Comprehension section reading each passage first, once and closely. Record your score, how many questions you finished, and how many you guessed on.
- Section two: questions first. Take a different timed section reading the question stems first. Record the same three numbers.
- Compare more than the score. Look at which method finished more questions, which one produced more main point misses, and which one left you rereading. A method that scores the same but finishes fewer questions is losing on time.
- Keep the winner for at least three more sections. One section is noisy. A method needs a few sections before you can trust the result, and switching methods every week prevents either one from becoming automatic.
If questions-first genuinely wins for you across several timed sections, keep it. Most students find the opposite: passage-first finishes more questions with fewer main point misses, because the first read does the work once instead of repeating it per question.
The routine to use instead
- Read once, closely, for structure. Main point, viewpoints, and how the paragraphs connect. Ignore most details on purpose; note where they live instead.
- Pause for one sentence. Summarize the passage in your own words before touching the questions. If you cannot, reread the first and last sentences of each paragraph.
- Answer from structure, verify the detail. Predict the kind of answer that fits your summary, then check the passage only where the question points you.
- Work in a sensible order. Main point questions first, then questions with line references, then vague inference questions last within a set.
- Review every miss the same day. Sort misses by question type and by cause. If most misses are main point questions, the first read needs work, not the answer choices.
As of October 2026, a Reading Comprehension section has four passage sets and roughly 26 to 28 questions in 35 minutes, which works out to about 8 to 9 minutes per set. That budget only works with one good read per passage. Questions-first reading spends that budget on hunting, and hunting is the slowest way to read a passage you have not yet understood.
Frequently asked questions
Should I read the passage or the questions first on LSAT Reading Comprehension?
Read the passage first, once and closely. Most questions test your understanding of the whole passage, including its main point and the author's position, which you cannot get from hunting for answers question by question.
Does reading the questions first ever help on the LSAT?
Only as a salvage plan when time has nearly run out. With a few minutes left on a final set, glance at the questions: answer global questions from a quick skim, or answer specific-reference questions from the lines they point to. It is a backup plan, not a study method.
Should I read the question stem first on Logical Reasoning?
Yes. Read the stem first, then the stimulus. The stem tells you the job, whether it asks you to strengthen, weaken, find a flaw, or something else, so you read the argument looking for what that job needs: the conclusion and the gap in its support. Logical Reasoning runs in the opposite order from Reading Comprehension, where the passage comes first.
What if I always run out of time on Reading Comprehension?
The usual cause is rereading: a weak first read forces a second read for almost every question. A structure-first first read, a one-sentence summary before the questions, and a skip-and-return rule for hard questions are the usual fixes. Answer every question, since there is no guessing penalty.
Does skimming the passage first and then rereading work?
Rarely. Skimming plus a careful reread is two reads, and the skim adds little: you still need the careful read to answer accurately. One careful read, with a summary before the questions, is usually faster overall.
How do I know which reading method works best for me?
Test both under timed conditions: one full 35-minute section per method, then compare score, questions finished, and main point misses. Keep the winner for at least three more sections before deciding, since a single section is noisy.