Mastering LSAT Logical Reasoning: A Ground-Up Guide

How LSAT Arguments Work: Conclusions, Premises, and Context

Logical Reasoning (LR) questions are, at their core, about evaluating arguments—how well a set of statements supports another statement. The LSAT is not testing whether you agree with the topic. It’s testing whether you can separate what the author claims from what the author supports, and then judge the quality of that support.

Arguments vs. Facts vs. Explanations

An argument is a set of claims where one or more statements (the premises) are offered to support another statement (the conclusion).

Not every paragraph is an argument. Two common “non-arguments” appear frequently:

• Set of facts: statements that describe a situation but don’t try to prove anything.
• Explanation: a statement (often surprising) followed by information that aims to explain why it happened, not to prove that it happened.

This matters because most LR question types (strengthen, weaken, flaw, assumption, etc.) require you to identify the conclusion and assess support. If you misclassify an explanation as an argument, you’ll start “attacking” the wrong thing.

Quick diagnostic:

• If the author is trying to convince you that something is true → likely an argument.
• If the author takes something as given and tells you why it occurred → likely an explanation.

Premises and Conclusions: The Basic Anatomy

• Premise: evidence, reasons, data, observations.
• Conclusion: what the author wants you to accept, believe, or do.

A key LSAT skill is noticing direction of support: premises support the conclusion; the conclusion does not support premises.

Common conclusion indicators

Words that often introduce conclusions:

• therefore, thus, hence
• so, consequently, as a result
• it follows that
• clearly, shows that, proves that

Common premise indicators

Words that often introduce premises:

• because, since, for
• given that
• after all
• in view of

Be careful: indicator words are helpful but not guaranteed. “Since” can be temporal (“since 2019”) rather than logical.

Subconclusions and Argument Chains

Many LR stimuli contain intermediate conclusions (often called subconclusions)—a claim that is supported by earlier premises and then used as a premise to support the main conclusion.

Why this matters:

• If you mistake a subconclusion for the main conclusion, you’ll pick answers that affect the wrong claim.
• Some question stems (like “the argument proceeds by…”) test your ability to map this structure.

Mini example (structure-focused):

The city’s bus ridership has increased 20% since fares were reduced. Therefore, reducing fares increases public transit use. So the city should reduce subway fares as well.

• Premise: bus ridership increased after bus fares were reduced.
• Subconclusion: reducing fares increases transit use.
• Main conclusion: the city should reduce subway fares.

Background and “Filler”

Stimuli often include context—information that sets the scene but does not do logical work. The LSAT sometimes uses background to distract you into treating it like evidence.

A practical method:

1. Identify the claim that needs support (often the conclusion).
2. Ask: “Which statements are being used to support that claim?” Those are premises.
3. Anything else is likely background.

Exam Focus

• Typical question patterns:
  • Identify the conclusion (“Which of the following most accurately expresses the main conclusion?”)
  • Describe structure (“The argument proceeds by…”)
  • Role questions (“The statement ‘…’ plays which role?”)
• Common mistakes:
  • Treating an explanation as an argument and trying to weaken it
  • Picking a “big idea” as the conclusion when the author’s actual conclusion is a recommendation
  • Ignoring a subconclusion and misidentifying what the premises ultimately support

Reading for Reasoning: A Reliable Process for Any LR Question

You don’t need a different reading style for every question type. What you need is a repeatable process that extracts what the test rewards: structure, gaps, and constraints.

Step 1: Paraphrase the Core

After reading the stimulus, you should be able to say (in your own words):

• Conclusion: “The author is trying to prove that…”
• Premises: “They support it by saying…”
• Assumption/GAP (if any): “This only works if…”

Not every argument has a big gap (some are deductive), but most LSAT arguments have at least one vulnerable connection.

Step 2: Predict What Would Help or Hurt

For many question types, you’re rewarded for making a quick prediction:

• Strengthen: what would make the premise-to-conclusion link more credible?
• Weaken: what would make that link less credible?
• Assumption: what must be true for the argument to work?

Predictions keep you from “shopping” among answer choices and getting seduced by irrelevant but plausible statements.

Step 3: Use the Question Stem as Your Task Definition

The stem tells you what “job” you’re doing:

• “Which of the following, if true, most strengthens…” → add support.
• “The argument’s reasoning is most vulnerable to criticism because…” → diagnose a flaw.
• “Which of the following must be true…” → derive a guaranteed consequence.

A major LR trap is mixing tasks. For example, choosing a choice that strengthens when the question asks for a necessary assumption.

Step 4: Evaluate Answer Choices Mechanically

Correct answers are correct for logical reasons—not because they “sound right.” A useful habit is to force each answer into a clear role:

• Does it affect the conclusion?
• Does it affect a premise?
• Does it connect them (assumption-style)?
• Is it out of scope?

If an answer doesn’t touch the argument’s core, it can’t be correct for strengthen/weaken/assumption/flaw questions.

Exam Focus

• Typical question patterns:
  • “If true” questions that require you to test new information against the argument
  • “Most strongly supported” questions that require cautious inference
  • “The reasoning is flawed because…” questions that require structure-first reading
• Common mistakes:
  • Reading for topic rather than logical function (premise vs background)
  • Failing to predict, then getting pulled toward impressive-sounding but irrelevant choices
  • Treating “if true” answer choices as if they must already be supported by the stimulus

Conditional Reasoning: If–Then Logic Without the Anxiety

Conditional reasoning appears constantly in LR—sometimes explicitly (“if,” “only if”), sometimes implicitly (“requires,” “depends on,” “guarantees”). You don’t need advanced formal logic; you need consistent translation and valid inference.

What Conditional Statements Mean

A conditional statement links a sufficient condition to a necessary condition:

• Sufficient condition: if this happens, it guarantees something else.
• Necessary condition: something that must be true if the sufficient condition occurs.

Form: “If A, then B.”

• A is sufficient for B.
• B is necessary for A.

Translating Common Indicator Phrases

Students often lose points by reversing conditions. Translation skill prevents that.

• “If A, then B” → A → B
• “A only if B” → A → B (only if introduces the necessary condition)
• “A if B” → B → A
• “A requires B” → A → B
• “A depends on B” → A → B
• “B is necessary for A” → A → B
• “B is sufficient for A” → B → A
• “Unless” often means: if not X, then Y. (“Y unless X” ≈ if not X, then Y.)

A reliable “only if” memory aid: ONLY introduces what must be there.

The Contrapositive (The Valid “Flip”)

From A → B, you can validly infer not B → not A. That is the contrapositive.

What you cannot infer:

• The inverse: not A → not B (invalid)
• The converse: B → A (invalid)

LSAT loves these mistakes because they mirror everyday sloppy reasoning.

Example:

If a chemical is corrosive, then it must be stored in a sealed container.

Valid:

• If it is not stored in a sealed container, then it is not corrosive.

Invalid:

• If it is stored in a sealed container, then it is corrosive.

Chaining Conditionals

If you have A → B and B → C, you can chain to get A → C. LR often asks you to combine rules.

Example:

• If a company is publicly traded, it must disclose quarterly earnings.
• If it discloses quarterly earnings, analysts can track its performance.
• Therefore, if a company is publicly traded, analysts can track its performance.

Conditional Reasoning in Disguise

Many stimuli never use “if,” but still create conditional relationships:

• “All,” “every,” “any,” “each” often imply conditional logic.
• “No,” “none” create universal negatives.

Example:

All certified divers have completed safety training.

Translation: Certified diver → completed safety training.

Common Pitfall: Confusing Necessary vs. Sufficient

A classic LSAT gap looks like this:

• Premise: A → B
• Conclusion: A

That’s affirming the consequent in disguise if the premise is actually B → A or if you assume B implies A.

Example flaw:

If the alarm is armed, the indicator light is on. The light is on, so the alarm is armed.

This wrongly assumes “light on” can only happen when the alarm is armed.

Exam Focus

• Typical question patterns:
  • Strengthen/weaken arguments that hinge on reversing a conditional
  • Must-be-true questions that require contraposition or chaining
  • Flaw questions featuring necessary/sufficient confusion
• Common mistakes:
  • Reversing “only if” and “if” clauses
  • Treating the converse as valid (“B, therefore A”)
  • Over-formalizing: turning every statement into symbols and losing track of meaning

Quantifiers and “Most/Some/All” Reasoning

The LSAT frequently uses quantifiers—words that describe amounts or proportions (all, most, many, some, none). These words control what you can validly infer.

The Big Quantifiers You Must Distinguish

• All: 100%
• None: 0%
• Some: at least one
• Most: more than half

These are not interchangeable. “Most” is especially tricky: it allows a large minority exception.

What You Can (and Cannot) Infer

From “All A are B”:

• You can infer: if something is A, it is B.
• You cannot infer: if something is B, it is A.

From “Some A are B”:

• You know at least one overlap exists.
• You cannot infer anything about all A’s, or all B’s.

From “Most A are B”:

• You know more than half of A’s are B.
• You cannot infer “Most B are A.”
• You also cannot infer “Some A are not B” (although that’s often likely, it’s not logically required—“most” could still allow all if “most” is used loosely in natural language, but on LSAT it’s treated as more-than-half, which still allows the possibility that all are B).

Quantifiers in Arguments: Where They Create Gaps

Quantifier mismatches are a frequent flaw:

• Premise talks about some, conclusion talks about all.
• Premise talks about most, conclusion talks about none.

Example:

Some of the city’s parks are unsafe at night. Therefore, the city’s parks are unsafe at night.

The argument illegally generalizes from “some” to “all (or effectively all).”

Negations You Should Be Comfortable With

Negation matters in must-be-true and in conditional contrapositive work.

• Negation of “all” is “not all” (which means some not).
• Negation of “some” is “none.”
• Negation of “most” is “not most” (which means half or fewer).

A subtle trap: “not all” does not mean “none.” It means at least one exception.

Exam Focus

• Typical question patterns:
  • Flaw/weakening based on overgeneralizing (some → all)
  • Must-be-true questions that require careful negation (“Which must be false/true if…”)
  • Principle questions involving “most” and “generally” language
• Common mistakes:
  • Treating “some” as “many” or “most”
  • Negating “all” as “none”
  • Assuming “most” implies a near-universal claim rather than just more-than-half

Assumptions: The Hidden Bridge Between Premises and Conclusions

Most LSAT arguments are not deductively airtight. They rely on something unstated—an assumption—to connect premises to conclusion. Understanding assumptions is central because many question types either ask for the assumption directly or manipulate it (strengthen/weaken).

What an Assumption Is

An assumption is a claim the argument needs in order for the premises to provide support for the conclusion.

A helpful way to think of it: premises are like stepping stones. The conclusion is the far bank. The assumption is the missing stone that makes the crossing possible.

Two Kinds of Assumption Questions

LR typically tests assumptions in two distinct ways:

1. Necessary Assumption: must be true for the argument to work.
   • If it’s false, the argument collapses.
2. Sufficient Assumption: if true, guarantees the conclusion.
   • It may be stronger than what the author strictly needed.

These differ in how you evaluate answer choices.

Necessary Assumptions and the Negation Test

For necessary-assumption questions, a powerful tool is the negation test:

1. Take an answer choice.
2. Negate it (make it as logically opposite as possible without going extreme).
3. Ask: does the argument fall apart?

If negating the choice destroys the reasoning, the choice is necessary.

Example (necessary assumption):

Premise: The new schedule reduces meeting time by 30%.
Conclusion: Therefore, the new schedule will increase employee productivity.

A likely assumption: reducing meeting time will free up time that can be used productively (and not be lost to confusion, coordination costs, etc.).

Negation test idea: “Reducing meeting time will not free up time for productive work.” If that were true, the conclusion is much less supported.

Sufficient Assumptions: The “Guarantee” Standard

Sufficient assumption answers often look more forceful. They fill the gap so completely that the conclusion must follow.

Example (sufficient assumption):

Premise: This medicine reduces symptom duration in clinical trials.
Conclusion: Therefore, it will reduce symptom duration for patients in this clinic.

A sufficient assumption might be: “Patients in this clinic are relevantly similar to those in the clinical trials, and the medicine will be administered under similar conditions.” That bridges general-to-specific.

Common Assumption Patterns (Where Gaps Usually Live)

1. Representative samples: a study group is assumed to represent the population.
2. Causal leaps: correlation is assumed to be causation.
3. Definitions/shifts in meaning: a term subtly changes between premise and conclusion.
4. Alternative causes: the argument assumes no other factor explains the result.
5. Feasibility/implementation: a recommendation assumes the plan can be carried out without unacceptable costs.

Worked Example: Finding the Gap

Stimulus:

Many residents who installed smart thermostats reported lower energy bills. Therefore, installing smart thermostats reduces energy consumption.

• Premise: people who installed reported lower bills.
• Conclusion: installing reduces energy consumption.
• Gap: lower bills might come from price changes, rebates, reporting error, or reduced rates—not consumption.

Possible necessary assumptions include:

• Energy prices were not lower for those residents for reasons unrelated to consumption.
• The reports are accurate.

Possible sufficient assumption:

• The only factor that changed was thermostat installation, and energy prices were constant.

Exam Focus

• Typical question patterns:
  • “The argument requires assuming which of the following?” (necessary)
  • “Which of the following, if assumed, allows the conclusion to be properly drawn?” (sufficient)
  • Strengthen/weaken questions where the correct answer targets the same gap as an assumption would
• Common mistakes:
  • Picking a statement that merely strengthens but is not required (for necessary assumption)
  • Negating answers incorrectly (turning “some” into “none,” etc.)
  • Confusing “relevant” with “required”: many true statements are irrelevant to the argument’s support

Strengthen and Weaken: Controlling the Argument’s Support

Strengthen and weaken questions are about adjusting the relationship between premises and conclusion. You’re not asked to prove the conclusion true or false; you’re asked to change how well the premises support it.

What It Means to Strengthen

To strengthen an argument is to make the conclusion more likely given the premises. Typical strengthen moves:

• Add a missing link (support an assumption).
• Rule out an alternative explanation.
• Confirm that evidence is reliable/representative.
• Provide an additional premise that points to the same conclusion.

What It Means to Weaken

To weaken an argument is to make the conclusion less likely given the premises. Typical weaken moves:

• Provide a counterexample.
• Offer an alternative cause or explanation.
• Show the evidence is flawed or unrepresentative.
• Show the plan has a cost or obstacle that undermines the recommendation.

The “New Information” Mindset

Most strengthen/weaken answers introduce new facts. You should treat them as true and ask: does this new fact touch the argument’s core?

If a choice is about the general topic but doesn’t interact with the premise-to-conclusion link, it won’t be correct.

Worked Example: Strengthen

Stimulus:

A company’s employee satisfaction increased after it introduced remote work. Therefore, remote work increases employee satisfaction.

Vulnerable point: causation. Maybe something else changed.

A strong strengthen answer would:

• Rule out other changes (e.g., “No other major workplace policies changed during that period”).
• Show a mechanism (e.g., “Employees reported that eliminating commuting time was the primary reason for improved satisfaction”).

Worked Example: Weaken

Same stimulus. A strong weaken answer would:

• Offer alternative cause: “At the same time, the company increased salaries significantly.”
• Show selection effect: “Only the employees who preferred remote work stayed; dissatisfied employees left.”

“Most Strongly Supports” vs. Strengthen

Be careful: “Which is most strongly supported by the statements above?” is an inference task (what follows from the stimulus). Strengthen is a help the argument task (introduce new support). Mixing these is a classic error.

Exam Focus

• Typical question patterns:
  • “Which of the following, if true, most strengthens/weakens the argument?”
  • Weaken questions targeting causal claims, sampling, or feasibility of a proposal
  • Strengthen questions that effectively ask you to support the necessary assumption
• Common mistakes:
  • Picking an answer that is merely related to the topic but doesn’t affect the reasoning
  • Attacking a premise when the question asks you to weaken the support (you often need to hit the link)
  • Overvaluing extreme language: strong-sounding choices are often wrong if they overshoot or are irrelevant

Flaw and Method of Reasoning: Naming What Went Wrong (or What Happened)

Flaw and method questions reward a structured understanding of arguments. They are less about “content” and more about “form.”

Flaw Questions: The Argument’s Weak Point

A flaw is a reasoning error—something that prevents the premises from adequately supporting the conclusion.

To do flaw questions well, you typically:

1. Identify premise and conclusion.
2. Ask what the argument assumes.
3. Describe the error in general terms (not topic-specific).

High-Frequency Flaw Patterns

1. Causation vs. correlation

   • Treats correlation as proof of causation.
   • Ignores alternative causes, reverse causation, or coincidence.

2. Sampling problems

   • Generalizes from an unrepresentative sample.
   • Uses too small a sample.

3. Equivocation (shifting meaning)

   • Uses a key term in two different senses.

4. Necessary vs. sufficient confusion

   • Assumes a necessary condition is sufficient, or vice versa.

5. Part-to-whole / whole-to-part

   • Concludes something about the whole from a part, or about each part from the whole.

6. False dilemma

   • Treats two options as exhaustive when other options exist.

7. Circular reasoning

   • The conclusion is essentially assumed in the premises.

Worked Example: Identify the Flaw

Stimulus:

Whenever the local team wins, the next day’s sales at downtown restaurants increase. Therefore, the team’s victory causes increased restaurant sales.

Flaw: correlation → causation. Also ignores alternative causes (e.g., weekend scheduling) and reverse causation makes less sense here but still conceptually possible.

A correct flaw answer would be phrased generally, like:

• “Confuses evidence of correlation with evidence of causation.”

Method of Reasoning Questions

Method questions ask what the argument does—its structure or strategy.

Common method descriptions:

• Draws a general conclusion from a specific example.
• Uses an analogy to support a conclusion.
• Appeals to an authority or expert.
• Rejects a proposal by pointing out an undesirable consequence.
• Supports a recommendation by citing a goal and claiming the recommendation achieves it.

Method questions are easier when you can summarize the argument as a “template.” For instance:

• “X is true in situation A. Situation B is similar. Therefore, X is true in situation B.” (analogy)

Exam Focus

• Typical question patterns:
  • “The reasoning is flawed because…” (identify the error)
  • “The argument proceeds by…” (describe the steps)
  • “Which describes the method of reasoning?” (structure template)
• Common mistakes:
  • Choosing an answer that criticizes the argument’s conclusion as false rather than its support as weak
  • Being too specific: correct flaw/method answers usually abstract away from the topic
  • Missing subtle term shifts (same word, different meaning)

Inference Questions: Must Be True, Most Strongly Supported, and Cannot Be True

Inference questions flip the usual LR task. Instead of evaluating an argument, you treat the stimulus as information and ask what it logically commits you to.

“Must Be True” vs. “Most Strongly Supported”

• Must Be True: the correct answer is guaranteed by the stimulus.
• Most Strongly Supported: the correct answer is not guaranteed, but it is the best-supported among the choices.

In practice, both reward cautious reasoning. Wrong answers often go slightly beyond what the text provides.

How to Approach Inference

1. Identify the firm statements (facts, conditionals, quantifiers).
2. Combine them conservatively (chain conditionals; intersect quantifiers carefully).
3. Avoid adding assumptions about typicality, motives, or “common sense” unless the stimulus forces it.

Common Inference Tools

• Conditional chaining and contraposition
• Quantifier logic (“all,” “some,” “none,” “most”)
• Elimination: many wrong answers are too strong, introduce new terms, or reverse relationships

Worked Example: Must Be True (Conditional)

Stimulus:

All members of the committee are certified. Some certified professionals are not members of the committee.

What must be true?

• It must be true that some certified professionals are members of the committee? Not necessarily—“all committee members are certified” does not guarantee the committee has any members.
• It must be true that if someone is not certified, they are not a committee member. Yes: contrapositive of “committee member → certified.”

So a must-be-true answer could be:

• “No one who lacks certification is a member of the committee.”

“Cannot Be True” / Must Be False

These ask for a statement that conflicts with the stimulus. Strategy:

• Translate key claims.
• For each answer, see whether it violates a rule or quantifier.

Exam Focus

• Typical question patterns:
  • “Which of the following must be true?” (deduction)
  • “Which is most strongly supported?” (best inference)
  • “Which cannot be true?” (find the contradiction)
• Common mistakes:
  • Treating “most strongly supported” like “must be true” and demanding certainty
  • Bringing in outside knowledge or typical assumptions
  • Overlooking that universal statements don’t imply existence (e.g., “All unicorns are white” doesn’t imply unicorns exist)

Causal Reasoning: How LSAT Arguments Explain (and Mis-explain) the World

Causal claims are everywhere in LR because they’re naturally vulnerable. The LSAT loves them because you can attack or support causation in a small number of predictable ways.

What a Causal Claim Looks Like

A causal conclusion asserts that one factor produces or influences another:

• “X causes Y”
• “X increases/decreases Y”
• “X leads to Y”
• “Y results from X”

Sometimes the stimulus presents causation as already accepted and asks you to reason from it, but often the argument is trying to prove causation.

The Three Classic Causal Threats

When an argument says “X causes Y” because X and Y occur together, you should immediately consider:

1. Alternative cause: something else causes Y (and maybe also X).
2. Reverse causation: Y causes X.
3. Coincidence / third variable: X and Y correlate due to chance or a separate factor.

These threats explain why causal arguments are so easy to weaken.

Strengthening Causal Claims

To strengthen “X causes Y,” you can:

• Rule out alternative causes (control variables).
• Show that when X changes, Y changes (especially in controlled settings).
• Provide a plausible mechanism linking X to Y.
• Show temporal order: X happens before Y.

Weakening Causal Claims

To weaken, you can:

• Provide an alternative cause.
• Suggest reverse causation.
• Show the correlation disappears in broader data.
• Show the causal direction is inconsistent (cases with X but no Y, or Y without X).

Causation in Policy Arguments

Many LR arguments recommend actions:

“We should do X because it will lead to Y (a goal).”

These combine causation with feasibility and cost-benefit reasoning. To weaken:

• X may not lead to Y.
• X may have unacceptable side effects.
• There may be a better alternative to achieve Y.

Worked Example: Causal Strengthen/Weaken

Stimulus:

After the city installed more streetlights, nighttime crime decreased. Therefore, installing streetlights reduces crime.

Strengthen answers might include:

• “No other significant anti-crime measures were introduced during that period.”
• “Comparable neighborhoods without new streetlights did not experience a crime decrease.”

Weaken answers might include:

• “During the same period, the city doubled police patrols at night.”
• “Crime moved to nearby unlit areas, leaving overall crime unchanged.” (this also targets how “crime decreased” might be measured)

Exam Focus

• Typical question patterns:
  • Strengthen/weaken arguments based on before-and-after observations
  • Flaw questions accusing correlation-causation confusion
  • Evaluate-the-argument questions asking what would most help determine causation
• Common mistakes:
  • Attacking causation when the stimulus is only describing an effect (explanation vs argument confusion)
  • Forgetting reverse causation as a possibility
  • Accepting self-reported data as if it were automatically reliable

Analogies and Comparisons: When Similarity Is (and Isn’t) Evidence

Arguments by analogy are persuasive in everyday life, which is exactly why the LSAT tests them—analogy is often informative but rarely conclusive.

What an Argument by Analogy Is

An argument by analogy claims:

• Situation A and situation B are similar in relevant ways.
• A has property P.
• Therefore, B likely has property P.

The key word is “relevant.” Similarity only supports a conclusion if the shared features matter to the property being inferred.

How to Strengthen an Analogy

You strengthen by showing:

• The two cases are similar in precisely the features that matter.
• Potential disanalogies (differences) don’t affect the outcome.

How to Weaken an Analogy

You weaken by:

• Identifying a relevant difference.
• Showing the similarity is superficial.
• Showing that A’s property P depends on a factor absent in B.

Worked Example: Weaken an Analogy

Stimulus:

A small town successfully reduced traffic by adding roundabouts. Therefore, a large city can reduce traffic by adding roundabouts.

A strong weaken answer:

• “The town’s intersections have far fewer lanes and lower traffic volume than the city’s intersections.”

That difference is relevant to whether roundabouts are feasible and effective.

Exam Focus

• Typical question patterns:
  • Flaw questions: “fails to establish that two situations are sufficiently similar”
  • Strengthen/weaken: choices that add or undermine relevant similarity
  • Method questions: “supports by drawing an analogy”
• Common mistakes:
  • Attacking the conclusion directly rather than the relevance of the comparison
  • Confusing “similar in any way” with “similar in the ways that matter”
  • Missing that some analogies are used only as illustrations (not as support)

Numbers, Surveys, and Sampling: How Evidence Can Mislead

LR frequently uses statistical language—polls, studies, percentages—not to test math, but to test reasoning about data quality.

Sampling and Representativeness

A sample is a subset used to infer something about a larger population. The inference is only as good as the sample is representative.

Common representativeness issues:

• Selection bias: the sample is chosen in a way that skews results.
• Response bias: only certain types of people respond.
• Small sample size: results may be unstable.
• Non-comparable groups: differences may be due to group differences, not the factor tested.

Surveys vs. Objective Measures

A lot of LR “study” premises are self-reports:

• “Participants reported feeling better.”

That can be useful but is vulnerable:

• People misremember.
• People want to please the researcher.
• Placebo effects can influence perception.

Percentage and Baseline Traps

A common LSAT move is to give a percentage without a baseline:

• “Complaints increased by 50%.”

That could mean from 2 to 3 complaints. The argument may treat it as dramatic without giving absolute numbers.

Another trap: confusing rate with total count.

• If the population grows, total incidents can rise even if the rate falls.

Worked Example: Sampling Flaw

Stimulus:

A website polled its users and found that 90% prefer the new layout. Therefore, most people prefer the new layout.

Flaw: website users may not represent “most people.” Also response bias: only engaged users answered.

A weaken answer could be:

• “The site’s users are mostly professional designers, who differ from typical internet users.”

A strengthen answer could be:

• “The users polled were randomly selected from a broad cross-section of the population.”

Exam Focus

• Typical question patterns:
  • Weaken/strengthen by attacking/supporting representativeness
  • Flaw questions about shifting from sample to population
  • Evaluate questions asking what would help determine whether the study supports the conclusion
• Common mistakes:
  • Treating any large percentage as decisive without asking “of whom?”
  • Ignoring that correlation can arise from selection effects
  • Missing baseline issues (percent change without absolute numbers)

Principles and Rules: Principle Questions and “Match the Reasoning”

Principle questions test whether you can apply an abstract rule to a concrete case—or extract a rule from a case.

Two Common Principle Tasks

1. Principle—Strengthen/Justify

   • You’re given an argument and asked for a principle that, if accepted, supports it.
   • The correct principle typically bridges the gap like a sufficient assumption, but in rule form.

2. Principle—Conform/Match

   • You’re given a principle and asked which scenario conforms to it.
   • This is like applying a rule: check conditions → see which case fits.

How to Handle Abstract Wording

Principles are often written with broad terms:

• “Whenever an action imposes significant risk on others, it should be regulated.”

To apply it, translate into a simple if–then:

• If action imposes significant risk on others → action should be regulated.

Then check each answer choice to see whether it satisfies the sufficient condition.

Worked Example: Principle That Justifies

Stimulus:

The restaurant should list allergens because customers have a right to know what risks they face.

A supporting principle might be:

• “Businesses should disclose information that is necessary for customers to avoid serious health risks.”

That principle connects the “right to know” premise to the “should list allergens” conclusion.

“Parallel Reasoning” and “Parallel Flaw”

These questions ask you to match the structure of an argument, sometimes including its flaw.

To do them:

1. Strip the stimulus to a skeleton (quantifiers, conditionals, causation, etc.).
2. Ignore topic; look for the same logical moves.
3. For parallel flaw, ensure the same error occurs (e.g., correlation → causation; necessary/sufficient mix-up).

Exam Focus

• Typical question patterns:
  • “Which principle most helps justify the argument?”
  • “Which situation conforms most closely to the principle?”
  • Parallel reasoning/flaw: “Which argument is most similar in its reasoning?”
• Common mistakes:
  • Choosing a principle that is on-topic but doesn’t actually justify the conclusion
  • Overlooking quantifier or conditional differences in parallel questions
  • Matching surface details instead of structure

Resolve, Explain, and Evaluate: Paradox Questions and Tests of Evidence

Some LR questions are built around a tension: two facts seem to conflict, or a result seems surprising. Your job is to reduce that tension.

Paradox / Resolve the Discrepancy

A paradox question gives you:

• Fact A
• Fact B
• A claim that A and B are hard to reconcile

The correct answer typically:

• Adds a missing fact that makes both possible.
• Clarifies a hidden distinction (different groups, time periods, definitions).

You are not strengthening an argument here; you are explaining how both facts can be true.

Worked Example: Resolving a Discrepancy

Stimulus:

A bookstore lowered prices, yet its profits decreased. This is surprising because lower prices usually increase sales.

A good resolution might be:

• “The price reduction was accompanied by a large increase in the store’s operating costs.”

Now lower prices could increase sales but still reduce profits due to higher costs.

Another resolution could be:

• “Sales increased only slightly, while revenue per item fell substantially.”

Evaluate the Argument

Evaluate questions ask for information that would help determine whether the conclusion is justified. The right answer is often phrased as a yes/no question in disguise—something where either possible answer would matter.

If the conclusion depends on whether X is true, then “Whether X is true” is good evaluation information.

Example:

Conclusion: The new tutoring program caused test scores to rise.

Good evaluation question:

• “Did any other major changes occur at the school during the same period?”

If yes, causation is less clear; if no, causation is more plausible.

Exam Focus

• Typical question patterns:
  • “Which of the following, if true, most helps to resolve the apparent discrepancy?”
  • “Which of the following would be most useful to determine whether the argument is sound?”
  • Sometimes framed as “explain” or “account for” a surprising outcome
• Common mistakes:
  • Treating paradox questions like weaken questions (trying to deepen the conflict)
  • Picking an answer that merely restates one side of the discrepancy
  • In evaluate questions, choosing info that matters only if it comes out one way (instead of being diagnostic either way)

Argument Parts and Perspective: Main Point, Role, and Point at Issue

Some LR questions focus less on evaluating and more on understanding what’s being said and how parts relate.

Main Point Questions

Main point asks: what is the overall conclusion?

These are easiest when you:

• Find recommendation or judgment language (“should,” “thus,” “therefore”).
• Distinguish main conclusion from evidence and subconclusions.

Beware common traps:

• An answer that is a premise stated strongly.
• An answer that is a broader theme than the conclusion.

Role of a Statement

Role questions point to a specific sentence and ask what it does:

• premise, intermediate conclusion, main conclusion
• objection, counterpoint, background, example

A strong way to answer is to ask two questions:

1. Does this statement support something else, or is it supported?
2. Is it the final thing the author wants you to accept?

Point at Issue / Agreement-Disagreement

These questions give two speakers and ask where they disagree.

Strategy:

1. For each speaker, paraphrase their position as yes/no on key claims.
2. Look for a statement one would accept and the other would reject.

Common trap: an answer that one speaker never addresses.

Worked Example: Point at Issue

Speaker 1: “The city should ban cars downtown because pollution levels are dangerous.”
Speaker 2: “Pollution is dangerous, but banning cars will harm local businesses.”

They agree pollution is dangerous; they disagree about whether the ban should be implemented.

Exam Focus

• Typical question patterns:
  • “The main conclusion is that…”
  • “The statement ‘…’ serves which function?”
  • “The speakers disagree about whether…”
• Common mistakes:
  • Confusing a strongly worded premise for a conclusion
  • Mislabeling a counterargument as the author’s view
  • In disagreement questions, picking an issue neither speaker clearly takes a position on

Common Logical Traps the LSAT Exploits (and How to Resist Them)

Many wrong answers are wrong in predictable ways. Learning these patterns helps you eliminate choices quickly without guessing.

Shell Games: Same Words, Different Meaning

An argument may subtly shift a key term:

• “effective” meaning “works at all” in the premise and “works better than alternatives” in the conclusion.

To guard against this, when you see a repeated term, ask: is it being used in the same sense?

Scope Shifts

The conclusion may change:

• group (some students → all people)
• time (last year → always)
• place (this city → every city)
• standard (better → best)

Many flaws and weaken answers exploit these shifts.

Normative vs. Descriptive Jumps

• Descriptive: what is (facts)
• Normative: what should be (values)

Arguments often jump from “is” to “should” without stating a value principle.

Example:

The policy will increase profits. Therefore, we should adopt it.

Assumption: increasing profits is the main goal and outweighs other considerations.

Extremes and Absolutes in Answer Choices

Words like “all,” “never,” “completely,” “guarantees” can be red flags—unless the stimulus itself is absolute. Correct answers can be strong, but they must be justified by the task.

• In must-be-true: strong language is often wrong unless forced.
• In sufficient assumption: strong language can be right because it must guarantee.

Irrelevant Comparisons

Arguments often compare two things but don’t establish the comparison is fair:

• different baselines
• different populations
• different conditions

Weakeners often point out the mismatch.

Exam Focus

• Typical question patterns:
  • Flaw and weaken answers that target scope shifts and term shifts
  • Assumption answers that repair descriptive-to-normative leaps
  • Parallel flaw answers that replicate the same scope error
• Common mistakes:
  • Treating extreme wording as automatically wrong (sometimes it’s exactly what’s needed)
  • Missing subtle scope shifts because the topic “feels consistent”
  • Ignoring value assumptions in “should” conclusions

Putting It Together: A Full Worked LR Example (End-to-End)

To see how these tools integrate, here’s an end-to-end walkthrough.

Stimulus

In a recent survey, employees who use standing desks reported fewer back problems than employees who do not. Therefore, providing standing desks to all employees will reduce back problems in the company.

Step 1: Identify conclusion and premises

• Premise: In a survey, standing-desk users reported fewer back problems.
• Conclusion: Providing standing desks to all employees will reduce back problems.

Step 2: Identify the gap

This is a causal/generalization argument.

• Correlation issue: standing desk use correlated with fewer problems.
• Alternative explanations: health-conscious employees may choose standing desks.
• Self-reporting reliability.
• Policy leap: “provide” doesn’t guarantee “use.”

Step 3: Apply to question types

If the question is Strengthen

Good answers would:

• rule out alternative causes (users similar to non-users)
• show standing desks preceded improvement
• show that when employees switched to standing desks, back problems decreased
• show employees will actually use them

If the question is Weaken

Good answers would:

• show selection bias (only already-healthy employees choose standing desks)
• show reverse causation (those with back problems avoid standing)
• show another change explains difference (desk users also have better chairs, more breaks)

If the question is Necessary Assumption

A likely necessary assumption is:

• If standing desks are provided, employees will use them enough to matter.

Negation: If standing desks are provided, employees will not use them (or not enough to matter). That severely undermines the conclusion—so it’s plausibly necessary.

If the question is Flaw

A correct flaw description might be:

• “Infers a causal conclusion from a correlation and fails to rule out alternative explanations.”

Exam Focus

• Typical question patterns:
  • Causal strengthen/weaken built on survey results
  • Assumption questions focused on implementation (“providing” vs “using”)
  • Flaw questions focused on correlation, selection bias, and self-reporting
• Common mistakes:
  • Overlooking the “policy step” (providing doesn’t ensure adoption)
  • Attacking the survey generally without connecting the criticism to the conclusion
  • Missing that the conclusion is about the whole company, not just surveyed employees