Critical Reasoning (CR) is arguably the highest-ROI topic to study for the GMAT Verbal section. The question types are predictable, the logic patterns repeat, and once you develop the right analytical framework, your accuracy improves fast. Here's the complete strategic approach.
The Anatomy of Every CR Argument
Before studying question types, you need to see every CR argument the same way. Every argument has three components:
- Conclusion — The main claim or recommendation the author is making
- Premises — The evidence or facts supporting the conclusion
- Gap (Assumption) — The unstated logical connection between premises and conclusion
Example:
> Company X's revenue increased 20% this year. Therefore, the company is more profitable than last year.
- Premise: Revenue increased 20%
- Conclusion: The company is more profitable
- Gap: Revenue increase doesn't automatically mean profit increase — costs could have risen 30%
Every wrong answer in CR exploits the gap. The gap is where the argument is vulnerable, and understanding it is the key to answering every question type correctly.
How to Identify the Conclusion
- Therefore, thus, hence, so, consequently
- "This suggests that…" / "It follows that…"
- Recommendations: "should," "must," "needs to"
- Predictions: "will," "is likely to"
If there are no indicator words, ask: "What is the author trying to convince me of?" That's the conclusion. Everything else is support.
Question Type 1: Weaken
What it asks: "Which of the following, if true, most weakens the argument?"
Strategy: Identify the gap. The correct answer attacks the gap — it shows that the conclusion doesn't necessarily follow from the premises even if the premises are true.
- Alternative cause: The premises describe a correlation, but there's another explanation
- Breaks the analogy: The argument compares two things, but they're relevantly different
- Implementation flaw: The plan seems good but has a practical obstacle
- Scope shift: What's true for the sample isn't true for the population
- Answers that weaken a premise (the argument assumes premises are true)
- Answers that are too extreme ("completely disproves" vs. "casts doubt")
- Answers that are irrelevant to the specific gap
Question Type 2: Strengthen
What it asks: "Which of the following, if true, most strengthens the argument?"
Strategy: The exact reverse of weaken. Find the gap and choose the answer that bridges it — making the conclusion more likely to follow from the premises.
- Eliminates an alternative cause: Shows that the stated explanation is the right one
- Confirms the analogy: Shows the comparison is valid
- Addresses a potential objection: Preemptively blocks a weakener
- Provides additional supporting evidence
Key insight: The correct answer doesn't need to prove the conclusion. It just needs to make it more likely. "Most strengthens" means "shifts the probability the most," not "makes it certain."
Question Type 3: Assumption
What it asks: "The argument above depends on which of the following assumptions?"
Strategy: The assumption IS the gap. It's the unstated belief that must be true for the conclusion to follow. Use the Negation Test: negate each answer choice. If negating it destroys the argument, it's a necessary assumption.
- Argument: "Sales fell because of the new competitor."
- Proposed assumption: "No other factors caused the sales decline."
- Negated: "Other factors DID cause the sales decline."
- Does this destroy the argument? Yes → It's a valid assumption.
- No alternative explanation exists
- A comparison is valid (same conditions, similar populations)
- A proposed plan is feasible
- A cited cause precedes the effect
- A sample is representative
- Assumptions that are too strong ("the ONLY cause" when the argument just says "a cause")
- Assumptions that restate the conclusion
- Assumptions that support the argument but aren't required for it
Question Type 4: Inference
What it asks: "Which of the following can be properly inferred from the statements above?"
Strategy: Unlike other CR types, inference questions don't have a conclusion to analyze. The passage gives you facts, and you must determine what follows logically. The correct answer will be directly supported — often conservatively phrased.
- The correct answer is ALWAYS supported by the passage — if you need outside knowledge, it's wrong
- Extreme language is almost always wrong ("all," "never," "must")
- Hedged language is often right ("some," "can," "may," "not necessarily")
- "Most strongly supported" means "most directly follows" — not "most interesting"
Common trap: Choosing an answer that's likely true in the real world but isn't supported by the specific passage. Inference questions test reading comprehension of the argument, not general knowledge.
Question Type 5: Evaluate
What it asks: "Which of the following would be most useful to evaluate the argument?"
Strategy: Identify the gap, then choose the answer that, depending on its actual answer, would either strengthen or weaken the argument. A good evaluate answer is a question whose "yes" answer strengthens and whose "no" answer weakens (or vice versa).
- If the answer to the proposed question is "yes" → does it affect the argument?
- If the answer is "no" → does it affect the argument?
- If both answers affect the argument in different directions → it's correct
- If neither or only one affects it → it's wrong
Example:
> "The city should build a new highway to reduce commute times."
>
> "Would the new highway attract additional drivers from other routes?"
> - If yes → commute times might not decrease → weakens
> - If no → the plan is more likely to work → strengthens
> - ✓ This is a valid evaluate answer
Question Type 6: Boldface / Role
What it asks: "The boldface portions play which of the following roles in the argument?"
- The main conclusion
- A sub-conclusion (intermediate conclusion that supports the main one)
- A premise (evidence supporting a conclusion)
- A counter-premise (evidence against the author's position)
- Background information
- Identify the main conclusion first (regardless of boldface)
- Determine each boldface portion's relationship to the main conclusion
- Match to the answer choice that correctly describes both roles
Universal CR Tips
Pre-phrase before reading answers
After reading the stimulus, predict what the correct answer should do BEFORE looking at the choices. This prevents the GMAT's attractive wrong answers from pulling you off track.
For weaken: "The correct answer should show that [gap] might not hold."
For strengthen: "The correct answer should show that [gap] is valid."
For assumption: "The argument assumes that [gap]."
Eliminate confidently
- Out of scope: Addresses something the argument doesn't discuss
- Wrong direction: Strengthens when the question asks to weaken (or vice versa)
- Too extreme: Uses absolute language that the argument doesn't require
Watch for scope and degree
- "some" vs. "most" vs. "all"
- "can" vs. "will" vs. "must"
- "contributed to" vs. "caused" vs. "was the primary cause of"
An answer might be wrong solely because it says "will always" when the argument only supports "can sometimes."
Timing
Aim for 2 minutes per CR question. Spend 60-75 seconds reading and understanding the argument (including identifying the conclusion, premises, and gap). Spend 45-60 seconds evaluating answer choices. If you're torn between two answers after 2.5 minutes, pick the less extreme one and move on.
Building Your CR Skills
Week 1: Learn to identify conclusions, premises, and gaps in every argument you see — not just GMAT problems. News articles, advertisements, and editorials all have logical structures.
Week 2-3: Practice by question type. Do batches of 10 weaken questions, then 10 strengthen, then 10 assumption. Build pattern recognition for each type.
Week 4+: Mix all question types. Time yourself. Review every error and classify why you got it wrong (misidentified the conclusion? Fell for a scope trap? Missed the gap?).
Critical Reasoning is a learnable skill. The logic is consistent, the patterns repeat, and with deliberate practice, most students see significant improvement within 2-3 weeks.