GMAT Critical Reasoning: Evaluate and Paradox
Last updated: May 2, 2026
Critical Reasoning: Evaluate and Paradox questions are one of the highest-leverage areas to study for the GMAT. This guide breaks down the rule, the elements you need to recognize, the named traps that catch most students, and a memory aid that scales to test day. Read it once, then practice the same sub-topic adaptively in the app.
The rule
Evaluate questions ask which piece of information would be most useful in judging whether an argument's conclusion is well-supported; the right answer is a question (or variable) whose two opposite answers swing the conclusion in opposite directions. Paradox questions present two facts that seem to contradict each other and ask which choice best explains how both can be true; the right answer reconciles the tension without denying either fact. In both, you treat the stimulus as a closed system and look for the choice that interacts with the gap between the premises and the surprising result.
Elements breakdown
Evaluate Setup
Identify the conclusion, the premises that support it, and the assumption that bridges them.
- Locate the conclusion sentence first
- Underline supporting premises
- Name the unstated bridge assumption
- Ask what would weaken or strengthen
The Variance Test
For Evaluate questions, plug each answer choice's two extreme outcomes into the argument.
- Imagine the answer is 'yes' then 'no'
- Check whether conclusion changes direction
- Pick choice with maximum swing
- Reject choices whose answers don't matter
Common examples:
- If a study cites a drug's success, ask whether the control group received the same care.
Paradox Setup
Pin down the two facts that appear to conflict and the missing context that would let both be true.
- Mark fact A and fact B clearly
- State why they seem incompatible
- Search for hidden third variable
- Reconcile without contradicting either fact
Answer Vetting
Filter choices through scope, direction, and explanatory power.
- Eliminate out-of-scope information
- Reject choices that deny a stated fact
- Reject choices that deepen the puzzle
- Confirm both facts can coexist
Common patterns and traps
The Both-Answers Swing
On Evaluate questions, the credited answer is a variable or question whose two opposite resolutions push the conclusion in opposite directions. If a 'yes' answer and a 'no' answer leave the argument in the same place, the choice is irrelevant. This is the single most reliable test on Evaluate.
A choice phrased as 'Whether X is true of Y' where imagining X true devastates the conclusion and imagining X false confirms it.
The Premise Echo
A common wrong-answer flavor that simply paraphrases something already stated in the stimulus. Because the information is already known, asking about it provides no new evaluative leverage. Students fall for this when the wording feels familiar and 'safe.'
A choice that restates a premise as a question, e.g., 'Whether sales rose after the launch' when the stimulus already says sales rose.
The Fact-Denying Resolution
On Paradox questions, a trap choice resolves the tension by implicitly contradicting one of the stated facts. The credited answer must keep both surprising facts intact and introduce a third variable. Anything that says 'actually fact A wasn't really true' is wrong.
A choice that explains away the puzzle by claiming the original measurement was inaccurate or the trend didn't really happen.
The Half-Resolution
A paradox choice that addresses only one half of the apparent contradiction and leaves the other half unexplained. The right answer must connect both sides; a half-resolution feels relevant but doesn't actually reconcile the tension.
A choice that explains why fact A occurred but says nothing about how fact B can simultaneously be true.
The Out-of-Scope Variable
A choice that introduces a real-sounding factor that the argument never invoked and that doesn't actually engage the conclusion's logic. These often sound 'business-y' or technical but operate outside the argument's frame. Test by asking whether resolving the variable would actually change the specific conclusion stated.
A choice mentioning industry-wide trends, future projections, or unrelated metrics that don't connect to the specific cause-and-effect link in the stimulus.
How it works
Imagine an argument that claims a city's new bike-share program reduced downtown car traffic because traffic counts fell 12% after launch. The unstated assumption is that nothing else changed at the same time. An Evaluate question would ask which inquiry would most help judge the claim — and the strongest choice would be something like 'whether a major employer simultaneously shifted to remote work,' because a 'yes' destroys the conclusion and a 'no' reinforces it. A Paradox version of the same scenario might note that bike-share usage was high yet downtown air quality worsened; the resolving answer might reveal that delivery-truck routes were rerouted through downtown during the same period. Notice how both question types reward the same instinct: find the variable the argument forgot to consider.
Worked examples
Which of the following would be most useful to know in evaluating Marta Reyes's conclusion?
- A Whether Tessen Logistics has the budget to install the driver-coaching app on the trucks in its other three regional fleets
- B Whether the average price of diesel fuel changed substantially during the quarter in which the coaching app was used ✓ Correct
- C Whether drivers in Tessen's other regional fleets are paid the same hourly wage as drivers in the fleet that used the coaching app
- D Whether Tessen Logistics has previously rolled out new operational technologies across all four of its regional fleets
- E Whether Marta Reyes designed the driver-coaching app herself or licensed it from an outside vendor
Why B is correct: The argument concludes that the app caused the fuel-cost drop and will replicate elsewhere. If diesel prices fell sharply during the same quarter, the savings might owe more to market conditions than to the app, and the cross-fleet projection collapses. If diesel prices held steady, the app's causal role is reinforced. That's the variance the Both-Answers Swing demands.
Why each wrong choice fails:
- A: Budget capacity addresses whether the rollout is feasible, not whether it would actually produce the predicted savings. The argument is about causation, not affordability. (The Out-of-Scope Variable)
- C: Driver wages don't connect to the causal claim that the app reduces fuel use through coaching feedback. Whether wages match or differ leaves the conclusion unmoved. (The Out-of-Scope Variable)
- D: A track record of cross-fleet rollouts speaks to organizational capability but not to whether this specific app caused this specific savings figure. Yes-or-no, the conclusion's logic doesn't shift. (The Out-of-Scope Variable)
- E: The app's authorship has no bearing on whether the app caused the fuel drop. This is a biographical detail dressed up as a relevant factor. (The Out-of-Scope Variable)
Which of the following, if true, most helps to explain the discrepancy described above?
- A The e-reader lending kiosk at the Halverton library can hold up to 40 different titles at one time, a smaller selection than the library's previous print collection.
- B Patrons at the Halverton library frequently report that they enjoy reading on e-readers as much as they enjoy reading printed books.
- C Each e-reader checked out from the kiosk contains, on average, twelve full-length novels that a single patron can read during one borrowing period. ✓ Correct
- D The Halverton library's annual operating budget did not increase in the year following the introduction of the e-reader kiosk.
- E Library patrons in towns near Halverton have continued to borrow printed novels at roughly the same rate they did in previous years.
Why C is correct: Both surprising facts must remain intact: total borrowings dropped sharply, yet the kiosk is constantly in use. If each borrowed e-reader holds twelve novels, a single check-out replaces twelve previous print check-outs. Heavy kiosk usage is consistent with much lower 'borrowing' counts because each transaction now bundles many books. Both facts coexist without contradiction.
Why each wrong choice fails:
- A: A smaller selection might explain why patrons borrow less, but it doesn't explain why the kiosk itself is constantly in use. It addresses only half the puzzle. (The Half-Resolution)
- B: Patron preference equivalence makes the drop in borrowings more puzzling, not less, and offers no mechanism for reconciling the two facts. It deepens the paradox rather than resolving it. (The Half-Resolution)
- D: A flat budget says nothing about why borrowings dropped while kiosk usage stayed high. It introduces a financial factor unrelated to either reported fact. (The Out-of-Scope Variable)
- E: What patrons in nearby towns do is irrelevant to the internal mechanics of Halverton's library. It doesn't explain how heavy kiosk use coexists with falling borrowings. (The Out-of-Scope Variable)
Which of the following would be most useful to determine in evaluating the officials' conclusion?
- A Whether patients on the split-dose schedule were, on average, more likely than patients on the single-dose schedule to also receive other respiratory vaccinations during the same winter ✓ Correct
- B Whether the seasonal influenza vaccine used in the clinical trial was the same formulation as the vaccine the regional health authority distributes to the general public
- C Whether the clinical trial included patients from age groups outside the range typically served by the regional health authority
- D Whether the cost per patient of the split-dose schedule is meaningfully higher than the cost per patient of the single-dose schedule
- E Whether the patients in the clinical trial were informed of which dosing schedule they had been assigned to
Why A is correct: The conclusion attributes the lower flu rate to stronger immunity from the split schedule itself. If split-dose patients were also more likely to receive other respiratory vaccines, those co-vaccinations could account for the 22% gap, and the immunity claim collapses. If they weren't, the causal attribution holds. That's a clean Both-Answers Swing on the actual cause-and-effect link.
Why each wrong choice fails:
- B: If the formulations differ, that complicates the rollout but doesn't directly test whether the split schedule itself produces stronger immunity than the single-dose schedule. It addresses generalizability rather than the causal mechanism. (The Out-of-Scope Variable)
- C: Age-range generalizability is a downstream concern about applying the result, not a test of whether the trial's outcome reflects stronger immunity. Knowing the answer doesn't swing the conclusion either way. (The Out-of-Scope Variable)
- D: Cost differences are policy considerations, not evidence about which schedule produces stronger immunity. The conclusion is about biological efficacy. (The Out-of-Scope Variable)
- E: Patient awareness of assignment matters for trial blinding but the stimulus already invites you to take the 22% figure at face value. Resolving this question doesn't directly target the immunity claim and is more of a methodological aside than a hinge variable. (The Premise Echo)
Memory aid
For Evaluate, run the 'Both-Answers Test': flip the choice to yes, then to no — if the conclusion doesn't move, it's wrong. For Paradox, run the 'Keep-Both-Facts Test': the right answer leaves both surprising facts intact and adds the missing context.
Key distinction
Evaluate asks what you would need to KNOW to judge the argument; Paradox asks what would EXPLAIN why two facts coexist. Evaluate answers are framed as questions or variables; Paradox answers are framed as new facts.
Summary
Find the assumption gap, then pick the choice whose two possible outcomes (Evaluate) or hidden third variable (Paradox) directly engages that gap without denying anything stated.
Practice critical reasoning: evaluate and paradox adaptively
Reading the rule is the start. Working GMAT-format questions on this sub-topic with adaptive selection, watching your mastery score climb in real time, and seeing the items you missed return on a spaced-repetition schedule — that's where score lift actually happens. Free for seven days. No credit card required.
Start your free 7-day trialFrequently asked questions
What is critical reasoning: evaluate and paradox on the GMAT?
Evaluate questions ask which piece of information would be most useful in judging whether an argument's conclusion is well-supported; the right answer is a question (or variable) whose two opposite answers swing the conclusion in opposite directions. Paradox questions present two facts that seem to contradict each other and ask which choice best explains how both can be true; the right answer reconciles the tension without denying either fact. In both, you treat the stimulus as a closed system and look for the choice that interacts with the gap between the premises and the surprising result.
How do I practice critical reasoning: evaluate and paradox questions?
The fastest way to improve on critical reasoning: evaluate and paradox is targeted, adaptive practice — working questions that focus on your specific weak spots within this sub-topic, getting immediate feedback, and revisiting items you missed on a spaced-repetition schedule. Neureto's adaptive engine does this automatically across the GMAT; start a free 7-day trial to see your sub-topic mastery climb in real time.
What's the most important distinction to remember for critical reasoning: evaluate and paradox?
Evaluate asks what you would need to KNOW to judge the argument; Paradox asks what would EXPLAIN why two facts coexist. Evaluate answers are framed as questions or variables; Paradox answers are framed as new facts.
Is there a memory aid for critical reasoning: evaluate and paradox questions?
For Evaluate, run the 'Both-Answers Test': flip the choice to yes, then to no — if the conclusion doesn't move, it's wrong. For Paradox, run the 'Keep-Both-Facts Test': the right answer leaves both surprising facts intact and adds the missing context.
What's a common trap on critical reasoning: evaluate and paradox questions?
Choosing answers that restate a premise
What's a common trap on critical reasoning: evaluate and paradox questions?
Picking explanations that deny one of the two paradox facts
Ready to drill these patterns?
Take a free GMAT assessment — about 25 minutes and Neureto will route more critical reasoning: evaluate and paradox questions your way until your sub-topic mastery score reflects real improvement, not luck. Free for seven days. No credit card required.
Start your free 7-day trial