Critical Reasoning Question Types for Placement Tests

Critical reasoning questions test one skill above all others: the ability to separate what the passage actually says from what you want it to say. That gap accounts for the majority of wrong answers.

These questions appear in AMCAT’s Logical Ability section, eLitmus, CoCubes, TCS NQT, Infosys InfyTQ, and company-specific tests across IT services, analytics, and consulting. Six question types cover nearly all of what you will encounter. Each has a different resolution structure. Getting them confused is the most common source of lost marks in the reasoning section.

Where Critical Reasoning Shows Up

Most placement tests organise their sections into quantitative aptitude, verbal ability, and logical or analytical reasoning. Critical reasoning sits inside the logical reasoning section: sometimes a labelled sub-section and sometimes mixed in with other deductive questions.

Here is where each question type tends to cluster:

Platform	CR Question Types Commonly Tested
AMCAT Logical Ability	Strengthen/weaken, assumption, cause-effect, syllogism
eLitmus pH Test	Inference, assumption, paragraph completion
CoCubes Reasoning	Statement-conclusion, course of action, syllogism
TCS NQT Reasoning	Syllogism, statement-conclusion, inference
Bank PO (SBI/IBPS)	All six types, at higher difficulty

Analytics-focused companies also place higher weight on reasoning. Mu Sigma’s analytical selection process includes structured case reasoning alongside its MuApt written test. The same reasoning skills that clear placement aptitude rounds become the foundation of day-to-day decision-science work.

Students targeting banking sector roles after engineering will find the critical reasoning weight is even higher: SBI PO and IBPS PO devote full sections to it at CAT-comparable difficulty.

Strengthen and Weaken Arguments

An argument has three parts: a premise (evidence), a conclusion (what the author claims follows), and the logical link between them. A strengthening statement makes that link more solid. A weakening statement creates a gap in it.

The question stem always makes this explicit: ‘Which of the following, if true, strengthens the argument?’ or ‘Which option, if true, most weakens the conclusion?‘

Worked Example

Argument: An IT company in Hyderabad launched a mentorship programme for junior engineers. In the six months after the launch, attrition among junior engineers fell.

• Q: Which statement, if true, STRENGTHENS the claim that the programme caused the reduction in attrition?
• Option A: ‘Several competitors in the same city also saw junior attrition fall during the same period due to a market-wide salary adjustment.’
• Option B: ‘Exit interviews from junior engineers who stayed cited the mentorship programme as the main reason for their decision.’
• Option C: ‘The mentorship programme was launched at the same time as a broader office-renovation project.‘

Resolution Steps

• Step 1: Identify the conclusion. The conclusion is that the mentorship programme caused the attrition drop.
• Step 2: Identify what the conclusion needs to be true. It needs a specific link between the programme and the decision to stay.
• Step 3: Test each option.
  • Option A introduces an alternative explanation (market salary adjustment) — this weakens, not strengthens.
  • Option B directly links the programme to the stay decision through exit interview data — this strengthens.
  • Option C introduces another concurrent change (renovation) — another alternative cause, which weakens.
• Answer: Option B. It provides direct evidence connecting the programme to the outcome claimed.

Assumption and Inference Questions

These are frequently confused. An assumption is a hidden premise the argument cannot function without. Remove it, and the argument collapses. An inference is a conclusion the passage allows you to draw from explicit information.

Assumptions go inward: what must be true for this argument to stand? Inferences go outward: what does this explicitly tell us?

Worked Example — Assumption

Argument: ‘All shortlisted candidates for the technical interview scored above the 70th percentile in the aptitude test. Ananya was shortlisted for the technical interview.’

• Implicit conclusion: Ananya scored above the 70th percentile.
• Q: Which of the following is an assumption the argument depends on?
• Option A: Ananya prepared for the aptitude test using paid coaching material.
• Option B: The company shortlists candidates based solely on aptitude percentile, with no other criteria.
• Option C: The 70th percentile cutoff is common across all companies using this test.

Resolution Steps

• Step 1: Map what we know explicitly: all shortlisted candidates scored above the 70th percentile; Ananya was shortlisted.
• Step 2: The conclusion follows if the first premise is exhaustive (no other shortlisting path exists).
• Step 3: Option A is irrelevant to the conclusion. Option C is about other companies — out of scope. Option B fills the gap: the argument works only if percentile score is the sole criterion used to shortlist.
• Answer: Option B. Without it, Ananya could have been shortlisted by some other route despite scoring below the cutoff.

Worked Example — Inference

Statement: ‘The college placed 210 students in software roles this year. Of these, 180 were from the Computer Science and IT departments.’

• Valid inference: More than 85% of the college’s software placements came from CSE and IT branches.
• Invalid inference: Students from other branches cannot get software roles. (The statement shows they can — 30 of them did.)

Cause-Effect and Correlation Questions

These questions give you two observations and ask whether one caused the other, whether the relationship is coincidental, or whether some third factor explains both.

The ground rule: unless the passage explicitly describes a mechanism linking two events, you cannot conclude causation. Correlated data proves only correlation.

Worked Example

• Statement 1: Enrollments in a Pune engineering college’s optional Python elective rose by 40 students this semester.

• Statement 2: The college’s median placement package improved over the same period.

• Q: What is the most defensible conclusion?

  • Option A: The Python elective caused the package improvement.
  • Option B: The two observations occurred simultaneously but no causal link is established by this data.
  • Option C: Students who took the Python elective received higher packages.

Resolution Steps

• Step 1: Are both statements facts? Yes. Enrollment count and median package are independently stated observations.
• Step 2: Does the passage describe any mechanism connecting them? No. No data links individual students’ elective status to their package outcome.
• Step 3: Option A asserts causation from correlation alone — incorrect. Option C goes further, adding a claim about individual students not supported by the passage. Option B accurately describes what the data allows.
• Answer: Option B. Simultaneous observations with no stated mechanism.

Syllogism: Deductive Reasoning Questions

A syllogism gives you two or three premise statements and asks what conclusion must logically follow. The answer must follow with certainty from the premises alone: not probably, not usually, but necessarily.

The fastest solving method: draw a simple Venn diagram. Represent each set as a circle, map the premises, and check whether the conclusion can be falsified by any valid arrangement.

Worked Example

• Premise 1: All candidates who pass the verbal section advance to the group discussion round.

• Premise 2: No candidate who fails the aptitude test passes the verbal section.

• Premise 3: Suresh passed the aptitude test and passed the verbal section.

• Q: Which conclusion follows?

  • Option A: Suresh advances to the group discussion round.
  • Option B: All candidates who pass the aptitude test advance to the group discussion round.
  • Option C: Suresh is the only candidate who passed both tests.

Resolution Steps

• Step 1: From Premise 1, passing the verbal section guarantees advancement to the GD round.
• Step 2: From Premise 3, Suresh passed the verbal section.
• Step 3: Applying Premise 1 to Suresh: he advances to the GD round. Option A follows with certainty.
• Step 4: Option B goes beyond Premise 1. Premise 1 does not guarantee advancement to every aptitude passer, only to verbal-section passers. Option B is invalid.
• Step 5: Option C makes a uniqueness claim. Nothing in the premises restricts passing to Suresh alone.
• Answer: Option A. It follows necessarily from Premises 1 and 3.

Statement-Conclusion and Course of Action

Two related sub-types that often appear in the same section.

Statement-conclusion questions ask whether a listed conclusion can be drawn from a given statement. The test: can the conclusion be false even if the statement is true? If yes, the conclusion does not follow.

Course-of-action questions give you a problem statement and ask which proposed action is appropriate, practical, and directly relevant to the stated problem.

Worked Example — Statement-Conclusion

Statement: ‘A telecom company’s customer complaint volume doubled in the third quarter compared to the second quarter.’

• Conclusion A: The company’s network quality deteriorated in Q3.
• Conclusion B: The company received more complaints in Q3 than in Q2.

Resolution Steps

• Step 1: Conclusion B is a direct restatement of the data. It cannot be false if the statement is true. Conclusion B follows.
• Step 2: Conclusion A introduces a specific explanation. The statement does not say why complaints doubled — billing errors, customer service wait times, or a price change could also explain it. Conclusion A does not follow from this statement alone.
• Answer: Conclusion B follows. Conclusion A does not.

Worked Example — Course of Action

• Problem: An edtech company’s live-class attendance dropped to 40% by week 3 of a cohort, against a target of 75%.
• Action A: Offer refunds to all enrolled students immediately to prevent reputational damage.
• Action B: Survey enrolled students to identify the root cause before making any changes.
• Action C: Cancel the remaining live sessions and convert the cohort to self-paced material.

Resolution Steps

• Step 1: The problem is an attendance shortfall. The cause is unknown.
• Step 2: Action A addresses a consequence (reputation) rather than the problem itself, and is premature before diagnosis.
• Step 3: Action C is a permanent structural change made without diagnostic information.
• Step 4: Action B directly targets the unknown: it gathers the information needed to make any other action effective.
• Answer: Action B follows. It is practical, addresses the core issue, and does not pre-empt a solution before the problem is understood.

Paragraph Completion and Contextual Inference

These questions give you an incomplete paragraph and ask you to choose the sentence that best completes it: logically, tonally, or as the most defensible next thought.

The correct answer must be (a) consistent with everything already said, (b) not a repetition of what was just said, and (c) not a leap into new territory the passage has not set up.

Worked Example

Incomplete paragraph: ‘Modern placement tests have shifted away from purely mechanical aptitude drills. Recruiters at analytics and consulting firms now include reasoning-based case questions to observe structured thinking under time pressure. This shift reflects…’

• Option A: …the growing importance of certifications in engineering curricula.
• Option B: …a recognition that the ability to reason about ambiguous data matters more than rote calculation speed.
• Option C: …that aptitude test scores are no longer used in shortlisting decisions.

Resolution Steps

• Step 1: The paragraph establishes two facts: tests have shifted toward reasoning questions; this is driven by what analytics and consulting recruiters want.
• Step 2: The completion should explain why that shift reflects something about hiring needs.
• Step 3: Option A introduces certifications, a new topic not set up by the paragraph.
• Step 4: Option C contradicts the paragraph’s premise (the tests still exist, they’ve just changed form).
• Step 5: Option B directly completes the logic: the shift reflects the value of reasoning ability over mechanical calculation speed.
• Answer: Option B.

Four Common Traps Across All Question Types

These traps appear across all six question types. Knowing the label does not prevent the error. You have to catch the specific form it takes in each question.

Correlation Is Not Causation

Two events occurring together does not establish that one caused the other. The cause-effect questions bank on this directly. It also surfaces in strengthen/weaken questions, where an option describing a concurrent event is tempting but wrong.

Reversed Cause-Effect

The passage may present A-then-B and imply A caused B. The trap is that B may have caused A, or both may share a common cause. Reverse the arrow in your head and check whether the reversed version is also consistent with the data.

Scope Shift

A conclusion that applies to a larger or different group than the one in the premise is a scope shift. If the statement is about ‘engineers in Chennai’, a conclusion about ‘engineers across India’ shifts the scope. This appears in inference and statement-conclusion questions.

Unstated Assumption Made Into Fact

Assumption questions test exactly this: the argument relies on an unstated premise. The trap is treating that premise as established fact just because the argument depends on it. The question asks you to name the assumption, not to accept it as proven.

---

When you work through reasoning questions regularly, a pattern emerges: the same structured thinking that clears aptitude rounds is what data analysts and product managers use when interpreting business data. The connection is not coincidental.

If analytics or data-adjacent roles are your placement target, TinkerLLM is an AI learning environment where you can apply these reasoning patterns to real AI tools at ₹299. Your target company’s test pattern will tell you which of the six question types to prioritise first.