Quantitative: Logical Counting
What it tests. Translating ambiguous English prose into solvable linear equations under time pressure; mental computation, since calculators are banned.
Worked example. A budget of $180,000 buys Component A at $1,200 and Component B at $1,500, where A units must be 15% higher than twice the number of B units. If the budget is exhausted, how many units of A were purchased?
Common traps. The distractor variable: misapplying the 15% modifier to the wrong side, or solving for B and circling that value without converting back to A.
How to handle it. Write the core equations immediately, simplify numbers (divide the budget equation by 100) and back-solve from the answer choices when algebra stalls.
Quantitative: Statements (Data Sufficiency)
What it tests. Executive meta-cognition: judging whether the data can answer the question without burning time computing it.
Worked example. What was the net profit margin for a consumer-goods client in FY2025, given (1) revenue of $450M, up 12%, and (2) operating expenses plus COGS of $390M with $15M of taxes and interest?
Common traps. The C-Trap: defaulting to both-statements-together without testing each in isolation, or forgetting that sufficient means a single unique answer.
How to handle it. Memorize the AD/BCE elimination matrix, keep statements strictly isolated, and stop the moment sufficiency is proven (do not finish the arithmetic).
Verbal: Understanding of Logical Text
What it tests. Formal logical parsing and semantic precision: separating explicit facts from unstated assumptions.
Worked example. A recommendation to shift sourcing to regional suppliers at 15% higher unit cost relies on which underlying assumption about cross-border delay and tariff losses?
Common traps. The Plausible Real-World Truth trap: picking an answer that sounds reasonable but is not structurally required by the argument.
How to handle it. Use the negation technique on assumption questions, map premises and conclusion first, and distrust extreme qualifiers like always, never or immediately.
Verbal: Reading a Passage
What it tests. Information filtering: extracting granular insight from a wall of text under time pressure.
Worked example. Categorize a claim about a specific paragraph as True, False, or Cannot Say based strictly on the text.
Common traps. The Extrapolation trap: marking likely-but-unproven claims True instead of Cannot Say, and misreading modifiers like most, some or frequently.
How to handle it. Read the question stems before the passage, jot a three-word summary per paragraph, and treat the passage as the only universe of truth.
Case Studies: Reading a Table of Numbers
What it tests. Structural data extraction and sequential calculation: isolating the right row/column intersection and running back-to-back math.
Worked example. Given per-plant output, cost per unit and defect rate, compute the monetary loss from defects at one facility and how much lower it would be at another plant defect rate.
Common traps. Unit mismatches (values in thousands or millions) and baseline errors (using the 2023 column instead of 2024).
How to handle it. Anchor your eye to the exact row and column, estimate first against the answer spread, and label intermediate steps so you can audit in seconds.
Case Studies: Reading a Graph
What it tests. Visual-spatial data processing and chart literacy: diagnosing trends and relationships between variables.
Worked example. Translate a visual slope into a CAGR, or identify which stacked-bar segment accelerated fastest over a timeframe.
Common traps. Dual Y-Axis Blindness: reading a point off the wrong axis (absolute value vs percentage), plus non-zero baselines that exaggerate small changes.
How to handle it. Audit axes and legends before touching the data, rely on numerical tick marks over visual height, and master fraction-to-percentage conversions.