2 · Mathematical Equations
Solve the system so every equation holds. Each letter is an integer between 1 and 20, and there is exactly one solution.
Exam format: 20 systems in 25 minutes, solved mentally — no rough paper.
Difficulty: low
What is B?
6 + A = 13
B − A = 5
First equation: A = 13 − 6 = 7. Substitute into the second: B − 7 = 5 → B = 12.
Difficulty: medium
What is B?
3 × C = A
A + C = 16
B = A + C − 6
Replace A with 3C in the second equation: 3C + C = 16 → C = 4, so A = 12. Then B = 12 + 4 − 6 = 10.
Difficulty: high
What is D?
A = 5 × B
C = B + 9
A − B + C − D = 12
D = 3 × B + 1
Substitute everything into the third equation: 5B − B + (B + 9) − (3B + 1) = 12 → 2B + 8 = 12 → B = 2. Then D = 3 × 2 + 1 = 7. (And A = 10, C = 11 — all within 1–20.)
Difficulty: low
What is A?
B ÷ 2 = A
A + B = 18
Multiply the first equation by 2: B = 2A. Substitute: A + 2A = 18 → A = 6 (and B = 12).
Difficulty: high
What is B?
A + 3 = B
D = 2 × A
C = D − B
C = 4
Since C = 4, the third equation gives D − B = 4. Substitute D = 2A and B = A + 3: 2A − A − 3 = 4 → A = 7. So B = 10 (and D = 14). Notice the habit: substitute everything into the one equation with a known value.
Difficulty: low
What is B?
A − 4 = B
A + B = 14
From the first equation A = B + 4, so (B + 4) + B = 14 → B = 5 (and A = 9).
Difficulty: medium
What is A?
C = A + B
A = 2 × B
C = 9
Substitute A = 2B into the first equation: 2B + B = 9 → B = 3, so A = 6.
Difficulty: medium
What is B?
B ÷ A = 4
B − A = 12
B = 4A, so 4A − A = 12 → A = 4 and B = 16.
Difficulty: medium
What is C?
A + B + C = 22
B = A + 2
C = 3 × A
Substitute everything into the first: A + (A + 2) + 3A = 22 → 5A = 20 → A = 4, so C = 12 (and B = 6).
Difficulty: high
What is A?
D − C = 5
C = 2 × B
A = B + C + D
B = 2
Start from the known value: B = 2 gives C = 4, then D = 9, so A = 2 + 4 + 9 = 15. Always anchor on the equation that already contains a number.
Difficulty: high
What is C?
C − B + D − A = 13
8 × A = C
3 × A = D
A + 5 = B
There is no equation here that hands you a number outright — the official materials' hardest equation exercise is built exactly this way, and the technique is always the same: pick the letter every other equation depends on and write everything in terms of it. Here that letter is A. C = 8A, D = 3A, B = A + 5. Substitute into the long equation: 8A − (A + 5) + 3A − A = 13, which is 9A − 5 = 13, so A = 2. Then C = 16 (and B = 7, D = 6 — all inside 1–20, as the format guarantees). 24 is 8 × 3 — multiplying the two coefficients instead of solving; 12 answers for D + wrong A; 18 comes from dropping the −5 when you collect terms.
3 · Latin Squares
Each of the letters A–E appears exactly once per row and once per column of the 5×5 grid. Which letter belongs in the ? field? Sometimes you must fill other cells in your head first.
Exam format: 20 grids in 25 minutes. Only elimination is needed — never guesswork.
Difficulty: low
Which letter replaces the question mark?
Look at column 1: it already contains E, A, D and C. The only letter missing is B.
Difficulty: medium
Which letter replaces the question mark?
Column 1 already has E and B, so the ? can only be A, C or D. Look at row 4 (A, E, C given): it still needs B and D, and its column-1 cell can't be B (column 1 already has one) — so that cell takes D. Now row 2: it already contains A, ruling A out for the ?. Only C remains.
Difficulty: high
Which letter replaces the question mark?
Column 3 already shows D, A and C — only B and E are missing, split between the ? and the bottom cell. Row 5 already contains E, so the bottom cell of column 3 must take B. That forces the ? to be E. Two steps, no guessing.
Difficulty: low
Which letter replaces the question mark?
Row 2 already contains E, A and B; column 4 adds E and D. Four letters excluded — only C fits.
Difficulty: low
Which letter replaces the question mark?
Column 5 contains B, C and A; row 5 adds B, A and E. The union rules out A, B, C and E — only D fits.
Difficulty: medium
Which letter replaces the question mark?
Row 4 still needs C and D, split between the ? and its column-3 cell. Column 3 already has a D (row 2), so the column-3 cell must take C — leaving D for the ?.
Difficulty: high
Which letter replaces the question mark?
Row 2 and column 2 together narrow the ? to A or C. Now test A: following the forced cells through columns 4 and 5 runs into a dead end — the grid can't complete. Only C works (verified unique by exhaustive check). On hard items, eliminate with the row-plus-column union first, then test the survivors.
Difficulty: high
Which letter replaces the question mark?
Row 5 and column 1 between them exclude only B and C, so three letters are still possible at the ? — you cannot read this one off. Turn it round and ask where a letter is still allowed. Look down column 1: C has exactly one legal free field there, so place it. That fills a cell in row 2, and now D has exactly one legal free field left in row 2 — place that too. Finally run the same test along row 5: every other free field in that row sits in a column that already contains a D, so the ? is the only place left for it. Answer: D. This is the “hidden single” move the official materials demonstrate — what you reach for when simple row-and-column elimination stalls, and the difference between a square you can solve and one you end up guessing.
4 · General Academic Module
You read a short academic text, then answer single-choice questions that apply it — 4 options each, one correct. 90 minutes for the whole module. This format is now confirmed in g.a.s.t.'s official GAM preparatory materials (published 16 July 2026), with topics spanning mathematics, natural sciences, engineering, business, economics, and social sciences. Warm up with five standalone questions, then try three passage sets in the official style below.
Applied reasoning
A standardised test reports results as percentile ranks: the percentage of all test takers whose score was equal to or lower than yours. In one sitting, 1,600 candidates take the test. Priya's score is higher than 1,120 of them and equal to none. What is Priya's percentile rank?
1,120 of 1,600 scored lower → 1,120 ÷ 1,600 = 0.70. Percentile rank = 70: she did as well as or better than 70% of all candidates. This is exactly how your dMAT certificate reports results (plus a 0–200 score per module).
Applied reasoning
A learning study reports: of 240 students who used spaced practice, 168 passed their exam; of 160 students who crammed, 96 passed. Which statement follows from this data?
Spaced practice caused the higher pass rate
Most students who passed had crammed
Fewer than half of the crammers passed
The pass rate was higher with spaced practice
168 ÷ 240 = 70% vs 96 ÷ 160 = 60% — the spaced-practice pass rate is higher, and that's all the data supports. "Caused" is the classic trap: the data shows association, not causation (maybe stronger students choose spaced practice). Passers: 168 > 96, so most passers did NOT cram; and 60% of crammers passing is more than half. Distinguishing what follows from what merely sounds plausible is the core General Academic Module skill.
Difficulty: Applied reasoning
A dMAT test city has 480 available seats across its centres, booked first-come-first-served. If about 5,000 candidates prefer that city, roughly what share of them will have to choose another city?
About 90%
About 10%
About 48%
Cannot be determined
480 of 5,000 preferences can be satisfied: 480 ÷ 5,000 = 9.6% get seats, so roughly 90% must book elsewhere. 'Cannot be determined' tempts because real demand is unknown — but the question says 'if about 5,000', making the calculation valid under its own assumption. (This is also why you register early.)
Difficulty: Applied reasoning
The dMAT score scale runs 0–200 with a fixed mean of 100 per module. Priya scores 118 on the core module and 96 on the subject module. Which statement follows?
She answered 118 core questions correctly
Her overall percentile must be above 50
Her core result is above average; her subject result is below average
Her subject performance failed the exam
The 0–200 scale has mean 100, so 118 is above and 96 below average — that is all the scale tells you. The score is a conversion, not a raw count (rules out B); the percentile depends on the whole distribution (C); and the dMAT has no pass/fail (D).
Difficulty: Applied reasoning
In a survey, 500 test takers who used official preparatory materials averaged 112, while 300 who didn't averaged 104. Which conclusion follows from the data alone?
On average, users of the official materials scored 8 points higher
Using the official materials raises a candidate's score by 8 points
Every user of the materials outscored every non-user
Most of the 800 candidates scored above 104
Only the difference in group averages follows from the data. 'Raises' claims causation (motivated students may self-select into preparing); averages say nothing about every individual (C); and a group mean doesn't locate the median (D).
5 · Passage sets — official GAM style
In the real subject module, each task is a text plus several questions. Read the passage once (no notes allowed in the exam!), then answer every question under it. One set each from business, natural sciences, and research methods — the topic areas named in the official materials.
PASSAGE A · BUSINESS ADMINISTRATION
Break-even analysis
A firm's fixed costs (rent, salaries, machinery) stay the same regardless of output, while variable costs are incurred per unit produced. If each unit sells at price p and costs v to make, the difference p − v is the contribution margin: what each sale contributes toward covering fixed costs F. The break-even quantity is the output at which total revenue exactly equals total cost: Q* = F ÷ (p − v). Below Q* the firm makes a loss; above it, a profit.
Difficulty: Passage · apply
A firm has fixed costs of ₹60,000 per month. Each unit sells for ₹50 and costs ₹30 to produce. What is the monthly break-even quantity?
1,200 units
2,000 units
4,000 units
3,000 units
Contribution margin = 50 − 30 = ₹20 per unit. Q* = 60,000 ÷ 20 = 3,000 units. The trap answers divide by the price alone (60,000 ÷ 50 = 1,200) or by the variable cost (60,000 ÷ 30 = 2,000) — both ignore that only the margin covers fixed costs.
Difficulty: Passage · concept
Which change lowers a firm's break-even quantity?
A rise in fixed costs
A rise in variable cost per unit
A rise in the selling price
A fall in the selling price
Q* = F ÷ (p − v). A higher selling price widens the contribution margin (the denominator), so fewer units are needed to cover fixed costs. Higher F raises Q*; higher v or lower p shrink the margin and raise Q* too.
Difficulty: Passage · transfer
Price and variable cost per unit both increase by exactly ₹5. What happens to the break-even quantity?
It rises
It stays the same
It falls
It cannot be determined
The contribution margin is (p + 5) − (v + 5) = p − v — unchanged. Since F is also unchanged, Q* is identical. The question tests whether you reason from the formula's structure rather than recalculating blindly — exactly the transfer skill the module measures.
PASSAGE B · NATURAL SCIENCES
Density and flotation
Density is mass per volume: ρ = m ÷ V. An object placed in a fluid floats if its average density is lower than the fluid's, and sinks if it is higher. A floating object displaces exactly enough fluid to support its weight, so the fraction of its volume below the surface equals the ratio of the densities: object ÷ fluid. Fresh water has a density of 1.0 g/cm³.
Difficulty: Passage · apply
A solid block has a mass of 400 g and a volume of 500 cm³. What fraction of the block is below the surface when it floats in fresh water?
Density = 400 ÷ 500 = 0.8 g/cm³. Submerged fraction = 0.8 ÷ 1.0 = 80%. The 20% option is the part above water — read what the question asks. 100% would mean the block barely sinks, which needs density ≥ 1.0.
Difficulty: Passage · concept
The same block is moved from fresh water into denser seawater (1.05 g/cm³). What happens?
It sinks deeper into the water
Exactly the same fraction is submerged
It sinks completely
It floats higher — a smaller fraction is submerged
Submerged fraction = ρobject ÷ ρfluid = 0.8 ÷ 1.05 ≈ 76%, less than the 80% in fresh water. A denser fluid supports the same weight with less displaced volume — the reason ships ride higher in the sea than in a river.
Difficulty: Passage · transfer
Objects A and B have equal mass, but A has twice the volume of B. Which statement follows?
A's density is twice B's
Their densities are equal
A's density is half of B's
It cannot be determined from the given information
ρ = m ÷ V: same numerator, double the denominator → half the density. No numbers are needed — the relationship alone answers it. "Cannot be determined" tempts because no masses are given, but the ratio is fully determined.
PASSAGE C · RESEARCH METHODS
Sampling and study design
Conclusions from a study are only as good as its design. A random sample gives every member of the population an equal chance of inclusion, so results can be generalised. A convenience or self-selected sample — whoever is nearby or chooses to respond — may differ systematically from the population. Observational data can show that two things are associated, but a causal claim requires an experiment in which researchers assign participants to conditions at random, so that no hidden third factor explains the difference.
Difficulty: Passage · apply
A university surveys students found in the library at 9 a.m. about campus-wide study habits. What is the main weakness?
The sample is too small to compute percentages
Library visitors at 9 a.m. may differ systematically from the student body
The question wording was leading
It proves nothing because correlation implies causation
This is a convenience sample: early-morning library users are plausibly more studious than average, so generalising to all students is unsafe. Nothing in the stem tells us the sample size or wording — don't import weaknesses the text doesn't state.
Difficulty: Passage · concept
What distinguishes a randomised experiment from an observational study?
Treatment is assigned at random by the researchers, which supports causal conclusions
It always includes more participants
It always runs for a longer period
It does not need a control group
Random assignment balances hidden factors across groups, so a difference in outcomes can be attributed to the treatment. Sample size and duration are separate design choices, and experiments typically rely on control groups rather than dropping them.
Difficulty: Passage · transfer
A fitness app finds that members who log meals lose more weight. What would be needed to support the claim that logging meals causes weight loss?
A much larger sample of app members
Random assignment of members to logging and non-logging groups
Repeating the same survey a year later
More precise measurement of weight
The association could reflect motivation: disciplined members both log meals and diet harder. Only random assignment removes that confound. A bigger sample, a repeat survey, or better scales make the association more precise — not causal.
Difficulty: Passage · transfer
A newspaper invites readers to vote online on a policy; 78% of the 100,000 votes oppose it. Why can't this result be generalised to the whole population?
The sample is too small
Percentages cannot be calculated from online votes
No reason — a large sample generalises automatically
Readers who choose to vote may differ systematically from the general population
100,000 is plenty — the flaw is self-selection: people with strong opinions (and this newspaper's readership) choose to vote. A biased sampling method isn't cured by size; a small random sample beats a huge self-selected one.