Probability Concepts and Problems for Aptitude Tests

Probability questions appear in every major campus placement aptitude test. Four reasoning patterns cover most of what gets asked: building the sample space, the addition rule, the multiplication rule, and conditional probability.

Formulas alone stall under time pressure. The reliable approach is understanding why each rule works: why the addition rule needs a subtraction term, when the denominator shrinks, and how conditional probability reframes the question. This article builds that intuition through worked examples across the classic problem types. The probability formulas reference guide covers each formula in condensed form for quick review.

Building the Sample Space

A sample space is the complete list of all possible outcomes for an experiment. Probability is then a ratio:

P(E) = number of favourable outcomes / total outcomes in sample space

Two conditions must hold for this to work: every outcome must be equally likely, and the list must be exhaustive.

Three standard experiments appear in most placement papers:

Experiment	Sample space	Size
Single coin toss	{H, T}	2
Single die roll	{1, 2, 3, 4, 5, 6}	6
Two coin tosses	{HH, HT, TH, TT}	4
Two dice	36 ordered pairs (1,1) to (6,6)	36

One common error: treating two-coin tosses as having three outcomes (both heads, mixed, both tails). The correct sample space has four entries because {HT} and {TH} are distinct outcomes. Collapsing them implies one is more likely than the other, which produces wrong answers.

Two worked problems using the table:

• Q: What is P(getting exactly one head in two coin tosses)?

  • Sample space: {HH, HT, TH, TT} — 4 equally likely outcomes
  • Favourable outcomes: {HT, TH} — 2 outcomes
  • Answer: P(exactly one head) = 2/4 = 1/2

• Q: What is P(rolling a prime number on a single die)?

  • Sample space: {1, 2, 3, 4, 5, 6} — 6 outcomes
  • Prime numbers in range: {2, 3, 5} — 3 outcomes
  • Answer: P(prime) = 3/6 = 1/2

Khan Academy’s basic probability article provides interactive practice problems to reinforce sample space construction.

The Addition Rule: OR Problems

“OR” questions ask for the probability that at least one of two events occurs. The general rule is:

P(A or B) = P(A) + P(B) minus P(A and B)

The subtraction term corrects for double-counting. If any outcome satisfies both A and B, adding P(A) and P(B) directly counts that outcome twice. Subtracting P(A and B) removes the duplicate.

The subtraction disappears only when the two events are mutually exclusive (meaning they cannot both occur at once). Sketch a quick Venn diagram: if the two circles do not overlap, skip the subtraction. If they overlap at all, the subtraction is required.

Three worked problems:

• Q: What is P(rolling odd or 6 on a single die)?

  • P(odd) = 3/6 — outcomes {1, 3, 5}
  • P(6) = 1/6 — outcome {6}
  • Check overlap: no number is both odd and equal to 6. Events are mutually exclusive.
  • Answer: P(odd or 6) = 3/6 + 1/6 = 4/6 = 2/3

• Q: What is P(rolling even or prime on a single die)?

  • P(even) = 3/6 — outcomes {2, 4, 6}
  • P(prime) = 3/6 — outcomes {2, 3, 5}
  • Check overlap: {2} is both even and prime. P(even and prime) = 1/6
  • Answer: P(even or prime) = 3/6 + 3/6 minus 1/6 = 5/6
  • Skipping the subtraction gives 3/6 + 3/6 = 1, which would mean the event is certain — but rolling a 1 satisfies neither condition, so the event is not certain.

• Q: What is P(drawing a king or a heart from a 52-card deck)?

  • P(king) = 4/52
  • P(heart) = 13/52
  • P(king and heart) = 1/52 — the king of hearts is the only card that is both
  • Answer: P(king or heart) = 4/52 + 13/52 minus 1/52 = 16/52 = 4/13
  • The answer is 4/13. Skipping the intersection gives 17/52, which overcounts the king of hearts.

The Multiplication Rule: AND Problems

“AND” questions require every listed event to occur. Individual probabilities multiply. Whether the denominator stays constant or shrinks depends on whether draws are with or without replacement.

Independent Events: Denominator Stays Constant

Each trial is unaffected by the others. A coin flip does not change the next flip. A drawn card returned to the deck leaves the total unchanged.

• Q: What is P(getting tails on two consecutive coin tosses)?
  • P(tails, first toss) = 1/2
  • P(tails, second toss) = 1/2 — independent
  • Answer: P(tails on both) = 1/2 × 1/2 = 1/4

Dependent Events: Denominator Shrinks

When each draw removes an item from the pool, the composition changes for the next draw.

• Q: A bag has 5 red and 3 green balls. What is P(drawing red first, then green) without replacement?
  • P(red first) = 5/8
  • After removing one red ball: 7 remain, 3 of which are green
  • P(green second | red first) = 3/7
  • Answer: P(red then green) = 5/8 × 3/7 = 15/56

The Complement Shortcut for At-Least-One Problems

Counting all “at least one success” cases directly often means summing several branches. The complement is faster.

• Q: What is P(getting at least one head in 3 coin tosses)?
  • P(no heads in 3 tosses) = 1/2 × 1/2 × 1/2 = 1/8
  • Answer: P(at least one head) = 1 minus 1/8 = 7/8
  • Direct check: {HHH, HHT, HTH, HTT, THH, THT, TTH, TTT} — 8 outcomes total, 7 contain at least one head. 7/8 confirmed.

Conditional Probability: Shrinking the Sample Space

Conditional probability answers a more specific question: given that one thing is already known to be true, what is the probability of something else?

The formula is P(A | B) = P(A and B) / P(B).

The mechanics: discard every outcome where B did not occur, then measure A within the surviving outcomes. The sample space shrinks from S to just the outcomes in B.

Two worked problems:

• Q: A bag has 4 red and 6 blue marbles. Two drawn without replacement. Given the first was red, what is P(the second is also red)?

  • After removing one red marble: 9 remain, 3 are red
  • P(second red | first red) = 3/9 = 1/3
  • Without the condition, P(second red) = 4/10 = 2/5. Knowing the first was red changes the answer.

• Q: A card is drawn from a standard 52-card deck. Given it is a face card, what is P(it is a king)?

  • Face cards (J, Q, K across 4 suits): 12 total
  • Kings among face cards: 4
  • P(king | face card) = 4/12 = 1/3
  • Without the condition, P(king) = 4/52 = 1/13. Knowing the card is a face card raises the probability from 1/13 to 1/3.

The conditional framework extends to Bayes reasoning: when new information arrives, it eliminates parts of the original sample space and rescales everything that remains. Most analytical-role placement tests include at least one question built on this logic.

Classic Problem Types in Placement Tests

Three problem types appear across most placement aptitude papers and are worth recognising on sight.

At-least-one problems are solved cleanest via complement, as shown above. Direct enumeration is possible but slow when the number of trials is large.

The birthday paradox is a useful calibration problem. With 23 people in a room, the probability that at least two share a birthday is approximately 50.7 percent, as Wikipedia’s birthday problem article documents. Most students guess much lower because they compare 23 against 365. The actual calculation compares pairs, not people:

• C(23, 2) = 253 unique pairs exist among 23 people
• Each pair independently has about a 1-in-365 chance of matching
• P(no shared birthday) = product of 364/365 × 363/365 × … × 343/365 (22 terms) ≈ 0.493
• P(at least one shared birthday) = 1 minus 0.493 ≈ 0.507

The same complement structure drives every at-least-one problem.

Card and ball problems require careful sample space enumeration. Always count ordered versus unordered draws correctly: drawing a red marble first and a blue marble second is a different event from blue first and red second when order matters.

Where probability appears in placement tests:

The campus placement evaluation test guide covers how each section of major aptitude tests is structured. Probability typically sits alongside time-and-work questions in the quantitative round. Analytics-oriented tests weight it more heavily. The Mu Sigma MuAPT test is a benchmark for conditional probability question difficulty, and D.E. Shaw’s campus process includes quantitative rounds where probability and combinatorics carry substantial marks.

The complement shortcut and conditional reasoning covered in this article connect directly to how language models work. Every token a model generates is chosen by exactly this logic: a probability distribution over possible next tokens, conditioned on everything that came before. TinkerLLM at ₹299 puts that distribution in your hands. Adjust temperature, inspect token probabilities, and watch Bayes-style conditioning reshape model output in real time.