Statistics › Section C · Probability › Topic C.2

Independent and mutually exclusive events

Does knowing one event happened change the chance of another? Conditional probability measures the change, independence is its absence, and mutual exclusivity is its extreme - three ideas this topic keeps firmly apart.

When knowing changes the odds

Does knowing one event happened change the chance of another? Sometimes yes, sometimes no, and much of the art of probability is telling the two cases apart. This topic builds conditional probability - the probability of one event given another - then defines independence (the case where knowing changes nothing), mutual exclusivity (the case where knowing changes everything), and the two sampling schemes, with and without replacement, that produce each.

Conditional probability

The probability of \(A\) given that \(B\) occurred restricts attention to \(B\)'s outcomes:

P(A \mid B) = \frac{P(A \text{ and } B)}{P(B)}

Suppose 30% of students take a music class and 12% take both music and drama. Among music students, the drama share is \(P(\text{drama} \mid \text{music}) = \frac{0.12}{0.30} = 0.40\). The number 0.40 is not the schoolwide drama rate - it is the rate inside the music classroom, because the given event, music, has become the whole sample space.

The multiplication rule

Rearranging the conditional formula gives the multiplication rule:

P(A \text{ and } B) = P(B) \times P(A \mid B)

The probability that two things both happen is the probability the first happens, times the probability the second happens given the first. Order does not matter mathematically - \(P(A) \times P(B \mid A)\) gives the same product - but the rule's power is sequencing: chains of events multiply their conditional probabilities step by step, which is how multi-stage problems are solved before tree diagrams make them visual in a later topic.

Independence

Two events are independent when knowing one occurred does not change the other's probability: \(P(A \mid B) = P(A)\). Under independence the multiplication rule simplifies to

P(A \text{ and } B) = P(A) \times P(B)

The statements \(P(A \mid B) = P(A)\), \(P(B \mid A) = P(B)\) and this product rule stand or fall together, so showing any one of them settles the question. Successive coin flips are independent: heads on the first flip leaves the second flip at 0.5. Drawing cards without replacement is not: removing an ace changes the chance that the next card is an ace. Independence is an assumption to check, never a default.

With and without replacement

Sampling with replacement returns each drawn item before the next draw, so every draw faces the same chances: the draws are independent. Sampling without replacement keeps each item out, so totals shrink and chances shift: the draws are dependent. From the marble jar (4 green, 3 yellow, 5 purple), drawing with replacement gives \(P(\text{green}) = \frac{4}{12}\) on every draw; without replacement, a green drawn first leaves \(\frac{3}{11}\) for the second. Always decide which scheme a problem describes before multiplying, because the two give different answers.

Mutually exclusive events

Two events are mutually exclusive when they cannot both occur: their overlap is empty, so \(P(A \text{ and } B) = 0\). On one die roll, "even" and "odd" are exclusive. For exclusive events, the probability that one or the other happens is just the sum, since there is no overlap to double-count - a rule formalised in the addition-rules topic. What exclusivity does not mean is independence; quite the opposite, as the next card shows.

Exclusive means maximally dependent

Mutually exclusive and independent are the two most confused ideas in probability, and they are near-opposites. If \(A\) and \(B\) are exclusive and \(A\) occurs, then \(B\)'s probability collapses to zero - knowing about one tells you everything about the other. Two events with positive probabilities therefore cannot be both mutually exclusive and independent: exclusivity is a form of maximal dependence. The pair to remember: exclusive events cannot happen together; independent events can, and neither notices the other.

Worked example: testing independence with data

In a survey of 200 students, 80 study French, 60 play in a band, and 24 do both. Are studying French and playing in a band independent? 1. \(P(F) = \frac{80}{200} = 0.40\). 2. \(P(B) = \frac{60}{200} = 0.30\). 3. If they were independent, \(P(F) \times P(B) = 0.40 \times 0.30 = 0.12\) would equal the observed \(P(\text{both}) = \frac{24}{200} = 0.12\). 4. It matches exactly, so in this sample the two events are independent. Check with a conditional: \(P(B \mid F) = \frac{24}{80} = 0.30\), the same as the overall band rate \(P(B)\), so knowing a student studies French does not change the band probability.

Worked example: two greens from the jar

Draw two marbles from the jar of 12 (4 green). 1. Without replacement: \(P(\text{both green}) = \frac{4}{12} \times \frac{3}{11} = \frac{12}{132} = \frac{1}{11} \approx 0.091\). 2. With replacement: \(\frac{4}{12} \times \frac{4}{12} = \frac{16}{144} = \frac{1}{9} \approx 0.111\). 3. Check both: without replacement the second factor uses 11 remaining marbles and only 3 greens, so it must be smaller than \(\frac{4}{12}\); returning the first marble restores the original chances, and \(\frac{1}{9} > \frac{1}{11}\) confirms the ordering. Forgetting to update the counts is the classic error.

Common mistakes to avoid

First, swapping \(P(A \mid B)\) and \(P(B \mid A)\): the probability of a positive test given disease is not the probability of disease given a positive test. Second, multiplying probabilities of dependent events as if they were independent - always check what each draw faces. Third, treating mutually exclusive as a synonym for independent: exclusive events are maximally dependent. Fourth, using \(P(A)P(B)\) without checking independence first. Fifth, after a without-replacement draw, forgetting to shrink both the total and the favourable count: 3 green among 11, not 4 among 12.

Key takeaways

Conditional probability restricts the world: \(P(A \mid B) = \frac{P(A \text{ and } B)}{P(B)}\). The multiplication rule \(P(A \text{ and } B) = P(B)P(A \mid B)\) chains events, and under independence it simplifies to \(P(A)P(B)\) - the defining property being \(P(A \mid B) = P(A)\). Sampling with replacement makes draws independent; without replacement makes them dependent. Mutually exclusive events share no outcomes, so \(P(A \text{ and } B) = 0\), and they are never independent when both probabilities are positive. Test independence by comparing the observed both-event probability against the product \(P(A)P(B)\).

Flashcards

Flashcard

P(A | B)

The probability of A given B

P(A and B) divided by P(B).

Test your understanding

Question 1 of 5

If P(A and B) = 0.15 and P(B) = 0.25, what is P(A | B)?

Question 2 of 5

Two events are independent when:

Question 3 of 5

On a single roll of one die, the events "even" and "odd" are:

Question 4 of 5

Can two events, each with positive probability, be both mutually exclusive and independent?

Question 5 of 5

A bag has 5 red and 5 blue counters. Two are drawn without replacement. What is the probability both are red?

← Previous topicNext topic →